wip

Bumps golang from 1.22.0 to 1.22.1.

---
updated-dependencies:
- dependency-name: golang
  dependency-type: direct:production
  update-type: version-update:semver-patch
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

.github/workflows/codeql.yml

@@ -38,11 +38,11 @@ jobs:
 
     steps:
     - name: Checkout repository
-      uses: actions/checkout@v3
+      uses: actions/checkout@v4
 
     # Initializes the CodeQL tools for scanning.
     - name: Initialize CodeQL
-      uses: github/codeql-action/init@v2
+      uses: github/codeql-action/init@v3
       with:
         languages: ${{ matrix.language }}
         # If you wish to specify custom queries, you can do so here or in a config file.
@@ -53,7 +53,7 @@ jobs:
     # Autobuild attempts to build any compiled languages  (C/C++, C#, or Java).
     # If this step fails, then you should remove it and run the build manually (see below)
     - name: Autobuild
-      uses: github/codeql-action/autobuild@v2
+      uses: github/codeql-action/autobuild@v3
 
     # ℹ️ Command-line programs to run using the OS shell.
     # 📚 See https://docs.github.com/en/actions/using-workflows/workflow-syntax-for-github-actions#jobsjob_idstepsrun
@@ -67,4 +67,4 @@ jobs:
     #   make release
 
     - name: Perform CodeQL Analysis
-      uses: github/codeql-action/analyze@v2
+      uses: github/codeql-action/analyze@v3

.github/workflows/docker-release.yml

@@ -14,16 +14,16 @@ jobs:
       IMAGE_NAME: mikefarah/yq
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
 
       - name: Set up QEMU
-        uses: docker/setup-qemu-action@v2
+        uses: docker/setup-qemu-action@v3
         with:
           platforms: all
 
       - name: Set up Docker Buildx
         id: buildx
-        uses: docker/setup-buildx-action@v2
+        uses: docker/setup-buildx-action@v3
         with:
           version: latest
 
@@ -45,7 +45,7 @@ jobs:
           docker buildx build \
             --label "org.opencontainers.image.authors=https://github.com/mikefarah/yq/graphs/contributors" \
             --label "org.opencontainers.image.created=$(date --rfc-3339=seconds)" \
-            --label "org.opencontainers.image.description=yq is a portable command-line YAML processor" \
+            --label "org.opencontainers.image.description=yq is a portable command-line data file processor" \
             --label "org.opencontainers.image.documentation=https://mikefarah.gitbook.io/yq/" \
             --label "org.opencontainers.image.licenses=MIT" \
             --label "org.opencontainers.image.revision=$(git rev-parse HEAD)" \
@@ -65,7 +65,7 @@ jobs:
           docker buildx build \
             --label "org.opencontainers.image.authors=https://github.com/mikefarah/yq/graphs/contributors" \
             --label "org.opencontainers.image.created=$(date --rfc-3339=seconds)" \
-            --label "org.opencontainers.image.description=yq is a portable command-line YAML processor" \
+            --label "org.opencontainers.image.description=yq is a portable command-line data file processor" \
             --label "org.opencontainers.image.documentation=https://mikefarah.gitbook.io/yq/" \
             --label "org.opencontainers.image.licenses=MIT" \
             --label "org.opencontainers.image.revision=$(git rev-parse HEAD)" \

.github/workflows/go.yml

@@ -11,13 +11,13 @@ jobs:
     steps:
 
     - name: Set up Go
-      uses: actions/setup-go@v4
+      uses: actions/setup-go@v5
       with:
         go-version: '^1.20'
       id: go
 
     - name: Check out code into the Go module directory
-      uses: actions/checkout@v3
+      uses: actions/checkout@v4
 
     - name: Get dependencies
       run: |

.github/workflows/release.yml

@@ -2,15 +2,15 @@ name: Release YQ
 on:
   push:
     tags:
-      - 'v*'
+      - 'v4.*'
       - 'draft-*'
 
 jobs:
   publishGitRelease:
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v3
-      - uses: actions/setup-go@v4
+      - uses: actions/checkout@v4
+      - uses: actions/setup-go@v5
         with:
           go-version: '^1.20'
           check-latest: true

.github/workflows/snap-release.yml

@@ -12,7 +12,7 @@ jobs:
       environment: snap
       runs-on: ubuntu-latest
       steps:
-        - uses: actions/checkout@v3
+        - uses: actions/checkout@v4
         - uses: snapcore/action-build@v1
           id: build
         - uses: snapcore/action-publish@v1

.github/workflows/test-yq.yml

@@ -13,7 +13,7 @@ jobs:
     runs-on: ubuntu-latest
 
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
       - name: Get test
         id: get_value
         uses: mikefarah/yq@master

.golangci.yml

@@ -20,12 +20,15 @@ linters:
     - unparam
 linters-settings:
   depguard:
-    list-type: blacklist
-    include-go-root: true
-    packages:
-      - io/ioutil
-    packages-with-error-message:
-      - io/ioutil: "The 'io/ioutil' package is deprecated. Use corresponding 'os' or 'io' functions instead."
+    rules:
+      prevent_unmaintained_packages:
+        list-mode: lax
+        files:
+          - $all
+          - "!$test"
+        deny:
+          - pkg: io/ioutil
+            desc: "replaced by io and os packages since Go 1.16: https://tip.golang.org/doc/go1.16#ioutil"
 issues:
   exclude-rules:
   - linters:

Dockerfile

@@ -1,4 +1,4 @@
-FROM golang:1.21.0 as builder
+FROM golang:1.22.1 as builder
 
 WORKDIR /go/src/mikefarah/yq
 

Dockerfile.dev

@@ -1,4 +1,4 @@
-FROM golang:1.21.0
+FROM golang:1.22.1
 
 COPY scripts/devtools.sh /opt/devtools.sh
 
@@ -6,5 +6,7 @@ RUN set -e -x \
     && /opt/devtools.sh
 ENV PATH=/go/bin:$PATH
 
+RUN apt-get update && apt-get install -y npm && npm install -g npx cspell@latest
+
 ENV CGO_ENABLED 0
 ENV GOPATH /go:/yq

Makefile

@@ -84,8 +84,12 @@ format: vendor
 	${ENGINERUN} bash ./scripts/format.sh
 
 
+.PHONY: spelling
+spelling: format
+	${ENGINERUN} bash ./scripts/spelling.sh
+
 .PHONY: secure
-secure: format
+secure: spelling
 	${ENGINERUN} bash ./scripts/secure.sh
 
 .PHONY: check

README.md

@@ -19,7 +19,7 @@ Pipe from STDIN:
 yq '.a.b[0].c' < file.yaml
 ```
 
-Update a yaml file, inplace
+Update a yaml file, in place
 ```bash
 yq -i '.a.b[0].c = "cool"' file.yaml
 ```
@@ -31,6 +31,10 @@ NAME=mike yq -i '.a.b[0].c = strenv(NAME)' file.yaml
 
 Merge multiple files
 ```bash
+# merge two files
+yq -n 'load("file1.yaml") * load("file2.yaml")'
+
+# merge using globs:
 # note the use of `ea` to evaluate all the files at once
 # instead of in sequence
 yq ea '. as $item ireduce ({}; . * $item )' path/to/*.yml
@@ -45,12 +49,17 @@ yq -i '
 ' file.yaml
 ```
 
+Find and update an item in an array:
+```bash
+yq '(.[] | select(.name == "foo") | .address) = "12 cat st"'
+```
+
 Convert JSON to YAML
 ```bash
 yq -Poy sample.json
 ```
 
-See the [documentation](https://mikefarah.gitbook.io/yq/) for more examples.
+See [recipes](https://mikefarah.gitbook.io/yq/recipes) for more examples and the [documentation](https://mikefarah.gitbook.io/yq/) for more information.
 
 Take a look at the discussions for [common questions](https://github.com/mikefarah/yq/discussions/categories/q-a), and [cool ideas](https://github.com/mikefarah/yq/discussions/categories/show-and-tell)
 
@@ -257,6 +266,9 @@ pacman -S go-yq
 ```
 
 ### Windows:
+
+Using [Chocolatey](https://chocolatey.org)
+
 [![Chocolatey](https://img.shields.io/chocolatey/v/yq.svg)](https://chocolatey.org/packages/yq)
 [![Chocolatey](https://img.shields.io/chocolatey/dt/yq.svg)](https://chocolatey.org/packages/yq)
 ```
@@ -264,12 +276,15 @@ choco install yq
 ```
 Supported by @chillum (https://chocolatey.org/packages/yq)
 
-and
+Using [scoop](https://scoop.sh/)
+```
+scoop install main/yq
+```
 
-### Winget
-winget install yq
-
-https://winget.run/pkg/MikeFarah/yq
+Using [winget](https://learn.microsoft.com/en-us/windows/package-manager/)
+```
+winget install --id MikeFarah.yq
+```
 
 ### Mac:
 Using [MacPorts](https://www.macports.org/)
@@ -299,7 +314,7 @@ https://pkgs.alpinelinux.org/package/edge/community/x86/yq
 - [Deeply data structures](https://mikefarah.gitbook.io/yq/operators/traverse-read)
 - [Sort keys](https://mikefarah.gitbook.io/yq/operators/sort-keys)
 - Manipulate yaml [comments](https://mikefarah.gitbook.io/yq/operators/comment-operators), [styling](https://mikefarah.gitbook.io/yq/operators/style), [tags](https://mikefarah.gitbook.io/yq/operators/tag) and [anchors and aliases](https://mikefarah.gitbook.io/yq/operators/anchor-and-alias-operators).
-- [Update inplace](https://mikefarah.gitbook.io/yq/v/v4.x/commands/evaluate#flags)
+- [Update in place](https://mikefarah.gitbook.io/yq/v/v4.x/commands/evaluate#flags)
 - [Complex expressions to select and update](https://mikefarah.gitbook.io/yq/operators/select#select-and-update-matching-values-in-map)
 - Keeps yaml formatting and comments when updating (though there are issues with whitespace)
 - [Decode/Encode base64 data](https://mikefarah.gitbook.io/yq/operators/encode-decode)
@@ -326,7 +341,7 @@ Examples:
 # yq defaults to 'eval' command if no command is specified. See "yq eval --help" for more examples.
 yq '.stuff' < myfile.yml # outputs the data at the "stuff" node from "myfile.yml"
 
-yq -i '.stuff = "foo"' myfile.yml # update myfile.yml inplace
+yq -i '.stuff = "foo"' myfile.yml # update myfile.yml in place
 
 
 Available Commands:
@@ -343,7 +358,7 @@ Flags:
       --header-preprocess             Slurp any header comments and separators before processing expression. (default true)
   -h, --help                          help for yq
   -I, --indent int                    sets indent level for output (default 2)
-  -i, --inplace                       update the file inplace of first file given.
+  -i, --inplace                       update the file in place of first file given.
   -p, --input-format string           [yaml|y|xml|x] parse format for input. Note that json is a subset of yaml. (default "yaml")
   -M, --no-colors                     force print with no colors
   -N, --no-doc                        Don't print document separators (---)

acceptance_tests/bad_args.sh

@@ -3,25 +3,25 @@
 testWriteInPlacePipeIn() {
   result=$(./yq e -i -n '.a' 2>&1)
   assertEquals 1 $?
-  assertEquals "Error: write inplace flag only applicable when giving an expression and at least one file" "$result"
+  assertEquals "Error: write in place flag only applicable when giving an expression and at least one file" "$result"
 }
 
 testWriteInPlacePipeInEvalall() {
   result=$(./yq ea -i -n '.a' 2>&1)
   assertEquals 1 $?
-  assertEquals "Error: write inplace flag only applicable when giving an expression and at least one file" "$result"
+  assertEquals "Error: write in place flag only applicable when giving an expression and at least one file" "$result"
 }
 
 testWriteInPlaceWithSplit() {
   result=$(./yq e -s "cat" -i '.a = "thing"' test.yml 2>&1)
   assertEquals 1 $?
-  assertEquals "Error: write inplace cannot be used with split file" "$result"
+  assertEquals "Error: write in place cannot be used with split file" "$result"
 }
 
 testWriteInPlaceWithSplitEvalAll() {
   result=$(./yq ea -s "cat" -i '.a = "thing"' test.yml 2>&1)
   assertEquals 1 $?
-  assertEquals "Error: write inplace cannot be used with split file" "$result"
+  assertEquals "Error: write in place cannot be used with split file" "$result"
 }
 
 testNullWithFiles() {

acceptance_tests/basic.sh

@@ -96,6 +96,17 @@ testBasicExpressionFromFile() {
   assertEquals '{"xyz":"meow","cool":"frog"}' "$X"
 }
 
+testBasicExpressionFromFileDos() {
+  ./yq -n ".xyz = 123" > test.yml
+  echo '.xyz = "meow" | .cool = "frog"' | sed 's/$'"/`echo \\\r`/" > instructions.txt
+  
+  X=$(./yq --from-file instructions.txt test.yml -o=j -I=0)
+  assertEquals '{"xyz":"meow","cool":"frog"}' "$X"
+
+  X=$(./yq ea --from-file instructions.txt test.yml -o=j -I=0)
+  assertEquals '{"xyz":"meow","cool":"frog"}' "$X"
+}
+
 testBasicGitHubAction() {
   ./yq -n ".a = 123" > test.yml
   X=$(cat /dev/null | ./yq test.yml)
@@ -143,7 +154,7 @@ testBasicCatWithFilesNoDash() {
 }
 
 # when the nullinput flag is used
-# dont automatically read STDIN (this breaks github actions)
+# don't automatically read STDIN (this breaks github actions)
 testBasicCreateFileGithubAction() {
   cat /dev/null | ./yq -n ".a = 123" > test.yml
 }
@@ -302,7 +313,7 @@ testBasicExitStatusNoEval() {
   assertEquals 1 "$?"
 }
 
-testBasicExtractFieldWithSeperator() {
+testBasicExtractFieldWithSeparator() {
     cat >test.yml <<EOL
 ---
 name: chart-name
@@ -312,7 +323,7 @@ EOL
   assertEquals "chart-name" "$X"
 }
 
-testBasicExtractMultipleFieldWithSeperator() {
+testBasicExtractMultipleFieldWithSeparator() {
     cat >test.yml <<EOL
 ---
 name: chart-name
@@ -351,6 +362,12 @@ EOM
   assertEquals "$expected" "$X"
 }
 
-
+testBasicClosedStdIn() {
+  cat >test.yml <<EOL
+a: 1
+EOL
+  X=$(./yq e '.a' test.yml <&-)
+  assertEquals "1" "$X"
+}
 
 source ./scripts/shunit2

acceptance_tests/flags.sh

@@ -0,0 +1,21 @@
+#!/bin/bash
+
+setUp() {
+  rm test*.yml || true
+  cat >test.yml <<EOL
+# comment
+EOL
+}
+
+testStringInterpolation() {
+    X=$(./yq -n '"Mike \(3 + 4)"')
+    assertEquals "Mike 7" "$X"
+}
+
+testNoStringInterpolation() {
+    X=$(./yq --string-interpolation=f -n '"Mike \(3 + 4)"')
+    assertEquals "Mike \(3 + 4)" "$X"
+}
+
+
+source ./scripts/shunit2

acceptance_tests/inputs-format.sh

@@ -63,6 +63,61 @@ EOM
   assertEquals "$expected" "$X"
 }
 
+testInputCSVCustomSeparator() {
+  cat >test.csv <<EOL
+fruit;yumLevel
+apple;5
+banana;4
+EOL
+
+  read -r -d '' expected << EOM
+- fruit: apple
+  yumLevel: 5
+- fruit: banana
+  yumLevel: 4
+EOM
+
+  X=$(./yq -p=csv --csv-separator ";" test.csv)
+  assertEquals "$expected" "$X"
+
+  X=$(./yq ea -p=csv --csv-separator ";" test.csv)
+  assertEquals "$expected" "$X"
+}
+
+testInputCSVNoAuto() {
+  cat >test.csv <<EOL
+thing1
+name: cat
+EOL
+
+  read -r -d '' expected << EOM
+- thing1: 'name: cat'
+EOM
+
+  X=$(./yq --csv-auto-parse=f test.csv -oy)
+  assertEquals "$expected" "$X"
+
+  X=$(./yq ea --csv-auto-parse=f test.csv -oy)
+  assertEquals "$expected" "$X"
+}
+
+testInputTSVNoAuto() {
+  cat >test.tsv <<EOL
+thing1
+name: cat
+EOL
+
+  read -r -d '' expected << EOM
+- thing1: 'name: cat'
+EOM
+
+  X=$(./yq --tsv-auto-parse=f test.tsv -oy)
+  assertEquals "$expected" "$X"
+
+  X=$(./yq ea --tsv-auto-parse=f test.tsv -oy)
+  assertEquals "$expected" "$X"
+}
+
 testInputCSVUTF8() {
   read -r -d '' expected << EOM
 - id: 1

acceptance_tests/leading-separator.sh

@@ -11,7 +11,7 @@ a: test
 EOL
 }
 
-testLeadingSeperatorWithDoc() {
+testLeadingSeparatorWithDoc() {
   cat >test.yml <<EOL
 # hi peeps
 # cool
@@ -35,7 +35,7 @@ EOM
 }
 
 
-testLeadingSeperatorWithNewlinesNewDoc() {
+testLeadingSeparatorWithNewlinesNewDoc() {
   cat >test.yml <<EOL
 # hi peeps
 # cool
@@ -62,7 +62,7 @@ EOM
   assertEquals "$expected" "$X"
 }
 
-testLeadingSeperatorWithNewlinesMoreComments() {
+testLeadingSeparatorWithNewlinesMoreComments() {
   cat >test.yml <<EOL
 # hi peeps
 # cool
@@ -92,7 +92,7 @@ EOM
 }
 
 
-testLeadingSeperatorWithDirective() {
+testLeadingSeparatorWithDirective() {
   cat >test.yml <<EOL
 %YAML 1.1
 ---
@@ -110,18 +110,18 @@ EOM
 }
 
 
-testLeadingSeperatorPipeIntoEvalSeq() {
+testLeadingSeparatorPipeIntoEvalSeq() {
   X=$(./yq e - < test.yml)
   expected=$(cat test.yml)
   assertEquals "$expected" "$X"
 }
 
-testLeadingSeperatorExtractField() {
+testLeadingSeparatorExtractField() {
   X=$(./yq e '.a' - < test.yml)
   assertEquals "test" "$X"
 }
 
-testLeadingSeperatorExtractFieldWithCommentsAfterSep() {
+testLeadingSeparatorExtractFieldWithCommentsAfterSep() {
   cat >test.yml <<EOL
 ---
 # hi peeps
@@ -132,7 +132,7 @@ EOL
   assertEquals "test" "$X"
 }
 
-testLeadingSeperatorExtractFieldWithCommentsBeforeSep() {
+testLeadingSeparatorExtractFieldWithCommentsBeforeSep() {
   cat >test.yml <<EOL
 # hi peeps
 # cool
@@ -144,7 +144,7 @@ EOL
 }
 
 
-testLeadingSeperatorExtractFieldMultiDoc() {
+testLeadingSeparatorExtractFieldMultiDoc() {
   cat >test.yml <<EOL
 ---
 a: test
@@ -161,13 +161,13 @@ EOM
   assertEquals "$expected" "$X"
 }
 
-testLeadingSeperatorExtractFieldMultiDocWithComments() {
+testLeadingSeparatorExtractFieldMultiDocWithComments() {
   cat >test.yml <<EOL
 # here
 ---
 # there
 a: test
-# whereever
+# wherever
 ---
 # you are
 a: test2
@@ -184,26 +184,26 @@ EOM
 }
 
 
-testLeadingSeperatorEvalSeq() {
+testLeadingSeparatorEvalSeq() {
   X=$(./yq e test.yml)
   expected=$(cat test.yml)
   assertEquals "$expected" "$X"
 }
 
-testLeadingSeperatorPipeIntoEvalAll() {
+testLeadingSeparatorPipeIntoEvalAll() {
   X=$(./yq ea - < test.yml)
   expected=$(cat test.yml)
   assertEquals "$expected" "$X"
 }
 
 
-testLeadingSeperatorEvalAll() {
+testLeadingSeparatorEvalAll() {
   X=$(./yq ea test.yml)
   expected=$(cat test.yml)
   assertEquals "$expected" "$X"
 }
 
-testLeadingSeperatorMultiDocEvalSimple() {
+testLeadingSeparatorMultiDocEvalSimple() {
   read -r -d '' expected << EOM
 ---
 a: test
@@ -217,7 +217,7 @@ EOM
   assertEquals "$expected" "$X"
 }
 
-testLeadingSeperatorMultiDocInOneFile() {
+testLeadingSeparatorMultiDocInOneFile() {
   cat >test.yml <<EOL
 ---
 # hi peeps
@@ -231,7 +231,7 @@ EOL
   assertEquals "$expected" "$X"
 }
 
-testLeadingSeperatorMultiDocInOneFileEvalAll() {
+testLeadingSeparatorMultiDocInOneFileEvalAll() {
   cat >test.yml <<EOL
 ---
 # hi peeps
@@ -245,7 +245,7 @@ EOL
   assertEquals "$expected" "$X"
 }
 
-testLeadingSeperatorMultiDocEvalComments() {
+testLeadingSeparatorMultiDocEvalComments() {
   cat >test.yml <<EOL
 # hi peeps
 # cool
@@ -273,7 +273,7 @@ EOM
   assertEquals "$expected" "$X"
 }
 
-testLeadingSeperatorMultiDocEvalCommentsTrailingSep() {
+testLeadingSeparatorMultiDocEvalCommentsTrailingSep() {
   cat >test.yml <<EOL
 # hi peeps
 # cool
@@ -305,7 +305,7 @@ EOM
   assertEquals "$expected" "$X"
 }
 
-testLeadingSeperatorMultiMultiDocEvalCommentsTrailingSep() {
+testLeadingSeparatorMultiMultiDocEvalCommentsTrailingSep() {
   cat >test.yml <<EOL
 # hi peeps
 # cool
@@ -345,7 +345,7 @@ EOM
   assertEquals "$expected" "$X"
 }
 
-testLeadingSeperatorMultiDocEvalCommentsLeadingSep() {
+testLeadingSeparatorMultiDocEvalCommentsLeadingSep() {
   cat >test.yml <<EOL
 ---
 # hi peeps
@@ -405,7 +405,7 @@ EOM
   assertEquals "$expected" "$X"
 }
 
-testLeadingSeperatorMultiDocEvalCommentsStripComments() {
+testLeadingSeparatorMultiDocEvalCommentsStripComments() {
   cat >test.yml <<EOL
 ---
 # hi peeps
@@ -428,7 +428,7 @@ EOM
   assertEquals "$expected" "$X"
 }
 
-testLeadingSeperatorMultiDocEvalCommentsLeadingSepNoDocFlag() {
+testLeadingSeparatorMultiDocEvalCommentsLeadingSepNoDocFlag() {
   cat >test.yml <<EOL
 ---
 # hi peeps
@@ -454,7 +454,7 @@ EOM
   assertEquals "$expected" "$X"
 }
 
-testLeadingSeperatorMultiDocEvalJsonFlag() {
+testLeadingSeparatorMultiDocEvalJsonFlag() {
   cat >test.yml <<EOL
 ---
 # hi peeps
@@ -483,7 +483,7 @@ EOM
   assertEquals "$expected" "$X"
 }
 
-testLeadingSeperatorMultiDocEvalAllJsonFlag() {
+testLeadingSeparatorMultiDocEvalAllJsonFlag() {
   cat >test.yml <<EOL
 ---
 # hi peeps
@@ -512,7 +512,7 @@ EOM
   assertEquals "$expected" "$X"
 }
 
-testLeadingSeperatorMultiDocEvalAll() {
+testLeadingSeparatorMultiDocEvalAll() {
   read -r -d '' expected << EOM
 ---
 a: test

acceptance_tests/load-file.sh

@@ -9,7 +9,7 @@ testLoadFileNotExist() {
 testLoadFileExpNotExist() {
   result=$(./yq e -n 'load(.a)' 2>&1)
   assertEquals 1 $?
-  assertEquals "Error: Filename expression returned nil" "$result"
+  assertEquals "Error: filename expression returned nil" "$result"
 }
 
 testStrLoadFileNotExist() {
@@ -21,7 +21,7 @@ testStrLoadFileNotExist() {
 testStrLoadFileExpNotExist() {
   result=$(./yq e -n 'strload(.a)' 2>&1)
   assertEquals 1 $?
-  assertEquals "Error: Filename expression returned nil" "$result"
+  assertEquals "Error: filename expression returned nil" "$result"
 }
 
 source ./scripts/shunit2

acceptance_tests/output-format.sh

@@ -72,6 +72,15 @@ EOL
   assertEquals "\"cat\"" "$X"
 }
 
+testOutputYamlRawOnRoot() {
+  cat >test.yml <<EOL
+'a'
+EOL
+
+  X=$(./yq e -r '.' test.yml)
+  assertEquals "a" "$X"
+}
+
 testOutputJsonRaw() {
   cat >test.yml <<EOL
 a: cat
@@ -189,6 +198,27 @@ EOM
   assertEquals "$expected" "$X"
 }
 
+testOutputCSVCustomSeparator() {
+  cat >test.yml <<EOL
+- fruit: apple
+  yumLevel: 5
+- fruit: banana
+  yumLevel: 4
+EOL
+
+  read -r -d '' expected << EOM
+fruit;yumLevel
+apple;5
+banana;4
+EOM
+
+  X=$(./yq -oc --csv-separator ";" test.yml)
+  assertEquals "$expected" "$X"
+
+  X=$(./yq ea -o=csv --csv-separator ";" test.yml)
+  assertEquals "$expected" "$X"
+}
+
 testOutputTSV() {
   cat >test.yml <<EOL
 - fruit: apple
@@ -271,4 +301,43 @@ EOM
   assertEquals "$expected" "$X"
 }
 
-source ./scripts/shunit2
+testLuaOutputPretty() {
+  cat >test.yml <<EOL
+animals:
+  cat: meow
+EOL
+
+  read -r -d '' expected << EOM
+return {
+	["animals"] = {
+		["cat"] = "meow";
+	};
+};
+EOM
+
+  X=$(./yq e --output-format=lua test.yml)
+  assertEquals "$expected" "$X"
+
+  X=$(./yq e --output-format=lua --prettyPrint test.yml)
+  assertEquals "$expected" "$X"
+
+}
+
+testLuaOutputSubset() {
+  cat >test.yml <<EOL
+animals:
+  cat: meow
+EOL
+
+  read -r -d '' expected << EOM
+return {
+	["cat"] = "meow";
+};
+EOM
+
+  X=$(./yq e --output-format=lua '.animals' test.yml)
+  assertEquals "$expected" "$X"
+
+}
+
+source ./scripts/shunit2

acceptance_tests/shebang.sh

@@ -0,0 +1,26 @@
+#!/bin/bash
+
+setUp() {
+  rm test*.yq || true
+  cat >test.yq <<EOL
+#!./yq
+.a.b
+EOL
+chmod +x test.yq
+
+rm test*.yml || true
+  cat >test.yml <<EOL
+a: {b: apple}
+EOL
+}
+
+testCanExecYqFile() {
+  read -r -d '' expected << EOM
+apple
+EOM
+   X=$(./test.yq test.yml)
+  assertEquals "$expected" "$X"
+}
+
+source ./scripts/shunit2
+

cmd/evaluate_all_command.go

@@ -12,8 +12,14 @@ func createEvaluateAllCommand() *cobra.Command {
 		Use:     "eval-all [expression] [yaml_file1]...",
 		Aliases: []string{"ea"},
 		Short:   "Loads _all_ yaml documents of _all_ yaml files and runs expression once",
+		ValidArgsFunction: func(cmd *cobra.Command, args []string, toComplete string) ([]string, cobra.ShellCompDirective) {
+			if len(args) == 0 {
+				return nil, cobra.ShellCompDirectiveNoFileComp
+			}
+			return nil, cobra.ShellCompDirectiveDefault
+		},
 		Example: `
-# Merge f2.yml into f1.yml (inplace)
+# Merge f2.yml into f1.yml (in place)
 yq eval-all --inplace 'select(fileIndex == 0) * select(fileIndex == 1)' f1.yml f2.yml
 ## the same command and expression using shortened names:
 yq ea -i 'select(fi == 0) * select(fi == 1)' f1.yml f2.yml
@@ -26,7 +32,7 @@ yq ea '. as $item ireduce ({}; . * $item )' file1.yml file2.yml ...
 ## use '-' as a filename to pipe from STDIN
 cat file2.yml | yq ea '.a.b' file1.yml - file3.yml
 `,
-		Long: `yq is a portable command-line YAML processor (https://github.com/mikefarah/yq/) 
+		Long: `yq is a portable command-line data file processor (https://github.com/mikefarah/yq/) 
 See https://mikefarah.gitbook.io/yq/ for detailed documentation and examples.
 
 ## Evaluate All ##
@@ -70,7 +76,7 @@ func evaluateAll(cmd *cobra.Command, args []string) (cmdError error) {
 		}()
 	}
 
-	format, err := yqlib.OutputFormatFromString(outputFormat)
+	format, err := yqlib.FormatFromString(outputFormat)
 	if err != nil {
 		return err
 	}

cmd/evaluate_sequence_command.go

@@ -13,6 +13,12 @@ func createEvaluateSequenceCommand() *cobra.Command {
 		Use:     "eval [expression] [yaml_file1]...",
 		Aliases: []string{"e"},
 		Short:   "(default) Apply the expression to each document in each yaml file in sequence",
+		ValidArgsFunction: func(cmd *cobra.Command, args []string, toComplete string) ([]string, cobra.ShellCompDirective) {
+			if len(args) == 0 {
+				return nil, cobra.ShellCompDirectiveNoFileComp
+			}
+			return nil, cobra.ShellCompDirectiveDefault
+		},
 		Example: `
 # Reads field under the given path for each file
 yq e '.a.b' f1.yml f2.yml 
@@ -28,10 +34,10 @@ cat file2.yml | yq e '.a.b' file1.yml - file3.yml
 ## Note that editing an empty file does not work.
 yq e -n '.a.b.c = "cat"' 
 
-# Update a file inplace
+# Update a file in place
 yq e '.a.b = "cool"' -i file.yaml 
 `,
-		Long: `yq is a portable command-line YAML processor (https://github.com/mikefarah/yq/) 
+		Long: `yq is a portable command-line data file processor (https://github.com/mikefarah/yq/) 
 See https://mikefarah.gitbook.io/yq/ for detailed documentation and examples.
 
 ## Evaluate Sequence ##
@@ -84,7 +90,7 @@ func evaluateSequence(cmd *cobra.Command, args []string) (cmdError error) {
 		}()
 	}
 
-	format, err := yqlib.OutputFormatFromString(outputFormat)
+	format, err := yqlib.FormatFromString(outputFormat)
 	if err != nil {
 		return err
 	}

cmd/root.go

@@ -1,21 +1,54 @@
 package cmd
 
 import (
+	"fmt"
 	"os"
+	"strings"
 
+	"github.com/mikefarah/yq/v4/pkg/properties"
+	"github.com/mikefarah/yq/v4/pkg/xml"
 	"github.com/mikefarah/yq/v4/pkg/yqlib"
 	"github.com/spf13/cobra"
 	logging "gopkg.in/op/go-logging.v1"
 )
 
+type runeValue rune
+
+func newRuneVar(p *rune) *runeValue {
+	return (*runeValue)(p)
+}
+
+func (r *runeValue) String() string {
+	return string(*r)
+}
+
+func (r *runeValue) Set(rawVal string) error {
+	val := strings.ReplaceAll(rawVal, "\\n", "\n")
+	val = strings.ReplaceAll(val, "\\t", "\t")
+	val = strings.ReplaceAll(val, "\\r", "\r")
+	val = strings.ReplaceAll(val, "\\f", "\f")
+	val = strings.ReplaceAll(val, "\\v", "\v")
+	if len(val) != 1 {
+		return fmt.Errorf("[%v] is not a valid character. Must be length 1 was %v", val, len(val))
+	}
+	*r = runeValue(rune(val[0]))
+	return nil
+}
+
+func (r *runeValue) Type() string {
+	return "char"
+}
+
 func New() *cobra.Command {
 	var rootCmd = &cobra.Command{
 		Use:   "yq",
-		Short: "yq is a lightweight and portable command-line YAML processor.",
-		Long: `yq is a portable command-line YAML processor (https://github.com/mikefarah/yq/) 
+		Short: "yq is a lightweight and portable command-line data file processor.",
+		Long: `yq is a portable command-line data file processor (https://github.com/mikefarah/yq/) 
 See https://mikefarah.gitbook.io/yq/ for detailed documentation and examples.`,
 		Example: `
-# yq defaults to 'eval' command if no command is specified. See "yq eval --help" for more examples.
+# yq tries to auto-detect the file format based off the extension, and defaults to YAML if it's unknown (or piping through STDIN)
+# Use the '-p/--input-format' flag to specify a format type.
+cat file.xml | yq -p xml
 
 # read the "stuff" node from "myfile.yml"
 yq '.stuff' < myfile.yml
@@ -24,7 +57,7 @@ yq '.stuff' < myfile.yml
 yq -i '.stuff = "foo"' myfile.yml
 
 # print contents of sample.json as idiomatic YAML
-yq -P sample.json
+yq -P -oy sample.json
 `,
 
 		RunE: func(cmd *cobra.Command, args []string) error {
@@ -37,18 +70,19 @@ yq -P sample.json
 		},
 		PersistentPreRunE: func(cmd *cobra.Command, args []string) error {
 			cmd.SetOut(cmd.OutOrStdout())
+			level := logging.WARNING
+			stringFormat := `[%{level}] %{color}%{time:15:04:05}%{color:reset} %{message}`
 
-			var format = logging.MustStringFormatter(
-				`%{color}%{time:15:04:05} %{shortfunc} [%{level:.4s}]%{color:reset} %{message}`,
-			)
+			if verbose {
+				level = logging.DEBUG
+				stringFormat = `[%{level:5.5s}] %{color}%{time:15:04:05}%{color:bold} %{shortfile:-33s} %{shortfunc:-25s}%{color:reset} %{message}`
+			}
+
+			var format = logging.MustStringFormatter(stringFormat)
 			var backend = logging.AddModuleLevel(
 				logging.NewBackendFormatter(logging.NewLogBackend(os.Stderr, "", 0), format))
 
-			if verbose {
-				backend.SetLevel(logging.DEBUG, "")
-			} else {
-				backend.SetLevel(logging.WARNING, "")
-			}
+			backend.SetLevel(level, "")
 
 			logging.SetBackend(backend)
 			yqlib.InitExpressionParser()
@@ -65,30 +99,78 @@ yq -P sample.json
 		panic(err)
 	}
 
-	rootCmd.PersistentFlags().StringVarP(&outputFormat, "output-format", "o", "auto", "[auto|a|yaml|y|json|j|props|p|xml|x|tsv|t|csv|c] output format type.")
-	rootCmd.PersistentFlags().StringVarP(&inputFormat, "input-format", "p", "auto", "[auto|a|yaml|y|props|p|xml|x|tsv|t|csv|c|toml] parse format for input. Note that json is a subset of yaml.")
+	rootCmd.PersistentFlags().StringVarP(&outputFormat, "output-format", "o", "auto", fmt.Sprintf("[auto|a|%v] output format type.", yqlib.GetAvailableOutputFormatString()))
+	var outputCompletions = []string{"auto"}
+	for _, formats := range yqlib.GetAvailableOutputFormats() {
+		outputCompletions = append(outputCompletions, formats.FormalName)
+	}
 
-	rootCmd.PersistentFlags().StringVar(&yqlib.ConfiguredXMLPreferences.AttributePrefix, "xml-attribute-prefix", yqlib.ConfiguredXMLPreferences.AttributePrefix, "prefix for xml attributes")
-	rootCmd.PersistentFlags().StringVar(&yqlib.ConfiguredXMLPreferences.ContentName, "xml-content-name", yqlib.ConfiguredXMLPreferences.ContentName, "name for xml content (if no attribute name is present).")
-	rootCmd.PersistentFlags().BoolVar(&yqlib.ConfiguredXMLPreferences.StrictMode, "xml-strict-mode", yqlib.ConfiguredXMLPreferences.StrictMode, "enables strict parsing of XML. See https://pkg.go.dev/encoding/xml for more details.")
-	rootCmd.PersistentFlags().BoolVar(&yqlib.ConfiguredXMLPreferences.KeepNamespace, "xml-keep-namespace", yqlib.ConfiguredXMLPreferences.KeepNamespace, "enables keeping namespace after parsing attributes")
-	rootCmd.PersistentFlags().BoolVar(&yqlib.ConfiguredXMLPreferences.UseRawToken, "xml-raw-token", yqlib.ConfiguredXMLPreferences.UseRawToken, "enables using RawToken method instead Token. Commonly disables namespace translations. See https://pkg.go.dev/encoding/xml#Decoder.RawToken for details.")
-	rootCmd.PersistentFlags().StringVar(&yqlib.ConfiguredXMLPreferences.ProcInstPrefix, "xml-proc-inst-prefix", yqlib.ConfiguredXMLPreferences.ProcInstPrefix, "prefix for xml processing instructions (e.g. <?xml version=\"1\"?>)")
-	rootCmd.PersistentFlags().StringVar(&yqlib.ConfiguredXMLPreferences.DirectiveName, "xml-directive-name", yqlib.ConfiguredXMLPreferences.DirectiveName, "name for xml directives (e.g. <!DOCTYPE thing cat>)")
-	rootCmd.PersistentFlags().BoolVar(&yqlib.ConfiguredXMLPreferences.SkipProcInst, "xml-skip-proc-inst", yqlib.ConfiguredXMLPreferences.SkipProcInst, "skip over process instructions (e.g. <?xml version=\"1\"?>)")
-	rootCmd.PersistentFlags().BoolVar(&yqlib.ConfiguredXMLPreferences.SkipDirectives, "xml-skip-directives", yqlib.ConfiguredXMLPreferences.SkipDirectives, "skip over directives (e.g. <!DOCTYPE thing cat>)")
+	if err = rootCmd.RegisterFlagCompletionFunc("output-format", cobra.FixedCompletions(outputCompletions, cobra.ShellCompDirectiveNoFileComp)); err != nil {
+		panic(err)
+	}
+	rootCmd.PersistentFlags().StringVarP(&inputFormat, "input-format", "p", "auto", fmt.Sprintf("[auto|a|%v] parse format for input.", yqlib.GetAvailableInputFormatString()))
+
+	var inputCompletions = []string{"auto"}
+	for _, formats := range yqlib.GetAvailableInputFormats() {
+		inputCompletions = append(inputCompletions, formats.FormalName)
+	}
+
+	if err = rootCmd.RegisterFlagCompletionFunc("input-format", cobra.FixedCompletions(inputCompletions, cobra.ShellCompDirectiveNoFileComp)); err != nil {
+		panic(err)
+	}
+
+	rootCmd.PersistentFlags().StringVar(&xml.ConfiguredXMLPreferences.AttributePrefix, "xml-attribute-prefix", xml.ConfiguredXMLPreferences.AttributePrefix, "prefix for xml attributes")
+	if err = rootCmd.RegisterFlagCompletionFunc("xml-attribute-prefix", cobra.NoFileCompletions); err != nil {
+		panic(err)
+	}
+	rootCmd.PersistentFlags().StringVar(&xml.ConfiguredXMLPreferences.ContentName, "xml-content-name", xml.ConfiguredXMLPreferences.ContentName, "name for xml content (if no attribute name is present).")
+	if err = rootCmd.RegisterFlagCompletionFunc("xml-content-name", cobra.NoFileCompletions); err != nil {
+		panic(err)
+	}
+	rootCmd.PersistentFlags().BoolVar(&xml.ConfiguredXMLPreferences.StrictMode, "xml-strict-mode", xml.ConfiguredXMLPreferences.StrictMode, "enables strict parsing of XML. See https://pkg.go.dev/encoding/xml for more details.")
+	rootCmd.PersistentFlags().BoolVar(&xml.ConfiguredXMLPreferences.KeepNamespace, "xml-keep-namespace", xml.ConfiguredXMLPreferences.KeepNamespace, "enables keeping namespace after parsing attributes")
+	rootCmd.PersistentFlags().BoolVar(&xml.ConfiguredXMLPreferences.UseRawToken, "xml-raw-token", xml.ConfiguredXMLPreferences.UseRawToken, "enables using RawToken method instead Token. Commonly disables namespace translations. See https://pkg.go.dev/encoding/xml#Decoder.RawToken for details.")
+	rootCmd.PersistentFlags().StringVar(&xml.ConfiguredXMLPreferences.ProcInstPrefix, "xml-proc-inst-prefix", xml.ConfiguredXMLPreferences.ProcInstPrefix, "prefix for xml processing instructions (e.g. <?xml version=\"1\"?>)")
+	if err = rootCmd.RegisterFlagCompletionFunc("xml-proc-inst-prefix", cobra.NoFileCompletions); err != nil {
+		panic(err)
+	}
+	rootCmd.PersistentFlags().StringVar(&xml.ConfiguredXMLPreferences.DirectiveName, "xml-directive-name", xml.ConfiguredXMLPreferences.DirectiveName, "name for xml directives (e.g. <!DOCTYPE thing cat>)")
+	if err = rootCmd.RegisterFlagCompletionFunc("xml-directive-name", cobra.NoFileCompletions); err != nil {
+		panic(err)
+	}
+	rootCmd.PersistentFlags().BoolVar(&xml.ConfiguredXMLPreferences.SkipProcInst, "xml-skip-proc-inst", xml.ConfiguredXMLPreferences.SkipProcInst, "skip over process instructions (e.g. <?xml version=\"1\"?>)")
+	rootCmd.PersistentFlags().BoolVar(&xml.ConfiguredXMLPreferences.SkipDirectives, "xml-skip-directives", xml.ConfiguredXMLPreferences.SkipDirectives, "skip over directives (e.g. <!DOCTYPE thing cat>)")
+
+	rootCmd.PersistentFlags().BoolVar(&yqlib.ConfiguredCsvPreferences.AutoParse, "csv-auto-parse", yqlib.ConfiguredCsvPreferences.AutoParse, "parse CSV YAML/JSON values")
+	rootCmd.PersistentFlags().Var(newRuneVar(&yqlib.ConfiguredCsvPreferences.Separator), "csv-separator", "CSV Separator character")
+
+	rootCmd.PersistentFlags().BoolVar(&yqlib.ConfiguredTsvPreferences.AutoParse, "tsv-auto-parse", yqlib.ConfiguredTsvPreferences.AutoParse, "parse TSV YAML/JSON values")
 
 	rootCmd.PersistentFlags().StringVar(&yqlib.ConfiguredLuaPreferences.DocPrefix, "lua-prefix", yqlib.ConfiguredLuaPreferences.DocPrefix, "prefix")
+	if err = rootCmd.RegisterFlagCompletionFunc("lua-prefix", cobra.NoFileCompletions); err != nil {
+		panic(err)
+	}
 	rootCmd.PersistentFlags().StringVar(&yqlib.ConfiguredLuaPreferences.DocSuffix, "lua-suffix", yqlib.ConfiguredLuaPreferences.DocSuffix, "suffix")
+	if err = rootCmd.RegisterFlagCompletionFunc("lua-suffix", cobra.NoFileCompletions); err != nil {
+		panic(err)
+	}
 	rootCmd.PersistentFlags().BoolVar(&yqlib.ConfiguredLuaPreferences.UnquotedKeys, "lua-unquoted", yqlib.ConfiguredLuaPreferences.UnquotedKeys, "output unquoted string keys (e.g. {foo=\"bar\"})")
 	rootCmd.PersistentFlags().BoolVar(&yqlib.ConfiguredLuaPreferences.Globals, "lua-globals", yqlib.ConfiguredLuaPreferences.Globals, "output keys as top-level global variables")
 
+	rootCmd.PersistentFlags().StringVar(&properties.ConfiguredPropertiesPreferences.KeyValueSeparator, "properties-separator", properties.ConfiguredPropertiesPreferences.KeyValueSeparator, "separator to use between keys and values")
+	rootCmd.PersistentFlags().BoolVar(&properties.ConfiguredPropertiesPreferences.UseArrayBrackets, "properties-array-brackets", properties.ConfiguredPropertiesPreferences.UseArrayBrackets, "use [x] in array paths (e.g. for SpringBoot)")
+
+	rootCmd.PersistentFlags().BoolVar(&yqlib.StringInterpolationEnabled, "string-interpolation", yqlib.StringInterpolationEnabled, "Toggles strings interpolation of \\(exp)")
+
 	rootCmd.PersistentFlags().BoolVarP(&nullInput, "null-input", "n", false, "Don't read input, simply evaluate the expression given. Useful for creating docs from scratch.")
 	rootCmd.PersistentFlags().BoolVarP(&noDocSeparators, "no-doc", "N", false, "Don't print document separators (---)")
 
 	rootCmd.PersistentFlags().IntVarP(&indent, "indent", "I", 2, "sets indent level for output")
+	if err = rootCmd.RegisterFlagCompletionFunc("indent", cobra.NoFileCompletions); err != nil {
+		panic(err)
+	}
 	rootCmd.Flags().BoolVarP(&version, "version", "V", false, "Print version information and quit")
-	rootCmd.PersistentFlags().BoolVarP(&writeInplace, "inplace", "i", false, "update the file inplace of first file given.")
+	rootCmd.PersistentFlags().BoolVarP(&writeInplace, "inplace", "i", false, "update the file in place of first file given.")
 	rootCmd.PersistentFlags().VarP(unwrapScalarFlag, "unwrapScalar", "r", "unwrap scalar, print the value with no quotes, colors or comments. Defaults to true for yaml")
 	rootCmd.PersistentFlags().Lookup("unwrapScalar").NoOptDefVal = "true"
 	rootCmd.PersistentFlags().BoolVarP(&nulSepOutput, "nul-output", "0", false, "Use NUL char to separate values. If unwrap scalar is also set, fail if unwrapped scalar contains NUL char.")
@@ -99,13 +181,28 @@ yq -P sample.json
 	rootCmd.PersistentFlags().BoolVarP(&forceColor, "colors", "C", false, "force print with colors")
 	rootCmd.PersistentFlags().BoolVarP(&forceNoColor, "no-colors", "M", false, "force print with no colors")
 	rootCmd.PersistentFlags().StringVarP(&frontMatter, "front-matter", "f", "", "(extract|process) first input as yaml front-matter. Extract will pull out the yaml content, process will run the expression against the yaml content, leaving the remaining data intact")
+	if err = rootCmd.RegisterFlagCompletionFunc("front-matter", cobra.FixedCompletions([]string{"extract", "process"}, cobra.ShellCompDirectiveNoFileComp)); err != nil {
+		panic(err)
+	}
 	rootCmd.PersistentFlags().StringVarP(&forceExpression, "expression", "", "", "forcibly set the expression argument. Useful when yq argument detection thinks your expression is a file.")
+	if err = rootCmd.RegisterFlagCompletionFunc("expression", cobra.NoFileCompletions); err != nil {
+		panic(err)
+	}
 	rootCmd.PersistentFlags().BoolVarP(&yqlib.ConfiguredYamlPreferences.LeadingContentPreProcessing, "header-preprocess", "", true, "Slurp any header comments and separators before processing expression.")
 
 	rootCmd.PersistentFlags().StringVarP(&splitFileExp, "split-exp", "s", "", "print each result (or doc) into a file named (exp). [exp] argument must return a string. You can use $index in the expression as the result counter.")
+	if err = rootCmd.RegisterFlagCompletionFunc("split-exp", cobra.NoFileCompletions); err != nil {
+		panic(err)
+	}
 	rootCmd.PersistentFlags().StringVarP(&splitFileExpFile, "split-exp-file", "", "", "Use a file to specify the split-exp expression.")
+	if err = rootCmd.MarkPersistentFlagFilename("split-exp-file"); err != nil {
+		panic(err)
+	}
 
 	rootCmd.PersistentFlags().StringVarP(&expressionFile, "from-file", "", "", "Load expression from specified file.")
+	if err = rootCmd.MarkPersistentFlagFilename("from-file"); err != nil {
+		panic(err)
+	}
 
 	rootCmd.AddCommand(
 		createEvaluateSequenceCommand(),

cmd/shell-completion.go

@@ -47,7 +47,7 @@ $ yq shell-completion fish > ~/.config/fish/completions/yq.fish
 		var err error = nil
 		switch args[0] {
 		case "bash":
-			err = cmd.Root().GenBashCompletion(os.Stdout)
+			err = cmd.Root().GenBashCompletionV2(os.Stdout, true)
 		case "zsh":
 			err = cmd.Root().GenZshCompletion(os.Stdout)
 		case "fish":

cmd/unwrap_flag.go

@@ -8,21 +8,21 @@ import (
 
 type boolFlag interface {
 	pflag.Value
-	IsExplicitySet() bool
+	IsExplicitlySet() bool
 	IsSet() bool
 }
 
 type unwrapScalarFlagStrc struct {
-	explicitySet bool
-	value        bool
+	explicitlySet bool
+	value         bool
 }
 
 func newUnwrapFlag() boolFlag {
 	return &unwrapScalarFlagStrc{value: true}
 }
 
-func (f *unwrapScalarFlagStrc) IsExplicitySet() bool {
-	return f.explicitySet
+func (f *unwrapScalarFlagStrc) IsExplicitlySet() bool {
+	return f.explicitlySet
 }
 
 func (f *unwrapScalarFlagStrc) IsSet() bool {
@@ -37,7 +37,7 @@ func (f *unwrapScalarFlagStrc) Set(value string) error {
 
 	v, err := strconv.ParseBool(value)
 	f.value = v
-	f.explicitySet = true
+	f.explicitlySet = true
 	return err
 }
 

cmd/utils.go

@@ -4,7 +4,10 @@ import (
 	"fmt"
 	"io"
 	"os"
+	"strings"
 
+	"github.com/mikefarah/yq/v4/pkg/properties"
+	"github.com/mikefarah/yq/v4/pkg/xml"
 	"github.com/mikefarah/yq/v4/pkg/yqlib"
 	"github.com/spf13/cobra"
 	"gopkg.in/op/go-logging.v1"
@@ -42,7 +45,7 @@ func initCommand(cmd *cobra.Command, args []string) (string, []string, error) {
 	}
 
 	if writeInplace && (len(args) == 0 || args[0] == "-") {
-		return "", nil, fmt.Errorf("write inplace flag only applicable when giving an expression and at least one file")
+		return "", nil, fmt.Errorf("write in place flag only applicable when giving an expression and at least one file")
 	}
 
 	if frontMatter != "" && len(args) == 0 {
@@ -50,7 +53,7 @@ func initCommand(cmd *cobra.Command, args []string) (string, []string, error) {
 	}
 
 	if writeInplace && splitFileExp != "" {
-		return "", nil, fmt.Errorf("write inplace cannot be used with split file")
+		return "", nil, fmt.Errorf("write in place cannot be used with split file")
 	}
 
 	if nullInput && len(args) > 0 {
@@ -63,9 +66,9 @@ func initCommand(cmd *cobra.Command, args []string) (string, []string, error) {
 	}
 	if inputFormat == "" || inputFormat == "auto" || inputFormat == "a" {
 
-		inputFormat = yqlib.FormatFromFilename(inputFilename)
+		inputFormat = yqlib.FormatStringFromFilename(inputFilename)
 
-		_, err := yqlib.InputFormatFromString(inputFormat)
+		_, err := yqlib.FormatFromString(inputFormat)
 		if err != nil {
 			// unknown file type, default to yaml
 			yqlib.GetLogger().Debug("Unknown file format extension '%v', defaulting to yaml", inputFormat)
@@ -85,14 +88,14 @@ func initCommand(cmd *cobra.Command, args []string) (string, []string, error) {
 		// before this was introduced, `yq -pcsv things.csv`
 		// would produce *yaml* output.
 		//
-		outputFormat = yqlib.FormatFromFilename(inputFilename)
+		outputFormat = yqlib.FormatStringFromFilename(inputFilename)
 		if inputFilename != "-" {
 			yqlib.GetLogger().Warning("yq default output is now 'auto' (based on the filename extension). Normally yq would output '%v', but for backwards compatibility 'yaml' has been set. Please use -oy to specify yaml, or drop the -p flag.", outputFormat)
 		}
 		outputFormat = "yaml"
 	}
 
-	outputFormatType, err := yqlib.OutputFormatFromString(outputFormat)
+	outputFormatType, err := yqlib.FormatFromString(outputFormat)
 
 	if err != nil {
 		return "", nil, err
@@ -100,57 +103,32 @@ func initCommand(cmd *cobra.Command, args []string) (string, []string, error) {
 	yqlib.GetLogger().Debug("Using input format %v", inputFormat)
 	yqlib.GetLogger().Debug("Using output format %v", outputFormat)
 
-	if outputFormatType == yqlib.YamlOutputFormat ||
-		outputFormatType == yqlib.PropsOutputFormat {
+	if outputFormatType == yqlib.YamlFormat ||
+		outputFormatType == properties.PropertiesFormat {
 		unwrapScalar = true
 	}
-	if unwrapScalarFlag.IsExplicitySet() {
+	if unwrapScalarFlag.IsExplicitlySet() {
 		unwrapScalar = unwrapScalarFlag.IsSet()
 	}
 
-	//copy preference form global setting
-	yqlib.ConfiguredYamlPreferences.UnwrapScalar = unwrapScalar
-
-	yqlib.ConfiguredYamlPreferences.PrintDocSeparators = !noDocSeparators
-
 	return expression, args, nil
 }
 
 func configureDecoder(evaluateTogether bool) (yqlib.Decoder, error) {
-	yqlibInputFormat, err := yqlib.InputFormatFromString(inputFormat)
+	format, err := yqlib.FormatFromString(inputFormat)
 	if err != nil {
 		return nil, err
 	}
-	yqlibDecoder, err := createDecoder(yqlibInputFormat, evaluateTogether)
+	yqlib.ConfiguredYamlPreferences.EvaluateTogether = evaluateTogether
+
+	yqlibDecoder := format.DecoderFactory()
 	if yqlibDecoder == nil {
 		return nil, fmt.Errorf("no support for %s input format", inputFormat)
 	}
-	return yqlibDecoder, err
+	return yqlibDecoder, nil
 }
 
-func createDecoder(format yqlib.InputFormat, evaluateTogether bool) (yqlib.Decoder, error) {
-	switch format {
-	case yqlib.XMLInputFormat:
-		return yqlib.NewXMLDecoder(yqlib.ConfiguredXMLPreferences), nil
-	case yqlib.PropertiesInputFormat:
-		return yqlib.NewPropertiesDecoder(), nil
-	case yqlib.JsonInputFormat:
-		return yqlib.NewJSONDecoder(), nil
-	case yqlib.CSVObjectInputFormat:
-		return yqlib.NewCSVObjectDecoder(','), nil
-	case yqlib.TSVObjectInputFormat:
-		return yqlib.NewCSVObjectDecoder('\t'), nil
-	case yqlib.TomlInputFormat:
-		return yqlib.NewTomlDecoder(), nil
-	case yqlib.YamlInputFormat:
-		prefs := yqlib.ConfiguredYamlPreferences
-		prefs.EvaluateTogether = evaluateTogether
-		return yqlib.NewYamlDecoder(prefs), nil
-	}
-	return nil, fmt.Errorf("invalid decoder: %v", format)
-}
-
-func configurePrinterWriter(format yqlib.PrinterOutputFormat, out io.Writer) (yqlib.PrinterWriter, error) {
+func configurePrinterWriter(format *yqlib.Format, out io.Writer) (yqlib.PrinterWriter, error) {
 
 	var printerWriter yqlib.PrinterWriter
 
@@ -168,39 +146,29 @@ func configurePrinterWriter(format yqlib.PrinterOutputFormat, out io.Writer) (yq
 }
 
 func configureEncoder() (yqlib.Encoder, error) {
-	yqlibOutputFormat, err := yqlib.OutputFormatFromString(outputFormat)
+	yqlibOutputFormat, err := yqlib.FormatFromString(outputFormat)
 	if err != nil {
 		return nil, err
 	}
-	yqlibEncoder, err := createEncoder(yqlibOutputFormat)
-	if yqlibEncoder == nil {
+	xml.ConfiguredXMLPreferences.Indent = indent
+	yqlib.ConfiguredYamlPreferences.Indent = indent
+	yqlib.ConfiguredJSONPreferences.Indent = indent
+
+	yqlib.ConfiguredYamlPreferences.UnwrapScalar = unwrapScalar
+	properties.ConfiguredPropertiesPreferences.UnwrapScalar = unwrapScalar
+	yqlib.ConfiguredJSONPreferences.UnwrapScalar = unwrapScalar
+
+	yqlib.ConfiguredYamlPreferences.ColorsEnabled = colorsEnabled
+	yqlib.ConfiguredJSONPreferences.ColorsEnabled = colorsEnabled
+
+	yqlib.ConfiguredYamlPreferences.PrintDocSeparators = !noDocSeparators
+
+	encoder := yqlibOutputFormat.EncoderFactory()
+
+	if encoder == nil {
 		return nil, fmt.Errorf("no support for %s output format", outputFormat)
 	}
-	return yqlibEncoder, err
-}
-
-func createEncoder(format yqlib.PrinterOutputFormat) (yqlib.Encoder, error) {
-	switch format {
-	case yqlib.JSONOutputFormat:
-		return yqlib.NewJSONEncoder(indent, colorsEnabled, unwrapScalar), nil
-	case yqlib.PropsOutputFormat:
-		return yqlib.NewPropertiesEncoder(unwrapScalar), nil
-	case yqlib.CSVOutputFormat:
-		return yqlib.NewCsvEncoder(','), nil
-	case yqlib.TSVOutputFormat:
-		return yqlib.NewCsvEncoder('\t'), nil
-	case yqlib.YamlOutputFormat:
-		return yqlib.NewYamlEncoder(indent, colorsEnabled, yqlib.ConfiguredYamlPreferences), nil
-	case yqlib.XMLOutputFormat:
-		return yqlib.NewXMLEncoder(indent, yqlib.ConfiguredXMLPreferences), nil
-	case yqlib.TomlOutputFormat:
-		return yqlib.NewTomlEncoder(), nil
-	case yqlib.ShellVariablesOutputFormat:
-		return yqlib.NewShellVariablesEncoder(), nil
-	case yqlib.LuaOutputFormat:
-		return yqlib.NewLuaEncoder(yqlib.ConfiguredLuaPreferences), nil
-	}
-	return nil, fmt.Errorf("invalid encoder: %v", format)
+	return encoder, err
 }
 
 // this is a hack to enable backwards compatibility with githubactions (which pipe /dev/null into everything)
@@ -224,8 +192,11 @@ func maybeFile(str string) bool {
 }
 
 func processStdInArgs(args []string) []string {
-	stat, _ := os.Stdin.Stat()
-	pipingStdin := (stat.Mode() & os.ModeCharDevice) == 0
+	stat, err := os.Stdin.Stat()
+	if err != nil {
+		yqlib.GetLogger().Debugf("error getting stdin: %v", err)
+	}
+	pipingStdin := stat != nil && (stat.Mode()&os.ModeCharDevice) == 0
 
 	// if we've been given a file, don't automatically
 	// read from stdin.
@@ -249,15 +220,25 @@ func processStdInArgs(args []string) []string {
 
 func processArgs(originalArgs []string) (string, []string, error) {
 	expression := forceExpression
+	args := processStdInArgs(originalArgs)
+	maybeFirstArgIsAFile := len(args) > 0 && maybeFile(args[0])
+
+	if expressionFile == "" && maybeFirstArgIsAFile && strings.HasSuffix(args[0], ".yq") {
+		// lets check if an expression file was given
+		yqlib.GetLogger().Debug("Assuming arg %v is an expression file", args[0])
+		expressionFile = args[0]
+		args = args[1:]
+	}
+
 	if expressionFile != "" {
 		expressionBytes, err := os.ReadFile(expressionFile)
 		if err != nil {
 			return "", nil, err
 		}
-		expression = string(expressionBytes)
+		//replace \r\n (windows) with good ol' unix file endings.
+		expression = strings.ReplaceAll(string(expressionBytes), "\r\n", "\n")
 	}
 
-	args := processStdInArgs(originalArgs)
 	yqlib.GetLogger().Debugf("processed args: %v", args)
 	if expression == "" && len(args) > 0 && args[0] != "-" && !maybeFile(args[0]) {
 		yqlib.GetLogger().Debug("assuming expression is '%v'", args[0])

cmd/version.go

@@ -11,7 +11,7 @@ var (
 	GitDescribe string
 
 	// Version is main version number that is being run at the moment.
-	Version = "v4.35.1"
+	Version = "v4.42.1"
 
 	// VersionPrerelease is a pre-release marker for the version. If this is "" (empty string)
 	// then it means that it is a final release. Otherwise, this is a pre-release

cspell.config.yaml

@@ -0,0 +1,14 @@
+---
+$schema: https://raw.githubusercontent.com/streetsidesoftware/cspell/main/cspell.schema.json
+version: '0.2'
+language: en-GB
+dictionaryDefinitions:
+  - name: project-words
+    path: './project-words.txt'
+    addWords: true
+dictionaries:
+  - project-words
+ignorePaths:
+  - 'vendor'
+  - 'bin'
+  - '/project-words.txt'

debian/changelog

@@ -127,7 +127,7 @@ yq (4.9.6) focal; urgency=medium
 
   * Added darwin/arm64 build, thanks @alecthomas
   * Incremented docker alpine base version, thanks @da6d6i7-bronga
-  * Bug fix: multine expression
+  * Bug fix: multiline expression
   * Bug fix: special character
 
  -- Roberto Mier Escandon <rmescandon@gmail.com>  Tue, 29 Jun 2021 21:32:14 +0200
@@ -135,7 +135,7 @@ yq (4.9.6) focal; urgency=medium
 yq (3.3.2) focal; urgency=medium
 
   * Bug fix: existStatus bug (#459)
-  * Automatically makes a os temp directory if it does not exist (#461)
+  * Automatically makes an os temp directory if it does not exist (#461)
 
  -- Roberto Mier Escandon <rmescandon@gmail.com>  Fri, 07 Aug 2020 18:53:01 +0200
 

debian/control

@@ -16,7 +16,7 @@ XSBC-Original-Maintainer: Roberto Mier Escandón <rmescandon@gmail.com>
 Package: yq
 Architecture: any
 Depends: ${shlibs:Depends}, ${misc:Depends}
-Description: lightweight and portable command-line YAML processor
+Description: lightweight and portable command-line data file processor
  .
  The aim of the project is to be the 
  [jq](https://github.com/stedolan/jq) or sed of yaml files.

examples/data.lua

@@ -0,0 +1,9 @@
+return {
+	["country"] = "Australia"; -- this place
+	["cities"] = {
+		"Sydney",
+		"Melbourne",
+		"Brisbane",
+		"Perth",
+	};
+};

examples/data1.yaml

@@ -1 +1,11 @@
-["foobar", "foobaz", "blarp"]
+# block comments come through
+person: # neither do comments on maps
+    name: Mike Wazowski # comments on values appear
+    pets: 
+    - cat # comments on array values appear
+    - dog # comments on array values appear
+    - things:
+        - frog
+    food: [pizza] # comments on arrays do not
+emptyArray: []
+emptyMap: []

examples/environment.yq

@@ -0,0 +1,5 @@
+#! yq
+.[] |(
+    ( select(kind == "scalar") | key + "='" + . + "'"),
+    ( select(kind == "seq") | key + "=(" + (map("'" + . + "'") | join(",")) + ")")
+)

examples/mike.xml

@@ -1 +1 @@
-<cat>3</cat>
+<zoo><thing><frog>boing</frog></thing></zoo>

examples/sample.yaml

@@ -1,11 +1,2 @@
-# Some doc
-
-a: true
-b:
-  c: 2
-  d: [3, 4, 5]
-  e:
-    - name: fred
-      value: 3
-    - name: sam
-      value: 4
+ # things
+a: apple

go.mod

@@ -2,21 +2,22 @@ module github.com/mikefarah/yq/v4
 
 require (
 	github.com/a8m/envsubst v1.4.2
-	github.com/alecthomas/participle/v2 v2.0.0
-	github.com/alecthomas/repr v0.2.0
+	github.com/alecthomas/participle/v2 v2.1.1
+	github.com/alecthomas/repr v0.4.0
 	github.com/dimchansky/utfbom v1.1.1
-	github.com/elliotchance/orderedmap v1.5.0
-	github.com/fatih/color v1.15.0
+	github.com/elliotchance/orderedmap v1.5.1
+	github.com/fatih/color v1.16.0
 	github.com/goccy/go-json v0.10.2
-	github.com/goccy/go-yaml v1.11.0
-	github.com/jinzhu/copier v0.3.5
+	github.com/goccy/go-yaml v1.11.3
+	github.com/jinzhu/copier v0.4.0
 	github.com/magiconair/properties v1.8.7
-	github.com/pelletier/go-toml/v2 v2.0.9
+	github.com/pelletier/go-toml/v2 v2.1.1
 	github.com/pkg/diff v0.0.0-20210226163009-20ebb0f2a09e
-	github.com/spf13/cobra v1.7.0
+	github.com/spf13/cobra v1.8.0
 	github.com/spf13/pflag v1.0.5
-	golang.org/x/net v0.14.0
-	golang.org/x/text v0.12.0
+	github.com/yuin/gopher-lua v1.1.1
+	golang.org/x/net v0.22.0
+	golang.org/x/text v0.14.0
 	gopkg.in/op/go-logging.v1 v1.0.0-20160211212156-b2cb9fa56473
 	gopkg.in/yaml.v3 v3.0.1
 )
@@ -24,9 +25,9 @@ require (
 require (
 	github.com/inconshreveable/mousetrap v1.1.0 // indirect
 	github.com/mattn/go-colorable v0.1.13 // indirect
-	github.com/mattn/go-isatty v0.0.17 // indirect
-	golang.org/x/sys v0.11.0 // indirect
-	golang.org/x/xerrors v0.0.0-20220609144429-65e65417b02f // indirect
+	github.com/mattn/go-isatty v0.0.20 // indirect
+	golang.org/x/sys v0.18.0 // indirect
+	golang.org/x/xerrors v0.0.0-20220907171357-04be3eba64a2 // indirect
 )
 
-go 1.20
+go 1.21

go.sum

@@ -1,45 +1,57 @@
 github.com/a8m/envsubst v1.4.2 h1:4yWIHXOLEJHQEFd4UjrWDrYeYlV7ncFWJOCBRLOZHQg=
 github.com/a8m/envsubst v1.4.2/go.mod h1:MVUTQNGQ3tsjOOtKCNd+fl8RzhsXcDvvAEzkhGtlsbY=
-github.com/alecthomas/assert/v2 v2.2.2 h1:Z/iVC0xZfWTaFNE6bA3z07T86hd45Xe2eLt6WVy2bbk=
-github.com/alecthomas/participle/v2 v2.0.0 h1:Fgrq+MbuSsJwIkw3fEj9h75vDP0Er5JzepJ0/HNHv0g=
-github.com/alecthomas/participle/v2 v2.0.0/go.mod h1:rAKZdJldHu8084ojcWevWAL8KmEU+AT+Olodb+WoN2Y=
-github.com/alecthomas/repr v0.2.0 h1:HAzS41CIzNW5syS8Mf9UwXhNH1J9aix/BvDRf1Ml2Yk=
-github.com/alecthomas/repr v0.2.0/go.mod h1:Fr0507jx4eOXV7AlPV6AVZLYrLIuIeSOWtW57eE/O/4=
-github.com/cpuguy83/go-md2man/v2 v2.0.2/go.mod h1:tgQtvFlXSQOSOSIRvRPT7W67SCa46tRHOmNcaadrF8o=
+github.com/alecthomas/assert/v2 v2.3.0 h1:mAsH2wmvjsuvyBvAmCtm7zFsBlb8mIHx5ySLVdDZXL0=
+github.com/alecthomas/assert/v2 v2.3.0/go.mod h1:pXcQ2Asjp247dahGEmsZ6ru0UVwnkhktn7S0bBDLxvQ=
+github.com/alecthomas/participle/v2 v2.1.1 h1:hrjKESvSqGHzRb4yW1ciisFJ4p3MGYih6icjJvbsmV8=
+github.com/alecthomas/participle/v2 v2.1.1/go.mod h1:Y1+hAs8DHPmc3YUFzqllV+eSQ9ljPTk0ZkPMtEdAx2c=
+github.com/alecthomas/repr v0.4.0 h1:GhI2A8MACjfegCPVq9f1FLvIBS+DrQ2KQBFZP1iFzXc=
+github.com/alecthomas/repr v0.4.0/go.mod h1:Fr0507jx4eOXV7AlPV6AVZLYrLIuIeSOWtW57eE/O/4=
+github.com/cpuguy83/go-md2man/v2 v2.0.3/go.mod h1:tgQtvFlXSQOSOSIRvRPT7W67SCa46tRHOmNcaadrF8o=
 github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
 github.com/davecgh/go-spew v1.1.1 h1:vj9j/u1bqnvCEfJOwUhtlOARqs3+rkHYY13jYWTU97c=
 github.com/davecgh/go-spew v1.1.1/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
 github.com/dimchansky/utfbom v1.1.1 h1:vV6w1AhK4VMnhBno/TPVCoK9U/LP0PkLCS9tbxHdi/U=
 github.com/dimchansky/utfbom v1.1.1/go.mod h1:SxdoEBH5qIqFocHMyGOXVAybYJdr71b1Q/j0mACtrfE=
-github.com/elliotchance/orderedmap v1.5.0 h1:1IsExUsjv5XNBD3ZdC7jkAAqLWOOKdbPTmkHx63OsBg=
-github.com/elliotchance/orderedmap v1.5.0/go.mod h1:wsDwEaX5jEoyhbs7x93zk2H/qv0zwuhg4inXhDkYqys=
-github.com/fatih/color v1.15.0 h1:kOqh6YHBtK8aywxGerMG2Eq3H6Qgoqeo13Bk2Mv/nBs=
-github.com/fatih/color v1.15.0/go.mod h1:0h5ZqXfHYED7Bhv2ZJamyIOUej9KtShiJESRwBDUSsw=
+github.com/elliotchance/orderedmap v1.5.1 h1:G1X4PYlljzimbdQ3RXmtIZiQ9d6aRQ3sH1nzjq5mECE=
+github.com/elliotchance/orderedmap v1.5.1/go.mod h1:wsDwEaX5jEoyhbs7x93zk2H/qv0zwuhg4inXhDkYqys=
+github.com/fatih/color v1.16.0 h1:zmkK9Ngbjj+K0yRhTVONQh1p/HknKYSlNT+vZCzyokM=
+github.com/fatih/color v1.16.0/go.mod h1:fL2Sau1YI5c0pdGEVCbKQbLXB6edEj1ZgiY4NijnWvE=
+github.com/go-playground/locales v0.13.0 h1:HyWk6mgj5qFqCT5fjGBuRArbVDfE4hi8+e8ceBS/t7Q=
+github.com/go-playground/locales v0.13.0/go.mod h1:taPMhCMXrRLJO55olJkUXHZBHCxTMfnGwq/HNwmWNS8=
+github.com/go-playground/universal-translator v0.17.0 h1:icxd5fm+REJzpZx7ZfpaD876Lmtgy7VtROAbHHXk8no=
+github.com/go-playground/universal-translator v0.17.0/go.mod h1:UkSxE5sNxxRwHyU+Scu5vgOQjsIJAF8j9muTVoKLVtA=
+github.com/go-playground/validator/v10 v10.4.1 h1:pH2c5ADXtd66mxoE0Zm9SUhxE20r7aM3F26W0hOn+GE=
+github.com/go-playground/validator/v10 v10.4.1/go.mod h1:nlOn6nFhuKACm19sB/8EGNn9GlaMV7XkbRSipzJ0Ii4=
 github.com/goccy/go-json v0.10.2 h1:CrxCmQqYDkv1z7lO7Wbh2HN93uovUHgrECaO5ZrCXAU=
 github.com/goccy/go-json v0.10.2/go.mod h1:6MelG93GURQebXPDq3khkgXZkazVtN9CRI+MGFi0w8I=
-github.com/goccy/go-yaml v1.11.0 h1:n7Z+zx8S9f9KgzG6KtQKf+kwqXZlLNR2F6018Dgau54=
-github.com/goccy/go-yaml v1.11.0/go.mod h1:H+mJrWtjPTJAHvRbV09MCK9xYwODM+wRTVFFTWckfng=
+github.com/goccy/go-yaml v1.11.3 h1:B3W9IdWbvrUu2OYQGwvU1nZtvMQJPBKgBUuweJjLj6I=
+github.com/goccy/go-yaml v1.11.3/go.mod h1:wKnAMd44+9JAAnGQpWVEgBzGt3YuTaQ4uXoHvE4m7WU=
+github.com/google/go-cmp v0.5.9 h1:O2Tfq5qg4qc4AmwVlvv0oLiVAGB7enBSJ2x2DqQFi38=
+github.com/google/go-cmp v0.5.9/go.mod h1:17dUlkBOakJ0+DkrSSNjCkIjxS6bF9zb3elmeNGIjoY=
 github.com/hexops/gotextdiff v1.0.3 h1:gitA9+qJrrTCsiCl7+kh75nPqQt1cx4ZkudSTLoUqJM=
+github.com/hexops/gotextdiff v1.0.3/go.mod h1:pSWU5MAI3yDq+fZBTazCSJysOMbxWL1BSow5/V2vxeg=
 github.com/inconshreveable/mousetrap v1.1.0 h1:wN+x4NVGpMsO7ErUn/mUI3vEoE6Jt13X2s0bqwp9tc8=
 github.com/inconshreveable/mousetrap v1.1.0/go.mod h1:vpF70FUmC8bwa3OWnCshd2FqLfsEA9PFc4w1p2J65bw=
-github.com/jinzhu/copier v0.3.5 h1:GlvfUwHk62RokgqVNvYsku0TATCF7bAHVwEXoBh3iJg=
-github.com/jinzhu/copier v0.3.5/go.mod h1:DfbEm0FYsaqBcKcFuvmOZb218JkPGtvSHsKg8S8hyyg=
+github.com/jinzhu/copier v0.4.0 h1:w3ciUoD19shMCRargcpm0cm91ytaBhDvuRpz1ODO/U8=
+github.com/jinzhu/copier v0.4.0/go.mod h1:DfbEm0FYsaqBcKcFuvmOZb218JkPGtvSHsKg8S8hyyg=
+github.com/leodido/go-urn v1.2.0 h1:hpXL4XnriNwQ/ABnpepYM/1vCLWNDfUNts8dX3xTG6Y=
+github.com/leodido/go-urn v1.2.0/go.mod h1:+8+nEpDfqqsY+g338gtMEUOtuK+4dEMhiQEgxpxOKII=
 github.com/magiconair/properties v1.8.7 h1:IeQXZAiQcpL9mgcAe1Nu6cX9LLw6ExEHKjN0VQdvPDY=
 github.com/magiconair/properties v1.8.7/go.mod h1:Dhd985XPs7jluiymwWYZ0G4Z61jb3vdS329zhj2hYo0=
 github.com/mattn/go-colorable v0.1.13 h1:fFA4WZxdEF4tXPZVKMLwD8oUnCTTo08duU7wxecdEvA=
 github.com/mattn/go-colorable v0.1.13/go.mod h1:7S9/ev0klgBDR4GtXTXX8a3vIGJpMovkB8vQcUbaXHg=
 github.com/mattn/go-isatty v0.0.16/go.mod h1:kYGgaQfpe5nmfYZH+SKPsOc2e4SrIfOl2e/yFXSvRLM=
-github.com/mattn/go-isatty v0.0.17 h1:BTarxUcIeDqL27Mc+vyvdWYSL28zpIhv3RoTdsLMPng=
-github.com/mattn/go-isatty v0.0.17/go.mod h1:kYGgaQfpe5nmfYZH+SKPsOc2e4SrIfOl2e/yFXSvRLM=
-github.com/pelletier/go-toml/v2 v2.0.9 h1:uH2qQXheeefCCkuBBSLi7jCiSmj3VRh2+Goq2N7Xxu0=
-github.com/pelletier/go-toml/v2 v2.0.9/go.mod h1:tJU2Z3ZkXwnxa4DPO899bsyIoywizdUvyaeZurnPPDc=
+github.com/mattn/go-isatty v0.0.20 h1:xfD0iDuEKnDkl03q4limB+vH+GxLEtL/jb4xVJSWWEY=
+github.com/mattn/go-isatty v0.0.20/go.mod h1:W+V8PltTTMOvKvAeJH7IuucS94S2C6jfK/D7dTCTo3Y=
+github.com/pelletier/go-toml/v2 v2.1.1 h1:LWAJwfNvjQZCFIDKWYQaM62NcYeYViCmWIwmOStowAI=
+github.com/pelletier/go-toml/v2 v2.1.1/go.mod h1:tJU2Z3ZkXwnxa4DPO899bsyIoywizdUvyaeZurnPPDc=
 github.com/pkg/diff v0.0.0-20210226163009-20ebb0f2a09e h1:aoZm08cpOy4WuID//EZDgcC4zIxODThtZNPirFr42+A=
 github.com/pkg/diff v0.0.0-20210226163009-20ebb0f2a09e/go.mod h1:pJLUxLENpZxwdsKMEsNbx1VGcRFpLqf3715MtcvvzbA=
 github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM=
 github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
 github.com/russross/blackfriday/v2 v2.1.0/go.mod h1:+Rmxgy9KzJVeS9/2gXHxylqXiyQDYRxCVz55jmeOWTM=
-github.com/spf13/cobra v1.7.0 h1:hyqWnYt1ZQShIddO5kBpj3vu05/++x6tJ6dg8EC572I=
-github.com/spf13/cobra v1.7.0/go.mod h1:uLxZILRyS/50WlhOIKD7W6V5bgeIt+4sICxh6uRMrb0=
+github.com/spf13/cobra v1.8.0 h1:7aJaZx1B85qltLMc546zn58BxxfZdR/W22ej9CFoEf0=
+github.com/spf13/cobra v1.8.0/go.mod h1:WXLWApfZ71AjXPya3WOlMsY9yMs7YeiHhFVlvLyhcho=
 github.com/spf13/pflag v1.0.5 h1:iy+VFUOCP1a+8yFto/drg2CJ5u0yRoB7fZw3DKv/JXA=
 github.com/spf13/pflag v1.0.5/go.mod h1:McXfInJRrz4CZXVZOBLb0bTZqETkiAhM9Iw0y3An2Bg=
 github.com/stretchr/objx v0.1.0/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME=
@@ -50,15 +62,20 @@ github.com/stretchr/testify v1.7.1/go.mod h1:6Fq8oRcR53rry900zMqJjRRixrwX3KX962/
 github.com/stretchr/testify v1.8.0/go.mod h1:yNjHg4UonilssWZ8iaSj1OCr/vHnekPRkoO+kdMU+MU=
 github.com/stretchr/testify v1.8.4 h1:CcVxjf3Q8PM0mHUKJCdn+eZZtm5yQwehR5yeSVQQcUk=
 github.com/stretchr/testify v1.8.4/go.mod h1:sz/lmYIOXD/1dqDmKjjqLyZ2RngseejIcXlSw2iwfAo=
-golang.org/x/net v0.14.0 h1:BONx9s002vGdD9umnlX1Po8vOZmrgH34qlHcD1MfK14=
-golang.org/x/net v0.14.0/go.mod h1:PpSgVXXLK0OxS0F31C1/tv6XNguvCrnXIDrFMspZIUI=
+github.com/yuin/gopher-lua v1.1.1 h1:kYKnWBjvbNP4XLT3+bPEwAXJx262OhaHDWDVOPjL46M=
+github.com/yuin/gopher-lua v1.1.1/go.mod h1:GBR0iDaNXjAgGg9zfCvksxSRnQx76gclCIb7kdAd1Pw=
+golang.org/x/crypto v0.21.0 h1:X31++rzVUdKhX5sWmSOFZxx8UW/ldWx55cbf08iNAMA=
+golang.org/x/crypto v0.21.0/go.mod h1:0BP7YvVV9gBbVKyeTG0Gyn+gZm94bibOW5BjDEYAOMs=
+golang.org/x/net v0.22.0 h1:9sGLhx7iRIHEiX0oAJ3MRZMUCElJgy7Br1nO+AMN3Tc=
+golang.org/x/net v0.22.0/go.mod h1:JKghWKKOSdJwpW2GEx0Ja7fmaKnMsbu+MWVZTokSYmg=
 golang.org/x/sys v0.0.0-20220811171246-fbc7d0a398ab/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
-golang.org/x/sys v0.11.0 h1:eG7RXZHdqOJ1i+0lgLgCpSXAp6M3LYlAo6osgSi0xOM=
-golang.org/x/sys v0.11.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
-golang.org/x/text v0.12.0 h1:k+n5B8goJNdU7hSvEtMUz3d1Q6D/XW4COJSJR6fN0mc=
-golang.org/x/text v0.12.0/go.mod h1:TvPlkZtksWOMsz7fbANvkp4WM8x/WCo/om8BMLbz+aE=
-golang.org/x/xerrors v0.0.0-20220609144429-65e65417b02f h1:uF6paiQQebLeSXkrTqHqz0MXhXXS1KgF41eUdBNvxK0=
-golang.org/x/xerrors v0.0.0-20220609144429-65e65417b02f/go.mod h1:K8+ghG5WaK9qNqU5K3HdILfMLy1f3aNYFI/wnl100a8=
+golang.org/x/sys v0.6.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
+golang.org/x/sys v0.18.0 h1:DBdB3niSjOA/O0blCZBqDefyWNYveAYMNF1Wum0DYQ4=
+golang.org/x/sys v0.18.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA=
+golang.org/x/text v0.14.0 h1:ScX5w1eTa3QqT8oi6+ziP7dTV1S2+ALU0bI+0zXKWiQ=
+golang.org/x/text v0.14.0/go.mod h1:18ZOQIKpY8NJVqYksKHtTdi31H5itFRjB5/qKTNYzSU=
+golang.org/x/xerrors v0.0.0-20220907171357-04be3eba64a2 h1:H2TDz8ibqkAF6YGhCdN3jS9O0/s90v0rJh3X/OLHEUk=
+golang.org/x/xerrors v0.0.0-20220907171357-04be3eba64a2/go.mod h1:K8+ghG5WaK9qNqU5K3HdILfMLy1f3aNYFI/wnl100a8=
 gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405 h1:yhCVgyC4o1eVCa2tZl7eS0r+SDo693bJlVdllGtEeKM=
 gopkg.in/check.v1 v0.0.0-20161208181325-20d25e280405/go.mod h1:Co6ibVJAznAaIkqp8huTwlJQCZ016jof/cbN4VW5Yz0=
 gopkg.in/op/go-logging.v1 v1.0.0-20160211212156-b2cb9fa56473 h1:6D+BvnJ/j6e222UW8s2qTSe3wGBtvo0MbVQG/c5k8RE=

how-it-works.md

@@ -29,7 +29,7 @@ It pipes the current, lets call it 'root' context through the `lhs` expression o
 cat
 ```
 
-Sidenote: this node holds not only its value 'cat', but comments and metadata too, including path and parent information.
+Side note: this node holds not only its value 'cat', but comments and metadata too, including path and parent information.
 
 The `=` operator then pipes the 'root' context through the `rhs` expression of `.b` to return the node
 

pkg/properties/decoder_properties.go

@@ -0,0 +1,117 @@
+package properties
+
+import (
+	"bytes"
+	"fmt"
+	"io"
+	"strconv"
+	"strings"
+
+	"github.com/magiconair/properties"
+	"github.com/mikefarah/yq/v4/pkg/yqlib"
+)
+
+type propertiesDecoder struct {
+	reader   io.Reader
+	finished bool
+	d        yqlib.DataTreeNavigator
+}
+
+func NewPropertiesDecoder() yqlib.Decoder {
+	return &propertiesDecoder{d: yqlib.NewDataTreeNavigator(), finished: false}
+}
+
+func (dec *propertiesDecoder) Init(reader io.Reader) error {
+	dec.reader = reader
+	dec.finished = false
+	return nil
+}
+
+func parsePropKey(key string) []interface{} {
+	pathStrArray := strings.Split(key, ".")
+	path := make([]interface{}, len(pathStrArray))
+	for i, pathStr := range pathStrArray {
+		num, err := strconv.ParseInt(pathStr, 10, 32)
+		if err == nil {
+			path[i] = num
+		} else {
+			path[i] = pathStr
+		}
+	}
+	return path
+}
+
+func (dec *propertiesDecoder) processComment(c string) string {
+	if c == "" {
+		return ""
+	}
+	return "# " + c
+}
+
+func (dec *propertiesDecoder) applyPropertyComments(context yqlib.Context, path []interface{}, comments []string) error {
+	rhsCandidateNode := &yqlib.CandidateNode{
+		Tag:         "!!str",
+		Value:       fmt.Sprintf("%v", path[len(path)-1]),
+		HeadComment: dec.processComment(strings.Join(comments, "\n")),
+		Kind:        yqlib.ScalarNode,
+	}
+	rhsCandidateNode.Tag = rhsCandidateNode.GuessTagFromCustomType()
+	return dec.d.DeeplyAssignKey(context, path, rhsCandidateNode)
+}
+
+func (dec *propertiesDecoder) applyProperty(context yqlib.Context, properties *properties.Properties, key string) error {
+	value, _ := properties.Get(key)
+	path := parsePropKey(key)
+
+	propertyComments := properties.GetComments(key)
+	if len(propertyComments) > 0 {
+		err := dec.applyPropertyComments(context, path, propertyComments)
+		if err != nil {
+			return nil
+		}
+	}
+
+	rhsNode := yqlib.CreateStringScalarNode(value)
+	rhsNode.Tag = rhsNode.GuessTagFromCustomType()
+
+	return dec.d.DeeplyAssign(context, path, rhsNode)
+}
+
+func (dec *propertiesDecoder) Decode() (*yqlib.CandidateNode, error) {
+	if dec.finished {
+		return nil, io.EOF
+	}
+	buf := new(bytes.Buffer)
+
+	if _, err := buf.ReadFrom(dec.reader); err != nil {
+		return nil, err
+	}
+	if buf.Len() == 0 {
+		dec.finished = true
+		return nil, io.EOF
+	}
+	properties, err := properties.LoadString(buf.String())
+	if err != nil {
+		return nil, err
+	}
+	properties.DisableExpansion = true
+
+	rootMap := &yqlib.CandidateNode{
+		Kind: yqlib.MappingNode,
+		Tag:  "!!map",
+	}
+
+	context := yqlib.Context{}
+	context = context.SingleChildContext(rootMap)
+
+	for _, key := range properties.Keys() {
+		if err := dec.applyProperty(context, properties, key); err != nil {
+			return nil, err
+		}
+
+	}
+	dec.finished = true
+
+	return rootMap, nil
+
+}

pkg/properties/encoder_properties.go

@@ -1,4 +1,4 @@
-package yqlib
+package properties
 
 import (
 	"bufio"
@@ -8,16 +8,16 @@ import (
 	"strings"
 
 	"github.com/magiconair/properties"
-	yaml "gopkg.in/yaml.v3"
+	"github.com/mikefarah/yq/v4/pkg/yqlib"
 )
 
 type propertiesEncoder struct {
-	unwrapScalar bool
+	prefs PropertiesPreferences
 }
 
-func NewPropertiesEncoder(unwrapScalar bool) Encoder {
+func NewPropertiesEncoder(prefs PropertiesPreferences) yqlib.Encoder {
 	return &propertiesEncoder{
-		unwrapScalar: unwrapScalar,
+		prefs: prefs,
 	}
 }
 
@@ -25,7 +25,7 @@ func (pe *propertiesEncoder) CanHandleAliases() bool {
 	return false
 }
 
-func (pe *propertiesEncoder) PrintDocumentSeparator(writer io.Writer) error {
+func (pe *propertiesEncoder) PrintDocumentSeparator(_ io.Writer) error {
 	return nil
 }
 
@@ -37,14 +37,14 @@ func (pe *propertiesEncoder) PrintLeadingContent(writer io.Writer, content strin
 		if errReading != nil && !errors.Is(errReading, io.EOF) {
 			return errReading
 		}
-		if strings.Contains(readline, "$yqDocSeperator$") {
+		if strings.Contains(readline, "$yqDocSeparator$") {
 
 			if err := pe.PrintDocumentSeparator(writer); err != nil {
 				return err
 			}
 
 		} else {
-			if err := writeString(writer, readline); err != nil {
+			if err := yqlib.WriteString(writer, readline); err != nil {
 				return err
 			}
 		}
@@ -52,7 +52,7 @@ func (pe *propertiesEncoder) PrintLeadingContent(writer io.Writer, content strin
 		if errors.Is(errReading, io.EOF) {
 			if readline != "" {
 				// the last comment we read didn't have a newline, put one in
-				if err := writeString(writer, "\n"); err != nil {
+				if err := yqlib.WriteString(writer, "\n"); err != nil {
 					return err
 				}
 			}
@@ -62,14 +62,15 @@ func (pe *propertiesEncoder) PrintLeadingContent(writer io.Writer, content strin
 	return nil
 }
 
-func (pe *propertiesEncoder) Encode(writer io.Writer, node *yaml.Node) error {
+func (pe *propertiesEncoder) Encode(writer io.Writer, node *yqlib.CandidateNode) error {
 
-	if node.Kind == yaml.ScalarNode {
-		return writeString(writer, node.Value+"\n")
+	if node.Kind == yqlib.ScalarNode {
+		return yqlib.WriteString(writer, node.Value+"\n")
 	}
 
-	mapKeysToStrings(node)
+	node.ConvertKeysToStrings()
 	p := properties.NewProperties()
+	p.WriteSeparator = pe.prefs.KeyValueSeparator
 	err := pe.doEncode(p, node, "", nil)
 	if err != nil {
 		return err
@@ -79,34 +80,32 @@ func (pe *propertiesEncoder) Encode(writer io.Writer, node *yaml.Node) error {
 	return err
 }
 
-func (pe *propertiesEncoder) doEncode(p *properties.Properties, node *yaml.Node, path string, keyNode *yaml.Node) error {
+func (pe *propertiesEncoder) doEncode(p *properties.Properties, node *yqlib.CandidateNode, path string, keyNode *yqlib.CandidateNode) error {
 
 	comments := ""
 	if keyNode != nil {
 		// include the key node comments if present
-		comments = headAndLineComment(keyNode)
+		comments = keyNode.CleanHeadAndLineComment()
 	}
-	comments = comments + headAndLineComment(node)
+	comments = comments + node.CleanHeadAndLineComment()
 	commentsWithSpaces := strings.ReplaceAll(comments, "\n", "\n ")
 	p.SetComments(path, strings.Split(commentsWithSpaces, "\n"))
 
 	switch node.Kind {
-	case yaml.ScalarNode:
+	case yqlib.ScalarNode:
 		var nodeValue string
-		if pe.unwrapScalar || !strings.Contains(node.Value, " ") {
+		if pe.prefs.UnwrapScalar || !strings.Contains(node.Value, " ") {
 			nodeValue = node.Value
 		} else {
 			nodeValue = fmt.Sprintf("%q", node.Value)
 		}
 		_, _, err := p.Set(path, nodeValue)
 		return err
-	case yaml.DocumentNode:
-		return pe.doEncode(p, node.Content[0], path, node)
-	case yaml.SequenceNode:
+	case yqlib.SequenceNode:
 		return pe.encodeArray(p, node.Content, path)
-	case yaml.MappingNode:
+	case yqlib.MappingNode:
 		return pe.encodeMap(p, node.Content, path)
-	case yaml.AliasNode:
+	case yqlib.AliasNode:
 		return pe.doEncode(p, node.Alias, path, nil)
 	default:
 		return fmt.Errorf("Unsupported node %v", node.Tag)
@@ -117,10 +116,17 @@ func (pe *propertiesEncoder) appendPath(path string, key interface{}) string {
 	if path == "" {
 		return fmt.Sprintf("%v", key)
 	}
+	switch key.(type) {
+	case int:
+		if pe.prefs.UseArrayBrackets {
+			return fmt.Sprintf("%v[%v]", path, key)
+		}
+
+	}
 	return fmt.Sprintf("%v.%v", path, key)
 }
 
-func (pe *propertiesEncoder) encodeArray(p *properties.Properties, kids []*yaml.Node, path string) error {
+func (pe *propertiesEncoder) encodeArray(p *properties.Properties, kids []*yqlib.CandidateNode, path string) error {
 	for index, child := range kids {
 		err := pe.doEncode(p, child, pe.appendPath(path, index), nil)
 		if err != nil {
@@ -130,7 +136,7 @@ func (pe *propertiesEncoder) encodeArray(p *properties.Properties, kids []*yaml.
 	return nil
 }
 
-func (pe *propertiesEncoder) encodeMap(p *properties.Properties, kids []*yaml.Node, path string) error {
+func (pe *propertiesEncoder) encodeMap(p *properties.Properties, kids []*yqlib.CandidateNode, path string) error {
 	for index := 0; index < len(kids); index = index + 2 {
 		key := kids[index]
 		value := kids[index+1]

pkg/properties/encoder_properties_test.go

@@ -0,0 +1,208 @@
+package properties
+
+import (
+	"bufio"
+	"bytes"
+	"strings"
+	"testing"
+
+	"github.com/mikefarah/yq/v4/test"
+)
+
+type keyValuePair struct {
+	key     string
+	value   string
+	comment string
+}
+
+func (kv *keyValuePair) String(unwrap bool, sep string) string {
+	builder := strings.Builder{}
+
+	if kv.comment != "" {
+		builder.WriteString(kv.comment)
+		builder.WriteString("\n")
+	}
+
+	builder.WriteString(kv.key)
+	builder.WriteString(sep)
+
+	if unwrap {
+		builder.WriteString(kv.value)
+	} else {
+		builder.WriteString("\"")
+		builder.WriteString(kv.value)
+		builder.WriteString("\"")
+	}
+
+	return builder.String()
+}
+
+type testProperties struct {
+	pairs []keyValuePair
+}
+
+func (tp *testProperties) String(unwrap bool, sep string) string {
+	kvs := []string{}
+
+	for _, kv := range tp.pairs {
+		kvs = append(kvs, kv.String(unwrap, sep))
+	}
+
+	return strings.Join(kvs, "\n")
+}
+
+func yamlToProps(sampleYaml string, unwrapScalar bool, separator string) string {
+	var output bytes.Buffer
+	writer := bufio.NewWriter(&output)
+
+	var propsEncoder = NewPropertiesEncoder(PropertiesPreferences{KeyValueSeparator: separator, UnwrapScalar: unwrapScalar})
+	inputs, err := readDocuments(strings.NewReader(sampleYaml), "sample.yml", 0, NewYamlDecoder(ConfiguredYamlPreferences))
+	if err != nil {
+		panic(err)
+	}
+	node := inputs.Front().Value.(*CandidateNode)
+	err = propsEncoder.Encode(writer, node)
+	if err != nil {
+		panic(err)
+	}
+	writer.Flush()
+
+	return strings.TrimSuffix(output.String(), "\n")
+}
+
+func doTest(t *testing.T, sampleYaml string, props testProperties, testUnwrapped, testWrapped bool) {
+	wraps := []bool{}
+	if testUnwrapped {
+		wraps = append(wraps, true)
+	}
+	if testWrapped {
+		wraps = append(wraps, false)
+	}
+
+	for _, unwrap := range wraps {
+		for _, sep := range []string{" = ", ";", "=", " "} {
+			var actualProps = yamlToProps(sampleYaml, unwrap, sep)
+			test.AssertResult(t, props.String(unwrap, sep), actualProps)
+		}
+	}
+}
+
+func TestPropertiesEncoderSimple(t *testing.T) {
+	var sampleYaml = `a: 'bob cool'`
+
+	doTest(
+		t, sampleYaml,
+		testProperties{
+			pairs: []keyValuePair{
+				{
+					key:   "a",
+					value: "bob cool",
+				},
+			},
+		},
+		true, true,
+	)
+}
+
+func TestPropertiesEncoderSimpleWithComments(t *testing.T) {
+	var sampleYaml = `a: 'bob cool' # line`
+
+	doTest(
+		t, sampleYaml,
+		testProperties{
+			pairs: []keyValuePair{
+				{
+					key:     "a",
+					value:   "bob cool",
+					comment: "# line",
+				},
+			},
+		},
+		true, true,
+	)
+}
+
+func TestPropertiesEncoderDeep(t *testing.T) {
+	var sampleYaml = `a: 
+  b: "bob cool"
+`
+
+	doTest(
+		t, sampleYaml,
+		testProperties{
+			pairs: []keyValuePair{
+				{
+					key:   "a.b",
+					value: "bob cool",
+				},
+			},
+		},
+		true, true,
+	)
+}
+
+func TestPropertiesEncoderDeepWithComments(t *testing.T) {
+	var sampleYaml = `a:  # a thing
+  b: "bob cool" # b thing
+`
+
+	doTest(
+		t, sampleYaml,
+		testProperties{
+			pairs: []keyValuePair{
+				{
+					key:     "a.b",
+					value:   "bob cool",
+					comment: "# b thing",
+				},
+			},
+		},
+		true, true,
+	)
+}
+
+func TestPropertiesEncoderArray_Unwrapped(t *testing.T) {
+	var sampleYaml = `a: 
+  b: [{c: dog}, {c: cat}]
+`
+
+	doTest(
+		t, sampleYaml,
+		testProperties{
+			pairs: []keyValuePair{
+				{
+					key:   "a.b.0.c",
+					value: "dog",
+				},
+				{
+					key:   "a.b.1.c",
+					value: "cat",
+				},
+			},
+		},
+		true, false,
+	)
+}
+
+func TestPropertiesEncoderArray_Wrapped(t *testing.T) {
+	var sampleYaml = `a: 
+  b: [{c: dog named jim}, {c: cat named jim}]
+`
+
+	doTest(
+		t, sampleYaml,
+		testProperties{
+			pairs: []keyValuePair{
+				{
+					key:   "a.b.0.c",
+					value: "dog named jim",
+				},
+				{
+					key:   "a.b.1.c",
+					value: "cat named jim",
+				},
+			},
+		},
+		false, true,
+	)
+}

pkg/properties/properties.go

@@ -0,0 +1,45 @@
+package properties
+
+import "github.com/mikefarah/yq/v4/pkg/yqlib"
+
+type PropertiesPreferences struct {
+	UnwrapScalar      bool
+	KeyValueSeparator string
+	UseArrayBrackets  bool
+}
+
+func NewDefaultPropertiesPreferences() PropertiesPreferences {
+	return PropertiesPreferences{
+		UnwrapScalar:      true,
+		KeyValueSeparator: " = ",
+		UseArrayBrackets:  false,
+	}
+}
+
+func (p *PropertiesPreferences) Copy() PropertiesPreferences {
+	return PropertiesPreferences{
+		UnwrapScalar:      p.UnwrapScalar,
+		KeyValueSeparator: p.KeyValueSeparator,
+		UseArrayBrackets:  p.UseArrayBrackets,
+	}
+}
+
+var PropertiesFormat = &yqlib.Format{"props", []string{"p", "properties"}, "properties",
+	func() yqlib.Encoder { return NewPropertiesEncoder(ConfiguredPropertiesPreferences) },
+	func() yqlib.Decoder { return NewPropertiesDecoder() },
+	nil,
+}
+
+var propertyYqRules = []*yqlib.ParticipleYqRule{
+	{"PropertiesDecode", `from_?props|@propsd`, yqlib.CreateDecodeOpYqAction(PropertiesFormat), 0},
+	{"PropsEncode", `to_?props|@props`, yqlib.CreateEncodeOpYqAction(PropertiesFormat, 2), 0},
+	{"LoadProperties", `load_?props`, yqlib.CreateLoadOpYqAction(NewPropertiesDecoder()), 0},
+}
+
+func RegisterPropertiesFormat() {
+	yqlib.RegisterFormat(PropertiesFormat)
+	yqlib.RegisterRules(propertyYqRules)
+
+}
+
+var ConfiguredPropertiesPreferences = NewDefaultPropertiesPreferences()

pkg/xml/decoder_xml.go

@@ -1,6 +1,6 @@
 //go:build !yq_noxml
 
-package yqlib
+package xml
 
 import (
 	"encoding/xml"
@@ -11,8 +11,8 @@ import (
 	"strings"
 	"unicode"
 
+	"github.com/mikefarah/yq/v4/pkg/yqlib"
 	"golang.org/x/net/html/charset"
-	yaml "gopkg.in/yaml.v3"
 )
 
 type xmlDecoder struct {
@@ -22,7 +22,7 @@ type xmlDecoder struct {
 	prefs        XmlPreferences
 }
 
-func NewXMLDecoder(prefs XmlPreferences) Decoder {
+func NewXMLDecoder(prefs XmlPreferences) yqlib.Decoder {
 	return &xmlDecoder{
 		finished: false,
 		prefs:    prefs,
@@ -36,14 +36,14 @@ func (dec *xmlDecoder) Init(reader io.Reader) error {
 	return nil
 }
 
-func (dec *xmlDecoder) createSequence(nodes []*xmlNode) (*yaml.Node, error) {
-	yamlNode := &yaml.Node{Kind: yaml.SequenceNode, Tag: "!!seq"}
+func (dec *xmlDecoder) createSequence(nodes []*xmlNode) (*yqlib.CandidateNode, error) {
+	yamlNode := &yqlib.CandidateNode{Kind: yqlib.SequenceNode, Tag: "!!seq"}
 	for _, child := range nodes {
 		yamlChild, err := dec.convertToYamlNode(child)
 		if err != nil {
 			return nil, err
 		}
-		yamlNode.Content = append(yamlNode.Content, yamlChild)
+		yamlNode.AddChild(yamlChild)
 	}
 
 	return yamlNode, nil
@@ -64,36 +64,36 @@ func (dec *xmlDecoder) processComment(c string) string {
 	return replacement
 }
 
-func (dec *xmlDecoder) createMap(n *xmlNode) (*yaml.Node, error) {
-	log.Debug("createMap: headC: %v, lineC: %v, footC: %v", n.HeadComment, n.LineComment, n.FootComment)
-	yamlNode := &yaml.Node{Kind: yaml.MappingNode, Tag: "!!map"}
+func (dec *xmlDecoder) createMap(n *xmlNode) (*yqlib.CandidateNode, error) {
+	yqlib.GetLogger().Debug("createMap: headC: %v, lineC: %v, footC: %v", n.HeadComment, n.LineComment, n.FootComment)
+	yamlNode := &yqlib.CandidateNode{Kind: yqlib.MappingNode, Tag: "!!map"}
 
 	if len(n.Data) > 0 {
-		log.Debugf("creating content node for map: %v", dec.prefs.ContentName)
+		yqlib.GetLogger().Debugf("creating content node for map: %v", dec.prefs.ContentName)
 		label := dec.prefs.ContentName
-		labelNode := createScalarNode(label, label)
+		labelNode := yqlib.CreateScalarNode(label, label)
 		labelNode.HeadComment = dec.processComment(n.HeadComment)
 		labelNode.LineComment = dec.processComment(n.LineComment)
 		labelNode.FootComment = dec.processComment(n.FootComment)
-		yamlNode.Content = append(yamlNode.Content, labelNode, dec.createValueNodeFromData(n.Data))
+		yamlNode.AddKeyValueChild(labelNode, dec.createValueNodeFromData(n.Data))
 	}
 
 	for i, keyValuePair := range n.Children {
 		label := keyValuePair.K
 		children := keyValuePair.V
-		labelNode := createScalarNode(label, label)
-		var valueNode *yaml.Node
+		labelNode := yqlib.CreateScalarNode(label, label)
+		var valueNode *yqlib.CandidateNode
 		var err error
 
 		if i == 0 {
-			log.Debugf("head comment here")
+			yqlib.GetLogger().Debugf("head comment here")
 			labelNode.HeadComment = dec.processComment(n.HeadComment)
 
 		}
-		log.Debugf("label=%v, i=%v, keyValuePair.FootComment: %v", label, i, keyValuePair.FootComment)
+		yqlib.GetLogger().Debugf("label=%v, i=%v, keyValuePair.FootComment: %v", label, i, keyValuePair.FootComment)
 		labelNode.FootComment = dec.processComment(keyValuePair.FootComment)
 
-		log.Debug("len of children in %v is %v", label, len(children))
+		yqlib.GetLogger().Debug("len of children in %v is %v", label, len(children))
 		if len(children) > 1 {
 			valueNode, err = dec.createSequence(children)
 			if err != nil {
@@ -106,7 +106,7 @@ func (dec *xmlDecoder) createMap(n *xmlNode) (*yaml.Node, error) {
 			if len(children[0].Children) == 0 && children[0].HeadComment != "" {
 				if len(children[0].Data) > 0 {
 
-					log.Debug("scalar comment hack, currentlabel [%v]", labelNode.HeadComment)
+					yqlib.GetLogger().Debug("scalar comment hack, currentlabel [%v]", labelNode.HeadComment)
 					labelNode.HeadComment = joinComments([]string{labelNode.HeadComment, strings.TrimSpace(children[0].HeadComment)}, "\n")
 					children[0].HeadComment = ""
 				} else {
@@ -120,39 +120,39 @@ func (dec *xmlDecoder) createMap(n *xmlNode) (*yaml.Node, error) {
 				return nil, err
 			}
 		}
-		yamlNode.Content = append(yamlNode.Content, labelNode, valueNode)
+		yamlNode.AddKeyValueChild(labelNode, valueNode)
 	}
 
 	return yamlNode, nil
 }
 
-func (dec *xmlDecoder) createValueNodeFromData(values []string) *yaml.Node {
+func (dec *xmlDecoder) createValueNodeFromData(values []string) *yqlib.CandidateNode {
 	switch len(values) {
 	case 0:
-		return createScalarNode(nil, "")
+		return yqlib.CreateScalarNode(nil, "")
 	case 1:
-		return createScalarNode(values[0], values[0])
+		return yqlib.CreateScalarNode(values[0], values[0])
 	default:
-		content := make([]*yaml.Node, 0)
+		content := make([]*yqlib.CandidateNode, 0)
 		for _, value := range values {
-			content = append(content, createScalarNode(value, value))
+			content = append(content, yqlib.CreateScalarNode(value, value))
 		}
-		return &yaml.Node{
-			Kind:    yaml.SequenceNode,
+		return &yqlib.CandidateNode{
+			Kind:    yqlib.SequenceNode,
 			Tag:     "!!seq",
 			Content: content,
 		}
 	}
 }
 
-func (dec *xmlDecoder) convertToYamlNode(n *xmlNode) (*yaml.Node, error) {
+func (dec *xmlDecoder) convertToYamlNode(n *xmlNode) (*yqlib.CandidateNode, error) {
 	if len(n.Children) > 0 {
 		return dec.createMap(n)
 	}
 
 	scalar := dec.createValueNodeFromData(n.Data)
 
-	log.Debug("scalar (%v), headC: %v, lineC: %v, footC: %v", scalar.Tag, n.HeadComment, n.LineComment, n.FootComment)
+	yqlib.GetLogger().Debug("scalar (%v), headC: %v, lineC: %v, footC: %v", scalar.Tag, n.HeadComment, n.LineComment, n.FootComment)
 	scalar.HeadComment = dec.processComment(n.HeadComment)
 	scalar.LineComment = dec.processComment(n.LineComment)
 	if scalar.Tag == "!!seq" {
@@ -165,7 +165,7 @@ func (dec *xmlDecoder) convertToYamlNode(n *xmlNode) (*yaml.Node, error) {
 	return scalar, nil
 }
 
-func (dec *xmlDecoder) Decode() (*CandidateNode, error) {
+func (dec *xmlDecoder) Decode() (*yqlib.CandidateNode, error) {
 	if dec.finished {
 		return nil, io.EOF
 	}
@@ -189,12 +189,7 @@ func (dec *xmlDecoder) Decode() (*CandidateNode, error) {
 	dec.readAnything = true
 	dec.finished = true
 
-	return &CandidateNode{
-		Node: &yaml.Node{
-			Kind:    yaml.DocumentNode,
-			Content: []*yaml.Node{firstNode},
-		},
-	}, nil
+	return firstNode, nil
 }
 
 type xmlNode struct {
@@ -217,17 +212,17 @@ func (n *xmlNode) AddChild(s string, c *xmlNode) {
 	if n.Children == nil {
 		n.Children = make([]*xmlChildrenKv, 0)
 	}
-	log.Debug("looking for %s", s)
+	yqlib.GetLogger().Debug("looking for %s", s)
 	// see if we can find an existing entry to add to
 	for _, childEntry := range n.Children {
 		if childEntry.K == s {
-			log.Debug("found it, appending an entry%s", s)
+			yqlib.GetLogger().Debug("found it, appending an entry%s", s)
 			childEntry.V = append(childEntry.V, c)
-			log.Debug("yay len of children in %v is %v", s, len(childEntry.V))
+			yqlib.GetLogger().Debug("yay len of children in %v is %v", s, len(childEntry.V))
 			return
 		}
 	}
-	log.Debug("not there, making a new one %s", s)
+	yqlib.GetLogger().Debug("not there, making a new one %s", s)
 	n.Children = append(n.Children, &xmlChildrenKv{K: s, V: []*xmlNode{c}})
 }
 
@@ -273,7 +268,7 @@ func (dec *xmlDecoder) decodeXML(root *xmlNode) error {
 
 		switch se := t.(type) {
 		case xml.StartElement:
-			log.Debug("start element %v", se.Name.Local)
+			yqlib.GetLogger().Debug("start element %v", se.Name.Local)
 			elem.state = "started"
 			// Build new a new current element and link it to its parent
 			elem = &element{
@@ -302,10 +297,14 @@ func (dec *xmlDecoder) decodeXML(root *xmlNode) error {
 			if len(newBit) > 0 {
 				elem.n.Data = append(elem.n.Data, newBit)
 				elem.state = "chardata"
-				log.Debug("chardata [%v] for %v", elem.n.Data, elem.label)
+				yqlib.GetLogger().Debug("chardata [%v] for %v", elem.n.Data, elem.label)
 			}
 		case xml.EndElement:
-			log.Debug("end element %v", elem.label)
+			if elem == nil {
+				yqlib.GetLogger().Debug("no element, probably bad xml")
+				continue
+			}
+			yqlib.GetLogger().Debug("end element %v", elem.label)
 			elem.state = "finished"
 			// And add it to its parent list
 			if elem.parent != nil {
@@ -321,10 +320,10 @@ func (dec *xmlDecoder) decodeXML(root *xmlNode) error {
 				applyFootComment(elem, commentStr)
 
 			} else if elem.state == "chardata" {
-				log.Debug("got a line comment for (%v) %v: [%v]", elem.state, elem.label, commentStr)
+				yqlib.GetLogger().Debug("got a line comment for (%v) %v: [%v]", elem.state, elem.label, commentStr)
 				elem.n.LineComment = joinComments([]string{elem.n.LineComment, commentStr}, " ")
 			} else {
-				log.Debug("got a head comment for (%v) %v: [%v]", elem.state, elem.label, commentStr)
+				yqlib.GetLogger().Debug("got a head comment for (%v) %v: [%v]", elem.state, elem.label, commentStr)
 				elem.n.HeadComment = joinComments([]string{elem.n.HeadComment, commentStr}, " ")
 			}
 
@@ -349,7 +348,7 @@ func applyFootComment(elem *element, commentStr string) {
 	if len(elem.n.Children) > 0 {
 		lastChildIndex := len(elem.n.Children) - 1
 		childKv := elem.n.Children[lastChildIndex]
-		log.Debug("got a foot comment, putting on last child for %v: [%v]", childKv.K, commentStr)
+		yqlib.GetLogger().Debug("got a foot comment, putting on last child for %v: [%v]", childKv.K, commentStr)
 		// if it's an array of scalars, put the foot comment on the scalar itself
 		if len(childKv.V) > 0 && len(childKv.V[0].Children) == 0 {
 			nodeToUpdate := childKv.V[len(childKv.V)-1]
@@ -358,7 +357,7 @@ func applyFootComment(elem *element, commentStr string) {
 			childKv.FootComment = joinComments([]string{elem.n.FootComment, commentStr}, " ")
 		}
 	} else {
-		log.Debug("got a foot comment for %v: [%v]", elem.label, commentStr)
+		yqlib.GetLogger().Debug("got a foot comment for %v: [%v]", elem.label, commentStr)
 		elem.n.FootComment = joinComments([]string{elem.n.FootComment, commentStr}, " ")
 	}
 }

pkg/xml/encoder_xml.go

@@ -1,6 +1,6 @@
 //go:build !yq_noxml
 
-package yqlib
+package xml
 
 import (
 	"encoding/xml"
@@ -9,7 +9,7 @@ import (
 	"regexp"
 	"strings"
 
-	yaml "gopkg.in/yaml.v3"
+	"github.com/mikefarah/yq/v4/pkg/yqlib"
 )
 
 type xmlEncoder struct {
@@ -19,10 +19,10 @@ type xmlEncoder struct {
 	leadingContent string
 }
 
-func NewXMLEncoder(indent int, prefs XmlPreferences) Encoder {
+func NewXMLEncoder(prefs XmlPreferences) yqlib.Encoder {
 	var indentString = ""
 
-	for index := 0; index < indent; index++ {
+	for index := 0; index < prefs.Indent; index++ {
 		indentString = indentString + " "
 	}
 	return &xmlEncoder{indentString, nil, prefs, ""}
@@ -32,28 +32,27 @@ func (e *xmlEncoder) CanHandleAliases() bool {
 	return false
 }
 
-func (e *xmlEncoder) PrintDocumentSeparator(writer io.Writer) error {
+func (e *xmlEncoder) PrintDocumentSeparator(_ io.Writer) error {
 	return nil
 }
 
-func (e *xmlEncoder) PrintLeadingContent(writer io.Writer, content string) error {
+func (e *xmlEncoder) PrintLeadingContent(_ io.Writer, content string) error {
 	e.leadingContent = content
 	return nil
 }
 
-func (e *xmlEncoder) Encode(writer io.Writer, node *yaml.Node) error {
+func (e *xmlEncoder) Encode(writer io.Writer, node *yqlib.CandidateNode) error {
 	encoder := xml.NewEncoder(writer)
 	// hack so we can manually add newlines to procInst and directives
 	e.writer = writer
 	encoder.Indent("", e.indentString)
 	var newLine xml.CharData = []byte("\n")
 
-	mapNode := unwrapDoc(node)
-	if mapNode.Tag == "!!map" {
+	if node.Tag == "!!map" {
 		// make sure <?xml .. ?> processing instructions are encoded first
-		for i := 0; i < len(mapNode.Content); i += 2 {
-			key := mapNode.Content[i]
-			value := mapNode.Content[i+1]
+		for i := 0; i < len(node.Content); i += 2 {
+			key := node.Content[i]
+			value := node.Content[i+1]
 
 			if key.Value == (e.prefs.ProcInstPrefix + "xml") {
 				name := strings.Replace(key.Value, e.prefs.ProcInstPrefix, "", 1)
@@ -62,7 +61,7 @@ func (e *xmlEncoder) Encode(writer io.Writer, node *yaml.Node) error {
 					return err
 				}
 				if _, err := e.writer.Write([]byte("\n")); err != nil {
-					log.Warning("Unable to write newline, skipping: %w", err)
+					yqlib.GetLogger().Warning("Unable to write newline, skipping: %w", err)
 				}
 			}
 		}
@@ -82,29 +81,12 @@ func (e *xmlEncoder) Encode(writer io.Writer, node *yaml.Node) error {
 	}
 
 	switch node.Kind {
-	case yaml.MappingNode:
+	case yqlib.MappingNode:
 		err := e.encodeTopLevelMap(encoder, node)
 		if err != nil {
 			return err
 		}
-	case yaml.DocumentNode:
-		err := e.encodeComment(encoder, headAndLineComment(node))
-		if err != nil {
-			return err
-		}
-		unwrappedNode := unwrapDoc(node)
-		if unwrappedNode.Kind != yaml.MappingNode {
-			return fmt.Errorf("cannot encode %v to XML - only maps can be encoded", unwrappedNode.Tag)
-		}
-		err = e.encodeTopLevelMap(encoder, unwrappedNode)
-		if err != nil {
-			return err
-		}
-		err = e.encodeComment(encoder, footComment(node))
-		if err != nil {
-			return err
-		}
-	case yaml.ScalarNode:
+	case yqlib.ScalarNode:
 		var charData xml.CharData = []byte(node.Value)
 		err := encoder.EncodeToken(charData)
 		if err != nil {
@@ -112,15 +94,15 @@ func (e *xmlEncoder) Encode(writer io.Writer, node *yaml.Node) error {
 		}
 		return encoder.Flush()
 	default:
-		return fmt.Errorf("unsupported type %v", node.Tag)
+		return fmt.Errorf("cannot encode %v to XML - only maps can be encoded", node.Tag)
 	}
 
 	return encoder.EncodeToken(newLine)
 
 }
 
-func (e *xmlEncoder) encodeTopLevelMap(encoder *xml.Encoder, node *yaml.Node) error {
-	err := e.encodeComment(encoder, headAndLineComment(node))
+func (e *xmlEncoder) encodeTopLevelMap(encoder *xml.Encoder, node *yqlib.CandidateNode) error {
+	err := e.encodeComment(encoder, node.CleanHeadAndLineComment())
 	if err != nil {
 		return err
 	}
@@ -129,12 +111,12 @@ func (e *xmlEncoder) encodeTopLevelMap(encoder *xml.Encoder, node *yaml.Node) er
 		value := node.Content[i+1]
 
 		start := xml.StartElement{Name: xml.Name{Local: key.Value}}
-		log.Debugf("comments of key %v", key.Value)
-		err := e.encodeComment(encoder, headAndLineComment(key))
+		yqlib.GetLogger().Debugf("comments of key %v", key.Value)
+		err := e.encodeComment(encoder, key.CleanHeadAndLineComment())
 		if err != nil {
 			return err
 		}
-		if headAndLineComment(key) != "" {
+		if key.CleanHeadAndLineComment() != "" {
 			var newLine xml.CharData = []byte("\n")
 			err = encoder.EncodeToken(newLine)
 			if err != nil {
@@ -142,7 +124,7 @@ func (e *xmlEncoder) encodeTopLevelMap(encoder *xml.Encoder, node *yaml.Node) er
 			}
 		}
 
-		if key.Value == (e.prefs.ProcInstPrefix + "xml") {
+		if key.Value == (e.prefs.ProcInstPrefix + "xml") { //nolint
 			// dont double process these.
 		} else if strings.HasPrefix(key.Value, e.prefs.ProcInstPrefix) {
 			name := strings.Replace(key.Value, e.prefs.ProcInstPrefix, "", 1)
@@ -151,7 +133,7 @@ func (e *xmlEncoder) encodeTopLevelMap(encoder *xml.Encoder, node *yaml.Node) er
 				return err
 			}
 			if _, err := e.writer.Write([]byte("\n")); err != nil {
-				log.Warning("Unable to write newline, skipping: %w", err)
+				yqlib.GetLogger().Warning("Unable to write newline, skipping: %w", err)
 			}
 		} else if key.Value == e.prefs.DirectiveName {
 			var directive xml.Directive = []byte(value.Value)
@@ -159,48 +141,48 @@ func (e *xmlEncoder) encodeTopLevelMap(encoder *xml.Encoder, node *yaml.Node) er
 				return err
 			}
 			if _, err := e.writer.Write([]byte("\n")); err != nil {
-				log.Warning("Unable to write newline, skipping: %w", err)
+				yqlib.GetLogger().Warning("Unable to write newline, skipping: %w", err)
 			}
 		} else {
 
-			log.Debugf("recursing")
+			yqlib.GetLogger().Debugf("recursing")
 
 			err = e.doEncode(encoder, value, start)
 			if err != nil {
 				return err
 			}
 		}
-		err = e.encodeComment(encoder, footComment(key))
+		err = e.encodeComment(encoder, key.CleanFootComment())
 		if err != nil {
 			return err
 		}
 	}
-	return e.encodeComment(encoder, footComment(node))
+	return e.encodeComment(encoder, node.CleanFootComment())
 }
 
-func (e *xmlEncoder) encodeStart(encoder *xml.Encoder, node *yaml.Node, start xml.StartElement) error {
+func (e *xmlEncoder) encodeStart(encoder *xml.Encoder, node *yqlib.CandidateNode, start xml.StartElement) error {
 	err := encoder.EncodeToken(start)
 	if err != nil {
 		return err
 	}
-	return e.encodeComment(encoder, headComment(node))
+	return e.encodeComment(encoder, node.CleanHeadComment())
 }
 
-func (e *xmlEncoder) encodeEnd(encoder *xml.Encoder, node *yaml.Node, start xml.StartElement) error {
+func (e *xmlEncoder) encodeEnd(encoder *xml.Encoder, node *yqlib.CandidateNode, start xml.StartElement) error {
 	err := encoder.EncodeToken(start.End())
 	if err != nil {
 		return err
 	}
-	return e.encodeComment(encoder, footComment(node))
+	return e.encodeComment(encoder, node.CleanFootComment())
 }
 
-func (e *xmlEncoder) doEncode(encoder *xml.Encoder, node *yaml.Node, start xml.StartElement) error {
+func (e *xmlEncoder) doEncode(encoder *xml.Encoder, node *yqlib.CandidateNode, start xml.StartElement) error {
 	switch node.Kind {
-	case yaml.MappingNode:
+	case yqlib.MappingNode:
 		return e.encodeMap(encoder, node, start)
-	case yaml.SequenceNode:
+	case yqlib.SequenceNode:
 		return e.encodeArray(encoder, node, start)
-	case yaml.ScalarNode:
+	case yqlib.ScalarNode:
 		err := e.encodeStart(encoder, node, start)
 		if err != nil {
 			return err
@@ -212,7 +194,7 @@ func (e *xmlEncoder) doEncode(encoder *xml.Encoder, node *yaml.Node, start xml.S
 			return err
 		}
 
-		if err = e.encodeComment(encoder, lineComment(node)); err != nil {
+		if err = e.encodeComment(encoder, node.CleanLineComment()); err != nil {
 			return err
 		}
 
@@ -227,13 +209,13 @@ var chompRegexp = regexp.MustCompile(`\n$`)
 
 func (e *xmlEncoder) encodeComment(encoder *xml.Encoder, commentStr string) error {
 	if commentStr != "" {
-		log.Debugf("got comment [%v]", commentStr)
+		yqlib.GetLogger().Debugf("got comment [%v]", commentStr)
 		// multi line string
 		if len(commentStr) > 2 && strings.Contains(commentStr[1:len(commentStr)-1], "\n") {
 			commentStr = chompRegexp.ReplaceAllString(commentStr, "")
-			log.Debugf("chompRegexp [%v]", commentStr)
+			yqlib.GetLogger().Debugf("chompRegexp [%v]", commentStr)
 			commentStr = xmlEncodeMultilineCommentRegex.ReplaceAllString(commentStr, "$1$2")
-			log.Debugf("processed multine [%v]", commentStr)
+			yqlib.GetLogger().Debugf("processed multiline [%v]", commentStr)
 			// if the first line is non blank, add a space
 			if commentStr[0] != '\n' && commentStr[0] != ' ' {
 				commentStr = " " + commentStr
@@ -245,9 +227,9 @@ func (e *xmlEncoder) encodeComment(encoder *xml.Encoder, commentStr string) erro
 
 		if !strings.HasSuffix(commentStr, " ") && !strings.HasSuffix(commentStr, "\n") {
 			commentStr = commentStr + " "
-			log.Debugf("added suffix [%v]", commentStr)
+			yqlib.GetLogger().Debugf("added suffix [%v]", commentStr)
 		}
-		log.Debugf("encoding comment [%v]", commentStr)
+		yqlib.GetLogger().Debugf("encoding comment [%v]", commentStr)
 
 		var comment xml.Comment = []byte(commentStr)
 		err := encoder.EncodeToken(comment)
@@ -258,9 +240,9 @@ func (e *xmlEncoder) encodeComment(encoder *xml.Encoder, commentStr string) erro
 	return nil
 }
 
-func (e *xmlEncoder) encodeArray(encoder *xml.Encoder, node *yaml.Node, start xml.StartElement) error {
+func (e *xmlEncoder) encodeArray(encoder *xml.Encoder, node *yqlib.CandidateNode, start xml.StartElement) error {
 
-	if err := e.encodeComment(encoder, headAndLineComment(node)); err != nil {
+	if err := e.encodeComment(encoder, node.CleanHeadAndLineComment()); err != nil {
 		return err
 	}
 
@@ -270,7 +252,7 @@ func (e *xmlEncoder) encodeArray(encoder *xml.Encoder, node *yaml.Node, start xm
 			return err
 		}
 	}
-	return e.encodeComment(encoder, footComment(node))
+	return e.encodeComment(encoder, node.CleanFootComment())
 }
 
 func (e *xmlEncoder) isAttribute(name string) bool {
@@ -280,8 +262,8 @@ func (e *xmlEncoder) isAttribute(name string) bool {
 		!strings.HasPrefix(name, e.prefs.ProcInstPrefix)
 }
 
-func (e *xmlEncoder) encodeMap(encoder *xml.Encoder, node *yaml.Node, start xml.StartElement) error {
-	log.Debug("its a map")
+func (e *xmlEncoder) encodeMap(encoder *xml.Encoder, node *yqlib.CandidateNode, start xml.StartElement) error {
+	yqlib.GetLogger().Debug("its a map")
 
 	//first find all the attributes and put them on the start token
 	for i := 0; i < len(node.Content); i += 2 {
@@ -289,7 +271,7 @@ func (e *xmlEncoder) encodeMap(encoder *xml.Encoder, node *yaml.Node, start xml.
 		value := node.Content[i+1]
 
 		if e.isAttribute(key.Value) {
-			if value.Kind == yaml.ScalarNode {
+			if value.Kind == yqlib.ScalarNode {
 				attributeName := strings.Replace(key.Value, e.prefs.AttributePrefix, "", 1)
 				start.Attr = append(start.Attr, xml.Attr{Name: xml.Name{Local: attributeName}, Value: value.Value})
 			} else {
@@ -308,7 +290,7 @@ func (e *xmlEncoder) encodeMap(encoder *xml.Encoder, node *yaml.Node, start xml.
 		key := node.Content[i]
 		value := node.Content[i+1]
 
-		err := e.encodeComment(encoder, headAndLineComment(key))
+		err := e.encodeComment(encoder, key.CleanHeadAndLineComment())
 		if err != nil {
 			return err
 		}
@@ -325,7 +307,7 @@ func (e *xmlEncoder) encodeMap(encoder *xml.Encoder, node *yaml.Node, start xml.
 			}
 		} else if key.Value == e.prefs.ContentName {
 			// directly encode the contents
-			err = e.encodeComment(encoder, headAndLineComment(value))
+			err = e.encodeComment(encoder, value.CleanHeadAndLineComment())
 			if err != nil {
 				return err
 			}
@@ -334,7 +316,7 @@ func (e *xmlEncoder) encodeMap(encoder *xml.Encoder, node *yaml.Node, start xml.
 			if err != nil {
 				return err
 			}
-			err = e.encodeComment(encoder, footComment(value))
+			err = e.encodeComment(encoder, value.CleanFootComment())
 			if err != nil {
 				return err
 			}
@@ -345,7 +327,7 @@ func (e *xmlEncoder) encodeMap(encoder *xml.Encoder, node *yaml.Node, start xml.
 				return err
 			}
 		}
-		err = e.encodeComment(encoder, footComment(key))
+		err = e.encodeComment(encoder, key.CleanFootComment())
 		if err != nil {
 			return err
 		}

pkg/xml/xml.go

@@ -0,0 +1,72 @@
+package xml
+
+import "github.com/mikefarah/yq/v4/pkg/yqlib"
+
+type XmlPreferences struct {
+	Indent          int
+	AttributePrefix string
+	ContentName     string
+	StrictMode      bool
+	KeepNamespace   bool
+	UseRawToken     bool
+	ProcInstPrefix  string
+	DirectiveName   string
+	SkipProcInst    bool
+	SkipDirectives  bool
+}
+
+func NewDefaultXmlPreferences() XmlPreferences {
+	return XmlPreferences{
+		Indent:          2,
+		AttributePrefix: "+@",
+		ContentName:     "+content",
+		StrictMode:      false,
+		KeepNamespace:   true,
+		UseRawToken:     true,
+		ProcInstPrefix:  "+p_",
+		DirectiveName:   "+directive",
+		SkipProcInst:    false,
+		SkipDirectives:  false,
+	}
+}
+
+func (p *XmlPreferences) Copy() XmlPreferences {
+	return XmlPreferences{
+		Indent:          p.Indent,
+		AttributePrefix: p.AttributePrefix,
+		ContentName:     p.ContentName,
+		StrictMode:      p.StrictMode,
+		KeepNamespace:   p.KeepNamespace,
+		UseRawToken:     p.UseRawToken,
+		ProcInstPrefix:  p.ProcInstPrefix,
+		DirectiveName:   p.DirectiveName,
+		SkipProcInst:    p.SkipProcInst,
+		SkipDirectives:  p.SkipDirectives,
+	}
+}
+
+var ConfiguredXMLPreferences = NewDefaultXmlPreferences()
+
+var XMLFormat = &yqlib.Format{"xml", []string{"x"}, "xml",
+	func() yqlib.Encoder { return NewXMLEncoder(ConfiguredXMLPreferences) },
+	func() yqlib.Decoder { return NewXMLDecoder(ConfiguredXMLPreferences) },
+	func(indent int) yqlib.Encoder {
+		prefs := ConfiguredXMLPreferences.Copy()
+		prefs.Indent = indent
+		return NewXMLEncoder(prefs)
+	},
+}
+
+var xmlYqRules = []*yqlib.ParticipleYqRule{
+	{"XMLEncodeWithIndent", `to_?xml\([0-9]+\)`, yqlib.CreateEncodeYqActionParsingIndent(XMLFormat), 0},
+	{"XmlDecode", `from_?xml|@xmld`, yqlib.CreateDecodeOpYqAction(XMLFormat), 0},
+	{"XMLEncode", `to_?xml`, yqlib.CreateEncodeOpYqAction(XMLFormat, 2), 0},
+	{"XMLEncodeNoIndent", `@xml`, yqlib.CreateEncodeOpYqAction(XMLFormat, 0), 0},
+	{"LoadXML", `load_?xml|xml_?load`, yqlib.CreateLoadOpYqAction(NewXMLDecoder(ConfiguredXMLPreferences)), 0},
+}
+
+func RegisterXmlFormat() {
+	yqlib.RegisterFormat(XMLFormat)
+	yqlib.RegisterRules(xmlYqRules)
+
+}

pkg/xml/xml_test.go

@@ -1,12 +1,13 @@
 //go:build !yq_noxml
 
-package yqlib
+package xml
 
 import (
 	"bufio"
 	"fmt"
 	"testing"
 
+	"github.com/mikefarah/yq/v4/pkg/yqlib"
 	"github.com/mikefarah/yq/v4/test"
 )
 
@@ -262,7 +263,13 @@ const expectedXmlWithProcInstAndDirectives = `<?xml version="1.0"?>
 </apple>
 `
 
-var xmlScenarios = []formatScenario{
+var xmlScenarios = []yqlib.FormatScenario{
+	{
+		skipDoc:     true,
+		description: "bad xml",
+		input:       `<?xml version="1.0" encoding="UTF-8"?></Child></Root>`,
+		expected:    "+p_xml: version=\"1.0\" encoding=\"UTF-8\"\n",
+	},
 	{
 		skipDoc:  true,
 		input:    "  <root>value<!-- comment--> </root>",
@@ -356,6 +363,19 @@ var xmlScenarios = []formatScenario{
 		input:          "<?xml version=\"1.0\" encoding=\"UTF-8\"?>\n<animal>cat</animal>\n<animal>goat</animal>",
 		expected:       "+p_xml: version=\"1.0\" encoding=\"UTF-8\"\nanimal:\n    - cat\n    - goat\n",
 	},
+	{
+		description:    "Parse xml: force as an array",
+		subdescription: "In XML, if your array has a single item, then yq doesn't know its an array. This is how you can consistently force it to be an array. This handles the 3 scenarios of having nothing in the array, having a single item and having multiple.",
+		input:          "<zoo><animal>cat</animal></zoo>",
+		expression:     ".zoo.animal |= ([] + .)",
+		expected:       "zoo:\n    animal:\n        - cat\n",
+	},
+	{
+		description: "Parse xml: force all as an array",
+		input:       "<zoo><thing><frog>boing</frog></thing></zoo>",
+		expression:  ".. |= [] + .",
+		expected:    "- zoo:\n    - thing:\n        - frog:\n            - boing\n",
+	},
 	{
 		description:    "Parse xml: attributes",
 		subdescription: "Attributes are converted to fields, with the default attribute prefix '+'. Use '--xml-attribute-prefix` to set your own.",
@@ -523,13 +543,6 @@ var xmlScenarios = []formatScenario{
 		expectedError: "cannot encode !!seq to XML - only maps can be encoded",
 		scenarioType:  "encode-error",
 	},
-	{
-		description:   "scalars cannot be encoded",
-		skipDoc:       true,
-		input:         "mike",
-		expectedError: "cannot encode !!str to XML - only maps can be encoded",
-		scenarioType:  "encode-error",
-	},
 	{
 		description:    "Encode xml: attributes with content",
 		subdescription: "Fields with the matching xml-content-name is assumed to be content.",
@@ -602,39 +615,41 @@ var xmlScenarios = []formatScenario{
 	},
 }
 
-func testXMLScenario(t *testing.T, s formatScenario) {
+func testXMLScenario(t *testing.T, s yqlib.FormatScenario) {
 	switch s.scenarioType {
 	case "", "decode":
-		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewXMLDecoder(ConfiguredXMLPreferences), NewYamlEncoder(4, false, ConfiguredYamlPreferences)), s.description)
+		yamlPrefs := yqlib.ConfiguredYamlPreferences.Copy()
+		yamlPrefs.Indent = 4
+		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewXMLDecoder(ConfiguredXMLPreferences), NewYamlEncoder(yamlPrefs)), s.description)
 	case "encode":
-		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewYamlDecoder(ConfiguredYamlPreferences), NewXMLEncoder(2, ConfiguredXMLPreferences)), s.description)
+		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewYamlDecoder(ConfiguredYamlPreferences), NewXMLEncoder(ConfiguredXMLPreferences)), s.description)
 	case "roundtrip":
-		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewXMLDecoder(ConfiguredXMLPreferences), NewXMLEncoder(2, ConfiguredXMLPreferences)), s.description)
+		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewXMLDecoder(ConfiguredXMLPreferences), NewXMLEncoder(ConfiguredXMLPreferences)), s.description)
 	case "decode-keep-ns":
 		prefs := NewDefaultXmlPreferences()
 		prefs.KeepNamespace = true
-		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewXMLDecoder(prefs), NewYamlEncoder(2, false, ConfiguredYamlPreferences)), s.description)
+		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewXMLDecoder(prefs), NewYamlEncoder(ConfiguredYamlPreferences)), s.description)
 	case "decode-raw-token":
 		prefs := NewDefaultXmlPreferences()
 		prefs.UseRawToken = true
-		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewXMLDecoder(prefs), NewYamlEncoder(2, false, ConfiguredYamlPreferences)), s.description)
+		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewXMLDecoder(prefs), NewYamlEncoder(ConfiguredYamlPreferences)), s.description)
 	case "decode-raw-token-off":
 		prefs := NewDefaultXmlPreferences()
 		prefs.UseRawToken = false
-		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewXMLDecoder(prefs), NewYamlEncoder(2, false, ConfiguredYamlPreferences)), s.description)
+		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewXMLDecoder(prefs), NewYamlEncoder(ConfiguredYamlPreferences)), s.description)
 	case "roundtrip-skip-directives":
 		prefs := NewDefaultXmlPreferences()
 		prefs.SkipDirectives = true
-		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewXMLDecoder(prefs), NewXMLEncoder(2, prefs)), s.description)
+		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewXMLDecoder(prefs), NewXMLEncoder(prefs)), s.description)
 	case "decode-error":
-		result, err := processFormatScenario(s, NewXMLDecoder(NewDefaultXmlPreferences()), NewYamlEncoder(2, false, ConfiguredYamlPreferences))
+		result, err := processFormatScenario(s, NewXMLDecoder(NewDefaultXmlPreferences()), NewYamlEncoder(ConfiguredYamlPreferences))
 		if err == nil {
 			t.Errorf("Expected error '%v' but it worked: %v", s.expectedError, result)
 		} else {
 			test.AssertResultComplexWithContext(t, s.expectedError, err.Error(), s.description)
 		}
 	case "encode-error":
-		result, err := processFormatScenario(s, NewYamlDecoder(ConfiguredYamlPreferences), NewXMLEncoder(2, NewDefaultXmlPreferences()))
+		result, err := processFormatScenario(s, NewYamlDecoder(ConfiguredYamlPreferences), NewXMLEncoder(NewDefaultXmlPreferences()))
 		if err == nil {
 			t.Errorf("Expected error '%v' but it worked: %v", s.expectedError, result)
 		} else {
@@ -646,8 +661,8 @@ func testXMLScenario(t *testing.T, s formatScenario) {
 	}
 }
 
-func documentXMLScenario(t *testing.T, w *bufio.Writer, i interface{}) {
-	s := i.(formatScenario)
+func documentXMLScenario(_ *testing.T, w *bufio.Writer, i interface{}) {
+	s := i.(FormatScenario)
 
 	if s.skipDoc {
 		return
@@ -664,14 +679,14 @@ func documentXMLScenario(t *testing.T, w *bufio.Writer, i interface{}) {
 	case "decode-raw-token-off":
 		documentXMLDecodeKeepNsRawTokenScenario(w, s)
 	case "roundtrip-skip-directives":
-		documentXMLSkipDirectrivesScenario(w, s)
+		documentXMLSkipDirectivesScenario(w, s)
 
 	default:
 		panic(fmt.Sprintf("unhandled scenario type %q", s.scenarioType))
 	}
 }
 
-func documentXMLDecodeScenario(w *bufio.Writer, s formatScenario) {
+func documentXMLDecodeScenario(w *bufio.Writer, s FormatScenario) {
 	writeOrPanic(w, fmt.Sprintf("## %v\n", s.description))
 
 	if s.subdescription != "" {
@@ -687,13 +702,13 @@ func documentXMLDecodeScenario(w *bufio.Writer, s formatScenario) {
 	if expression == "" {
 		expression = "."
 	}
-	writeOrPanic(w, fmt.Sprintf("```bash\nyq -p=xml '%v' sample.xml\n```\n", expression))
+	writeOrPanic(w, fmt.Sprintf("```bash\nyq -oy '%v' sample.xml\n```\n", expression))
 	writeOrPanic(w, "will output\n")
 
-	writeOrPanic(w, fmt.Sprintf("```yaml\n%v```\n\n", mustProcessFormatScenario(s, NewXMLDecoder(ConfiguredXMLPreferences), NewYamlEncoder(2, false, ConfiguredYamlPreferences))))
+	writeOrPanic(w, fmt.Sprintf("```yaml\n%v```\n\n", mustProcessFormatScenario(s, NewXMLDecoder(ConfiguredXMLPreferences), NewYamlEncoder(ConfiguredYamlPreferences))))
 }
 
-func documentXMLDecodeKeepNsScenario(w *bufio.Writer, s formatScenario) {
+func documentXMLDecodeKeepNsScenario(w *bufio.Writer, s FormatScenario) {
 	writeOrPanic(w, fmt.Sprintf("## %v\n", s.description))
 
 	if s.subdescription != "" {
@@ -705,19 +720,19 @@ func documentXMLDecodeKeepNsScenario(w *bufio.Writer, s formatScenario) {
 	writeOrPanic(w, fmt.Sprintf("```xml\n%v\n```\n", s.input))
 
 	writeOrPanic(w, "then\n")
-	writeOrPanic(w, "```bash\nyq -p=xml -o=xml --xml-keep-namespace=false '.' sample.xml\n```\n")
+	writeOrPanic(w, "```bash\nyq --xml-keep-namespace=false '.' sample.xml\n```\n")
 	writeOrPanic(w, "will output\n")
 	prefs := NewDefaultXmlPreferences()
 	prefs.KeepNamespace = false
-	writeOrPanic(w, fmt.Sprintf("```xml\n%v```\n\n", mustProcessFormatScenario(s, NewXMLDecoder(prefs), NewXMLEncoder(2, prefs))))
+	writeOrPanic(w, fmt.Sprintf("```xml\n%v```\n\n", mustProcessFormatScenario(s, NewXMLDecoder(prefs), NewXMLEncoder(prefs))))
 
 	prefsWithout := NewDefaultXmlPreferences()
 	prefs.KeepNamespace = true
 	writeOrPanic(w, "instead of\n")
-	writeOrPanic(w, fmt.Sprintf("```xml\n%v```\n\n", mustProcessFormatScenario(s, NewXMLDecoder(prefsWithout), NewXMLEncoder(2, prefsWithout))))
+	writeOrPanic(w, fmt.Sprintf("```xml\n%v```\n\n", mustProcessFormatScenario(s, NewXMLDecoder(prefsWithout), NewXMLEncoder(prefsWithout))))
 }
 
-func documentXMLDecodeKeepNsRawTokenScenario(w *bufio.Writer, s formatScenario) {
+func documentXMLDecodeKeepNsRawTokenScenario(w *bufio.Writer, s FormatScenario) {
 	writeOrPanic(w, fmt.Sprintf("## %v\n", s.description))
 
 	if s.subdescription != "" {
@@ -729,22 +744,22 @@ func documentXMLDecodeKeepNsRawTokenScenario(w *bufio.Writer, s formatScenario)
 	writeOrPanic(w, fmt.Sprintf("```xml\n%v\n```\n", s.input))
 
 	writeOrPanic(w, "then\n")
-	writeOrPanic(w, "```bash\nyq -p=xml -o=xml --xml-raw-token=false '.' sample.xml\n```\n")
+	writeOrPanic(w, "```bash\nyq --xml-raw-token=false '.' sample.xml\n```\n")
 	writeOrPanic(w, "will output\n")
 
 	prefs := NewDefaultXmlPreferences()
 	prefs.UseRawToken = false
 
-	writeOrPanic(w, fmt.Sprintf("```xml\n%v```\n\n", mustProcessFormatScenario(s, NewXMLDecoder(prefs), NewXMLEncoder(2, prefs))))
+	writeOrPanic(w, fmt.Sprintf("```xml\n%v```\n\n", mustProcessFormatScenario(s, NewXMLDecoder(prefs), NewXMLEncoder(prefs))))
 
 	prefsWithout := NewDefaultXmlPreferences()
 	prefsWithout.UseRawToken = true
 
 	writeOrPanic(w, "instead of\n")
-	writeOrPanic(w, fmt.Sprintf("```xml\n%v```\n\n", mustProcessFormatScenario(s, NewXMLDecoder(prefsWithout), NewXMLEncoder(2, prefsWithout))))
+	writeOrPanic(w, fmt.Sprintf("```xml\n%v```\n\n", mustProcessFormatScenario(s, NewXMLDecoder(prefsWithout), NewXMLEncoder(prefsWithout))))
 }
 
-func documentXMLEncodeScenario(w *bufio.Writer, s formatScenario) {
+func documentXMLEncodeScenario(w *bufio.Writer, s FormatScenario) {
 	writeOrPanic(w, fmt.Sprintf("## %v\n", s.description))
 
 	if s.subdescription != "" {
@@ -756,13 +771,13 @@ func documentXMLEncodeScenario(w *bufio.Writer, s formatScenario) {
 	writeOrPanic(w, fmt.Sprintf("```yaml\n%v\n```\n", s.input))
 
 	writeOrPanic(w, "then\n")
-	writeOrPanic(w, "```bash\nyq -o=xml '.' sample.yml\n```\n")
+	writeOrPanic(w, "```bash\nyq -o=xml sample.yml\n```\n")
 	writeOrPanic(w, "will output\n")
 
-	writeOrPanic(w, fmt.Sprintf("```xml\n%v```\n\n", mustProcessFormatScenario(s, NewYamlDecoder(ConfiguredYamlPreferences), NewXMLEncoder(2, ConfiguredXMLPreferences))))
+	writeOrPanic(w, fmt.Sprintf("```xml\n%v```\n\n", mustProcessFormatScenario(s, NewYamlDecoder(ConfiguredYamlPreferences), NewXMLEncoder(ConfiguredXMLPreferences))))
 }
 
-func documentXMLRoundTripScenario(w *bufio.Writer, s formatScenario) {
+func documentXMLRoundTripScenario(w *bufio.Writer, s FormatScenario) {
 	writeOrPanic(w, fmt.Sprintf("## %v\n", s.description))
 
 	if s.subdescription != "" {
@@ -774,13 +789,13 @@ func documentXMLRoundTripScenario(w *bufio.Writer, s formatScenario) {
 	writeOrPanic(w, fmt.Sprintf("```xml\n%v\n```\n", s.input))
 
 	writeOrPanic(w, "then\n")
-	writeOrPanic(w, "```bash\nyq -p=xml -o=xml '.' sample.xml\n```\n")
+	writeOrPanic(w, "```bash\nyq '.' sample.xml\n```\n")
 	writeOrPanic(w, "will output\n")
 
-	writeOrPanic(w, fmt.Sprintf("```xml\n%v```\n\n", mustProcessFormatScenario(s, NewXMLDecoder(ConfiguredXMLPreferences), NewXMLEncoder(2, ConfiguredXMLPreferences))))
+	writeOrPanic(w, fmt.Sprintf("```xml\n%v```\n\n", mustProcessFormatScenario(s, NewXMLDecoder(ConfiguredXMLPreferences), NewXMLEncoder(ConfiguredXMLPreferences))))
 }
 
-func documentXMLSkipDirectrivesScenario(w *bufio.Writer, s formatScenario) {
+func documentXMLSkipDirectivesScenario(w *bufio.Writer, s FormatScenario) {
 	writeOrPanic(w, fmt.Sprintf("## %v\n", s.description))
 
 	if s.subdescription != "" {
@@ -792,12 +807,12 @@ func documentXMLSkipDirectrivesScenario(w *bufio.Writer, s formatScenario) {
 	writeOrPanic(w, fmt.Sprintf("```xml\n%v\n```\n", s.input))
 
 	writeOrPanic(w, "then\n")
-	writeOrPanic(w, "```bash\nyq -p=xml -o=xml --xml-skip-directives '.' sample.xml\n```\n")
+	writeOrPanic(w, "```bash\nyq --xml-skip-directives '.' sample.xml\n```\n")
 	writeOrPanic(w, "will output\n")
 	prefs := NewDefaultXmlPreferences()
 	prefs.SkipDirectives = true
 
-	writeOrPanic(w, fmt.Sprintf("```xml\n%v```\n\n", mustProcessFormatScenario(s, NewXMLDecoder(prefs), NewXMLEncoder(2, prefs))))
+	writeOrPanic(w, fmt.Sprintf("```xml\n%v```\n\n", mustProcessFormatScenario(s, NewXMLDecoder(prefs), NewXMLEncoder(prefs))))
 }
 
 func TestXMLScenarios(t *testing.T) {

pkg/yqlib/all_at_once_evaluator.go

@@ -2,8 +2,6 @@ package yqlib
 
 import (
 	"container/list"
-
-	yaml "gopkg.in/yaml.v3"
 )
 
 // A yaml expression evaluator that runs the expression once against all files/nodes in memory.
@@ -11,7 +9,7 @@ type Evaluator interface {
 	EvaluateFiles(expression string, filenames []string, printer Printer, decoder Decoder) error
 
 	// EvaluateNodes takes an expression and one or more yaml nodes, returning a list of matching candidate nodes
-	EvaluateNodes(expression string, nodes ...*yaml.Node) (*list.List, error)
+	EvaluateNodes(expression string, nodes ...*CandidateNode) (*list.List, error)
 
 	// EvaluateCandidateNodes takes an expression and list of candidate nodes, returning a list of matching candidate nodes
 	EvaluateCandidateNodes(expression string, inputCandidateNodes *list.List) (*list.List, error)
@@ -26,10 +24,10 @@ func NewAllAtOnceEvaluator() Evaluator {
 	return &allAtOnceEvaluator{treeNavigator: NewDataTreeNavigator()}
 }
 
-func (e *allAtOnceEvaluator) EvaluateNodes(expression string, nodes ...*yaml.Node) (*list.List, error) {
+func (e *allAtOnceEvaluator) EvaluateNodes(expression string, nodes ...*CandidateNode) (*list.List, error) {
 	inputCandidates := list.New()
 	for _, node := range nodes {
-		inputCandidates.PushBack(&CandidateNode{Node: node})
+		inputCandidates.PushBack(node)
 	}
 	return e.EvaluateCandidateNodes(expression, inputCandidates)
 }
@@ -65,13 +63,7 @@ func (e *allAtOnceEvaluator) EvaluateFiles(expression string, filenames []string
 	}
 
 	if allDocuments.Len() == 0 {
-		candidateNode := &CandidateNode{
-			Document:       0,
-			Filename:       "",
-			Node:           &yaml.Node{Kind: yaml.DocumentNode, Content: []*yaml.Node{{Tag: "!!null", Kind: yaml.ScalarNode}}},
-			FileIndex:      0,
-			LeadingContent: "",
-		}
+		candidateNode := CreateScalarNode(nil, "")
 		allDocuments.PushBack(candidateNode)
 	}
 

pkg/yqlib/all_at_once_evaluator_test.go

@@ -1,6 +1,8 @@
 package yqlib
 
 import (
+	"bufio"
+	"strings"
 	"testing"
 
 	"github.com/mikefarah/yq/v4/test"
@@ -18,7 +20,7 @@ var evaluateNodesScenario = []expressionScenario{
 		document:   `a: hello`,
 		expression: `.`,
 		expected: []string{
-			"D0, P[], (doc)::a: hello\n",
+			"D0, P[], (!!map)::a: hello\n",
 		},
 	},
 	{
@@ -32,9 +34,23 @@ var evaluateNodesScenario = []expressionScenario{
 
 func TestAllAtOnceEvaluateNodes(t *testing.T) {
 	var evaluator = NewAllAtOnceEvaluator()
+	// logging.SetLevel(logging.DEBUG, "")
 	for _, tt := range evaluateNodesScenario {
-		node := test.ParseData(tt.document)
-		list, _ := evaluator.EvaluateNodes(tt.expression, &node)
+		decoder := NewYamlDecoder(ConfiguredYamlPreferences)
+		reader := bufio.NewReader(strings.NewReader(tt.document))
+		err := decoder.Init(reader)
+		if err != nil {
+			t.Error(err)
+			return
+		}
+		candidateNode, errorReading := decoder.Decode()
+
+		if errorReading != nil {
+			t.Error(errorReading)
+			return
+		}
+
+		list, _ := evaluator.EvaluateNodes(tt.expression, candidateNode)
 		test.AssertResultComplex(t, tt.expected, resultsToString(t, list))
 	}
 }

pkg/yqlib/candidate_node.go

@@ -3,51 +3,189 @@ package yqlib
 import (
 	"container/list"
 	"fmt"
+	"strconv"
 	"strings"
-
-	"github.com/jinzhu/copier"
-	yaml "gopkg.in/yaml.v3"
 )
 
+type Kind uint32
+
+const (
+	SequenceNode Kind = 1 << iota
+	MappingNode
+	ScalarNode
+	AliasNode
+)
+
+type Style uint32
+
+const (
+	TaggedStyle Style = 1 << iota
+	DoubleQuotedStyle
+	SingleQuotedStyle
+	LiteralStyle
+	FoldedStyle
+	FlowStyle
+)
+
+func CreateStringScalarNode(stringValue string) *CandidateNode {
+	var node = &CandidateNode{Kind: ScalarNode}
+	node.Value = stringValue
+	node.Tag = "!!str"
+	return node
+}
+
+func CreateScalarNode(value interface{}, stringValue string) *CandidateNode {
+	var node = &CandidateNode{Kind: ScalarNode}
+	node.Value = stringValue
+
+	switch value.(type) {
+	case float32, float64:
+		node.Tag = "!!float"
+	case int, int64, int32:
+		node.Tag = "!!int"
+	case bool:
+		node.Tag = "!!bool"
+	case string:
+		node.Tag = "!!str"
+	case nil:
+		node.Tag = "!!null"
+	}
+	return node
+}
+
 type CandidateNode struct {
-	Node   *yaml.Node     // the actual node
+	Kind  Kind
+	Style Style
+
+	Tag     string
+	Value   string
+	Anchor  string
+	Alias   *CandidateNode
+	Content []*CandidateNode
+
+	HeadComment string
+	LineComment string
+	FootComment string
+
 	Parent *CandidateNode // parent node
-	Key    *yaml.Node     // node key, if this is a value from a map (or index in an array)
+	Key    *CandidateNode // node key, if this is a value from a map (or index in an array)
 
-	LeadingContent  string
-	TrailingContent string
+	LeadingContent string
 
-	Path      []interface{} /// the path we took to get to this node
-	Document  uint          // the document index of this node
-	Filename  string
-	FileIndex int
+	document uint // the document index of this node
+	filename string
+
+	Line   int
+	Column int
+
+	fileIndex int
 	// when performing op against all nodes given, this will treat all the nodes as one
 	// (e.g. top level cross document merge). This property does not propagate to child nodes.
 	EvaluateTogether bool
 	IsMapKey         bool
 }
 
+func (n *CandidateNode) CreateChild() *CandidateNode {
+	return &CandidateNode{
+		Parent: n,
+	}
+}
+
+func (n *CandidateNode) SetDocument(idx uint) {
+	n.document = idx
+}
+
+func (n *CandidateNode) GetDocument() uint {
+	// defer to parent
+	if n.Parent != nil {
+		return n.Parent.GetDocument()
+	}
+	return n.document
+}
+
+func (n *CandidateNode) SetFilename(name string) {
+	n.filename = name
+}
+
+func (n *CandidateNode) GetFilename() string {
+	if n.Parent != nil {
+		return n.Parent.GetFilename()
+	}
+	return n.filename
+}
+
+func (n *CandidateNode) SetFileIndex(idx int) {
+	n.fileIndex = idx
+}
+
+func (n *CandidateNode) GetFileIndex() int {
+	if n.Parent != nil {
+		return n.Parent.GetFileIndex()
+	}
+	return n.fileIndex
+}
+
 func (n *CandidateNode) GetKey() string {
 	keyPrefix := ""
 	if n.IsMapKey {
-		keyPrefix = "key-"
+		keyPrefix = fmt.Sprintf("key-%v-", n.Value)
 	}
-	return fmt.Sprintf("%v%v - %v", keyPrefix, n.Document, n.Path)
+	key := ""
+	if n.Key != nil {
+		key = n.Key.Value
+	}
+	return fmt.Sprintf("%v%v - %v", keyPrefix, n.GetDocument(), key)
 }
 
-func (n *CandidateNode) GetNiceTag() string {
-	return unwrapDoc(n.Node).Tag
+func (n *CandidateNode) getParsedKey() interface{} {
+	if n.IsMapKey {
+		return n.Value
+	}
+	if n.Key == nil {
+		return nil
+	}
+	if n.Key.Tag == "!!str" {
+		return n.Key.Value
+	}
+	index, err := parseInt(n.Key.Value)
+	if err != nil {
+		return n.Key.Value
+	}
+	return index
+
+}
+
+func (n *CandidateNode) GetPath() []interface{} {
+	key := n.getParsedKey()
+	if n.Parent != nil && key != nil {
+		return append(n.Parent.GetPath(), key)
+	}
+
+	if key != nil {
+		return []interface{}{key}
+	}
+	return make([]interface{}, 0)
 }
 
 func (n *CandidateNode) GetNicePath() string {
-	if n.Path != nil && len(n.Path) >= 0 {
-		pathStr := make([]string, len(n.Path))
-		for i, v := range n.Path {
-			pathStr[i] = fmt.Sprintf("%v", v)
+	var sb strings.Builder
+	path := n.GetPath()
+	for i, element := range path {
+		elementStr := fmt.Sprintf("%v", element)
+		switch element.(type) {
+		case int:
+			sb.WriteString("[" + elementStr + "]")
+		default:
+			if i == 0 {
+				sb.WriteString(elementStr)
+			} else if strings.ContainsRune(elementStr, '.') {
+				sb.WriteString("[" + elementStr + "]")
+			} else {
+				sb.WriteString("." + elementStr)
+			}
 		}
-		return strings.Join(pathStr, ".")
 	}
-	return ""
+	return sb.String()
 }
 
 func (n *CandidateNode) AsList() *list.List {
@@ -56,134 +194,268 @@ func (n *CandidateNode) AsList() *list.List {
 	return elMap
 }
 
-func (n *CandidateNode) CreateChildInMap(key *yaml.Node, node *yaml.Node) *CandidateNode {
-	var value interface{}
-	if key != nil {
-		value = key.Value
+func (n *CandidateNode) SetParent(parent *CandidateNode) {
+	n.Parent = parent
+}
+
+func (n *CandidateNode) AddKeyValueChild(rawKey *CandidateNode, rawValue *CandidateNode) (*CandidateNode, *CandidateNode) {
+	key := rawKey.Copy()
+	key.SetParent(n)
+	key.IsMapKey = true
+
+	value := rawValue.Copy()
+	value.SetParent(n)
+	value.IsMapKey = false // force this, incase we are creating a value from a key
+	value.Key = key
+
+	n.Content = append(n.Content, key, value)
+	return key, value
+}
+
+func (n *CandidateNode) AddChild(rawChild *CandidateNode) {
+	value := rawChild.Copy()
+	value.SetParent(n)
+	if value.Key != nil {
+		value.Key.SetParent(n)
+	} else {
+		index := len(n.Content)
+		keyNode := CreateScalarNode(index, fmt.Sprintf("%v", index))
+		keyNode.SetParent(n)
+		value.Key = keyNode
 	}
-	return &CandidateNode{
-		Node:      node,
-		Path:      n.createChildPath(value),
-		Parent:    n,
-		Key:       key,
-		Document:  n.Document,
-		Filename:  n.Filename,
-		FileIndex: n.FileIndex,
+	n.Content = append(n.Content, value)
+}
+
+func (n *CandidateNode) AddChildren(children []*CandidateNode) {
+	if n.Kind == MappingNode {
+		for i := 0; i < len(children); i += 2 {
+			key := children[i]
+			value := children[i+1]
+			n.AddKeyValueChild(key, value)
+		}
+
+	} else {
+		for _, rawChild := range children {
+			n.AddChild(rawChild)
+		}
 	}
 }
 
-func (n *CandidateNode) CreateChildInArray(index int, node *yaml.Node) *CandidateNode {
-	return &CandidateNode{
-		Node:      node,
-		Path:      n.createChildPath(index),
-		Parent:    n,
-		Key:       &yaml.Node{Kind: yaml.ScalarNode, Value: fmt.Sprintf("%v", index), Tag: "!!int"},
-		Document:  n.Document,
-		Filename:  n.Filename,
-		FileIndex: n.FileIndex,
+func (n *CandidateNode) GetValueRep() (interface{}, error) {
+	log.Debugf("GetValueRep for %v value: %v", n.GetNicePath(), n.Value)
+	realTag := n.GuessTagFromCustomType()
+
+	switch realTag {
+	case "!!int":
+		_, val, err := parseInt64(n.Value)
+		return val, err
+	case "!!float":
+		// need to test this
+		return strconv.ParseFloat(n.Value, 64)
+	case "!!bool":
+		return isTruthyNode(n), nil
+	case "!!null":
+		return nil, nil
 	}
+
+	return n.Value, nil
 }
 
-func (n *CandidateNode) CreateReplacement(node *yaml.Node) *CandidateNode {
-	return &CandidateNode{
-		Node:      node,
-		Path:      n.createChildPath(nil),
-		Parent:    n.Parent,
-		Key:       n.Key,
-		IsMapKey:  n.IsMapKey,
-		Document:  n.Document,
-		Filename:  n.Filename,
-		FileIndex: n.FileIndex,
+func (n *CandidateNode) GuessTagFromCustomType() string {
+	if strings.HasPrefix(n.Tag, "!!") {
+		return n.Tag
+	} else if n.Value == "" {
+		log.Debug("GuessTagFromCustomType: node has no value to guess the type with")
+		return n.Tag
 	}
+	dataBucket, errorReading := parseSnippet(n.Value)
+
+	if errorReading != nil {
+		log.Debug("GuessTagFromCustomType: could not guess underlying tag type %v", errorReading)
+		return n.Tag
+	}
+	guessedTag := dataBucket.Tag
+	log.Info("im guessing the tag %v is a %v", n.Tag, guessedTag)
+	return guessedTag
 }
 
-func (n *CandidateNode) CreateReplacementWithDocWrappers(node *yaml.Node) *CandidateNode {
-	replacement := n.CreateReplacement(node)
+func (n *CandidateNode) CreateReplacement(kind Kind, tag string, value string) *CandidateNode {
+	node := &CandidateNode{
+		Kind:  kind,
+		Tag:   tag,
+		Value: value,
+	}
+	return n.CopyAsReplacement(node)
+}
+
+func (n *CandidateNode) CopyAsReplacement(replacement *CandidateNode) *CandidateNode {
+	newCopy := replacement.Copy()
+	newCopy.Parent = n.Parent
+
+	if n.IsMapKey {
+		newCopy.Key = n
+	} else {
+		newCopy.Key = n.Key
+	}
+
+	return newCopy
+}
+
+func (n *CandidateNode) CreateReplacementWithComments(kind Kind, tag string, style Style) *CandidateNode {
+	replacement := n.CreateReplacement(kind, tag, "")
 	replacement.LeadingContent = n.LeadingContent
-	replacement.TrailingContent = n.TrailingContent
+	replacement.HeadComment = n.HeadComment
+	replacement.LineComment = n.LineComment
+	replacement.FootComment = n.FootComment
+	replacement.Style = style
 	return replacement
 }
 
-func (n *CandidateNode) createChildPath(path interface{}) []interface{} {
-	if path == nil {
-		newPath := make([]interface{}, len(n.Path))
-		copy(newPath, n.Path)
-		return newPath
-	}
-
-	//don't use append as they may actually modify the path of the orignal node!
-	newPath := make([]interface{}, len(n.Path)+1)
-	copy(newPath, n.Path)
-	newPath[len(n.Path)] = path
-	return newPath
+func (n *CandidateNode) Copy() *CandidateNode {
+	return n.doCopy(true)
 }
 
-func (n *CandidateNode) Copy() (*CandidateNode, error) {
-	clone := &CandidateNode{}
-	err := copier.Copy(clone, n)
-	if err != nil {
-		return nil, err
+func (n *CandidateNode) CopyWithoutContent() *CandidateNode {
+	return n.doCopy(false)
+}
+
+func (n *CandidateNode) doCopy(cloneContent bool) *CandidateNode {
+	var content []*CandidateNode
+
+	var copyKey *CandidateNode
+	if n.Key != nil {
+		copyKey = n.Key.Copy()
 	}
-	clone.Node = deepClone(n.Node)
-	return clone, nil
+
+	clone := &CandidateNode{
+		Kind:  n.Kind,
+		Style: n.Style,
+
+		Tag:    n.Tag,
+		Value:  n.Value,
+		Anchor: n.Anchor,
+
+		// ok not to clone this,
+		// as its a reference to somewhere else.
+		Alias:   n.Alias,
+		Content: content,
+
+		HeadComment: n.HeadComment,
+		LineComment: n.LineComment,
+		FootComment: n.FootComment,
+
+		Parent: n.Parent,
+		Key:    copyKey,
+
+		LeadingContent: n.LeadingContent,
+
+		document:  n.document,
+		filename:  n.filename,
+		fileIndex: n.fileIndex,
+
+		Line:   n.Line,
+		Column: n.Column,
+
+		EvaluateTogether: n.EvaluateTogether,
+		IsMapKey:         n.IsMapKey,
+	}
+
+	if cloneContent {
+		clone.AddChildren(n.Content)
+	}
+
+	return clone
 }
 
 // updates this candidate from the given candidate node
 func (n *CandidateNode) UpdateFrom(other *CandidateNode, prefs assignPreferences) {
-
+	if n == other {
+		log.Debugf("UpdateFrom, no need to update from myself.")
+		return
+	}
 	// if this is an empty map or empty array, use the style of other node.
-	if (n.Node.Kind != yaml.ScalarNode && len(n.Node.Content) == 0) ||
+	if (n.Kind != ScalarNode && len(n.Content) == 0) ||
 		// if the tag has changed (e.g. from str to bool)
-		(guessTagFromCustomType(n.Node) != guessTagFromCustomType(other.Node)) {
-		n.Node.Style = other.Node.Style
+		(n.GuessTagFromCustomType() != other.GuessTagFromCustomType()) {
+		n.Style = other.Style
 	}
 
-	n.Node.Content = deepCloneContent(other.Node.Content)
-	n.Node.Kind = other.Node.Kind
-	n.Node.Value = other.Node.Value
+	n.Content = make([]*CandidateNode, 0)
+	n.Kind = other.Kind
+	n.AddChildren(other.Content)
+
+	n.Value = other.Value
 
 	n.UpdateAttributesFrom(other, prefs)
 
 }
 
 func (n *CandidateNode) UpdateAttributesFrom(other *CandidateNode, prefs assignPreferences) {
-	log.Debug("UpdateAttributesFrom: n: %v other: %v", n.GetKey(), other.GetKey())
-	if n.Node.Kind != other.Node.Kind {
+	log.Debug("UpdateAttributesFrom: n: %v other: %v", NodeToString(n), NodeToString(other))
+	if n.Kind != other.Kind {
 		// clear out the contents when switching to a different type
 		// e.g. map to array
-		n.Node.Content = make([]*yaml.Node, 0)
-		n.Node.Value = ""
+		n.Content = make([]*CandidateNode, 0)
+		n.Value = ""
 	}
-	n.Node.Kind = other.Node.Kind
+	n.Kind = other.Kind
 
 	// don't clobber custom tags...
-	if prefs.ClobberCustomTags || strings.HasPrefix(n.Node.Tag, "!!") || n.Node.Tag == "" {
-		n.Node.Tag = other.Node.Tag
+	if prefs.ClobberCustomTags || strings.HasPrefix(n.Tag, "!!") || n.Tag == "" {
+		n.Tag = other.Tag
 	}
 
-	n.Node.Alias = other.Node.Alias
+	n.Alias = other.Alias
 
 	if !prefs.DontOverWriteAnchor {
-		n.Node.Anchor = other.Node.Anchor
+		n.Anchor = other.Anchor
 	}
 
 	// merge will pickup the style of the new thing
 	// when autocreating nodes
 
-	if n.Node.Style == 0 {
-		n.Node.Style = other.Node.Style
+	if n.Style == 0 {
+		n.Style = other.Style
 	}
 
-	if other.Node.FootComment != "" {
-		n.Node.FootComment = other.Node.FootComment
+	if other.FootComment != "" {
+		n.FootComment = other.FootComment
 	}
-	if other.TrailingContent != "" {
-		n.TrailingContent = other.TrailingContent
+	if other.HeadComment != "" {
+		n.HeadComment = other.HeadComment
 	}
-	if other.Node.HeadComment != "" {
-		n.Node.HeadComment = other.Node.HeadComment
+	if other.LineComment != "" {
+		n.LineComment = other.LineComment
 	}
-	if other.Node.LineComment != "" {
-		n.Node.LineComment = other.Node.LineComment
+}
+
+func (n *CandidateNode) CleanHeadAndLineComment() string {
+	return n.CleanHeadComment() + n.CleanLineComment()
+}
+
+func (n *CandidateNode) CleanHeadComment() string {
+	return strings.Replace(n.HeadComment, "#", "", 1)
+}
+
+func (n *CandidateNode) CleanLineComment() string {
+	return strings.Replace(n.LineComment, "#", "", 1)
+}
+
+func (n *CandidateNode) CleanFootComment() string {
+	return strings.Replace(n.FootComment, "#", "", 1)
+}
+
+func (n *CandidateNode) ConvertKeysToStrings() {
+
+	if n.Kind == MappingNode {
+		for index, child := range n.Content {
+			if index%2 == 0 { // its a map key
+				child.Tag = "!!str"
+			}
+		}
+	}
+
+	for _, child := range n.Content {
+		child.ConvertKeysToStrings()
 	}
 }

pkg/yqlib/candidate_node_goccy_yaml.go

@@ -0,0 +1,180 @@
+package yqlib
+
+import (
+	"fmt"
+	"strings"
+
+	yaml "github.com/goccy/go-yaml"
+	"github.com/goccy/go-yaml/ast"
+	goccyToken "github.com/goccy/go-yaml/token"
+)
+
+func (o *CandidateNode) goccyDecodeIntoChild(childNode ast.Node, cm yaml.CommentMap) (*CandidateNode, error) {
+	newChild := o.CreateChild()
+
+	err := newChild.UnmarshalGoccyYAML(childNode, cm)
+	return newChild, err
+}
+
+func (o *CandidateNode) UnmarshalGoccyYAML(node ast.Node, cm yaml.CommentMap) error {
+	log.Debugf("UnmarshalYAML %v", node)
+	log.Debugf("UnmarshalYAML %v", node.Type().String())
+	log.Debugf("UnmarshalYAML Node Value: %v", node.String())
+	log.Debugf("UnmarshalYAML Node GetComment: %v", node.GetComment())
+
+	if node.GetComment() != nil {
+		commentMapComments := cm[node.GetPath()]
+		for _, comment := range node.GetComment().Comments {
+			// need to use the comment map to find the position :/
+			log.Debugf("%v has a comment of [%v]", node.GetPath(), comment.Token.Value)
+			for _, commentMapComment := range commentMapComments {
+				commentMapValue := strings.Join(commentMapComment.Texts, "\n")
+				if commentMapValue == comment.Token.Value {
+					log.Debug("found a matching entry in comment map")
+					// we found the comment in the comment map,
+					// now we can process the position
+					switch commentMapComment.Position {
+					case yaml.CommentHeadPosition:
+						o.HeadComment = comment.String()
+						log.Debug("its a head comment %v", comment.String())
+					case yaml.CommentLinePosition:
+						o.LineComment = comment.String()
+						log.Debug("its a line comment %v", comment.String())
+					case yaml.CommentFootPosition:
+						o.FootComment = comment.String()
+						log.Debug("its a foot comment %v", comment.String())
+					}
+				}
+			}
+
+		}
+	}
+
+	o.Value = node.String()
+	switch node.Type() {
+	case ast.IntegerType:
+		o.Kind = ScalarNode
+		o.Tag = "!!int"
+	case ast.FloatType:
+		o.Kind = ScalarNode
+		o.Tag = "!!float"
+	case ast.BoolType:
+		o.Kind = ScalarNode
+		o.Tag = "!!bool"
+	case ast.NullType:
+		o.Kind = ScalarNode
+		o.Tag = "!!null"
+		o.Value = node.GetToken().Value
+	case ast.StringType:
+		o.Kind = ScalarNode
+		o.Tag = "!!str"
+		switch node.GetToken().Type {
+		case goccyToken.SingleQuoteType:
+			o.Style = SingleQuotedStyle
+		case goccyToken.DoubleQuoteType:
+			o.Style = DoubleQuotedStyle
+		}
+		o.Value = node.(*ast.StringNode).Value
+		log.Debugf("string value %v", node.(*ast.StringNode).Value)
+	case ast.LiteralType:
+		o.Kind = ScalarNode
+		o.Tag = "!!str"
+		o.Style = LiteralStyle
+		astLiteral := node.(*ast.LiteralNode)
+		log.Debugf("astLiteral.Start.Type %v", astLiteral.Start.Type)
+		if astLiteral.Start.Type == goccyToken.FoldedType {
+			log.Debugf("folded Type %v", astLiteral.Start.Type)
+			o.Style = FoldedStyle
+		}
+		log.Debug("start value: %v ", node.(*ast.LiteralNode).Start.Value)
+		log.Debug("start value: %v ", node.(*ast.LiteralNode).Start.Type)
+		// TODO: here I could put the original value with line breaks
+		// to solve the multiline > problem
+		o.Value = astLiteral.Value.Value
+	case ast.TagType:
+		if err := o.UnmarshalGoccyYAML(node.(*ast.TagNode).Value, cm); err != nil {
+			return err
+		}
+		o.Tag = node.(*ast.TagNode).Start.Value
+	case ast.MappingType:
+		log.Debugf("UnmarshalYAML -  a mapping node")
+		o.Kind = MappingNode
+		o.Tag = "!!map"
+
+		mappingNode := node.(*ast.MappingNode)
+		if mappingNode.IsFlowStyle {
+			o.Style = FlowStyle
+		}
+		for _, mappingValueNode := range mappingNode.Values {
+			err := o.goccyProcessMappingValueNode(mappingValueNode, cm)
+			if err != nil {
+				return ast.ErrInvalidAnchorName
+			}
+		}
+		if mappingNode.FootComment != nil {
+			log.Debugf("mapping node has a foot comment of: %v", mappingNode.FootComment)
+			o.FootComment = mappingNode.FootComment.String()
+		}
+	case ast.MappingValueType:
+		log.Debugf("UnmarshalYAML -  a mapping node")
+		o.Kind = MappingNode
+		o.Tag = "!!map"
+		mappingValueNode := node.(*ast.MappingValueNode)
+		err := o.goccyProcessMappingValueNode(mappingValueNode, cm)
+		if err != nil {
+			return ast.ErrInvalidAnchorName
+		}
+	case ast.SequenceType:
+		log.Debugf("UnmarshalYAML -  a sequence node")
+		o.Kind = SequenceNode
+		o.Tag = "!!seq"
+		sequenceNode := node.(*ast.SequenceNode)
+		if sequenceNode.IsFlowStyle {
+			o.Style = FlowStyle
+		}
+		astSeq := sequenceNode.Values
+		o.Content = make([]*CandidateNode, len(astSeq))
+		for i := 0; i < len(astSeq); i++ {
+			keyNode := o.CreateChild()
+			keyNode.IsMapKey = true
+			keyNode.Tag = "!!int"
+			keyNode.Kind = ScalarNode
+			keyNode.Value = fmt.Sprintf("%v", i)
+
+			valueNode, err := o.goccyDecodeIntoChild(astSeq[i], cm)
+			if err != nil {
+				return err
+			}
+
+			valueNode.Key = keyNode
+			o.Content[i] = valueNode
+		}
+
+	default:
+		log.Debugf("UnmarshalYAML -  node idea of the type!!")
+	}
+	log.Debugf("KIND: %v", o.Kind)
+	return nil
+}
+
+func (o *CandidateNode) goccyProcessMappingValueNode(mappingEntry *ast.MappingValueNode, cm yaml.CommentMap) error {
+	log.Debug("UnmarshalYAML MAP KEY entry %v", mappingEntry.Key)
+	keyNode, err := o.goccyDecodeIntoChild(mappingEntry.Key, cm)
+	if err != nil {
+		return err
+	}
+	keyNode.IsMapKey = true
+
+	log.Debug("UnmarshalYAML MAP VALUE entry %v", mappingEntry.Value)
+	valueNode, err := o.goccyDecodeIntoChild(mappingEntry.Value, cm)
+	if err != nil {
+		return err
+	}
+
+	if mappingEntry.FootComment != nil {
+		valueNode.FootComment = mappingEntry.FootComment.String()
+	}
+	o.AddKeyValueChild(keyNode, valueNode)
+
+	return nil
+}

pkg/yqlib/candidate_node_test.go

@@ -0,0 +1,162 @@
+package yqlib
+
+import (
+	"fmt"
+	"testing"
+
+	"github.com/mikefarah/yq/v4/test"
+)
+
+type valueRepScenario struct {
+	input    string
+	tag      string
+	expected interface{}
+}
+
+var valueRepScenarios = []valueRepScenario{
+	{
+		input:    `"cat"`,
+		expected: `"cat"`,
+	},
+	{
+		input:    `3`,
+		expected: int64(3),
+	},
+	{
+		input:    `3.1`,
+		expected: float64(3.1),
+	},
+	{
+		input:    `true`,
+		expected: true,
+	},
+	{
+		input:    `y`,
+		tag:      "!!bool",
+		expected: true,
+	},
+	{
+		tag:      "!!null",
+		expected: nil,
+	},
+}
+
+func TestCandidateNodeGetValueRepScenarios(t *testing.T) {
+	for _, tt := range valueRepScenarios {
+		node := CandidateNode{Value: tt.input, Tag: tt.tag}
+		actual, err := node.GetValueRep()
+		if err != nil {
+			t.Error(err)
+			return
+		}
+		test.AssertResult(t, tt.expected, actual)
+	}
+}
+
+func TestCandidateNodeChildWhenParentUpdated(t *testing.T) {
+	parent := CandidateNode{}
+	child := parent.CreateChild()
+	parent.SetDocument(1)
+	parent.SetFileIndex(2)
+	parent.SetFilename("meow")
+	test.AssertResultWithContext(t, "meow", child.GetFilename(), "filename")
+	test.AssertResultWithContext(t, 2, child.GetFileIndex(), "file index")
+	test.AssertResultWithContext(t, uint(1), child.GetDocument(), "document index")
+}
+
+type CreateScalarNodeScenario struct {
+	value       interface{}
+	stringValue string
+	expectedTag string
+}
+
+var createScalarScenarios = []CreateScalarNodeScenario{
+	{
+		value:       "mike",
+		stringValue: "mike",
+		expectedTag: "!!str",
+	},
+	{
+		value:       3,
+		stringValue: "3",
+		expectedTag: "!!int",
+	},
+	{
+		value:       3.1,
+		stringValue: "3.1",
+		expectedTag: "!!float",
+	},
+	{
+		value:       true,
+		stringValue: "true",
+		expectedTag: "!!bool",
+	},
+	{
+		value:       nil,
+		stringValue: "~",
+		expectedTag: "!!null",
+	},
+}
+
+func TestCreateScalarNodeScenarios(t *testing.T) {
+	for _, tt := range createScalarScenarios {
+		actual := CreateScalarNode(tt.value, tt.stringValue)
+		test.AssertResultWithContext(t, tt.stringValue, actual.Value, fmt.Sprintf("Value for: Value: [%v], String: %v", tt.value, tt.stringValue))
+		test.AssertResultWithContext(t, tt.expectedTag, actual.Tag, fmt.Sprintf("Value for: Value: [%v], String: %v", tt.value, tt.stringValue))
+	}
+}
+
+func TestGetKeyForMapValue(t *testing.T) {
+	key := CreateStringScalarNode("yourKey")
+	n := CandidateNode{Key: key, Value: "meow", document: 3}
+	test.AssertResult(t, "3 - yourKey", n.GetKey())
+}
+
+func TestGetKeyForMapKey(t *testing.T) {
+	key := CreateStringScalarNode("yourKey")
+	key.IsMapKey = true
+	key.document = 3
+	test.AssertResult(t, "key-yourKey-3 - ", key.GetKey())
+}
+
+func TestGetKeyForValue(t *testing.T) {
+	n := CandidateNode{Value: "meow", document: 3}
+	test.AssertResult(t, "3 - ", n.GetKey())
+}
+
+func TestGetParsedKeyForMapKey(t *testing.T) {
+	key := CreateStringScalarNode("yourKey")
+	key.IsMapKey = true
+	key.document = 3
+	test.AssertResult(t, "yourKey", key.getParsedKey())
+}
+
+func TestGetParsedKeyForLooseValue(t *testing.T) {
+	n := CandidateNode{Value: "meow", document: 3}
+	test.AssertResult(t, nil, n.getParsedKey())
+}
+
+func TestGetParsedKeyForMapValue(t *testing.T) {
+	key := CreateStringScalarNode("yourKey")
+	n := CandidateNode{Key: key, Value: "meow", document: 3}
+	test.AssertResult(t, "yourKey", n.getParsedKey())
+}
+
+func TestGetParsedKeyForArrayValue(t *testing.T) {
+	key := CreateScalarNode(4, "4")
+	n := CandidateNode{Key: key, Value: "meow", document: 3}
+	test.AssertResult(t, 4, n.getParsedKey())
+}
+
+func TestCandidateNodeAddKeyValueChild(t *testing.T) {
+	key := CandidateNode{Value: "cool", IsMapKey: true}
+	node := CandidateNode{}
+
+	// if we use a key in a new node as a value, it should no longer be marked as a key
+
+	_, keyIsValueNow := node.AddKeyValueChild(&CandidateNode{Value: "newKey"}, &key)
+
+	test.AssertResult(t, keyIsValueNow.IsMapKey, false)
+	test.AssertResult(t, key.IsMapKey, true)
+
+}

pkg/yqlib/candidate_node_yaml.go

@@ -0,0 +1,218 @@
+package yqlib
+
+import (
+	"fmt"
+
+	yaml "gopkg.in/yaml.v3"
+)
+
+func MapYamlStyle(original yaml.Style) Style {
+	switch original {
+	case yaml.TaggedStyle:
+		return TaggedStyle
+	case yaml.DoubleQuotedStyle:
+		return DoubleQuotedStyle
+	case yaml.SingleQuotedStyle:
+		return SingleQuotedStyle
+	case yaml.LiteralStyle:
+		return LiteralStyle
+	case yaml.FoldedStyle:
+		return FoldedStyle
+	case yaml.FlowStyle:
+		return FlowStyle
+	case 0:
+		return 0
+	}
+	return Style(original)
+}
+
+func MapToYamlStyle(original Style) yaml.Style {
+	switch original {
+	case TaggedStyle:
+		return yaml.TaggedStyle
+	case DoubleQuotedStyle:
+		return yaml.DoubleQuotedStyle
+	case SingleQuotedStyle:
+		return yaml.SingleQuotedStyle
+	case LiteralStyle:
+		return yaml.LiteralStyle
+	case FoldedStyle:
+		return yaml.FoldedStyle
+	case FlowStyle:
+		return yaml.FlowStyle
+	case 0:
+		return 0
+	}
+	return yaml.Style(original)
+}
+
+func (o *CandidateNode) copyFromYamlNode(node *yaml.Node, anchorMap map[string]*CandidateNode) {
+	o.Style = MapYamlStyle(node.Style)
+
+	o.Tag = node.Tag
+	o.Value = node.Value
+	o.Anchor = node.Anchor
+
+	if o.Anchor != "" {
+		anchorMap[o.Anchor] = o
+		log.Debug("set anchor %v to %v", o.Anchor, NodeToString(o))
+	}
+
+	// its a single alias
+	if node.Alias != nil && node.Alias.Anchor != "" {
+		o.Alias = anchorMap[node.Alias.Anchor]
+		log.Debug("set alias to %v", NodeToString(anchorMap[node.Alias.Anchor]))
+	}
+	o.HeadComment = node.HeadComment
+	o.LineComment = node.LineComment
+	o.FootComment = node.FootComment
+
+	o.Line = node.Line
+	o.Column = node.Column
+}
+
+func (o *CandidateNode) copyToYamlNode(node *yaml.Node) {
+	node.Style = MapToYamlStyle(o.Style)
+
+	node.Tag = o.Tag
+	node.Value = o.Value
+	node.Anchor = o.Anchor
+
+	// node.Alias = TODO - find Alias in our own structure
+	// might need to be a post process thing
+
+	node.HeadComment = o.HeadComment
+
+	node.LineComment = o.LineComment
+	node.FootComment = o.FootComment
+
+	node.Line = o.Line
+	node.Column = o.Column
+}
+
+func (o *CandidateNode) decodeIntoChild(childNode *yaml.Node, anchorMap map[string]*CandidateNode) (*CandidateNode, error) {
+	newChild := o.CreateChild()
+
+	// null yaml.Nodes to not end up calling UnmarshalYAML
+	// so we call it explicitly
+	if childNode.Tag == "!!null" {
+		newChild.Kind = ScalarNode
+		newChild.copyFromYamlNode(childNode, anchorMap)
+		return newChild, nil
+	}
+
+	err := newChild.UnmarshalYAML(childNode, anchorMap)
+	return newChild, err
+}
+
+func (o *CandidateNode) UnmarshalYAML(node *yaml.Node, anchorMap map[string]*CandidateNode) error {
+	log.Debugf("UnmarshalYAML %v", node.Tag)
+	switch node.Kind {
+	case yaml.AliasNode:
+		log.Debug("UnmarshalYAML - alias from yaml: %v", o.Tag)
+		o.Kind = AliasNode
+		o.copyFromYamlNode(node, anchorMap)
+		return nil
+	case yaml.ScalarNode:
+		log.Debugf("UnmarshalYAML -  a scalar")
+		o.Kind = ScalarNode
+		o.copyFromYamlNode(node, anchorMap)
+		return nil
+	case yaml.MappingNode:
+		log.Debugf("UnmarshalYAML -  a mapping node")
+		o.Kind = MappingNode
+		o.copyFromYamlNode(node, anchorMap)
+		o.Content = make([]*CandidateNode, len(node.Content))
+		for i := 0; i < len(node.Content); i += 2 {
+
+			keyNode, err := o.decodeIntoChild(node.Content[i], anchorMap)
+			if err != nil {
+				return err
+			}
+
+			keyNode.IsMapKey = true
+
+			valueNode, err := o.decodeIntoChild(node.Content[i+1], anchorMap)
+			if err != nil {
+				return err
+			}
+
+			valueNode.Key = keyNode
+
+			o.Content[i] = keyNode
+			o.Content[i+1] = valueNode
+		}
+		log.Debugf("UnmarshalYAML -  finished mapping node")
+		return nil
+	case yaml.SequenceNode:
+		log.Debugf("UnmarshalYAML -  a sequence: %v", len(node.Content))
+		o.Kind = SequenceNode
+
+		o.copyFromYamlNode(node, anchorMap)
+		log.Debugf("node Style: %v", node.Style)
+		log.Debugf("o Style: %v", o.Style)
+		o.Content = make([]*CandidateNode, len(node.Content))
+		for i := 0; i < len(node.Content); i++ {
+			keyNode := o.CreateChild()
+			keyNode.IsMapKey = true
+			keyNode.Tag = "!!int"
+			keyNode.Kind = ScalarNode
+			keyNode.Value = fmt.Sprintf("%v", i)
+
+			valueNode, err := o.decodeIntoChild(node.Content[i], anchorMap)
+			if err != nil {
+				return err
+			}
+
+			valueNode.Key = keyNode
+			o.Content[i] = valueNode
+		}
+		return nil
+	case 0:
+		// not sure when this happens
+		o.copyFromYamlNode(node, anchorMap)
+		log.Debugf("UnmarshalYAML -  err.. %v", NodeToString(o))
+		return nil
+	default:
+		return fmt.Errorf("orderedMap: invalid yaml node")
+	}
+}
+
+func (o *CandidateNode) MarshalYAML() (*yaml.Node, error) {
+	log.Debug("MarshalYAML to yaml: %v", o.Tag)
+	switch o.Kind {
+	case AliasNode:
+		log.Debug("MarshalYAML - alias to yaml: %v", o.Tag)
+		target := &yaml.Node{Kind: yaml.AliasNode}
+		o.copyToYamlNode(target)
+		return target, nil
+	case ScalarNode:
+		log.Debug("MarshalYAML - scalar: %v", o.Value)
+		target := &yaml.Node{Kind: yaml.ScalarNode}
+		o.copyToYamlNode(target)
+		return target, nil
+	case MappingNode, SequenceNode:
+		targetKind := yaml.MappingNode
+		if o.Kind == SequenceNode {
+			targetKind = yaml.SequenceNode
+		}
+		target := &yaml.Node{Kind: targetKind}
+		o.copyToYamlNode(target)
+		log.Debugf("original style: %v", o.Style)
+		log.Debugf("original: %v, tag: %v, style: %v, kind: %v", NodeToString(o), target.Tag, target.Style, target.Kind == yaml.SequenceNode)
+		target.Content = make([]*yaml.Node, len(o.Content))
+		for i := 0; i < len(o.Content); i++ {
+
+			child, err := o.Content[i].MarshalYAML()
+
+			if err != nil {
+				return nil, err
+			}
+			target.Content[i] = child
+		}
+		return target, nil
+	}
+	target := &yaml.Node{}
+	o.copyToYamlNode(target)
+	return target, nil
+}

pkg/yqlib/candidiate_node_json.go

@@ -0,0 +1,177 @@
+package yqlib
+
+import (
+	"bytes"
+	"errors"
+	"fmt"
+	"io"
+
+	"github.com/goccy/go-json"
+)
+
+func (o *CandidateNode) setScalarFromJson(value interface{}) error {
+	o.Kind = ScalarNode
+	switch rawData := value.(type) {
+	case nil:
+		o.Tag = "!!null"
+		o.Value = "null"
+	case float64, float32:
+		o.Value = fmt.Sprintf("%v", value)
+		o.Tag = "!!float"
+		// json decoder returns ints as float.
+		if value == float64(int64(rawData.(float64))) {
+			// aha it's an int disguised as a float
+			o.Tag = "!!int"
+			o.Value = fmt.Sprintf("%v", int64(value.(float64)))
+		}
+	case int, int64, int32:
+		o.Value = fmt.Sprintf("%v", value)
+		o.Tag = "!!int"
+	case bool:
+		o.Value = fmt.Sprintf("%v", value)
+		o.Tag = "!!bool"
+	case string:
+		o.Value = rawData
+		o.Tag = "!!str"
+	default:
+		return fmt.Errorf("unrecognised type :( %v", rawData)
+	}
+	return nil
+}
+
+func (o *CandidateNode) UnmarshalJSON(data []byte) error {
+	log.Debug("UnmarshalJSON")
+	switch data[0] {
+	case '{':
+		log.Debug("UnmarshalJSON -  its a map!")
+		// its a map
+		o.Kind = MappingNode
+		o.Tag = "!!map"
+
+		dec := json.NewDecoder(bytes.NewReader(data))
+		_, err := dec.Token() // open object
+		if err != nil {
+			return err
+		}
+
+		// cycle through k/v
+		var tok json.Token
+		for tok, err = dec.Token(); err == nil; tok, err = dec.Token() {
+			// we can expect two types: string or Delim. Delim automatically means
+			// that it is the closing bracket of the object, whereas string means
+			// that there is another key.
+			if _, ok := tok.(json.Delim); ok {
+				break
+			}
+
+			childKey := o.CreateChild()
+			childKey.IsMapKey = true
+			childKey.Value = tok.(string)
+			childKey.Kind = ScalarNode
+			childKey.Tag = "!!str"
+
+			childValue := o.CreateChild()
+			childValue.Key = childKey
+
+			if err := dec.Decode(childValue); err != nil {
+				return err
+			}
+			o.Content = append(o.Content, childKey, childValue)
+		}
+		// unexpected error
+		if err != nil && !errors.Is(err, io.EOF) {
+			return err
+		}
+		return nil
+	case '[':
+		o.Kind = SequenceNode
+		o.Tag = "!!seq"
+		log.Debug("UnmarshalJSON -  its an array!")
+		var children []*CandidateNode
+		if err := json.Unmarshal(data, &children); err != nil {
+			return err
+		}
+		// now we put the children into the content, and set a key value for them
+		for i, child := range children {
+
+			if child == nil {
+				// need to represent it as a null scalar
+				child = CreateScalarNode(nil, "null")
+			}
+			childKey := o.CreateChild()
+			childKey.Kind = ScalarNode
+			childKey.Tag = "!!int"
+			childKey.Value = fmt.Sprintf("%v", i)
+			childKey.IsMapKey = true
+
+			child.Parent = o
+			child.Key = childKey
+			o.Content = append(o.Content, child)
+		}
+		return nil
+	}
+	log.Debug("UnmarshalJSON -  its a scalar!")
+	// otherwise, must be a scalar
+	var scalar interface{}
+	err := json.Unmarshal(data, &scalar)
+
+	if err != nil {
+		return err
+	}
+	log.Debug("UnmarshalJSON -  scalar is %v", scalar)
+
+	return o.setScalarFromJson(scalar)
+
+}
+
+func (o *CandidateNode) MarshalJSON() ([]byte, error) {
+	log.Debugf("MarshalJSON %v", NodeToString(o))
+	buf := new(bytes.Buffer)
+	enc := json.NewEncoder(buf)
+	enc.SetIndent("", " ")
+	enc.SetEscapeHTML(false) // do not escape html chars e.g. &, <, >
+
+	switch o.Kind {
+	case AliasNode:
+		log.Debugf("MarshalJSON AliasNode")
+		err := enc.Encode(o.Alias)
+		return buf.Bytes(), err
+	case ScalarNode:
+		log.Debugf("MarshalJSON ScalarNode")
+		value, err := o.GetValueRep()
+		if err != nil {
+			return buf.Bytes(), err
+		}
+		err = enc.Encode(value)
+		return buf.Bytes(), err
+	case MappingNode:
+		log.Debugf("MarshalJSON MappingNode")
+		buf.WriteByte('{')
+		for i := 0; i < len(o.Content); i += 2 {
+			if err := enc.Encode(o.Content[i].Value); err != nil {
+				return nil, err
+			}
+			buf.WriteByte(':')
+			if err := enc.Encode(o.Content[i+1]); err != nil {
+				return nil, err
+			}
+			if i != len(o.Content)-2 {
+				buf.WriteByte(',')
+			}
+		}
+		buf.WriteByte('}')
+		return buf.Bytes(), nil
+	case SequenceNode:
+		log.Debugf("MarshalJSON SequenceNode, %v, len: %v", o.Content, len(o.Content))
+		var err error
+		if len(o.Content) == 0 {
+			buf.WriteString("[]")
+		} else {
+			err = enc.Encode(o.Content)
+		}
+		return buf.Bytes(), err
+	default:
+		err := enc.Encode(nil)
+		return buf.Bytes(), err
+	}
+}

pkg/yqlib/context.go

@@ -5,7 +5,6 @@ import (
 	"fmt"
 	"time"
 
-	"github.com/jinzhu/copier"
 	logging "gopkg.in/op/go-logging.v1"
 )
 
@@ -58,13 +57,19 @@ func (n *Context) SetVariable(name string, value *list.List) {
 func (n *Context) ChildContext(results *list.List) Context {
 	clone := Context{DontAutoCreate: n.DontAutoCreate, datetimeLayout: n.datetimeLayout}
 	clone.Variables = make(map[string]*list.List)
-	if len(n.Variables) > 0 {
-		err := copier.Copy(&clone.Variables, n.Variables)
-		if err != nil {
-			log.Error("Error cloning context :(")
-			panic(err)
+	for variableKey, originalValueList := range n.Variables {
+
+		variableCopyList := list.New()
+		for el := originalValueList.Front(); el != nil; el = el.Next() {
+			// note that we dont make a copy of the candidate node
+			// this is so the 'ref' operator can work correctly.
+			clonedNode := el.Value.(*CandidateNode)
+			variableCopyList.PushBack(clonedNode)
 		}
+
+		clone.Variables[variableKey] = variableCopyList
 	}
+
 	clone.MatchingNodes = results
 	return clone
 }
@@ -78,35 +83,18 @@ func (n *Context) ToString() string {
 }
 
 func (n *Context) DeepClone() Context {
-	clone := Context{}
-	err := copier.Copy(&clone, n)
-	// copier doesn't do lists properly for some reason
-	clone.MatchingNodes = list.New()
+
+	clonedContent := list.New()
 	for el := n.MatchingNodes.Front(); el != nil; el = el.Next() {
-		clonedNode, err := el.Value.(*CandidateNode).Copy()
-		if err != nil {
-			log.Error("Error cloning context :(")
-			panic(err)
-		}
-		clone.MatchingNodes.PushBack(clonedNode)
+		clonedNode := el.Value.(*CandidateNode).Copy()
+		clonedContent.PushBack(clonedNode)
 	}
 
-	if err != nil {
-		log.Error("Error cloning context :(")
-		panic(err)
-	}
-	return clone
+	return n.ChildContext(clonedContent)
 }
 
 func (n *Context) Clone() Context {
-	clone := Context{}
-	err := copier.Copy(&clone, n)
-
-	if err != nil {
-		log.Error("Error cloning context :(")
-		panic(err)
-	}
-	return clone
+	return n.ChildContext(n.MatchingNodes)
 }
 
 func (n *Context) ReadOnlyClone() Context {

pkg/yqlib/context_test.go

@@ -0,0 +1,51 @@
+package yqlib
+
+import (
+	"container/list"
+	"testing"
+
+	"github.com/mikefarah/yq/v4/test"
+)
+
+func TestChildContext(t *testing.T) {
+
+	expectedOriginal := make(map[string]*list.List)
+	expectedOriginal["dog"] = list.New()
+	expectedOriginal["dog"].PushBack(&CandidateNode{Value: "woof"})
+
+	originalVariables := make(map[string]*list.List)
+	originalVariables["dog"] = list.New()
+	originalVariables["dog"].PushBack(&CandidateNode{Value: "woof"})
+
+	original := Context{
+		DontAutoCreate: true,
+		datetimeLayout: "cat",
+		Variables:      originalVariables,
+	}
+
+	newResults := list.New()
+	newResults.PushBack(&CandidateNode{Value: "bar"})
+
+	clone := original.ChildContext(newResults)
+	test.AssertResultComplex(t, originalVariables, clone.Variables)
+
+	clone.Variables["dog"].PushBack("bark")
+	// ensure this is a separate copy
+	test.AssertResultComplex(t, 1, originalVariables["dog"].Len())
+
+}
+
+func TestChildContextNoVariables(t *testing.T) {
+
+	original := Context{
+		DontAutoCreate: true,
+		datetimeLayout: "cat",
+	}
+
+	newResults := list.New()
+	newResults.PushBack(&CandidateNode{Value: "bar"})
+
+	clone := original.ChildContext(newResults)
+	test.AssertResultComplex(t, make(map[string]*list.List), clone.Variables)
+
+}

pkg/yqlib/csv.go

@@ -0,0 +1,23 @@
+package yqlib
+
+type CsvPreferences struct {
+	Separator rune
+	AutoParse bool
+}
+
+func NewDefaultCsvPreferences() CsvPreferences {
+	return CsvPreferences{
+		Separator: ',',
+		AutoParse: true,
+	}
+}
+
+func NewDefaultTsvPreferences() CsvPreferences {
+	return CsvPreferences{
+		Separator: '\t',
+		AutoParse: true,
+	}
+}
+
+var ConfiguredCsvPreferences = NewDefaultCsvPreferences()
+var ConfiguredTsvPreferences = NewDefaultTsvPreferences()

pkg/yqlib/csv_test.go

@@ -12,6 +12,11 @@ const csvSimple = `name,numberOfCats,likesApples,height
 Gary,1,true,168.8
 Samantha's Rabbit,2,false,-188.8
 `
+
+const csvSimpleWithObject = `name,numberOfCats,likesApples,height,facts
+Gary,1,true,168.8,cool: true
+Samantha's Rabbit,2,false,-188.8,tall: indeed
+`
 const csvMissing = `name,numberOfCats,likesApples,height
 ,null,,168.8
 `
@@ -39,6 +44,31 @@ const expectedYamlFromCSV = `- name: Gary
   likesApples: false
   height: -188.8
 `
+const expectedYamlFromCSVWithObject = `- name: Gary
+  numberOfCats: 1
+  likesApples: true
+  height: 168.8
+  facts:
+    cool: true
+- name: Samantha's Rabbit
+  numberOfCats: 2
+  likesApples: false
+  height: -188.8
+  facts:
+    tall: indeed
+`
+
+const expectedYamlFromCSVNoParsing = `- name: Gary
+  numberOfCats: 1
+  likesApples: true
+  height: 168.8
+  facts: 'cool: true'
+- name: Samantha's Rabbit
+  numberOfCats: 2
+  likesApples: false
+  height: -188.8
+  facts: 'tall: indeed'
+`
 
 const expectedYamlFromCSVMissingData = `- name: Gary
   numberOfCats: 1
@@ -64,7 +94,7 @@ const tsvTestExpectedSimpleCsv = `i	like	csv
 because	excel	is	cool
 `
 
-var csvScenarios = []formatScenario{
+var csvScenarios = []FormatScenario{
 	{
 		description:  "Encode CSV simple",
 		input:        csvTestSimpleYaml,
@@ -119,12 +149,42 @@ var csvScenarios = []formatScenario{
 		expected:     csvMissing,
 		scenarioType: "roundtrip-csv",
 	},
+	{
+		description:  "decode csv key",
+		skipDoc:      true,
+		input:        csvSimple,
+		expression:   ".[0].name | key",
+		expected:     "name\n",
+		scenarioType: "decode-csv",
+	},
+	{
+		description:  "decode csv parent",
+		skipDoc:      true,
+		input:        csvSimple,
+		expression:   ".[0].name | parent | .height",
+		expected:     "168.8\n",
+		scenarioType: "decode-csv",
+	},
 	{
 		description:    "Parse CSV into an array of objects",
-		subdescription: "First row is assumed to be the header row.",
-		input:          csvSimple,
-		expected:       expectedYamlFromCSV,
-		scenarioType:   "decode-csv-object",
+		subdescription: "First row is assumed to be the header row. By default, entries with YAML/JSON formatting will be parsed!",
+		input:          csvSimpleWithObject,
+		expected:       expectedYamlFromCSVWithObject,
+		scenarioType:   "decode-csv",
+	},
+	{
+		description:  "Decode CSV line breaks",
+		skipDoc:      true,
+		input:        "heading1\n\"some data\nwith a line break\"\n",
+		expected:     "- heading1: |-\n    some data\n    with a line break\n",
+		scenarioType: "decode-csv",
+	},
+	{
+		description:    "Parse CSV into an array of objects, no auto-parsing",
+		subdescription: "First row is assumed to be the header row. Entries with YAML/JSON will be left as strings.",
+		input:          csvSimpleWithObject,
+		expected:       expectedYamlFromCSVNoParsing,
+		scenarioType:   "decode-csv-no-auto",
 	},
 	{
 		description:  "Scalar roundtrip",
@@ -150,24 +210,26 @@ var csvScenarios = []formatScenario{
 	},
 }
 
-func testCSVScenario(t *testing.T, s formatScenario) {
+func testCSVScenario(t *testing.T, s FormatScenario) {
 	switch s.scenarioType {
 	case "encode-csv":
-		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewYamlDecoder(ConfiguredYamlPreferences), NewCsvEncoder(',')), s.description)
+		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewYamlDecoder(ConfiguredYamlPreferences), NewCsvEncoder(ConfiguredCsvPreferences)), s.description)
 	case "encode-tsv":
-		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewYamlDecoder(ConfiguredYamlPreferences), NewCsvEncoder('\t')), s.description)
-	case "decode-csv-object":
-		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewCSVObjectDecoder(','), NewYamlEncoder(2, false, ConfiguredYamlPreferences)), s.description)
+		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewYamlDecoder(ConfiguredYamlPreferences), NewCsvEncoder(ConfiguredTsvPreferences)), s.description)
+	case "decode-csv":
+		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewCSVObjectDecoder(ConfiguredCsvPreferences), NewYamlEncoder(ConfiguredYamlPreferences)), s.description)
+	case "decode-csv-no-auto":
+		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewCSVObjectDecoder(CsvPreferences{Separator: ',', AutoParse: false}), NewYamlEncoder(ConfiguredYamlPreferences)), s.description)
 	case "decode-tsv-object":
-		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewCSVObjectDecoder('\t'), NewYamlEncoder(2, false, ConfiguredYamlPreferences)), s.description)
+		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewCSVObjectDecoder(ConfiguredTsvPreferences), NewYamlEncoder(ConfiguredYamlPreferences)), s.description)
 	case "roundtrip-csv":
-		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewCSVObjectDecoder(','), NewCsvEncoder(',')), s.description)
+		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewCSVObjectDecoder(ConfiguredCsvPreferences), NewCsvEncoder(ConfiguredCsvPreferences)), s.description)
 	default:
 		panic(fmt.Sprintf("unhandled scenario type %q", s.scenarioType))
 	}
 }
 
-func documentCSVDecodeObjectScenario(w *bufio.Writer, s formatScenario, formatType string) {
+func documentCSVDecodeObjectScenario(w *bufio.Writer, s FormatScenario, formatType string) {
 	writeOrPanic(w, fmt.Sprintf("## %v\n", s.description))
 
 	if s.subdescription != "" {
@@ -188,11 +250,36 @@ func documentCSVDecodeObjectScenario(w *bufio.Writer, s formatScenario, formatTy
 	}
 
 	writeOrPanic(w, fmt.Sprintf("```yaml\n%v```\n\n",
-		mustProcessFormatScenario(s, NewCSVObjectDecoder(separator), NewYamlEncoder(s.indent, false, ConfiguredYamlPreferences))),
+		mustProcessFormatScenario(s, NewCSVObjectDecoder(CsvPreferences{Separator: separator, AutoParse: true}), NewYamlEncoder(ConfiguredYamlPreferences))),
 	)
 }
 
-func documentCSVEncodeScenario(w *bufio.Writer, s formatScenario, formatType string) {
+func documentCSVDecodeObjectNoAutoScenario(w *bufio.Writer, s FormatScenario, formatType string) {
+	writeOrPanic(w, fmt.Sprintf("## %v\n", s.description))
+
+	if s.subdescription != "" {
+		writeOrPanic(w, s.subdescription)
+		writeOrPanic(w, "\n\n")
+	}
+
+	writeOrPanic(w, fmt.Sprintf("Given a sample.%v file of:\n", formatType))
+	writeOrPanic(w, fmt.Sprintf("```%v\n%v\n```\n", formatType, s.input))
+
+	writeOrPanic(w, "then\n")
+	writeOrPanic(w, fmt.Sprintf("```bash\nyq -p=%v --csv-auto-parse=f sample.%v\n```\n", formatType, formatType))
+	writeOrPanic(w, "will output\n")
+
+	separator := ','
+	if formatType == "tsv" {
+		separator = '\t'
+	}
+
+	writeOrPanic(w, fmt.Sprintf("```yaml\n%v```\n\n",
+		mustProcessFormatScenario(s, NewCSVObjectDecoder(CsvPreferences{Separator: separator, AutoParse: false}), NewYamlEncoder(ConfiguredYamlPreferences))),
+	)
+}
+
+func documentCSVEncodeScenario(w *bufio.Writer, s FormatScenario, formatType string) {
 	writeOrPanic(w, fmt.Sprintf("## %v\n", s.description))
 
 	if s.subdescription != "" {
@@ -218,13 +305,14 @@ func documentCSVEncodeScenario(w *bufio.Writer, s formatScenario, formatType str
 	if formatType == "tsv" {
 		separator = '\t'
 	}
-
+	csvPrefs := NewDefaultCsvPreferences()
+	csvPrefs.Separator = separator
 	writeOrPanic(w, fmt.Sprintf("```%v\n%v```\n\n", formatType,
-		mustProcessFormatScenario(s, NewYamlDecoder(ConfiguredYamlPreferences), NewCsvEncoder(separator))),
+		mustProcessFormatScenario(s, NewYamlDecoder(ConfiguredYamlPreferences), NewCsvEncoder(csvPrefs))),
 	)
 }
 
-func documentCSVRoundTripScenario(w *bufio.Writer, s formatScenario, formatType string) {
+func documentCSVRoundTripScenario(w *bufio.Writer, s FormatScenario, formatType string) {
 	writeOrPanic(w, fmt.Sprintf("## %v\n", s.description))
 
 	if s.subdescription != "" {
@@ -251,13 +339,16 @@ func documentCSVRoundTripScenario(w *bufio.Writer, s formatScenario, formatType
 		separator = '\t'
 	}
 
+	csvPrefs := NewDefaultCsvPreferences()
+	csvPrefs.Separator = separator
+
 	writeOrPanic(w, fmt.Sprintf("```%v\n%v```\n\n", formatType,
-		mustProcessFormatScenario(s, NewCSVObjectDecoder(separator), NewCsvEncoder(separator))),
+		mustProcessFormatScenario(s, NewCSVObjectDecoder(CsvPreferences{Separator: separator, AutoParse: true}), NewCsvEncoder(csvPrefs))),
 	)
 }
 
-func documentCSVScenario(t *testing.T, w *bufio.Writer, i interface{}) {
-	s := i.(formatScenario)
+func documentCSVScenario(_ *testing.T, w *bufio.Writer, i interface{}) {
+	s := i.(FormatScenario)
 	if s.skipDoc {
 		return
 	}
@@ -266,8 +357,10 @@ func documentCSVScenario(t *testing.T, w *bufio.Writer, i interface{}) {
 		documentCSVEncodeScenario(w, s, "csv")
 	case "encode-tsv":
 		documentCSVEncodeScenario(w, s, "tsv")
-	case "decode-csv-object":
+	case "decode-csv":
 		documentCSVDecodeObjectScenario(w, s, "csv")
+	case "decode-csv-no-auto":
+		documentCSVDecodeObjectNoAutoScenario(w, s, "csv")
 	case "decode-tsv-object":
 		documentCSVDecodeObjectScenario(w, s, "tsv")
 	case "roundtrip-csv":

pkg/yqlib/data_tree_navigator.go

@@ -4,16 +4,16 @@ import (
 	"fmt"
 
 	logging "gopkg.in/op/go-logging.v1"
-	yaml "gopkg.in/yaml.v3"
 )
 
 type DataTreeNavigator interface {
-	// given the context and a expressionNode,
+	// given the context and an expressionNode,
 	// this will process the against the given expressionNode and return
 	// a new context of matching candidates
 	GetMatchingNodes(context Context, expressionNode *ExpressionNode) (Context, error)
 
-	DeeplyAssign(context Context, path []interface{}, rhsNode *yaml.Node) error
+	DeeplyAssign(context Context, path []interface{}, rhsNode *CandidateNode) error
+	DeeplyAssignKey(context Context, path []interface{}, rhsNode *CandidateNode) error
 }
 
 type dataTreeNavigator struct {
@@ -23,20 +23,34 @@ func NewDataTreeNavigator() DataTreeNavigator {
 	return &dataTreeNavigator{}
 }
 
-func (d *dataTreeNavigator) DeeplyAssign(context Context, path []interface{}, rhsNode *yaml.Node) error {
+func (d *dataTreeNavigator) DeeplyAssign(context Context, path []interface{}, rhsCandidateNode *CandidateNode) error {
+	return d.deeplyAssign(context, path, rhsCandidateNode, false)
+}
 
-	rhsCandidateNode := &CandidateNode{
-		Path: path,
-		Node: rhsNode,
-	}
+func (d *dataTreeNavigator) DeeplyAssignKey(context Context, path []interface{}, rhsCandidateNode *CandidateNode) error {
+	return d.deeplyAssign(context, path, rhsCandidateNode, true)
+}
+
+func (d *dataTreeNavigator) deeplyAssign(context Context, path []interface{}, rhsCandidateNode *CandidateNode, targetKey bool) error {
 
 	assignmentOp := &Operation{OperationType: assignOpType, Preferences: assignPreferences{}}
 
+	if rhsCandidateNode.Kind == MappingNode {
+		log.Debug("DeeplyAssign: deeply merging object")
+		// if the rhs is a map, we need to deeply merge it in.
+		// otherwise we'll clobber any existing fields
+		assignmentOp = &Operation{OperationType: multiplyAssignOpType, Preferences: multiplyPreferences{
+			AppendArrays:  true,
+			TraversePrefs: traversePreferences{DontFollowAlias: true},
+			AssignPrefs:   assignPreferences{},
+		}}
+	}
+
 	rhsOp := &Operation{OperationType: valueOpType, CandidateNode: rhsCandidateNode}
 
 	assignmentOpNode := &ExpressionNode{
 		Operation: assignmentOp,
-		LHS:       createTraversalTree(path, traversePreferences{}, false),
+		LHS:       createTraversalTree(path, traversePreferences{}, targetKey),
 		RHS:       &ExpressionNode{Operation: rhsOp},
 	}
 
@@ -55,7 +69,6 @@ func (d *dataTreeNavigator) GetMatchingNodes(context Context, expressionNode *Ex
 			log.Debug(NodeToString(el.Value.(*CandidateNode)))
 		}
 	}
-	log.Debug(">>")
 	handler := expressionNode.Operation.OperationType.Handler
 	if handler != nil {
 		return handler(d, context, expressionNode)

pkg/yqlib/decoder.go

@@ -1,63 +1,10 @@
 package yqlib
 
 import (
-	"fmt"
 	"io"
-	"strings"
-)
-
-type InputFormat uint
-
-const (
-	YamlInputFormat = 1 << iota
-	XMLInputFormat
-	PropertiesInputFormat
-	Base64InputFormat
-	JsonInputFormat
-	CSVObjectInputFormat
-	TSVObjectInputFormat
-	TomlInputFormat
-	UriInputFormat
 )
 
 type Decoder interface {
 	Init(reader io.Reader) error
 	Decode() (*CandidateNode, error)
 }
-
-func InputFormatFromString(format string) (InputFormat, error) {
-	switch format {
-	case "yaml", "yml", "y":
-		return YamlInputFormat, nil
-	case "xml", "x":
-		return XMLInputFormat, nil
-	case "properties", "props", "p":
-		return PropertiesInputFormat, nil
-	case "json", "ndjson", "j":
-		return JsonInputFormat, nil
-	case "csv", "c":
-		return CSVObjectInputFormat, nil
-	case "tsv", "t":
-		return TSVObjectInputFormat, nil
-	case "toml":
-		return TomlInputFormat, nil
-	default:
-		return 0, fmt.Errorf("unknown format '%v' please use [yaml|json|props|csv|tsv|xml|toml]", format)
-	}
-}
-
-func FormatFromFilename(filename string) string {
-
-	if filename != "" {
-		GetLogger().Debugf("checking file extension '%s' for auto format detection", filename)
-		nPos := strings.LastIndex(filename, ".")
-		if nPos > -1 {
-			format := filename[nPos+1:]
-			GetLogger().Debugf("detected format '%s'", format)
-			return format
-		}
-	}
-
-	GetLogger().Debugf("using default inputFormat 'yaml'")
-	return "yaml"
-}

pkg/yqlib/decoder_base64.go

@@ -5,8 +5,6 @@ import (
 	"encoding/base64"
 	"io"
 	"strings"
-
-	yaml "gopkg.in/yaml.v3"
 )
 
 type base64Padder struct {
@@ -70,11 +68,5 @@ func (dec *base64Decoder) Decode() (*CandidateNode, error) {
 		}
 	}
 	dec.readAnything = true
-	return &CandidateNode{
-		Node: &yaml.Node{
-			Kind:  yaml.ScalarNode,
-			Tag:   "!!str",
-			Value: buf.String(),
-		},
-	}, nil
+	return CreateStringScalarNode(buf.String()), nil
 }

pkg/yqlib/decoder_csv_object.go

@@ -6,44 +6,42 @@ import (
 	"io"
 
 	"github.com/dimchansky/utfbom"
-	yaml "gopkg.in/yaml.v3"
 )
 
 type csvObjectDecoder struct {
-	separator rune
-	reader    csv.Reader
-	finished  bool
+	prefs    CsvPreferences
+	reader   csv.Reader
+	finished bool
 }
 
-func NewCSVObjectDecoder(separator rune) Decoder {
-	return &csvObjectDecoder{separator: separator}
+func NewCSVObjectDecoder(prefs CsvPreferences) Decoder {
+	return &csvObjectDecoder{prefs: prefs}
 }
 
 func (dec *csvObjectDecoder) Init(reader io.Reader) error {
 	cleanReader, enc := utfbom.Skip(reader)
 	log.Debugf("Detected encoding: %s\n", enc)
 	dec.reader = *csv.NewReader(cleanReader)
-	dec.reader.Comma = dec.separator
+	dec.reader.Comma = dec.prefs.Separator
 	dec.finished = false
 	return nil
 }
 
-func (dec *csvObjectDecoder) convertToYamlNode(content string) *yaml.Node {
+func (dec *csvObjectDecoder) convertToNode(content string) *CandidateNode {
 	node, err := parseSnippet(content)
-	if err != nil {
-		return createScalarNode(content, content)
+	// if we're not auto-parsing, then we wont put in parsed objects or arrays
+	// but we still parse scalars
+	if err != nil || (!dec.prefs.AutoParse && node.Kind != ScalarNode) {
+		return CreateScalarNode(content, content)
 	}
 	return node
 }
 
-func (dec *csvObjectDecoder) createObject(headerRow []string, contentRow []string) *yaml.Node {
-	objectNode := &yaml.Node{Kind: yaml.MappingNode, Tag: "!!map"}
+func (dec *csvObjectDecoder) createObject(headerRow []string, contentRow []string) *CandidateNode {
+	objectNode := &CandidateNode{Kind: MappingNode, Tag: "!!map"}
 
 	for i, header := range headerRow {
-		objectNode.Content = append(
-			objectNode.Content,
-			createScalarNode(header, header),
-			dec.convertToYamlNode(contentRow[i]))
+		objectNode.AddKeyValueChild(CreateScalarNode(header, header), dec.convertToNode(contentRow[i]))
 	}
 	return objectNode
 }
@@ -58,13 +56,13 @@ func (dec *csvObjectDecoder) Decode() (*CandidateNode, error) {
 		return nil, err
 	}
 
-	rootArray := &yaml.Node{Kind: yaml.SequenceNode, Tag: "!!seq"}
+	rootArray := &CandidateNode{Kind: SequenceNode, Tag: "!!seq"}
 
 	contentRow, err := dec.reader.Read()
 
 	for err == nil && len(contentRow) > 0 {
 		log.Debugf("Adding contentRow: %v", contentRow)
-		rootArray.Content = append(rootArray.Content, dec.createObject(headerRow, contentRow))
+		rootArray.AddChild(dec.createObject(headerRow, contentRow))
 		contentRow, err = dec.reader.Read()
 		log.Debugf("Read next contentRow: %v, %v", contentRow, err)
 	}
@@ -72,10 +70,5 @@ func (dec *csvObjectDecoder) Decode() (*CandidateNode, error) {
 		return nil, err
 	}
 
-	return &CandidateNode{
-		Node: &yaml.Node{
-			Kind:    yaml.DocumentNode,
-			Content: []*yaml.Node{rootArray},
-		},
-	}, nil
+	return rootArray, nil
 }

pkg/yqlib/decoder_goccy_yaml.go

@@ -0,0 +1,46 @@
+//go:build !yq_noyaml
+
+//
+// NOTE this is still a WIP - not yet ready.
+//
+
+package yqlib
+
+import (
+	"io"
+
+	yaml "github.com/goccy/go-yaml"
+	"github.com/goccy/go-yaml/ast"
+)
+
+type goccyYamlDecoder struct {
+	decoder yaml.Decoder
+	cm      yaml.CommentMap
+}
+
+func NewGoccyYAMLDecoder() Decoder {
+	return &goccyYamlDecoder{}
+}
+
+func (dec *goccyYamlDecoder) Init(reader io.Reader) error {
+	dec.cm = yaml.CommentMap{}
+	dec.decoder = *yaml.NewDecoder(reader, yaml.CommentToMap(dec.cm), yaml.UseOrderedMap())
+	return nil
+}
+
+func (dec *goccyYamlDecoder) Decode() (*CandidateNode, error) {
+
+	var ast ast.Node
+
+	err := dec.decoder.Decode(&ast)
+	if err != nil {
+		return nil, err
+	}
+
+	candidateNode := &CandidateNode{}
+	if err := candidateNode.UnmarshalGoccyYAML(ast, dec.cm); err != nil {
+		return nil, err
+	}
+
+	return candidateNode, nil
+}

pkg/yqlib/decoder_json.go

@@ -3,11 +3,9 @@
 package yqlib
 
 import (
-	"fmt"
 	"io"
 
 	"github.com/goccy/go-json"
-	yaml "gopkg.in/yaml.v3"
 )
 
 type jsonDecoder struct {
@@ -25,68 +23,11 @@ func (dec *jsonDecoder) Init(reader io.Reader) error {
 
 func (dec *jsonDecoder) Decode() (*CandidateNode, error) {
 
-	var dataBucket orderedMap
-	log.Debug("going to decode")
+	var dataBucket CandidateNode
 	err := dec.decoder.Decode(&dataBucket)
 	if err != nil {
 		return nil, err
 	}
-	node, err := dec.convertToYamlNode(&dataBucket)
 
-	if err != nil {
-		return nil, err
-	}
-
-	return &CandidateNode{
-		Node: &yaml.Node{
-			Kind:    yaml.DocumentNode,
-			Content: []*yaml.Node{node},
-		},
-	}, nil
-}
-
-func (dec *jsonDecoder) convertToYamlNode(data *orderedMap) (*yaml.Node, error) {
-	if data == nil {
-		return createScalarNode(nil, "null"), nil
-	}
-	if data.kv == nil {
-		switch rawData := data.altVal.(type) {
-		case nil:
-			return createScalarNode(nil, "null"), nil
-		case float64, float32:
-			// json decoder returns ints as float.
-			return parseSnippet(fmt.Sprintf("%v", rawData))
-		case int, int64, int32, string, bool:
-			return createScalarNode(rawData, fmt.Sprintf("%v", rawData)), nil
-		case []*orderedMap:
-			return dec.parseArray(rawData)
-		default:
-			return nil, fmt.Errorf("unrecognised type :( %v", rawData)
-		}
-	}
-
-	var yamlMap = &yaml.Node{Kind: yaml.MappingNode, Tag: "!!map"}
-	for _, keyValuePair := range data.kv {
-		yamlValue, err := dec.convertToYamlNode(&keyValuePair.V)
-		if err != nil {
-			return nil, err
-		}
-		yamlMap.Content = append(yamlMap.Content, createScalarNode(keyValuePair.K, keyValuePair.K), yamlValue)
-	}
-	return yamlMap, nil
-
-}
-
-func (dec *jsonDecoder) parseArray(dataArray []*orderedMap) (*yaml.Node, error) {
-
-	var yamlMap = &yaml.Node{Kind: yaml.SequenceNode, Tag: "!!seq"}
-
-	for _, value := range dataArray {
-		yamlValue, err := dec.convertToYamlNode(value)
-		if err != nil {
-			return nil, err
-		}
-		yamlMap.Content = append(yamlMap.Content, yamlValue)
-	}
-	return yamlMap, nil
+	return &dataBucket, nil
 }

pkg/yqlib/decoder_lua.go

@@ -0,0 +1,165 @@
+//go:build !yq_nolua
+
+package yqlib
+
+import (
+	"fmt"
+	"io"
+	"math"
+
+	lua "github.com/yuin/gopher-lua"
+)
+
+type luaDecoder struct {
+	reader   io.Reader
+	finished bool
+	prefs    LuaPreferences
+}
+
+func NewLuaDecoder(prefs LuaPreferences) Decoder {
+	return &luaDecoder{
+		prefs: prefs,
+	}
+}
+
+func (dec *luaDecoder) Init(reader io.Reader) error {
+	dec.reader = reader
+	return nil
+}
+
+func (dec *luaDecoder) convertToYamlNode(ls *lua.LState, lv lua.LValue) *CandidateNode {
+	switch lv.Type() {
+	case lua.LTNil:
+		return &CandidateNode{
+			Kind:  ScalarNode,
+			Tag:   "!!null",
+			Value: "",
+		}
+	case lua.LTBool:
+		return &CandidateNode{
+			Kind:  ScalarNode,
+			Tag:   "!!bool",
+			Value: lv.String(),
+		}
+	case lua.LTNumber:
+		n := float64(lua.LVAsNumber(lv))
+		// various special case floats
+		if math.IsNaN(n) {
+			return &CandidateNode{
+				Kind:  ScalarNode,
+				Tag:   "!!float",
+				Value: ".nan",
+			}
+		}
+		if math.IsInf(n, 1) {
+			return &CandidateNode{
+				Kind:  ScalarNode,
+				Tag:   "!!float",
+				Value: ".inf",
+			}
+		}
+		if math.IsInf(n, -1) {
+			return &CandidateNode{
+				Kind:  ScalarNode,
+				Tag:   "!!float",
+				Value: "-.inf",
+			}
+		}
+
+		// does it look like an integer?
+		if n == float64(int(n)) {
+			return &CandidateNode{
+				Kind:  ScalarNode,
+				Tag:   "!!int",
+				Value: lv.String(),
+			}
+		}
+
+		return &CandidateNode{
+			Kind:  ScalarNode,
+			Tag:   "!!float",
+			Value: lv.String(),
+		}
+	case lua.LTString:
+		return &CandidateNode{
+			Kind:  ScalarNode,
+			Tag:   "!!str",
+			Value: lv.String(),
+		}
+	case lua.LTFunction:
+		return &CandidateNode{
+			Kind:  ScalarNode,
+			Tag:   "tag:lua.org,2006,function",
+			Value: lv.String(),
+		}
+	case lua.LTTable:
+		// Simultaneously create a sequence and a map, pick which one to return
+		// based on whether all keys were consecutive integers
+		i := 1
+		yaml_sequence := &CandidateNode{
+			Kind: SequenceNode,
+			Tag:  "!!seq",
+		}
+		yaml_map := &CandidateNode{
+			Kind: MappingNode,
+			Tag:  "!!map",
+		}
+		t := lv.(*lua.LTable)
+		k, v := ls.Next(t, lua.LNil)
+		for k != lua.LNil {
+			if ki, ok := k.(lua.LNumber); i != 0 && ok && math.Mod(float64(ki), 1) == 0 && int(ki) == i {
+				i++
+			} else {
+				i = 0
+			}
+			newKey := dec.convertToYamlNode(ls, k)
+
+			yv := dec.convertToYamlNode(ls, v)
+			yaml_map.AddKeyValueChild(newKey, yv)
+
+			if i != 0 {
+				yaml_sequence.AddChild(yv)
+			}
+			k, v = ls.Next(t, k)
+		}
+		if i != 0 {
+			return yaml_sequence
+		}
+		return yaml_map
+	default:
+		return &CandidateNode{
+			Kind:        ScalarNode,
+			LineComment: fmt.Sprintf("Unhandled Lua type: %s", lv.Type().String()),
+			Tag:         "!!null",
+			Value:       lv.String(),
+		}
+	}
+}
+
+func (dec *luaDecoder) decideTopLevelNode(ls *lua.LState) *CandidateNode {
+	if ls.GetTop() == 0 {
+		// no items were explicitly returned, encode the globals table instead
+		return dec.convertToYamlNode(ls, ls.Get(lua.GlobalsIndex))
+	}
+	return dec.convertToYamlNode(ls, ls.Get(1))
+}
+
+func (dec *luaDecoder) Decode() (*CandidateNode, error) {
+	if dec.finished {
+		return nil, io.EOF
+	}
+	ls := lua.NewState(lua.Options{SkipOpenLibs: true})
+	defer ls.Close()
+	fn, err := ls.Load(dec.reader, "@input")
+	if err != nil {
+		return nil, err
+	}
+	ls.Push(fn)
+	err = ls.PCall(0, lua.MultRet, nil)
+	if err != nil {
+		return nil, err
+	}
+	firstNode := dec.decideTopLevelNode(ls)
+	dec.finished = true
+	return firstNode, nil
+}

pkg/yqlib/decoder_properties.go

@@ -1,145 +0,0 @@
-package yqlib
-
-import (
-	"bytes"
-	"fmt"
-	"io"
-	"strconv"
-	"strings"
-
-	"github.com/magiconair/properties"
-	"gopkg.in/yaml.v3"
-)
-
-type propertiesDecoder struct {
-	reader   io.Reader
-	finished bool
-	d        DataTreeNavigator
-}
-
-func NewPropertiesDecoder() Decoder {
-	return &propertiesDecoder{d: NewDataTreeNavigator(), finished: false}
-}
-
-func (dec *propertiesDecoder) Init(reader io.Reader) error {
-	dec.reader = reader
-	dec.finished = false
-	return nil
-}
-
-func parsePropKey(key string) []interface{} {
-	pathStrArray := strings.Split(key, ".")
-	path := make([]interface{}, len(pathStrArray))
-	for i, pathStr := range pathStrArray {
-		num, err := strconv.ParseInt(pathStr, 10, 32)
-		if err == nil {
-			path[i] = num
-		} else {
-			path[i] = pathStr
-		}
-	}
-	return path
-}
-
-func (dec *propertiesDecoder) processComment(c string) string {
-	if c == "" {
-		return ""
-	}
-	return "# " + c
-}
-
-func (dec *propertiesDecoder) applyPropertyComments(context Context, path []interface{}, comments []string) error {
-	assignmentOp := &Operation{OperationType: assignOpType, Preferences: assignPreferences{}}
-
-	rhsCandidateNode := &CandidateNode{
-		Path: path,
-		Node: &yaml.Node{
-			Tag:         "!!str",
-			Value:       fmt.Sprintf("%v", path[len(path)-1]),
-			HeadComment: dec.processComment(strings.Join(comments, "\n")),
-			Kind:        yaml.ScalarNode,
-		},
-	}
-
-	rhsCandidateNode.Node.Tag = guessTagFromCustomType(rhsCandidateNode.Node)
-
-	rhsOp := &Operation{OperationType: referenceOpType, CandidateNode: rhsCandidateNode}
-
-	assignmentOpNode := &ExpressionNode{
-		Operation: assignmentOp,
-		LHS:       createTraversalTree(path, traversePreferences{}, true),
-		RHS:       &ExpressionNode{Operation: rhsOp},
-	}
-
-	_, err := dec.d.GetMatchingNodes(context, assignmentOpNode)
-	return err
-}
-
-func (dec *propertiesDecoder) applyProperty(context Context, properties *properties.Properties, key string) error {
-	value, _ := properties.Get(key)
-	path := parsePropKey(key)
-
-	propertyComments := properties.GetComments(key)
-	if len(propertyComments) > 0 {
-		err := dec.applyPropertyComments(context, path, propertyComments)
-		if err != nil {
-			return nil
-		}
-	}
-
-	rhsNode := &yaml.Node{
-		Value: value,
-		Tag:   "!!str",
-		Kind:  yaml.ScalarNode,
-	}
-
-	rhsNode.Tag = guessTagFromCustomType(rhsNode)
-
-	return dec.d.DeeplyAssign(context, path, rhsNode)
-}
-
-func (dec *propertiesDecoder) Decode() (*CandidateNode, error) {
-	if dec.finished {
-		return nil, io.EOF
-	}
-	buf := new(bytes.Buffer)
-
-	if _, err := buf.ReadFrom(dec.reader); err != nil {
-		return nil, err
-	}
-	if buf.Len() == 0 {
-		dec.finished = true
-		return nil, io.EOF
-	}
-	properties, err := properties.LoadString(buf.String())
-	if err != nil {
-		return nil, err
-	}
-	properties.DisableExpansion = true
-
-	rootMap := &CandidateNode{
-		Node: &yaml.Node{
-			Kind: yaml.MappingNode,
-			Tag:  "!!map",
-		},
-	}
-
-	context := Context{}
-	context = context.SingleChildContext(rootMap)
-
-	for _, key := range properties.Keys() {
-		if err := dec.applyProperty(context, properties, key); err != nil {
-			return nil, err
-		}
-
-	}
-	dec.finished = true
-
-	return &CandidateNode{
-		Node: &yaml.Node{
-			Kind:    yaml.DocumentNode,
-			Content: []*yaml.Node{rootMap.Node},
-		},
-	}, nil
-
-}

pkg/yqlib/decoder_test.go

@@ -7,7 +7,7 @@ import (
 	"strings"
 )
 
-type formatScenario struct {
+type FormatScenario struct {
 	input          string
 	indent         int
 	expression     string
@@ -19,7 +19,7 @@ type formatScenario struct {
 	expectedError  string
 }
 
-func processFormatScenario(s formatScenario, decoder Decoder, encoder Encoder) (string, error) {
+func processFormatScenario(s FormatScenario, decoder Decoder, encoder Encoder) (string, error) {
 	var output bytes.Buffer
 	writer := bufio.NewWriter(&output)
 
@@ -27,11 +27,14 @@ func processFormatScenario(s formatScenario, decoder Decoder, encoder Encoder) (
 		decoder = NewYamlDecoder(ConfiguredYamlPreferences)
 	}
 
+	log.Debugf("reading docs")
 	inputs, err := readDocuments(strings.NewReader(s.input), "sample.yml", 0, decoder)
 	if err != nil {
 		return "", err
 	}
 
+	log.Debugf("done reading the documents")
+
 	expression := s.expression
 	if expression == "" {
 		expression = "."
@@ -45,6 +48,8 @@ func processFormatScenario(s formatScenario, decoder Decoder, encoder Encoder) (
 
 	context, err := NewDataTreeNavigator().GetMatchingNodes(Context{MatchingNodes: inputs}, exp)
 
+	log.Debugf("Going to print: %v", NodesToString(context.MatchingNodes))
+
 	if err != nil {
 		return "", err
 	}
@@ -59,7 +64,7 @@ func processFormatScenario(s formatScenario, decoder Decoder, encoder Encoder) (
 	return output.String(), nil
 }
 
-func mustProcessFormatScenario(s formatScenario, decoder Decoder, encoder Encoder) string {
+func mustProcessFormatScenario(s FormatScenario, decoder Decoder, encoder Encoder) string {
 
 	result, err := processFormatScenario(s, decoder, encoder)
 	if err != nil {

pkg/yqlib/decoder_toml.go

@@ -11,7 +11,6 @@ import (
 	"time"
 
 	toml "github.com/pelletier/go-toml/v2/unstable"
-	yaml "gopkg.in/yaml.v3"
 )
 
 type tomlDecoder struct {
@@ -37,10 +36,9 @@ func (dec *tomlDecoder) Init(reader io.Reader) error {
 	}
 	dec.parser.Reset(buf.Bytes())
 	dec.rootMap = &CandidateNode{
-		Node: &yaml.Node{
-			Kind: yaml.MappingNode,
-			Tag:  "!!map",
-		}}
+		Kind: MappingNode,
+		Tag:  "!!map",
+	}
 	return nil
 }
 
@@ -58,12 +56,13 @@ func (dec *tomlDecoder) getFullPath(tomlNode *toml.Node) []interface{} {
 func (dec *tomlDecoder) processKeyValueIntoMap(rootMap *CandidateNode, tomlNode *toml.Node) error {
 	value := tomlNode.Value()
 	path := dec.getFullPath(value.Next())
-	log.Debug("!!!processKeyValueIntoMap: %v", path)
+	log.Debug("processKeyValueIntoMap: %v", path)
 
 	valueNode, err := dec.decodeNode(value)
 	if err != nil {
 		return err
 	}
+
 	context := Context{}
 	context = context.SingleChildContext(rootMap)
 
@@ -71,14 +70,14 @@ func (dec *tomlDecoder) processKeyValueIntoMap(rootMap *CandidateNode, tomlNode
 }
 
 func (dec *tomlDecoder) decodeKeyValuesIntoMap(rootMap *CandidateNode, tomlNode *toml.Node) (bool, error) {
-	log.Debug("!! DECODE_KV_INTO_MAP -- processing first (current) entry")
+	log.Debug("decodeKeyValuesIntoMap -- processing first (current) entry")
 	if err := dec.processKeyValueIntoMap(rootMap, tomlNode); err != nil {
 		return false, err
 	}
 
 	for dec.parser.NextExpression() {
 		nextItem := dec.parser.Expression()
-		log.Debug("!! DECODE_KV_INTO_MAP -- next exp, its a %v", nextItem.Kind)
+		log.Debug("decodeKeyValuesIntoMap -- next exp, its a %v", nextItem.Kind)
 
 		if nextItem.Kind == toml.KeyValue {
 			if err := dec.processKeyValueIntoMap(rootMap, nextItem); err != nil {
@@ -86,18 +85,17 @@ func (dec *tomlDecoder) decodeKeyValuesIntoMap(rootMap *CandidateNode, tomlNode
 			}
 		} else {
 			// run out of key values
-			log.Debug("! DECODE_KV_INTO_MAP - ok we are done in decodeKeyValuesIntoMap, gota a %v", nextItem.Kind)
-			log.Debug("! DECODE_KV_INTO_MAP - processAgainstCurrentExp = true!")
+			log.Debug("done in decodeKeyValuesIntoMap, gota a %v", nextItem.Kind)
 			return true, nil
 		}
 	}
-	log.Debug("! DECODE_KV_INTO_MAP - no more things to read in")
+	log.Debug("no more things to read in")
 	return false, nil
 }
 
-func (dec *tomlDecoder) createInlineTableMap(tomlNode *toml.Node) (*yaml.Node, error) {
-	content := make([]*yaml.Node, 0)
-	log.Debug("!! createInlineTableMap")
+func (dec *tomlDecoder) createInlineTableMap(tomlNode *toml.Node) (*CandidateNode, error) {
+	content := make([]*CandidateNode, 0)
+	log.Debug("createInlineTableMap")
 
 	iterator := tomlNode.Children()
 	for iterator.Next() {
@@ -107,28 +105,26 @@ func (dec *tomlDecoder) createInlineTableMap(tomlNode *toml.Node) (*yaml.Node, e
 		}
 
 		keyValues := &CandidateNode{
-			Node: &yaml.Node{
-				Kind: yaml.MappingNode,
-				Tag:  "!!map",
-			},
+			Kind: MappingNode,
+			Tag:  "!!map",
 		}
 
 		if err := dec.processKeyValueIntoMap(keyValues, child); err != nil {
 			return nil, err
 		}
 
-		content = append(content, keyValues.Node.Content...)
+		content = append(content, keyValues.Content...)
 	}
 
-	return &yaml.Node{
-		Kind:    yaml.MappingNode,
+	return &CandidateNode{
+		Kind:    MappingNode,
 		Tag:     "!!map",
 		Content: content,
 	}, nil
 }
 
-func (dec *tomlDecoder) createArray(tomlNode *toml.Node) (*yaml.Node, error) {
-	content := make([]*yaml.Node, 0)
+func (dec *tomlDecoder) createArray(tomlNode *toml.Node) (*CandidateNode, error) {
+	content := make([]*CandidateNode, 0)
 	iterator := tomlNode.Children()
 	for iterator.Next() {
 		child := iterator.Node()
@@ -139,43 +135,43 @@ func (dec *tomlDecoder) createArray(tomlNode *toml.Node) (*yaml.Node, error) {
 		content = append(content, yamlNode)
 	}
 
-	return &yaml.Node{
-		Kind:    yaml.SequenceNode,
+	return &CandidateNode{
+		Kind:    SequenceNode,
 		Tag:     "!!seq",
 		Content: content,
 	}, nil
 
 }
 
-func (dec *tomlDecoder) createStringScalar(tomlNode *toml.Node) (*yaml.Node, error) {
+func (dec *tomlDecoder) createStringScalar(tomlNode *toml.Node) (*CandidateNode, error) {
 	content := string(tomlNode.Data)
-	return createScalarNode(content, content), nil
+	return CreateScalarNode(content, content), nil
 }
 
-func (dec *tomlDecoder) createBoolScalar(tomlNode *toml.Node) (*yaml.Node, error) {
+func (dec *tomlDecoder) createBoolScalar(tomlNode *toml.Node) (*CandidateNode, error) {
 	content := string(tomlNode.Data)
-	return createScalarNode(content == "true", content), nil
+	return CreateScalarNode(content == "true", content), nil
 }
 
-func (dec *tomlDecoder) createIntegerScalar(tomlNode *toml.Node) (*yaml.Node, error) {
+func (dec *tomlDecoder) createIntegerScalar(tomlNode *toml.Node) (*CandidateNode, error) {
 	content := string(tomlNode.Data)
 	_, num, err := parseInt64(content)
-	return createScalarNode(num, content), err
+	return CreateScalarNode(num, content), err
 }
 
-func (dec *tomlDecoder) createDateTimeScalar(tomlNode *toml.Node) (*yaml.Node, error) {
+func (dec *tomlDecoder) createDateTimeScalar(tomlNode *toml.Node) (*CandidateNode, error) {
 	content := string(tomlNode.Data)
 	val, err := parseDateTime(time.RFC3339, content)
-	return createScalarNode(val, content), err
+	return CreateScalarNode(val, content), err
 }
 
-func (dec *tomlDecoder) createFloatScalar(tomlNode *toml.Node) (*yaml.Node, error) {
+func (dec *tomlDecoder) createFloatScalar(tomlNode *toml.Node) (*CandidateNode, error) {
 	content := string(tomlNode.Data)
 	num, err := strconv.ParseFloat(content, 64)
-	return createScalarNode(num, content), err
+	return CreateScalarNode(num, content), err
 }
 
-func (dec *tomlDecoder) decodeNode(tomlNode *toml.Node) (*yaml.Node, error) {
+func (dec *tomlDecoder) decodeNode(tomlNode *toml.Node) (*CandidateNode, error) {
 	switch tomlNode.Kind {
 	case toml.Key, toml.String:
 		return dec.createStringScalar(tomlNode)
@@ -241,81 +237,75 @@ func (dec *tomlDecoder) Decode() (*CandidateNode, error) {
 	// must have finished
 	dec.finished = true
 
-	if len(dec.rootMap.Node.Content) == 0 {
+	if len(dec.rootMap.Content) == 0 {
 		return nil, io.EOF
 	}
 
-	return &CandidateNode{
-		Node: &yaml.Node{
-			Kind:    yaml.DocumentNode,
-			Content: []*yaml.Node{dec.rootMap.Node},
-		},
-	}, deferredError
+	return dec.rootMap, deferredError
 
 }
 
 func (dec *tomlDecoder) processTopLevelNode(currentNode *toml.Node) (bool, error) {
 	var runAgainstCurrentExp bool
 	var err error
-	log.Debug("!!!!!!!!!!!!Going to process %v state is current %v", currentNode.Kind, NodeToString(dec.rootMap))
+	log.Debug("processTopLevelNode: Going to process %v state is current %v", currentNode.Kind, NodeToString(dec.rootMap))
 	if currentNode.Kind == toml.Table {
-		runAgainstCurrentExp, err = dec.processTable((currentNode))
+		runAgainstCurrentExp, err = dec.processTable(currentNode)
 	} else if currentNode.Kind == toml.ArrayTable {
-		runAgainstCurrentExp, err = dec.processArrayTable((currentNode))
+		runAgainstCurrentExp, err = dec.processArrayTable(currentNode)
 	} else {
 		runAgainstCurrentExp, err = dec.decodeKeyValuesIntoMap(dec.rootMap, currentNode)
 	}
 
-	log.Debug("!!!!!!!!!!!!DONE Processing state is now %v", NodeToString(dec.rootMap))
+	log.Debug("processTopLevelNode: DONE Processing state is now %v", NodeToString(dec.rootMap))
 	return runAgainstCurrentExp, err
 }
 
 func (dec *tomlDecoder) processTable(currentNode *toml.Node) (bool, error) {
-	log.Debug("!!! processing table")
+	log.Debug("Enter processTable")
 	fullPath := dec.getFullPath(currentNode.Child())
-	log.Debug("!!!fullpath: %v", fullPath)
-
-	hasValue := dec.parser.NextExpression()
-	if !hasValue {
-		return false, fmt.Errorf("error retrieving table %v value: %w", fullPath, dec.parser.Error())
-	}
+	log.Debug("fullpath: %v", fullPath)
 
 	tableNodeValue := &CandidateNode{
-		Node: &yaml.Node{
-			Kind: yaml.MappingNode,
-			Tag:  "!!map",
-		},
+		Kind:    MappingNode,
+		Tag:     "!!map",
+		Content: make([]*CandidateNode, 0),
 	}
 
-	tableValue := dec.parser.Expression()
-	if tableValue.Kind != toml.KeyValue {
-		log.Debug("got an empty table, returning")
-		return true, nil
+	var tableValue *toml.Node
+	runAgainstCurrentExp := false
+	var err error
+	hasValue := dec.parser.NextExpression()
+	// check to see if there is any table data
+	if hasValue {
+		tableValue = dec.parser.Expression()
+		// next expression is not table data, so we are done
+		if tableValue.Kind != toml.KeyValue {
+			log.Debug("got an empty table, returning")
+			return true, nil
+		}
+
+		runAgainstCurrentExp, err = dec.decodeKeyValuesIntoMap(tableNodeValue, tableValue)
+		if err != nil && !errors.Is(io.EOF, err) {
+			return false, err
+		}
 	}
 
-	runAgainstCurrentExp, err := dec.decodeKeyValuesIntoMap(tableNodeValue, tableValue)
-	log.Debugf("table node err: %w", err)
-	if err != nil && !errors.Is(io.EOF, err) {
-		return false, err
-	}
 	c := Context{}
-
 	c = c.SingleChildContext(dec.rootMap)
-	err = dec.d.DeeplyAssign(c, fullPath, tableNodeValue.Node)
+	err = dec.d.DeeplyAssign(c, fullPath, tableNodeValue)
 	if err != nil {
 		return false, err
 	}
 	return runAgainstCurrentExp, nil
 }
 
-func (dec *tomlDecoder) arrayAppend(context Context, path []interface{}, rhsNode *yaml.Node) error {
+func (dec *tomlDecoder) arrayAppend(context Context, path []interface{}, rhsNode *CandidateNode) error {
+	log.Debug("arrayAppend to path: %v,%v", path, NodeToString(rhsNode))
 	rhsCandidateNode := &CandidateNode{
-		Path: path,
-		Node: &yaml.Node{
-			Kind:    yaml.SequenceNode,
-			Tag:     "!!seq",
-			Content: []*yaml.Node{rhsNode},
-		},
+		Kind:    SequenceNode,
+		Tag:     "!!seq",
+		Content: []*CandidateNode{rhsNode},
 	}
 
 	assignmentOp := &Operation{OperationType: addAssignOpType}
@@ -333,9 +323,9 @@ func (dec *tomlDecoder) arrayAppend(context Context, path []interface{}, rhsNode
 }
 
 func (dec *tomlDecoder) processArrayTable(currentNode *toml.Node) (bool, error) {
-	log.Debug("!!! processing table")
+	log.Debug("Entering processArrayTable")
 	fullPath := dec.getFullPath(currentNode.Child())
-	log.Debug("!!!fullpath: %v", fullPath)
+	log.Debug("Fullpath: %v", fullPath)
 
 	// need to use the array append exp to add another entry to
 	// this array: fullpath += [ thing ]
@@ -346,10 +336,8 @@ func (dec *tomlDecoder) processArrayTable(currentNode *toml.Node) (bool, error)
 	}
 
 	tableNodeValue := &CandidateNode{
-		Node: &yaml.Node{
-			Kind: yaml.MappingNode,
-			Tag:  "!!map",
-		},
+		Kind: MappingNode,
+		Tag:  "!!map",
 	}
 
 	tableValue := dec.parser.Expression()
@@ -363,7 +351,7 @@ func (dec *tomlDecoder) processArrayTable(currentNode *toml.Node) (bool, error)
 	c = c.SingleChildContext(dec.rootMap)
 
 	// += function
-	err = dec.arrayAppend(c, fullPath, tableNodeValue.Node)
+	err = dec.arrayAppend(c, fullPath, tableNodeValue)
 
 	return runAgainstCurrentExp, err
 }

pkg/yqlib/decoder_uri.go

@@ -4,8 +4,6 @@ import (
 	"bytes"
 	"io"
 	"net/url"
-
-	yaml "gopkg.in/yaml.v3"
 )
 
 type uriDecoder struct {
@@ -50,11 +48,5 @@ func (dec *uriDecoder) Decode() (*CandidateNode, error) {
 		return nil, err
 	}
 	dec.readAnything = true
-	return &CandidateNode{
-		Node: &yaml.Node{
-			Kind:  yaml.ScalarNode,
-			Tag:   "!!str",
-			Value: newValue,
-		},
-	}, nil
+	return CreateStringScalarNode(newValue), nil
 }

pkg/yqlib/decoder_yaml.go

@@ -20,8 +20,12 @@ type yamlDecoder struct {
 	leadingContent string
 	bufferRead     bytes.Buffer
 
-	readAnything bool
-	firstFile    bool
+	// anchor map persists over multiple documents for convenience.
+	anchorMap map[string]*CandidateNode
+
+	readAnything  bool
+	firstFile     bool
+	documentIndex uint
 }
 
 func NewYamlDecoder(prefs YamlPreferences) Decoder {
@@ -33,7 +37,7 @@ func (dec *yamlDecoder) processReadStream(reader *bufio.Reader) (io.Reader, stri
 	var yamlDirectiveLineRegEx = regexp.MustCompile(`^\s*%YA`)
 	var sb strings.Builder
 	for {
-		peekBytes, err := reader.Peek(3)
+		peekBytes, err := reader.Peek(4)
 		if errors.Is(err, io.EOF) {
 			// EOF are handled else where..
 			return reader, sb.String(), nil
@@ -47,9 +51,17 @@ func (dec *yamlDecoder) processReadStream(reader *bufio.Reader) (io.Reader, stri
 			} else if err != nil {
 				return reader, sb.String(), err
 			}
-		} else if string(peekBytes) == "---" {
+		} else if string(peekBytes) == "--- " {
+			_, err := reader.ReadString(' ')
+			sb.WriteString("$yqDocSeparator$\n")
+			if errors.Is(err, io.EOF) {
+				return reader, sb.String(), nil
+			} else if err != nil {
+				return reader, sb.String(), err
+			}
+		} else if string(peekBytes) == "---\n" {
 			_, err := reader.ReadString('\n')
-			sb.WriteString("$yqDocSeperator$\n")
+			sb.WriteString("$yqDocSeparator$\n")
 			if errors.Is(err, io.EOF) {
 				return reader, sb.String(), nil
 			} else if err != nil {
@@ -93,12 +105,15 @@ func (dec *yamlDecoder) Init(reader io.Reader) error {
 	dec.readAnything = false
 	dec.decoder = *yaml.NewDecoder(readerToUse)
 	dec.firstFile = false
+	dec.documentIndex = 0
+	dec.anchorMap = make(map[string]*CandidateNode)
 	return nil
 }
 
 func (dec *yamlDecoder) Decode() (*CandidateNode, error) {
-	var dataBucket yaml.Node
-	err := dec.decoder.Decode(&dataBucket)
+	var yamlNode yaml.Node
+	err := dec.decoder.Decode(&yamlNode)
+
 	if errors.Is(err, io.EOF) && dec.leadingContent != "" && !dec.readAnything {
 		// force returning an empty node with a comment.
 		dec.readAnything = true
@@ -116,28 +131,27 @@ func (dec *yamlDecoder) Decode() (*CandidateNode, error) {
 		return nil, err
 	}
 
-	candidateNode := &CandidateNode{
-		Node: &dataBucket,
+	candidateNode := CandidateNode{document: dec.documentIndex}
+	// don't bother with the DocumentNode
+	err = candidateNode.UnmarshalYAML(yamlNode.Content[0], dec.anchorMap)
+	if err != nil {
+		return nil, err
 	}
 
+	candidateNode.HeadComment = yamlNode.HeadComment + candidateNode.HeadComment
+	candidateNode.FootComment = yamlNode.FootComment + candidateNode.FootComment
+
 	if dec.leadingContent != "" {
 		candidateNode.LeadingContent = dec.leadingContent
 		dec.leadingContent = ""
 	}
 	dec.readAnything = true
-	// move document comments into candidate node
-	// otherwise unwrap drops them.
-	candidateNode.TrailingContent = dataBucket.FootComment
-	dataBucket.FootComment = ""
-	return candidateNode, nil
+	dec.documentIndex++
+	return &candidateNode, nil
 }
 
 func (dec *yamlDecoder) blankNodeWithComment() *CandidateNode {
-	return &CandidateNode{
-		Document:       0,
-		Filename:       "",
-		Node:           &yaml.Node{Kind: yaml.DocumentNode, Content: []*yaml.Node{{Tag: "!!null", Kind: yaml.ScalarNode}}},
-		FileIndex:      0,
-		LeadingContent: dec.leadingContent,
-	}
+	node := CreateScalarNode(nil, "")
+	node.LeadingContent = dec.leadingContent
+	return node
 }

pkg/yqlib/doc/operators/boolean-operators.md

@@ -4,7 +4,7 @@ The `or` and `and` operators take two parameters and return a boolean result.
 
 `not` flips a boolean from true to false, or vice versa. 
 
-`any` will return `true` if there are any `true` values in a array sequence, and `all` will return true if _all_ elements in an array are true.
+`any` will return `true` if there are any `true` values in an array sequence, and `all` will return true if _all_ elements in an array are true.
 
 `any_c(condition)` and `all_c(condition)` are like `any` and `all` but they take a condition expression that is used against each element to determine if it's `true`. Note: in `jq` you can simply pass a condition to `any` or `all` and it simply works - `yq` isn't that clever..yet
 

pkg/yqlib/doc/operators/column.md

@@ -2,6 +2,8 @@
 
 Returns the column of the matching node. Starts from 1, 0 indicates there was no column data.
 
+Column is the number of characters that precede that node on the line it starts.
+
 ## Returns column of _value_ node
 Given a sample.yml file of:
 ```yaml

pkg/yqlib/doc/operators/comment-operators.md

@@ -187,7 +187,6 @@ yq '. head_comment="single"' sample.yml
 will output
 ```yaml
 # single
-
 a: cat
 ```
 

pkg/yqlib/doc/operators/create-collect-into-object.md

@@ -66,6 +66,7 @@ will output
 ```yaml
 Mike: cat
 Mike: dog
+---
 Rosey: monkey
 Rosey: sheep
 ```

pkg/yqlib/doc/operators/document-index.md

@@ -85,6 +85,7 @@ will output
 ```yaml
 match: cat
 doc: 0
+---
 match: frog
 doc: 1
 ```

pkg/yqlib/doc/operators/file-operators.md

@@ -54,7 +54,6 @@ yq eval-all 'file_index' sample.yml another.yml
 will output
 ```yaml
 0
----
 1
 ```
 

pkg/yqlib/doc/operators/group-by.md

@@ -26,7 +26,7 @@ will output
     bar: 100
 ```
 
-## Group by field, with nuls
+## Group by field, with nulls
 Given a sample.yml file of:
 ```yaml
 - cat: dog

pkg/yqlib/doc/operators/headers/Main.md

@@ -1,5 +1,5 @@
 # NAME
-  *yq* is a portable command-line YAML processor
+  *yq* is a portable command-line data file processor
 
 # SYNOPSIS 
 
@@ -11,7 +11,7 @@ eval-all/ea - Loads all yaml documents of all yaml files and runs expression onc
 
 # DESCRIPTION
 
-a lightweight and portable command-line YAML processor. `yq` uses [jq](https://github.com/stedolan/jq) like syntax but works with yaml files as well as json. It doesn't yet support everything `jq` does - but it does support the most common operations and functions, and more is being added continuously.
+a lightweight and portable command-line data file processor. `yq` uses [jq](https://github.com/stedolan/jq) like syntax but works with yaml, json, xml, csv, properties and TOML files. It doesn't yet support everything `jq` does - but it does support the most common operations and functions, and more is being added continuously.
 
 This documentation is also available at https://mikefarah.gitbook.io/yq/
 # QUICK GUIDE 
@@ -26,7 +26,7 @@ yq '.a.b[0].c' file.yaml
 cat file.yaml | yq '.a.b[0].c'
 ```
 
-## Update a yaml file, inplace
+## Update a yaml file, in place
 ```bash
 yq -i '.a.b[0].c = "cool"' file.yaml
 ```

pkg/yqlib/doc/operators/headers/boolean-operators.md

@@ -4,7 +4,7 @@ The `or` and `and` operators take two parameters and return a boolean result.
 
 `not` flips a boolean from true to false, or vice versa. 
 
-`any` will return `true` if there are any `true` values in a array sequence, and `all` will return true if _all_ elements in an array are true.
+`any` will return `true` if there are any `true` values in an array sequence, and `all` will return true if _all_ elements in an array are true.
 
 `any_c(condition)` and `all_c(condition)` are like `any` and `all` but they take a condition expression that is used against each element to determine if it's `true`. Note: in `jq` you can simply pass a condition to `any` or `all` and it simply works - `yq` isn't that clever..yet
 

pkg/yqlib/doc/operators/headers/column.md

@@ -1,3 +1,5 @@
 # Column
 
 Returns the column of the matching node. Starts from 1, 0 indicates there was no column data.
+
+Column is the number of characters that precede that node on the line it starts.

pkg/yqlib/doc/operators/headers/multiply-merge.md

@@ -3,7 +3,9 @@
 Like the multiple operator in jq, depending on the operands, this multiply operator will do different things. Currently numbers, arrays and objects are supported.
 
 ## Objects and arrays - merging
-Objects are merged deeply matching on matching keys. By default, array values override and are not deeply merged.
+Objects are merged _deeply_ matching on matching keys. By default, array values override and are not deeply merged.
+
+You can use the add operator `+`, to shallow merge objects, see more info [here](https://mikefarah.gitbook.io/yq/operators/add).
 
 Note that when merging objects, this operator returns the merged object (not the parent). This will be clearer in the examples below.
 
@@ -16,6 +18,7 @@ You can control how objects are merged by using one or more of the following fla
 - `n` only merge _new_ fields
 - `c` clobber custom tags
 
+To perform a shallow merge only, use the add operator `+`, see more info [here](https://mikefarah.gitbook.io/yq/operators/add).
 
 ### Merge two files together
 This uses the load operator to merge file2 into file1.

pkg/yqlib/doc/operators/headers/style.md

@@ -1,3 +1,5 @@
 # Style
 
-The style operator can be used to get or set the style of nodes (e.g. string style, yaml style)
+The style operator can be used to get or set the style of nodes (e.g. string style, yaml style).
+Use this to control the formatting of the document in yaml.
+

pkg/yqlib/doc/operators/headers/to_number.md

@@ -0,0 +1,2 @@
+# To Number
+Parses the input as a number. yq will try to parse values as an int first, failing that it will try float. Values that already ints or floats will be left alone.

pkg/yqlib/doc/operators/kind.md

@@ -0,0 +1,77 @@
+
+## Get kind
+Given a sample.yml file of:
+```yaml
+a: cat
+b: 5
+c: 3.2
+e: true
+f: []
+g: {}
+h: null
+```
+then
+```bash
+yq '.. | kind' sample.yml
+```
+will output
+```yaml
+map
+scalar
+scalar
+scalar
+scalar
+seq
+map
+scalar
+```
+
+## Get kind, ignores custom tags
+Unlike tag, kind is not affected by custom tags.
+
+Given a sample.yml file of:
+```yaml
+a: !!thing cat
+b: !!foo {}
+c: !!bar []
+```
+then
+```bash
+yq '.. | kind' sample.yml
+```
+will output
+```yaml
+map
+scalar
+map
+seq
+```
+
+## Add comments only to scalars
+An example of how you can use kind
+
+Given a sample.yml file of:
+```yaml
+a:
+  b: 5
+  c: 3.2
+e: true
+f: []
+g: {}
+h: null
+```
+then
+```bash
+yq '(.. | select(kind == "scalar")) line_comment = "this is a scalar"' sample.yml
+```
+will output
+```yaml
+a:
+  b: 5 # this is a scalar
+  c: 3.2 # this is a scalar
+e: true # this is a scalar
+f: []
+g: {}
+h: null # this is a scalar
+```
+

pkg/yqlib/doc/operators/multiply-merge.md

@@ -3,7 +3,9 @@
 Like the multiple operator in jq, depending on the operands, this multiply operator will do different things. Currently numbers, arrays and objects are supported.
 
 ## Objects and arrays - merging
-Objects are merged deeply matching on matching keys. By default, array values override and are not deeply merged.
+Objects are merged _deeply_ matching on matching keys. By default, array values override and are not deeply merged.
+
+You can use the add operator `+`, to shallow merge objects, see more info [here](https://mikefarah.gitbook.io/yq/operators/add).
 
 Note that when merging objects, this operator returns the merged object (not the parent). This will be clearer in the examples below.
 
@@ -16,6 +18,7 @@ You can control how objects are merged by using one or more of the following fla
 - `n` only merge _new_ fields
 - `c` clobber custom tags
 
+To perform a shallow merge only, use the add operator `+`, see more info [here](https://mikefarah.gitbook.io/yq/operators/add).
 
 ### Merge two files together
 This uses the load operator to merge file2 into file1.
@@ -519,7 +522,7 @@ will output
 some: thing
 ```
 
-## Merging an null with an array
+## Merging a null with an array
 Running
 ```bash
 yq --null-input 'null * ["some"]'

pkg/yqlib/doc/operators/parent.md

@@ -37,6 +37,43 @@ fruit: banana
 name: sam
 ```
 
+## N-th parent
+You can optionally supply the number of levels to go up for the parent, the default being 1.
+
+Given a sample.yml file of:
+```yaml
+a:
+  b:
+    c: cat
+```
+then
+```bash
+yq '.a.b.c | parent(2)' sample.yml
+```
+will output
+```yaml
+b:
+  c: cat
+```
+
+## N-th parent - another level
+Given a sample.yml file of:
+```yaml
+a:
+  b:
+    c: cat
+```
+then
+```bash
+yq '.a.b.c | parent(3)' sample.yml
+```
+will output
+```yaml
+a:
+  b:
+    c: cat
+```
+
 ## No parent
 Given a sample.yml file of:
 ```yaml

pkg/yqlib/doc/operators/pick.md

@@ -13,7 +13,7 @@ myMap:
   cat: meow
   dog: bark
   thing: hamster
-  hamster: squeek
+  hamster: squeak
 ```
 then
 ```bash
@@ -22,7 +22,7 @@ yq '.myMap |= pick(["hamster", "cat", "goat"])' sample.yml
 will output
 ```yaml
 myMap:
-  hamster: squeek
+  hamster: squeak
   cat: meow
 ```
 

pkg/yqlib/doc/operators/string-operators.md

@@ -56,6 +56,40 @@ IFS= read -rd '' output < <(cat my_file)
 output=$output ./yq '.data.values = strenv(output)' first.yml
 ```
 
+## Interpolation
+Given a sample.yml file of:
+```yaml
+value: things
+another: stuff
+```
+then
+```bash
+yq '.message = "I like \(.value) and \(.another)"' sample.yml
+```
+will output
+```yaml
+value: things
+another: stuff
+message: I like things and stuff
+```
+
+## Interpolation - not a string
+Given a sample.yml file of:
+```yaml
+value:
+  an: apple
+```
+then
+```bash
+yq '.message = "I like \(.value)"' sample.yml
+```
+will output
+```yaml
+value:
+  an: apple
+message: 'I like an: apple'
+```
+
 ## To up (upper) case
 Works with unicode characters
 
@@ -383,3 +417,32 @@ will output
 - word
 ```
 
+## To string
+Note that you may want to force `yq` to leave scalar values wrapped by passing in `--unwrapScalar=false` or `-r=f`
+
+Given a sample.yml file of:
+```yaml
+- 1
+- true
+- null
+- ~
+- cat
+- an: object
+- - array
+  - 2
+```
+then
+```bash
+yq '.[] |= to_string' sample.yml
+```
+will output
+```yaml
+- "1"
+- "true"
+- "null"
+- "~"
+- cat
+- "an: object"
+- "- array\n- 2"
+```
+

pkg/yqlib/doc/operators/style.md

@@ -1,6 +1,8 @@
 # Style
 
-The style operator can be used to get or set the style of nodes (e.g. string style, yaml style)
+The style operator can be used to get or set the style of nodes (e.g. string style, yaml style).
+Use this to control the formatting of the document in yaml.
+
 
 ## Update and set style of a particular node (simple)
 Given a sample.yml file of:
@@ -45,6 +47,12 @@ a: cat
 b: 5
 c: 3.2
 e: true
+f:
+  - 1
+  - 2
+  - 3
+g:
+  something: cool
 ```
 then
 ```bash
@@ -57,6 +65,12 @@ a: !!str cat
 b: !!int 5
 c: !!float 3.2
 e: !!bool true
+f: !!seq
+  - !!int 1
+  - !!int 2
+  - !!int 3
+g: !!map
+  something: !!str cool
 ```
 
 ## Set double quote style
@@ -66,6 +80,12 @@ a: cat
 b: 5
 c: 3.2
 e: true
+f:
+  - 1
+  - 2
+  - 3
+g:
+  something: cool
 ```
 then
 ```bash
@@ -77,6 +97,12 @@ a: "cat"
 b: "5"
 c: "3.2"
 e: "true"
+f:
+  - "1"
+  - "2"
+  - "3"
+g:
+  something: "cool"
 ```
 
 ## Set double quote style on map keys too
@@ -86,6 +112,12 @@ a: cat
 b: 5
 c: 3.2
 e: true
+f:
+  - 1
+  - 2
+  - 3
+g:
+  something: cool
 ```
 then
 ```bash
@@ -97,6 +129,12 @@ will output
 "b": "5"
 "c": "3.2"
 "e": "true"
+"f":
+  - "1"
+  - "2"
+  - "3"
+"g":
+  "something": "cool"
 ```
 
 ## Set single quote style
@@ -106,6 +144,12 @@ a: cat
 b: 5
 c: 3.2
 e: true
+f:
+  - 1
+  - 2
+  - 3
+g:
+  something: cool
 ```
 then
 ```bash
@@ -117,6 +161,12 @@ a: 'cat'
 b: '5'
 c: '3.2'
 e: 'true'
+f:
+  - '1'
+  - '2'
+  - '3'
+g:
+  something: 'cool'
 ```
 
 ## Set literal quote style
@@ -126,6 +176,12 @@ a: cat
 b: 5
 c: 3.2
 e: true
+f:
+  - 1
+  - 2
+  - 3
+g:
+  something: cool
 ```
 then
 ```bash
@@ -141,6 +197,16 @@ c: |-
   3.2
 e: |-
   true
+f:
+  - |-
+    1
+  - |-
+    2
+  - |-
+    3
+g:
+  something: |-
+    cool
 ```
 
 ## Set folded quote style
@@ -150,6 +216,12 @@ a: cat
 b: 5
 c: 3.2
 e: true
+f:
+  - 1
+  - 2
+  - 3
+g:
+  something: cool
 ```
 then
 ```bash
@@ -165,6 +237,16 @@ c: >-
   3.2
 e: >-
   true
+f:
+  - >-
+    1
+  - >-
+    2
+  - >-
+    3
+g:
+  something: >-
+    cool
 ```
 
 ## Set flow quote style
@@ -174,6 +256,12 @@ a: cat
 b: 5
 c: 3.2
 e: true
+f:
+  - 1
+  - 2
+  - 3
+g:
+  something: cool
 ```
 then
 ```bash
@@ -181,7 +269,7 @@ yq '.. style="flow"' sample.yml
 ```
 will output
 ```yaml
-{a: cat, b: 5, c: 3.2, e: true}
+{a: cat, b: 5, c: 3.2, e: true, f: [1, 2, 3], g: {something: cool}}
 ```
 
 ## Reset style - or pretty print
@@ -189,10 +277,7 @@ Set empty (default) quote style, note the usage of `...` to match keys too. Note
 
 Given a sample.yml file of:
 ```yaml
-a: cat
-"b": 5
-'c': 3.2
-"e": true
+{a: cat, "b": 5, 'c': 3.2, "e": true,  f: [1,2,3], "g": { something: "cool"} }
 ```
 then
 ```bash
@@ -204,6 +289,12 @@ a: cat
 b: 5
 c: 3.2
 e: true
+f:
+  - 1
+  - 2
+  - 3
+g:
+  something: cool
 ```
 
 ## Set style relatively with assign-update

pkg/yqlib/doc/operators/to_number.md

@@ -0,0 +1,49 @@
+# To Number
+Parses the input as a number. yq will try to parse values as an int first, failing that it will try float. Values that already ints or floats will be left alone.
+
+## Converts strings to numbers
+Given a sample.yml file of:
+```yaml
+- "3"
+- "3.1"
+- "-1e3"
+```
+then
+```bash
+yq '.[] | to_number' sample.yml
+```
+will output
+```yaml
+3
+3.1
+-1e3
+```
+
+## Doesn't change numbers
+Given a sample.yml file of:
+```yaml
+- 3
+- 3.1
+- -1e3
+```
+then
+```bash
+yq '.[] | to_number' sample.yml
+```
+will output
+```yaml
+3
+3.1
+-1e3
+```
+
+## Cannot convert null
+Running
+```bash
+yq --null-input '.a.b | to_number'
+```
+will output
+```bash
+Error: cannot convert node value [null] at path a.b of tag !!null to number
+```
+

pkg/yqlib/doc/usage/convert.md

@@ -14,7 +14,7 @@ Given a sample.json file of:
 ```
 then
 ```bash
-yq -P '.' sample.json
+yq -p=json sample.json
 ```
 will output
 ```yaml
@@ -30,7 +30,7 @@ Given a sample.json file of:
 ```
 then
 ```bash
-yq -P '.' sample.json
+yq -p=json sample.json
 ```
 will output
 ```yaml
@@ -125,9 +125,7 @@ will output
 {"whatever":"cat"}
 ```
 
-## Roundtrip NDJSON
-Unfortunately the json encoder strips leading spaces of values.
-
+## Roundtrip JSON Lines / NDJSON
 Given a sample.json file of:
 ```json
 {"this": "is a multidoc json file"}
@@ -147,7 +145,7 @@ will output
 ```
 
 ## Roundtrip multi-document JSON
-The NDJSON parser can also handle multiple multi-line json documents in a single file!
+The parser can also handle multiple multi-line json documents in a single file (despite this not being in the JSON Lines / NDJSON spec). Typically you would have one entire JSON document per line, but the parser also supports multiple multi-line json documents
 
 Given a sample.json file of:
 ```json
@@ -229,7 +227,7 @@ will output
 {"a number":4}
 ```
 
-## Decode NDJSON
+## Decode JSON Lines / NDJSON
 Given a sample.json file of:
 ```json
 {"this": "is a multidoc json file"}
@@ -246,7 +244,7 @@ will output
 this: is a multidoc json file
 ---
 each:
-    - line is a valid json document
+  - line is a valid json document
 ---
 a number: 4
 ```

pkg/yqlib/doc/usage/csv-tsv.md

@@ -2,7 +2,7 @@
 Encode/Decode/Roundtrip CSV and TSV files.
 
 ## Encode 
-Currently supports arrays of homogenous flat objects, that is: no nesting and it assumes the _first_ object has all the keys required:
+Currently supports arrays of homogeneous flat objects, that is: no nesting and it assumes the _first_ object has all the keys required:
 
 ```yaml
 - name: Bobo
@@ -136,13 +136,13 @@ Samantha's Rabbit,,-188.8
 ```
 
 ## Parse CSV into an array of objects
-First row is assumed to be the header row.
+First row is assumed to be the header row. By default, entries with YAML/JSON formatting will be parsed!
 
 Given a sample.csv file of:
 ```csv
-name,numberOfCats,likesApples,height
-Gary,1,true,168.8
-Samantha's Rabbit,2,false,-188.8
+name,numberOfCats,likesApples,height,facts
+Gary,1,true,168.8,cool: true
+Samantha's Rabbit,2,false,-188.8,tall: indeed
 
 ```
 then
@@ -155,10 +155,42 @@ will output
   numberOfCats: 1
   likesApples: true
   height: 168.8
+  facts:
+    cool: true
 - name: Samantha's Rabbit
   numberOfCats: 2
   likesApples: false
   height: -188.8
+  facts:
+    tall: indeed
+```
+
+## Parse CSV into an array of objects, no auto-parsing
+First row is assumed to be the header row. Entries with YAML/JSON will be left as strings.
+
+Given a sample.csv file of:
+```csv
+name,numberOfCats,likesApples,height,facts
+Gary,1,true,168.8,cool: true
+Samantha's Rabbit,2,false,-188.8,tall: indeed
+
+```
+then
+```bash
+yq -p=csv --csv-auto-parse=f sample.csv
+```
+will output
+```yaml
+- name: Gary
+  numberOfCats: 1
+  likesApples: true
+  height: 168.8
+  facts: 'cool: true'
+- name: Samantha's Rabbit
+  numberOfCats: 2
+  likesApples: false
+  height: -188.8
+  facts: 'tall: indeed'
 ```
 
 ## Parse TSV into an array of objects

pkg/yqlib/doc/usage/formatting-expressions.md

@@ -0,0 +1,100 @@
+# Formatting Expressions
+
+`From version v4.41+`
+
+You can put expressions into `.yq` files, use whitespace and comments to break up complex expressions and explain what's going on.
+
+## Using expression files and comments
+Note that you can execute the file directly - but make sure you make the expression file executable.
+
+Given a sample.yaml file of:
+```yaml
+a:
+  b: old
+```
+And an 'update.yq' expression file of:
+```bash
+#! yq
+
+# This is a yq expression that updates the map
+# for several great reasons outlined here.
+
+.a.b = "new" # line comment here
+| .a.c = "frog"
+
+# Now good things will happen.
+```
+then
+```bash
+./update.yq sample.yaml
+```
+will output
+```yaml
+a:
+  b: new
+  c: frog
+```
+
+## Flags in expression files
+You can specify flags on the shebang line, this only works when executing the file directly.
+
+Given a sample.yaml file of:
+```yaml
+a:
+  b: old
+```
+And an 'update.yq' expression file of:
+```bash
+#! yq -oj
+
+# This is a yq expression that updates the map
+# for several great reasons outlined here.
+
+.a.b = "new" # line comment here
+| .a.c = "frog"
+
+# Now good things will happen.
+```
+then
+```bash
+./update.yq sample.yaml
+```
+will output
+```yaml
+{
+  "a": {
+    "b": "new",
+    "c": "frog"
+  }
+}
+```
+
+## Commenting out yq expressions
+Note that `c` is no longer set to 'frog'. In this example we're calling yq directly and passing the expression file into `--from-file`, this is no different from executing the expression file directly.
+
+Given a sample.yaml file of:
+```yaml
+a:
+  b: old
+```
+And an 'update.yq' expression file of:
+```bash
+#! yq
+# This is a yq expression that updates the map
+# for several great reasons outlined here.
+
+.a.b = "new" # line comment here
+# | .a.c = "frog"
+
+# Now good things will happen.
+```
+then
+```bash
+yq --from-file update.yq sample.yml
+```
+will output
+```yaml
+a:
+  b: new
+```
+

pkg/yqlib/doc/usage/headers/csv-tsv.md

@@ -2,7 +2,7 @@
 Encode/Decode/Roundtrip CSV and TSV files.
 
 ## Encode 
-Currently supports arrays of homogenous flat objects, that is: no nesting and it assumes the _first_ object has all the keys required:
+Currently supports arrays of homogeneous flat objects, that is: no nesting and it assumes the _first_ object has all the keys required:
 
 ```yaml
 - name: Bobo

pkg/yqlib/doc/usage/headers/formatting-expressions.md

@@ -0,0 +1,5 @@
+# Formatting Expressions
+
+`From version v4.41+`
+
+You can put expressions into `.yq` files, use whitespace and comments to break up complex expressions and explain what's going on.

pkg/yqlib/doc/usage/headers/recipes.md

@@ -0,0 +1,5 @@
+# Recipes
+
+These examples are intended to show how you can use multiple operators together so you get an idea of how you can perform complex data manipulation.
+
+Please see the details [operator docs](https://mikefarah.gitbook.io/yq/operators) for details on each individual operator.

pkg/yqlib/doc/usage/lua.md

@@ -1,5 +1,33 @@
 
-## Basic example
+## Basic input example
+Given a sample.lua file of:
+```lua
+return {
+	["country"] = "Australia"; -- this place
+	["cities"] = {
+		"Sydney",
+		"Melbourne",
+		"Brisbane",
+		"Perth",
+	};
+};
+
+```
+then
+```bash
+yq -oy '.' sample.lua
+```
+will output
+```yaml
+country: Australia
+cities:
+  - Sydney
+  - Melbourne
+  - Brisbane
+  - Perth
+```
+
+## Basic output example
 Given a sample.yml file of:
 ```yaml
 ---
@@ -101,6 +129,8 @@ numbers:
   - octal: 0o30
   - float: 123.45
   - infinity: .inf
+    plus_infinity: +.inf
+    minus_infinity: -.inf
   - not: .nan
 
 ```
@@ -134,6 +164,8 @@ return {
 		},
 		{
 			["infinity"] = (1/0);
+			["plus_infinity"] = (1/0);
+			["minus_infinity"] = (-1/0);
 		},
 		{
 			["not"] = (0/0);

pkg/yqlib/doc/usage/properties.md

@@ -14,6 +14,8 @@ person: # neither do comments on maps
     name: Mike Wazowski # comments on values appear
     pets: 
     - cat # comments on array values appear
+    - nested:
+        - list entry
     food: [pizza] # comments on arrays do not
 emptyArray: []
 emptyMap: []
@@ -31,9 +33,76 @@ person.name = Mike Wazowski
 
 # comments on array values appear
 person.pets.0 = cat
+person.pets.1.nested.0 = list entry
 person.food.0 = pizza
 ```
 
+## Encode properties with array brackets
+Declare the --properties-array-brackets flag to give array paths in brackets (e.g. SpringBoot).
+
+Given a sample.yml file of:
+```yaml
+# block comments come through
+person: # neither do comments on maps
+    name: Mike Wazowski # comments on values appear
+    pets: 
+    - cat # comments on array values appear
+    - nested:
+        - list entry
+    food: [pizza] # comments on arrays do not
+emptyArray: []
+emptyMap: []
+
+```
+then
+```bash
+yq -o=props --properties-array-brackets sample.yml
+```
+will output
+```properties
+# block comments come through
+# comments on values appear
+person.name = Mike Wazowski
+
+# comments on array values appear
+person.pets[0] = cat
+person.pets[1].nested[0] = list entry
+person.food[0] = pizza
+```
+
+## Encode properties - custom separator
+Use the --properties-customer-separator flag to specify your own key/value separator.
+
+Given a sample.yml file of:
+```yaml
+# block comments come through
+person: # neither do comments on maps
+    name: Mike Wazowski # comments on values appear
+    pets: 
+    - cat # comments on array values appear
+    - nested:
+        - list entry
+    food: [pizza] # comments on arrays do not
+emptyArray: []
+emptyMap: []
+
+```
+then
+```bash
+yq -o=props --properties-customer-separator=" :@ " sample.yml
+```
+will output
+```properties
+# block comments come through
+# comments on values appear
+person.name :@ Mike Wazowski
+
+# comments on array values appear
+person.pets.0 :@ cat
+person.pets.1.nested.0 :@ list entry
+person.food.0 :@ pizza
+```
+
 ## Encode properties: scalar encapsulation
 Note that string values with blank characters in them are encapsulated with double quotes
 
@@ -44,6 +113,8 @@ person: # neither do comments on maps
     name: Mike Wazowski # comments on values appear
     pets: 
     - cat # comments on array values appear
+    - nested:
+        - list entry
     food: [pizza] # comments on arrays do not
 emptyArray: []
 emptyMap: []
@@ -61,6 +132,7 @@ person.name = "Mike Wazowski"
 
 # comments on array values appear
 person.pets.0 = cat
+person.pets.1.nested.0 = "list entry"
 person.food.0 = pizza
 ```
 
@@ -72,6 +144,8 @@ person: # neither do comments on maps
     name: Mike Wazowski # comments on values appear
     pets: 
     - cat # comments on array values appear
+    - nested:
+        - list entry
     food: [pizza] # comments on arrays do not
 emptyArray: []
 emptyMap: []
@@ -85,6 +159,7 @@ will output
 ```properties
 person.name = Mike Wazowski
 person.pets.0 = cat
+person.pets.1.nested.0 = list entry
 person.food.0 = pizza
 ```
 
@@ -98,6 +173,8 @@ person: # neither do comments on maps
     name: Mike Wazowski # comments on values appear
     pets: 
     - cat # comments on array values appear
+    - nested:
+        - list entry
     food: [pizza] # comments on arrays do not
 emptyArray: []
 emptyMap: []
@@ -115,6 +192,7 @@ person.name = Mike Wazowski
 
 # comments on array values appear
 person.pets.0 = cat
+person.pets.1.nested.0 = list entry
 person.food.0 = pizza
 emptyArray = 
 emptyMap = 
@@ -129,6 +207,7 @@ person.name = Mike Wazowski
 
 # comments on array values appear
 person.pets.0 = cat
+person.pets.1.nested.0 = list entry
 person.food.0 = pizza
 
 ```
@@ -139,14 +218,16 @@ yq -p=props sample.properties
 will output
 ```yaml
 person:
-    # block comments come through
-    # comments on values appear
-    name: Mike Wazowski
-    pets:
-        # comments on array values appear
-        - cat
-    food:
-        - pizza
+  # block comments come through
+  # comments on values appear
+  name: Mike Wazowski
+  pets:
+    # comments on array values appear
+    - cat
+    - nested:
+        - list entry
+  food:
+    - pizza
 ```
 
 ## Decode properties - array should be a map
@@ -163,7 +244,7 @@ yq -p=props '.things |= array_to_map' sample.properties
 will output
 ```yaml
 things:
-    10: mike
+  10: mike
 ```
 
 ## Roundtrip
@@ -175,6 +256,7 @@ person.name = Mike Wazowski
 
 # comments on array values appear
 person.pets.0 = cat
+person.pets.1.nested.0 = list entry
 person.food.0 = pizza
 
 ```
@@ -190,6 +272,7 @@ person.name = Mike Wazowski
 
 # comments on array values appear
 person.pets.0 = dog
+person.pets.1.nested.0 = list entry
 person.food.0 = pizza
 ```
 

pkg/yqlib/doc/usage/recipes.md

@@ -0,0 +1,264 @@
+# Recipes
+
+These examples are intended to show how you can use multiple operators together so you get an idea of how you can perform complex data manipulation.
+
+Please see the details [operator docs](https://mikefarah.gitbook.io/yq/operators) for details on each individual operator.
+
+## Find items in an array
+We have an array and we want to find the elements with a particular name.
+
+Given a sample.yml file of:
+```yaml
+- name: Foo
+  numBuckets: 0
+- name: Bar
+  numBuckets: 0
+```
+then
+```bash
+yq '.[] | select(.name == "Foo")' sample.yml
+```
+will output
+```yaml
+name: Foo
+numBuckets: 0
+```
+
+### Explanation:
+- `.[]` splats the array, and puts all the items in the context.
+- These items are then piped (`|`) into `select(.name == "Foo")` which will select all the nodes that have a name property set to 'Foo'.
+- See the [select](https://mikefarah.gitbook.io/yq/operators/select) operator for more information.
+
+## Find and update items in an array
+We have an array and we want to _update_ the elements with a particular name.
+
+Given a sample.yml file of:
+```yaml
+- name: Foo
+  numBuckets: 0
+- name: Bar
+  numBuckets: 0
+```
+then
+```bash
+yq '(.[] | select(.name == "Foo") | .numBuckets) |= . + 1' sample.yml
+```
+will output
+```yaml
+- name: Foo
+  numBuckets: 1
+- name: Bar
+  numBuckets: 0
+```
+
+### Explanation:
+- Following from the example above`.[]` splats the array, selects filters the items.
+- We then pipe (`|`) that into `.numBuckets`, which will select that field from all the matching items
+- Splat, select and the field are all in brackets, that whole expression is passed to the `|=` operator as the left hand side expression, with `. + 1` as the right hand side expression.
+- `|=` is the operator that updates fields relative to their own value, which is referenced as dot (`.`).
+- The expression `. + 1` increments the numBuckets counter.
+- See the [assign](https://mikefarah.gitbook.io/yq/operators/assign-update) and [add](https://mikefarah.gitbook.io/yq/operators/add) operators for more information.
+
+## Deeply prune a tree
+Say we are only interested in child1 and child2, and want to filter everything else out.
+
+Given a sample.yml file of:
+```yaml
+parentA:
+  - bob
+parentB:
+  child1: i am child1
+  child3: hiya
+parentC:
+  childX: cool
+  child2: me child2
+```
+then
+```bash
+yq '(
+  .. | # recurse through all the nodes
+  select(has("child1") or has("child2")) | # match parents that have either child1 or child2
+  (.child1, .child2) | # select those children
+  select(.) # filter out nulls
+) as $i ireduce({};  # using that set of nodes, create a new result map
+  setpath($i | path; $i) # and put in each node, using its original path
+)' sample.yml
+```
+will output
+```yaml
+parentB:
+  child1: i am child1
+parentC:
+  child2: me child2
+```
+
+### Explanation:
+- Find all the matching child1 and child2 nodes
+- Using ireduce, create a new map using just those nodes
+- Set each node into the new map using its original path
+
+## Multiple or complex updates to items in an array
+We have an array and we want to _update_ the elements with a particular name in reference to its type.
+
+Given a sample.yml file of:
+```yaml
+myArray:
+  - name: Foo
+    type: cat
+  - name: Bar
+    type: dog
+```
+then
+```bash
+yq 'with(.myArray[]; .name = .name + " - " + .type)' sample.yml
+```
+will output
+```yaml
+myArray:
+  - name: Foo - cat
+    type: cat
+  - name: Bar - dog
+    type: dog
+```
+
+### Explanation:
+- The with operator will effectively loop through each given item in the first given expression, and run the second expression against it.
+- `.myArray[]` splats the array in `myArray`. So `with` will run against each item in that array
+- `.name = .name + " - " + .type` this expression is run against every item, updating the name to be a concatenation of the original name as well as the type.
+- See the [with](https://mikefarah.gitbook.io/yq/operators/with) operator for more information and examples.
+
+## Sort an array by a field
+Given a sample.yml file of:
+```yaml
+myArray:
+  - name: Foo
+    numBuckets: 1
+  - name: Bar
+    numBuckets: 0
+```
+then
+```bash
+yq '.myArray |= sort_by(.numBuckets)' sample.yml
+```
+will output
+```yaml
+myArray:
+  - name: Bar
+    numBuckets: 0
+  - name: Foo
+    numBuckets: 1
+```
+
+### Explanation:
+- We want to resort `.myArray`.
+- `sort_by` works by piping an array into it, and it pipes out a sorted array.
+- So, we use `|=` to update `.myArray`. This is the same as doing `.myArray = (.myArray | sort_by(.numBuckets))`
+
+## Filter, flatten, sort and unique
+Lets find the unique set of names from the document.
+
+Given a sample.yml file of:
+```yaml
+- type: foo
+  names:
+    - Fred
+    - Catherine
+- type: bar
+  names:
+    - Zelda
+- type: foo
+  names: Fred
+- type: foo
+  names: Ava
+```
+then
+```bash
+yq '[.[] | select(.type == "foo") | .names] | flatten | sort | unique' sample.yml
+```
+will output
+```yaml
+- Ava
+- Catherine
+- Fred
+```
+
+### Explanation:
+- `.[] | select(.type == "foo") | .names` will select the array elements of type "foo"
+- Splat `.[]` will unwrap the array and match all the items. We need to do this so we can work on the child items, for instance, filter items out using the `select` operator.
+- But we still want the final results back into an array. So after we're doing working on the children, we wrap everything back into an array using square brackets around the expression. `[.[] | select(.type == "foo") | .names]`
+- Now have have an array of all the 'names' values. Which includes arrays of strings as well as strings on their own.
+- Pipe `|` this array through `flatten`. This will flatten nested arrays. So now we have a flat list of all the name value strings
+- Next we pipe `|` that through `sort` and then `unique` to get a sorted, unique list of the names!
+- See the [flatten](https://mikefarah.gitbook.io/yq/operators/flatten), [sort](https://mikefarah.gitbook.io/yq/operators/sort) and [unique](https://mikefarah.gitbook.io/yq/operators/unique) for more information and examples.
+
+## Export as environment variables (script), or any custom format
+Given a yaml document, lets output a script that will configure environment variables with that data. This same approach can be used for exporting into custom formats.
+
+Given a sample.yml file of:
+```yaml
+var0: string0
+var1: string1
+fruit:
+  - apple
+  - banana
+  - peach
+```
+then
+```bash
+yq '.[] |(
+	( select(kind == "scalar") | key + "='\''" + . + "'\''"),
+	( select(kind == "seq") | key + "=(" + (map("'\''" + . + "'\''") | join(",")) + ")")
+)' sample.yml
+```
+will output
+```yaml
+var0='string0'
+var1='string1'
+fruit=('apple','banana','peach')
+```
+
+### Explanation:
+- `.[]` matches all top level elements
+- We need a string expression for each of the different types that will produce the bash syntax, we'll use the union operator, to join them together
+- Scalars, we just need the key and quoted value: `( select(kind == "scalar") | key + "='" + . + "'")`
+- Sequences (or arrays) are trickier, we need to quote each value and `join` them with `,`: `map("'" + . + "'") | join(",")`
+
+## Custom format with nested data
+Like the previous example, but lets handle nested data structures. In this custom example, we're going to join the property paths with _. The important thing to keep in mind is that our expression is not recursive (despite the data structure being so). Instead we match _all_ elements on the tree and operate on them.
+
+Given a sample.yml file of:
+```yaml
+simple: string0
+simpleArray:
+  - apple
+  - banana
+  - peach
+deep:
+  property: value
+  array:
+    - cat
+```
+then
+```bash
+yq '.. |(
+	( select(kind == "scalar" and parent | kind != "seq") | (path | join("_")) + "='\''" + . + "'\''"),
+	( select(kind == "seq") | (path | join("_")) + "=(" + (map("'\''" + . + "'\''") | join(",")) + ")")
+)' sample.yml
+```
+will output
+```yaml
+simple='string0'
+deep_property='value'
+simpleArray=('apple','banana','peach')
+deep_array=('cat')
+```
+
+### Explanation:
+- You'll need to understand how the previous example works to understand this extension.
+- `..` matches _all_ elements, instead of `.[]` from the previous example that just matches top level elements.
+- Like before, we need a string expression for each of the different types that will produce the bash syntax, we'll use the union operator, to join them together
+- This time, however, our expression matches every node in the data structure.
+- We only want to print scalars that are not in arrays (because we handle the separately), so well add `and parent | kind != "seq"` to the select operator expression for scalars
+- We don't just want the key any more, we want the full path. So instead of `key` we have `path | join("_")`
+- The expression for sequences follows the same logic
+

pkg/yqlib/doc/usage/shellvariables.md

@@ -34,7 +34,7 @@ Given a sample.yml file of:
 ascii_=_symbols: replaced with _
 "ascii_	_controls": dropped (this example uses \t)
 nonascii_א_characters: dropped
-effrot_expeñded_tò_preserve_accented_latin_letters: moderate (via unicode NFKD)
+effort_expeñded_tò_preserve_accented_latin_letters: moderate (via unicode NFKD)
 
 ```
 then
@@ -46,7 +46,7 @@ will output
 ascii___symbols='replaced with _'
 ascii__controls='dropped (this example uses \t)'
 nonascii__characters=dropped
-effrot_expended_to_preserve_accented_latin_letters='moderate (via unicode NFKD)'
+effort_expended_to_preserve_accented_latin_letters='moderate (via unicode NFKD)'
 ```
 
 ## Encode shell variables: empty values, arrays and maps