Bumping version

Added DELPATHS operator
Added SETPATH operator
2026-08-24 08:22:13 +08:00 · 2022-10-05 20:15:02 +11:00 · 2022-10-05 20:09:53 +11:00 · 2022-10-05 14:12:08 +11:00 · 2022-09-30 11:30:31 +10:00 · 2022-09-30 11:26:36 +10:00
79 changed files with 2180 additions and 201 deletions
@@ -10,10 +10,10 @@ jobs:
    runs-on: ubuntu-latest
    steps:

-    - name: Set up Go 1.18
+    - name: Set up Go 1.19
      uses: actions/setup-go@v3
      with:
-        go-version: 1.18
+        go-version: 1.19
      id: go

    - name: Check out code into the Go module directory
@@ -12,7 +12,7 @@ jobs:
      - uses: actions/checkout@v3
      - uses: actions/setup-go@v3
        with:
-          go-version: '^1.18'
+          go-version: '^1.19'
      - name: Compile man page markup
        id: gen-man-page-md
        run: |
@@ -56,4 +56,9 @@ _build
 debian/files

 # intellij
-/.idea
+/.idea
+
+# vscode
+.vscode
+
+yq3
@@ -1,4 +1,4 @@
-FROM golang:1.18.4 as builder
+FROM golang:1.19.1 as builder

 WORKDIR /go/src/mikefarah/yq

@@ -1,4 +1,4 @@
-FROM golang:1.18.4
+FROM golang:1.19.1

 COPY scripts/devtools.sh /opt/devtools.sh

@@ -3,7 +3,7 @@
 ![Build](https://github.com/mikefarah/yq/workflows/Build/badge.svg)  ![Docker Pulls](https://img.shields.io/docker/pulls/mikefarah/yq.svg) ![Github Releases (by Release)](https://img.shields.io/github/downloads/mikefarah/yq/total.svg) ![Go Report](https://goreportcard.com/badge/github.com/mikefarah/yq) ![CodeQL](https://github.com/mikefarah/yq/workflows/CodeQL/badge.svg)


-a lightweight and portable command-line YAML, JSON and XML processor. `yq` uses [jq](https://github.com/stedolan/jq) like syntax but works with yaml files as well as json and xml. 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 YAML, JSON and XML processor. `yq` uses [jq](https://github.com/stedolan/jq) like syntax but works with yaml files as well as json, xml, properties, csv and tsv. It doesn't yet support everything `jq` does - but it does support the most common operations and functions, and more is being added continuously.

 yq is written in go - so you can download a dependency free binary for your platform and you are good to go! If you prefer there are a variety of package managers that can be used as well as Docker and Podman, all listed below.

@@ -73,6 +73,8 @@ yq -P sample.json

 See the [documentation](https://mikefarah.gitbook.io/yq/) for more examples.

+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)
+
 ## Install

 ### [Download the latest binary](https://github.com/mikefarah/yq/releases/latest)
@@ -207,12 +209,27 @@ RUN apk add --no-cache tzdata
 USER yq
 ```

+#### Podman with SELinux
+
+If you are using podman with SELinux, you will need to set the shared volume flag `:z` on the volume mount:
+
+```
+-v "${PWD}":/workdir:z
+```
+
 ### GitHub Action
 ```
  - name: Set foobar to cool
    uses: mikefarah/yq@master
    with:
      cmd: yq -i '.foo.bar = "cool"' 'config.yml'
+  - name: Get an entry with a variable that might contain dots or spaces
+    id: get_username
+    uses: mikefarah/yq@master
+    with:
+      cmd: yq '.all.children.["${{ matrix.ip_address }}"].username' ops/inventories/production.yml
+  - name: Reuse a variable obtained in another step
+    run: echo ${{ steps.get_username.outputs.result }}
 ```

 See https://mikefarah.gitbook.io/yq/usage/github-action for more.
@@ -290,10 +307,10 @@ Supported by @rmescandon (https://launchpad.net/~rmescandon/+archive/ubuntu/yq)
 - 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)
 - [Load content from other files](https://mikefarah.gitbook.io/yq/operators/load)
- [Convert to/from json](https://mikefarah.gitbook.io/yq/v/v4.x/usage/convert)
+- [Convert to/from json/ndjson](https://mikefarah.gitbook.io/yq/v/v4.x/usage/convert)
 - [Convert to/from xml](https://mikefarah.gitbook.io/yq/v/v4.x/usage/xml)
 - [Convert to/from properties](https://mikefarah.gitbook.io/yq/v/v4.x/usage/properties)
- [Convert to csv/tsv](https://mikefarah.gitbook.io/yq/usage/csv-tsv)
+- [Convert to/from csv/tsv](https://mikefarah.gitbook.io/yq/usage/csv-tsv)
 - [General shell completion scripts (bash/zsh/fish/powershell)](https://mikefarah.gitbook.io/yq/v/v4.x/commands/shell-completion)
 - [Reduce](https://mikefarah.gitbook.io/yq/operators/reduce) to merge multiple files or sum an array or other fancy things.
 - [Github Action](https://mikefarah.gitbook.io/yq/usage/github-action) to use in your automated pipeline (thanks @devorbitus)
@@ -0,0 +1,43 @@
+#!/bin/bash
+
+setUp() {
+  rm test*.yml || true
+  
+}
+
+testLineCountFirstLineComment() {
+  cat >test.yml <<EOL
+#test123 
+abc: 123 
+test123: 123123 
+#comment 
+lalilu: lalilu
+EOL
+
+  X=$(./yq '.lalilu | line' --header-preprocess=false < test.yml)
+  assertEquals "5" "$X"
+}
+
+testArrayOfDocs() {
+  cat >test.yml <<EOL
+---
+# leading comment doc 1
+a: 1
+---
+# leading comment doc 2
+a: 2
+EOL
+
+read -r -d '' expected << EOM
+- # leading comment doc 1
+  a: 1
+- # leading comment doc 2
+  a: 2
+EOM
+
+  X=$(./yq ea '[.]' --header-preprocess=false < test.yml)
+  assertEquals "$expected" "$X"
+
+}
+
+source ./scripts/shunit2
@@ -3,6 +3,8 @@
 setUp() {
  rm test*.yml 2>/dev/null || true
  rm test*.properties 2>/dev/null || true
+  rm test*.csv 2>/dev/null || true
+  rm test*.tsv 2>/dev/null || true
  rm test*.xml 2>/dev/null || true
 }

@@ -40,6 +42,51 @@ EOM
  assertEquals "$expected" "$X"
 }

+testInputCSV() {
+  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=$(./yq e -p=csv test.csv)
+  assertEquals "$expected" "$X"
+
+  X=$(./yq ea -p=csv test.csv)
+  assertEquals "$expected" "$X"
+}
+
+testInputTSV() {
+  cat >test.tsv <<EOL
+fruit	yumLevel
+apple	5
+banana	4
+EOL
+
+  read -r -d '' expected << EOM
+- fruit: apple
+  yumLevel: 5
+- fruit: banana
+  yumLevel: 4
+EOM
+
+  X=$(./yq e -p=t test.tsv)
+  assertEquals "$expected" "$X"
+
+  X=$(./yq ea -p=t test.tsv)
+  assertEquals "$expected" "$X"
+}
+
+
+
+
 testInputXml() {
  cat >test.yml <<EOL
 <cat legs="4">BiBi</cat>
@@ -1,5 +1,8 @@
 #!/bin/bash

+
+# examples where header-preprocess is required
+
 setUp() {
  rm test*.yml || true
  cat >test.yml <<EOL
@@ -102,6 +102,48 @@ EOM
  assertEquals "$expected" "$X"
 }

+testOutputCSV() {
+  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=$(./yq -o=c test.yml)
+  assertEquals "$expected" "$X"
+
+  X=$(./yq ea -o=csv test.yml)
+  assertEquals "$expected" "$X"
+}
+
+testOutputTSV() {
+  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=$(./yq -o=t test.yml)
+  assertEquals "$expected" "$X"
+
+  X=$(./yq ea -o=tsv test.yml)
+  assertEquals "$expected" "$X"
+}
+
 testOutputXml() {
  cat >test.yml <<EOL
 a: {b: {c: ["cat"]}}
@@ -68,6 +68,10 @@ func configureDecoder() (yqlib.Decoder, error) {
 		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
 	}

 	return yqlib.NewYamlDecoder(), nil
@@ -11,7 +11,7 @@ var (
 	GitDescribe string

 	// Version is main version number that is being run at the moment.
-	Version = "4.26.1"
+	Version = "4.28.1"

 	// 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
@@ -1,12 +1,5 @@
-# --------------------------------------------------
-# It's a test with comment
-# --------------------------------------------------
-test:
-  - name: a
-  - name: b
-  - name: c
-groups:
-  - name: a
-  - name: b
-  - name: c
-  - name: c
+volumes:
+  beehive-conf:
+    driver: local
+services:
+  - beehive
@@ -0,0 +1,3 @@
+name,numberOfCats,likesApples,height
+Gary,1,true,168.8
+Samantha's Rabbit,2,false,-188.8
@@ -2,7 +2,8 @@ FROM mikefarah/yq:4

 COPY entrypoint.sh /entrypoint.sh

-# this seems to be the default user in github actions
-USER 1001
+# github action recommendation is to run as root.
+# https://docs.github.com/en/actions/creating-actions/dockerfile-support-for-github-actions#user
+USER root

 ENTRYPOINT ["/entrypoint.sh"]
@@ -2,11 +2,11 @@ module github.com/mikefarah/yq/v4

 require (
 	github.com/a8m/envsubst v1.3.0
-	github.com/alecthomas/participle/v2 v2.0.0-beta.4
+	github.com/alecthomas/participle/v2 v2.0.0-beta.5
 	github.com/alecthomas/repr v0.1.0
-	github.com/elliotchance/orderedmap v1.4.0
+	github.com/elliotchance/orderedmap v1.5.0
 	github.com/fatih/color v1.13.0
-	github.com/goccy/go-json v0.9.10
+	github.com/goccy/go-json v0.9.11
 	github.com/goccy/go-yaml v1.9.5
 	github.com/jinzhu/copier v0.3.5
 	github.com/magiconair/properties v1.8.6
@@ -27,4 +27,4 @@ require (
 	golang.org/x/xerrors v0.0.0-20220609144429-65e65417b02f // indirect
 )

-go 1.18
+go 1.19
@@ -1,15 +1,15 @@
 github.com/a8m/envsubst v1.3.0 h1:GmXKmVssap0YtlU3E230W98RWtWCyIZzjtf1apWWyAg=
 github.com/a8m/envsubst v1.3.0/go.mod h1:MVUTQNGQ3tsjOOtKCNd+fl8RzhsXcDvvAEzkhGtlsbY=
 github.com/alecthomas/assert/v2 v2.0.3 h1:WKqJODfOiQG0nEJKFKzDIG3E29CN2/4zR9XGJzKIkbg=
-github.com/alecthomas/participle/v2 v2.0.0-beta.4 h1:ublfGBm+x+p2j7KotHhrUMbKtejT7M0Gv1Mt1u3absw=
-github.com/alecthomas/participle/v2 v2.0.0-beta.4/go.mod h1:RC764t6n4L8D8ITAJv0qdokritYSNR3wV5cVwmIEaMM=
+github.com/alecthomas/participle/v2 v2.0.0-beta.5 h1:y6dsSYVb1G5eK6mgmy+BgI3Mw35a3WghArZ/Hbebrjo=
+github.com/alecthomas/participle/v2 v2.0.0-beta.5/go.mod h1:RC764t6n4L8D8ITAJv0qdokritYSNR3wV5cVwmIEaMM=
 github.com/alecthomas/repr v0.1.0 h1:ENn2e1+J3k09gyj2shc0dHr/yjaWSHRlrJ4DPMevDqE=
 github.com/alecthomas/repr v0.1.0/go.mod h1:2kn6fqh/zIyPLmm3ugklbEi5hg5wS435eygvNfaDQL8=
 github.com/cpuguy83/go-md2man/v2 v2.0.2/go.mod h1:tgQtvFlXSQOSOSIRvRPT7W67SCa46tRHOmNcaadrF8o=
 github.com/davecgh/go-spew v1.1.0 h1:ZDRjVQ15GmhC3fiQ8ni8+OwkZQO4DARzQgrnXU1Liz8=
 github.com/davecgh/go-spew v1.1.0/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
-github.com/elliotchance/orderedmap v1.4.0 h1:wZtfeEONCbx6in1CZyE6bELEt/vFayMvsxqI5SgsR+A=
-github.com/elliotchance/orderedmap v1.4.0/go.mod h1:wsDwEaX5jEoyhbs7x93zk2H/qv0zwuhg4inXhDkYqys=
+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 v1.10.0/go.mod h1:ELkj/draVOlAH/xkhN6mQ50Qd0MPOk5AAr3maGEBuJM=
 github.com/fatih/color v1.13.0 h1:8LOYc1KYPPmyKMuN8QV2DNRWNbLo6LZ0iLs8+mlH53w=
 github.com/fatih/color v1.13.0/go.mod h1:kLAiJbzzSOZDVNGyDpeOxJ47H46qBXwg5ILebYFFOfk=
@@ -17,8 +17,8 @@ github.com/go-playground/assert/v2 v2.0.1/go.mod h1:VDjEfimB/XKnb+ZQfWdccd7VUvSc
 github.com/go-playground/locales v0.13.0/go.mod h1:taPMhCMXrRLJO55olJkUXHZBHCxTMfnGwq/HNwmWNS8=
 github.com/go-playground/universal-translator v0.17.0/go.mod h1:UkSxE5sNxxRwHyU+Scu5vgOQjsIJAF8j9muTVoKLVtA=
 github.com/go-playground/validator/v10 v10.4.1/go.mod h1:nlOn6nFhuKACm19sB/8EGNn9GlaMV7XkbRSipzJ0Ii4=
-github.com/goccy/go-json v0.9.10 h1:hCeNmprSNLB8B8vQKWl6DpuH0t60oEs+TAk9a7CScKc=
-github.com/goccy/go-json v0.9.10/go.mod h1:6MelG93GURQebXPDq3khkgXZkazVtN9CRI+MGFi0w8I=
+github.com/goccy/go-json v0.9.11 h1:/pAaQDLHEoCq/5FFmSKBswWmK6H0e8g4159Kc/X/nqk=
+github.com/goccy/go-json v0.9.11/go.mod h1:6MelG93GURQebXPDq3khkgXZkazVtN9CRI+MGFi0w8I=
 github.com/goccy/go-yaml v1.9.5 h1:Eh/+3uk9kLxG4koCX6lRMAPS1OaMSAi+FJcya0INdB0=
 github.com/goccy/go-yaml v1.9.5/go.mod h1:U/jl18uSupI5rdI2jmuCswEA2htH9eXfferR3KfscvA=
 github.com/hexops/gotextdiff v1.0.3 h1:gitA9+qJrrTCsiCl7+kh75nPqQt1cx4ZkudSTLoUqJM=
@@ -37,7 +37,7 @@ The `=` operator then pipes the 'root' context through the `rhs` expression of `
 dog
 ```

-Both sides have now been evaluated, so now the operator copies across the value from the RHS (`.b`) to the the LHS (`.a`), and it returns the now updated context:
+Both sides have now been evaluated, so now the operator copies across the value from the RHS (`.b`) to the LHS (`.a`), and it returns the now updated context:

 ```yaml
 a: dog
@@ -157,7 +157,7 @@ func (n *CandidateNode) UpdateAttributesFrom(other *CandidateNode, prefs assignP
 	n.Node.Kind = other.Node.Kind

 	// don't clobber custom tags...
-	if strings.HasPrefix(n.Node.Tag, "!!") || n.Node.Tag == "" {
+	if prefs.ClobberCustomTags || strings.HasPrefix(n.Node.Tag, "!!") || n.Node.Tag == "" {
 		n.Node.Tag = other.Node.Tag
 	}

@@ -0,0 +1,273 @@
+package yqlib
+
+import (
+	"bufio"
+	"fmt"
+	"testing"
+
+	"github.com/mikefarah/yq/v4/test"
+)
+
+const csvSimple = `name,numberOfCats,likesApples,height
+Gary,1,true,168.8
+Samantha's Rabbit,2,false,-188.8
+`
+
+const expectedUpdatedSimpleCsv = `name,numberOfCats,likesApples,height
+Gary,3,true,168.8
+Samantha's Rabbit,2,false,-188.8
+`
+
+const csvSimpleShort = `Name,Number of Cats
+Gary,1
+Samantha's Rabbit,2
+`
+
+const tsvSimple = `name	numberOfCats	likesApples	height
+Gary	1	true	168.8
+Samantha's Rabbit	2	false	-188.8
+`
+
+const expectedYamlFromCSV = `- name: Gary
+  numberOfCats: 1
+  likesApples: true
+  height: 168.8
+- name: Samantha's Rabbit
+  numberOfCats: 2
+  likesApples: false
+  height: -188.8
+`
+
+const expectedYamlFromCSVMissingData = `- name: Gary
+  numberOfCats: 1
+  height: 168.8
+- name: Samantha's Rabbit
+  height: -188.8
+  likesApples: false
+`
+
+const csvSimpleMissingData = `name,numberOfCats,height
+Gary,1,168.8
+Samantha's Rabbit,,-188.8
+`
+
+const csvTestSimpleYaml = `- [i, like, csv]
+- [because, excel, is, cool]`
+
+const expectedSimpleCsv = `i,like,csv
+because,excel,is,cool
+`
+
+const tsvTestExpectedSimpleCsv = `i	like	csv
+because	excel	is	cool
+`
+
+var csvScenarios = []formatScenario{
+	{
+		description:  "Encode CSV simple",
+		input:        csvTestSimpleYaml,
+		expected:     expectedSimpleCsv,
+		scenarioType: "encode-csv",
+	},
+	{
+		description:  "Encode TSV simple",
+		input:        csvTestSimpleYaml,
+		expected:     tsvTestExpectedSimpleCsv,
+		scenarioType: "encode-tsv",
+	},
+	{
+		description:  "Encode Empty",
+		skipDoc:      true,
+		input:        `[]`,
+		expected:     "",
+		scenarioType: "encode-csv",
+	},
+	{
+		description:  "Comma in value",
+		skipDoc:      true,
+		input:        `["comma, in, value", things]`,
+		expected:     "\"comma, in, value\",things\n",
+		scenarioType: "encode-csv",
+	},
+	{
+		description:  "Encode array of objects to csv",
+		input:        expectedYamlFromCSV,
+		expected:     csvSimple,
+		scenarioType: "encode-csv",
+	},
+	{
+		description:    "Encode array of objects to custom csv format",
+		subdescription: "Add the header row manually, then the we convert each object into an array of values - resulting in an array of arrays. Pick the columns and call the header whatever you like.",
+		input:          expectedYamlFromCSV,
+		expected:       csvSimpleShort,
+		expression:     `[["Name", "Number of Cats"]] +  [.[] | [.name, .numberOfCats ]]`,
+		scenarioType:   "encode-csv",
+	},
+	{
+		description:    "Encode array of objects to csv - missing fields behaviour",
+		subdescription: "First entry is used to determine the headers, and it is missing 'likesApples', so it is not included in the csv. Second entry does not have 'numberOfCats' so that is blank",
+		input:          expectedYamlFromCSVMissingData,
+		expected:       csvSimpleMissingData,
+		scenarioType:   "encode-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",
+	},
+	{
+		description:    "Parse TSV into an array of objects",
+		subdescription: "First row is assumed to be the header row.",
+		input:          tsvSimple,
+		expected:       expectedYamlFromCSV,
+		scenarioType:   "decode-tsv-object",
+	},
+	{
+		description:  "Round trip",
+		input:        csvSimple,
+		expected:     expectedUpdatedSimpleCsv,
+		expression:   `(.[] | select(.name == "Gary") | .numberOfCats) = 3`,
+		scenarioType: "roundtrip-csv",
+	},
+}
+
+func testCSVScenario(t *testing.T, s formatScenario) {
+	switch s.scenarioType {
+	case "encode-csv":
+		test.AssertResultWithContext(t, s.expected, processFormatScenario(s, NewYamlDecoder(), NewCsvEncoder(',')), s.description)
+	case "encode-tsv":
+		test.AssertResultWithContext(t, s.expected, processFormatScenario(s, NewYamlDecoder(), NewCsvEncoder('\t')), s.description)
+	case "decode-csv-object":
+		test.AssertResultWithContext(t, s.expected, processFormatScenario(s, NewCSVObjectDecoder(','), NewYamlEncoder(2, false, true, true)), s.description)
+	case "decode-tsv-object":
+		test.AssertResultWithContext(t, s.expected, processFormatScenario(s, NewCSVObjectDecoder('\t'), NewYamlEncoder(2, false, true, true)), s.description)
+	case "roundtrip-csv":
+		test.AssertResultWithContext(t, s.expected, processFormatScenario(s, NewCSVObjectDecoder(','), NewCsvEncoder(',')), s.description)
+	default:
+		panic(fmt.Sprintf("unhandled scenario type %q", s.scenarioType))
+	}
+}
+
+func documentCSVDecodeObjectScenario(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 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",
+		processFormatScenario(s, NewCSVObjectDecoder(separator), NewYamlEncoder(s.indent, false, true, true))),
+	)
+}
+
+func documentCSVEncodeScenario(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, "Given a sample.yml file of:\n")
+	writeOrPanic(w, fmt.Sprintf("```yaml\n%v\n```\n", s.input))
+
+	writeOrPanic(w, "then\n")
+
+	expression := s.expression
+
+	if expression != "" {
+		writeOrPanic(w, fmt.Sprintf("```bash\nyq -o=%v '%v' sample.yml\n```\n", formatType, expression))
+	} else {
+		writeOrPanic(w, fmt.Sprintf("```bash\nyq -o=%v sample.yml\n```\n", formatType))
+	}
+	writeOrPanic(w, "will output\n")
+
+	separator := ','
+	if formatType == "tsv" {
+		separator = '\t'
+	}
+
+	writeOrPanic(w, fmt.Sprintf("```%v\n%v```\n\n", formatType,
+		processFormatScenario(s, NewYamlDecoder(), NewCsvEncoder(separator))),
+	)
+}
+
+func documentCSVRoundTripScenario(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")
+
+	expression := s.expression
+
+	if expression != "" {
+		writeOrPanic(w, fmt.Sprintf("```bash\nyq -p=%v -o=%v '%v' sample.%v\n```\n", formatType, formatType, expression, formatType))
+	} else {
+		writeOrPanic(w, fmt.Sprintf("```bash\nyq -p=%v -o=%v sample.%v\n```\n", formatType, formatType, formatType))
+	}
+	writeOrPanic(w, "will output\n")
+
+	separator := ','
+	if formatType == "tsv" {
+		separator = '\t'
+	}
+
+	writeOrPanic(w, fmt.Sprintf("```%v\n%v```\n\n", formatType,
+		processFormatScenario(s, NewCSVObjectDecoder(separator), NewCsvEncoder(separator))),
+	)
+}
+
+func documentCSVScenario(t *testing.T, w *bufio.Writer, i interface{}) {
+	s := i.(formatScenario)
+	if s.skipDoc {
+		return
+	}
+	switch s.scenarioType {
+	case "encode-csv":
+		documentCSVEncodeScenario(w, s, "csv")
+	case "encode-tsv":
+		documentCSVEncodeScenario(w, s, "tsv")
+	case "decode-csv-object":
+		documentCSVDecodeObjectScenario(w, s, "csv")
+	case "decode-tsv-object":
+		documentCSVDecodeObjectScenario(w, s, "tsv")
+	case "roundtrip-csv":
+		documentCSVRoundTripScenario(w, s, "csv")
+
+	default:
+		panic(fmt.Sprintf("unhandled scenario type %q", s.scenarioType))
+	}
+}
+
+func TestCSVScenarios(t *testing.T) {
+	for _, tt := range csvScenarios {
+		testCSVScenario(t, tt)
+	}
+	genericScenarios := make([]interface{}, len(csvScenarios))
+	for i, s := range csvScenarios {
+		genericScenarios[i] = s
+	}
+	documentScenarios(t, "usage", "csv-tsv", genericScenarios, documentCSVScenario)
+}
@@ -15,6 +15,8 @@ const (
 	PropertiesInputFormat
 	Base64InputFormat
 	JsonInputFormat
+	CSVObjectInputFormat
+	TSVObjectInputFormat
 )

 type Decoder interface {
@@ -32,6 +34,10 @@ func InputFormatFromString(format string) (InputFormat, error) {
 		return PropertiesInputFormat, nil
 	case "json", "ndjson", "j":
 		return JsonInputFormat, nil
+	case "csv", "c":
+		return CSVObjectInputFormat, nil
+	case "tsv", "t":
+		return TSVObjectInputFormat, nil
 	default:
 		return 0, fmt.Errorf("unknown format '%v' please use [yaml|xml|props]", format)
 	}
@@ -0,0 +1,77 @@
+package yqlib
+
+import (
+	"encoding/csv"
+	"errors"
+	"io"
+
+	yaml "gopkg.in/yaml.v3"
+)
+
+type csvObjectDecoder struct {
+	separator rune
+	reader    csv.Reader
+	finished  bool
+}
+
+func NewCSVObjectDecoder(separator rune) Decoder {
+	return &csvObjectDecoder{separator: separator}
+}
+
+func (dec *csvObjectDecoder) Init(reader io.Reader) {
+	dec.reader = *csv.NewReader(reader)
+	dec.reader.Comma = dec.separator
+	dec.finished = false
+}
+
+func (dec *csvObjectDecoder) convertToYamlNode(content string) *yaml.Node {
+	node, err := parseSnippet(content)
+	if err != nil {
+		return createScalarNode(content, content)
+	}
+	return node
+}
+
+func (dec *csvObjectDecoder) createObject(headerRow []string, contentRow []string) *yaml.Node {
+	objectNode := &yaml.Node{Kind: yaml.MappingNode, Tag: "!!map"}
+
+	for i, header := range headerRow {
+		objectNode.Content = append(
+			objectNode.Content,
+			createScalarNode(header, header),
+			dec.convertToYamlNode(contentRow[i]))
+	}
+	return objectNode
+}
+
+func (dec *csvObjectDecoder) Decode(rootYamlNode *yaml.Node) error {
+	if dec.finished {
+		return io.EOF
+	}
+	headerRow, err := dec.reader.Read()
+	log.Debugf(": headerRow%v", headerRow)
+	if err != nil {
+		return err
+	}
+
+	rootArray := &yaml.Node{Kind: yaml.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))
+		contentRow, err = dec.reader.Read()
+		log.Debugf("Read next contentRow: %v, %v", contentRow, err)
+	}
+	if !errors.Is(err, io.EOF) {
+		return err
+	}
+
+	log.Debugf("finished, contentRow%v", contentRow)
+	log.Debugf("err: %v", err)
+
+	rootYamlNode.Kind = yaml.DocumentNode
+	rootYamlNode.Content = []*yaml.Node{rootArray}
+	return nil
+}
@@ -22,13 +22,13 @@ func (dec *jsonDecoder) Init(reader io.Reader) {

 func (dec *jsonDecoder) Decode(rootYamlNode *yaml.Node) error {

-	var dataBucket interface{}
+	var dataBucket orderedMap
 	log.Debug("going to decode")
 	err := dec.decoder.Decode(&dataBucket)
 	if err != nil {
 		return err
 	}
-	node, err := dec.convertToYamlNode(dataBucket)
+	node, err := dec.convertToYamlNode(&dataBucket)

 	if err != nil {
 		return err
@@ -38,39 +38,36 @@ func (dec *jsonDecoder) Decode(rootYamlNode *yaml.Node) error {
 	return nil
 }

-func (dec *jsonDecoder) convertToYamlNode(data interface{}) (*yaml.Node, error) {
-	switch data := data.(type) {
-	case nil:
-		return createScalarNode(nil, "null"), nil
-	case float64, float32:
-		// json decoder returns ints as float.
-		return parseSnippet(fmt.Sprintf("%v", data))
-	case int, int64, int32, string, bool:
-		return createScalarNode(data, fmt.Sprintf("%v", data)), nil
-	case map[string]interface{}:
-		return dec.parseMap(data)
-	case []interface{}:
-		return dec.parseArray(data)
-	default:
-		return nil, fmt.Errorf("unrecognised type :(")
+func (dec *jsonDecoder) convertToYamlNode(data *orderedMap) (*yaml.Node, error) {
+	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)
+		}
 	}
-}
-
-func (dec *jsonDecoder) parseMap(dataMap map[string]interface{}) (*yaml.Node, error) {

 	var yamlMap = &yaml.Node{Kind: yaml.MappingNode}
-
-	for key, value := range dataMap {
-		yamlValue, err := dec.convertToYamlNode(value)
+	for _, keyValuePair := range data.kv {
+		yamlValue, err := dec.convertToYamlNode(&keyValuePair.V)
 		if err != nil {
 			return nil, err
 		}
-		yamlMap.Content = append(yamlMap.Content, createScalarNode(key, fmt.Sprintf("%v", key)), yamlValue)
+		yamlMap.Content = append(yamlMap.Content, createScalarNode(keyValuePair.K, keyValuePair.K), yamlValue)
 	}
 	return yamlMap, nil
+
 }

-func (dec *jsonDecoder) parseArray(dataArray []interface{}) (*yaml.Node, error) {
+func (dec *jsonDecoder) parseArray(dataArray []*orderedMap) (*yaml.Node, error) {

 	var yamlMap = &yaml.Node{Kind: yaml.SequenceNode}

@@ -79,3 +79,36 @@ will output
 cat
 ```

+## Update or create - entity exists
+This initialises `a` if it's not present
+
+Given a sample.yml file of:
+```yaml
+a: 1
+```
+then
+```bash
+yq '(.a // (.a = 0)) += 1' sample.yml
+```
+will output
+```yaml
+a: 2
+```
+
+## Update or create - entity does not exist
+This initialises `a` if it's not present
+
+Given a sample.yml file of:
+```yaml
+b: camel
+```
+then
+```bash
+yq '(.a // (.a = 0)) += 1' sample.yml
+```
+will output
+```yaml
+b: camel
+a: 1
+```
+
@@ -7,6 +7,11 @@ Which will assign the LHS node values to the RHS node values. The RHS expression

 ### relative form: `|=`
 This will do a similar thing to the plain form, however, the RHS expression is run against _the LHS nodes_. This is useful for updating values based on old values, e.g. increment.
+
+
+### Flags
+- `c` clobber custom tags
+
 {% hint style="warning" %}
 Note that versions prior to 4.18 require the 'eval/e' command to be specified.&#x20;

@@ -235,3 +240,37 @@ a:
    - bogs
 ```

+## Custom types are maintained by default
+Given a sample.yml file of:
+```yaml
+a: !cat meow
+b: !dog woof
+```
+then
+```bash
+yq '.a = .b' sample.yml
+```
+will output
+```yaml
+a: !cat woof
+b: !dog woof
+```
+
+## Custom types: clovver
+Use the `c` option to clobber custom tags
+
+Given a sample.yml file of:
+```yaml
+a: !cat meow
+b: !dog woof
+```
+then
+```bash
+yq '.a =c .b' sample.yml
+```
+will output
+```yaml
+a: !dog woof
+b: !dog woof
+```
+
@@ -1,6 +1,6 @@
 # Comment Operators

-Use these comment operators to set or retrieve comments.
+Use these comment operators to set or retrieve comments. Note that line comments on maps/arrays are actually set on the _key_ node as opposed to the _value_ (map/array). See below for examples.

 Like the `=` and `|=` assign operators, the same syntax applies when updating comments:

@@ -17,6 +17,8 @@ Note that versions prior to 4.18 require the 'eval/e' command to be specified.&#
 {% endhint %}

 ## Set line comment
+Set the comment on the key node for more reliability (see below).
+
 Given a sample.yml file of:
 ```yaml
 a: cat
@@ -30,6 +32,24 @@ will output
 a: cat # single
 ```

+## Set line comment of a maps/arrays
+For maps and arrays, you need to set the line comment on the _key_ node. This will also work for scalars.
+
+Given a sample.yml file of:
+```yaml
+a:
+  b: things
+```
+then
+```bash
+yq '(.a | key) line_comment="single"' sample.yml
+```
+will output
+```yaml
+a: # single
+  b: things
+```
+
 ## Use update assign to perform relative updates
 Given a sample.yml file of:
 ```yaml
@@ -46,6 +66,121 @@ a: cat # cat
 b: dog # dog
 ```

+## Where is the comment - map key example
+The underlying yaml parser can assign comments in a document to surprising nodes. Use an expression like this to find where you comment is. 'p' indicates the path, 'isKey' is if the node is a map key (as opposed to a map value).
+From this, you can see the 'hello-world-comment' is actually on the 'hello' key
+
+Given a sample.yml file of:
+```yaml
+hello: # hello-world-comment
+  message: world
+```
+then
+```bash
+yq '[... | {"p": path | join("."), "isKey": is_key, "hc": headComment, "lc": lineComment, "fc": footComment}]' sample.yml
+```
+will output
+```yaml
+- p: ""
+  isKey: false
+  hc: ""
+  lc: ""
+  fc: ""
+- p: hello
+  isKey: true
+  hc: ""
+  lc: hello-world-comment
+  fc: ""
+- p: hello
+  isKey: false
+  hc: ""
+  lc: ""
+  fc: ""
+- p: hello.message
+  isKey: true
+  hc: ""
+  lc: ""
+  fc: ""
+- p: hello.message
+  isKey: false
+  hc: ""
+  lc: ""
+  fc: ""
+```
+
+## Retrieve comment - map key example
+From the previous example, we know that the comment is on the 'hello' _key_ as a lineComment
+
+Given a sample.yml file of:
+```yaml
+hello: # hello-world-comment
+  message: world
+```
+then
+```bash
+yq '.hello | key | line_comment' sample.yml
+```
+will output
+```yaml
+hello-world-comment
+```
+
+## Where is the comment - array example
+The underlying yaml parser can assign comments in a document to surprising nodes. Use an expression like this to find where you comment is. 'p' indicates the path, 'isKey' is if the node is a map key (as opposed to a map value).
+From this, you can see the 'under-name-comment' is actually on the first child
+
+Given a sample.yml file of:
+```yaml
+name:
+  # under-name-comment
+  - first-array-child
+```
+then
+```bash
+yq '[... | {"p": path | join("."), "isKey": is_key, "hc": headComment, "lc": lineComment, "fc": footComment}]' sample.yml
+```
+will output
+```yaml
+- p: ""
+  isKey: false
+  hc: ""
+  lc: ""
+  fc: ""
+- p: name
+  isKey: true
+  hc: ""
+  lc: ""
+  fc: ""
+- p: name
+  isKey: false
+  hc: ""
+  lc: ""
+  fc: ""
+- p: name.0
+  isKey: false
+  hc: under-name-comment
+  lc: ""
+  fc: ""
+```
+
+## Retrieve comment - array example
+From the previous example, we know that the comment is on the first child as a headComment
+
+Given a sample.yml file of:
+```yaml
+name:
+  # under-name-comment
+  - first-array-child
+```
+then
+```bash
+yq '.name[0] | headComment' sample.yml
+```
+will output
+```yaml
+under-name-comment
+```
+
 ## Set head comment
 Given a sample.yml file of:
 ```yaml
@@ -62,6 +197,25 @@ will output
 a: cat
 ```

+## Set head comment of a map entry
+Given a sample.yml file of:
+```yaml
+f: foo
+a:
+  b: cat
+```
+then
+```bash
+yq '(.a | key) head_comment="single"' sample.yml
+```
+will output
+```yaml
+f: foo
+# single
+a:
+  b: cat
+```
+
 ## Set foot comment, using an expression
 Given a sample.yml file of:
 ```yaml
@@ -9,16 +9,16 @@ These operators are useful to process yaml documents that have stringified embed

 | Format | Decode (from string) | Encode (to string) |
 | --- | -- | --|
-| Yaml | from_yaml | to_yaml(i)/@yaml |
-| JSON | from_json | to_json(i)/@json |
-| Properties | from_props  | to_props/@props |
-| CSV |  | to_csv/@csv |
-| TSV |  | to_tsv/@tsv |
-| XML | from_xml | to_xml(i)/@xml |
+| Yaml | from_yaml/@yamld | to_yaml(i)/@yaml |
+| JSON | from_json/@jsond | to_json(i)/@json |
+| Properties | from_props/@propsd  | to_props/@props |
+| CSV | from_csv/@csvd | to_csv/@csv |
+| TSV | from_tsv/@tsvd | to_tsv/@tsv |
+| XML | from_xml/@xmld | to_xml(i)/@xml |
 | Base64 | @base64d | @base64 |


-CSV and TSV format both accept either a single array or scalars (representing a single row), or an array of array of scalars (representing multiple rows). 
+See CSV and TSV [documentation](https://mikefarah.gitbook.io/yq/usage/csv-tsv) for accepted formats.

 XML uses the `--xml-attribute-prefix` and `xml-content-name` flags to identify attributes and content fields.

@@ -132,7 +132,7 @@ a: |-
 ```
 then
 ```bash
-yq '.a |= from_props' sample.yml
+yq '.a |= @propsd' sample.yml
 ```
 will output
 ```yaml
@@ -141,6 +141,42 @@ a:
  dogs: cool as well
 ```

+## Decode csv encoded string
+Given a sample.yml file of:
+```yaml
+a: |-
+  cats,dogs
+  great,cool as well
+```
+then
+```bash
+yq '.a |= @csvd' sample.yml
+```
+will output
+```yaml
+a:
+  - cats: great
+    dogs: cool as well
+```
+
+## Decode tsv encoded string
+Given a sample.yml file of:
+```yaml
+a: |-
+  cats	dogs
+  great	cool as well
+```
+then
+```bash
+yq '.a |= @tsvd' sample.yml
+```
+will output
+```yaml
+a:
+  - cats: great
+    dogs: cool as well
+```
+
 ## Encode value as yaml string
 Indent defaults to 2

@@ -6,4 +6,8 @@ This operator is used to update node values. It can be used in either the:
 Which will assign the LHS node values to the RHS node values. The RHS expression is run against the matching nodes in the pipeline.

 ### relative form: `|=`
-This will do a similar thing to the plain form, however, the RHS expression is run against _the LHS nodes_. This is useful for updating values based on old values, e.g. increment.
+This will do a similar thing to the plain form, however, the RHS expression is run against _the LHS nodes_. This is useful for updating values based on old values, e.g. increment.
+
+
+### Flags
+- `c` clobber custom tags
@@ -1,6 +1,6 @@
 # Comment Operators

-Use these comment operators to set or retrieve comments.
+Use these comment operators to set or retrieve comments. Note that line comments on maps/arrays are actually set on the _key_ node as opposed to the _value_ (map/array). See below for examples.

 Like the `=` and `|=` assign operators, the same syntax applies when updating comments:

@@ -9,16 +9,16 @@ These operators are useful to process yaml documents that have stringified embed

 | Format | Decode (from string) | Encode (to string) |
 | --- | -- | --|
-| Yaml | from_yaml | to_yaml(i)/@yaml |
-| JSON | from_json | to_json(i)/@json |
-| Properties | from_props  | to_props/@props |
-| CSV |  | to_csv/@csv |
-| TSV |  | to_tsv/@tsv |
-| XML | from_xml | to_xml(i)/@xml |
+| Yaml | from_yaml/@yamld | to_yaml(i)/@yaml |
+| JSON | from_json/@jsond | to_json(i)/@json |
+| Properties | from_props/@propsd  | to_props/@props |
+| CSV | from_csv/@csvd | to_csv/@csv |
+| TSV | from_tsv/@tsvd | to_tsv/@tsv |
+| XML | from_xml/@xmld | to_xml(i)/@xml |
 | Base64 | @base64d | @base64 |


-CSV and TSV format both accept either a single array or scalars (representing a single row), or an array of array of scalars (representing multiple rows). 
+See CSV and TSV [documentation](https://mikefarah.gitbook.io/yq/usage/csv-tsv) for accepted formats.

 XML uses the `--xml-attribute-prefix` and `xml-content-name` flags to identify attributes and content fields.

@@ -14,6 +14,7 @@ You can control how objects are merged by using one or more of the following fla
 - `d` deeply merge arrays
 - `?` only merge _existing_ fields
 - `n` only merge _new_ fields
+- `c` clobber custom tags


 ### Merge two files together
@@ -1,5 +1,8 @@
 # Path

-The path operator can be used to get the traversal paths of matching nodes in an expression. The path is returned as an array, which if traversed in order will lead to the matching node.
+The `path` operator can be used to get the traversal paths of matching nodes in an expression. The path is returned as an array, which if traversed in order will lead to the matching node.

 You can get the key/index of matching nodes by using the `path` operator to return the path array then piping that through `.[-1]` to get the last element of that array, the key.
+
+Use `setpath` to set a value to the path array returned by `path`, and similarly `delpaths` for an array of path arrays.
+
@@ -118,3 +118,43 @@ will output
 comment on key
 ```

+## Check node is a key
+Given a sample.yml file of:
+```yaml
+a:
+  b:
+    - cat
+  c: frog
+```
+then
+```bash
+yq '[... | { "p": path | join("."), "isKey": is_key, "tag": tag }]' sample.yml
+```
+will output
+```yaml
+- p: ""
+  isKey: false
+  tag: '!!map'
+- p: a
+  isKey: true
+  tag: '!!str'
+- p: a
+  isKey: false
+  tag: '!!map'
+- p: a.b
+  isKey: true
+  tag: '!!str'
+- p: a.b
+  isKey: false
+  tag: '!!seq'
+- p: a.b.0
+  isKey: false
+  tag: '!!str'
+- p: a.c
+  isKey: true
+  tag: '!!str'
+- p: a.c
+  isKey: false
+  tag: '!!str'
+```
+
@@ -14,6 +14,7 @@ You can control how objects are merged by using one or more of the following fla
 - `d` deeply merge arrays
 - `?` only merge _existing_ fields
 - `n` only merge _new_ fields
+- `c` clobber custom tags


 ### Merge two files together
@@ -481,3 +482,26 @@ b: !goat
  dog: woof
 ```

+## Custom types: clobber tags
+Use the `c` option to clobber custom tags. Note that the second tag is now used
+
+Given a sample.yml file of:
+```yaml
+a: !horse
+  cat: meow
+b: !goat
+  dog: woof
+```
+then
+```bash
+yq '.a *=c .b' sample.yml
+```
+will output
+```yaml
+a: !goat
+  cat: meow
+  dog: woof
+b: !goat
+  dog: woof
+```
+
@@ -1,9 +1,12 @@
 # Path

-The path operator can be used to get the traversal paths of matching nodes in an expression. The path is returned as an array, which if traversed in order will lead to the matching node.
+The `path` operator can be used to get the traversal paths of matching nodes in an expression. The path is returned as an array, which if traversed in order will lead to the matching node.

 You can get the key/index of matching nodes by using the `path` operator to return the path array then piping that through `.[-1]` to get the last element of that array, the key.

+Use `setpath` to set a value to the path array returned by `path`, and similarly `delpaths` for an array of path arrays.
+
+
 {% hint style="warning" %}
 Note that versions prior to 4.18 require the 'eval/e' command to be specified.&#x20;

@@ -98,3 +101,114 @@ will output
  value: frog
 ```

+## Set path
+Given a sample.yml file of:
+```yaml
+a:
+  b: cat
+```
+then
+```bash
+yq 'setpath(["a", "b"]; "things")' sample.yml
+```
+will output
+```yaml
+a:
+  b: things
+```
+
+## Set on empty document
+Running
+```bash
+yq --null-input 'setpath(["a", "b"]; "things")'
+```
+will output
+```yaml
+a:
+  b: things
+```
+
+## Set array path
+Given a sample.yml file of:
+```yaml
+a:
+  - cat
+  - frog
+```
+then
+```bash
+yq 'setpath(["a", 0]; "things")' sample.yml
+```
+will output
+```yaml
+a:
+  - things
+  - frog
+```
+
+## Set array path empty
+Running
+```bash
+yq --null-input 'setpath(["a", 0]; "things")'
+```
+will output
+```yaml
+a:
+  - things
+```
+
+## Delete path
+Notice delpaths takes an _array_ of paths.
+
+Given a sample.yml file of:
+```yaml
+a:
+  b: cat
+  c: dog
+  d: frog
+```
+then
+```bash
+yq 'delpaths([["a", "c"], ["a", "d"]])' sample.yml
+```
+will output
+```yaml
+a:
+  b: cat
+```
+
+## Delete array path
+Given a sample.yml file of:
+```yaml
+a:
+  - cat
+  - frog
+```
+then
+```bash
+yq 'delpaths([["a", 0]])' sample.yml
+```
+will output
+```yaml
+a:
+  - frog
+```
+
+## Delete - wrong parameter
+delpaths does not work with a single path array
+
+Given a sample.yml file of:
+```yaml
+a:
+  - cat
+  - frog
+```
+then
+```bash
+yq 'delpaths(["a", 0])' sample.yml
+```
+will output
+```bash
+Error: DELPATHS: expected entry [0] to be a sequence, but its a !!str. Note that delpaths takes an array of path arrays, e.g. [["a", "b"]]
+```
+
@@ -112,6 +112,26 @@ will output
 cat; meow; 1; ; true
 ```

+## Trim strings
+Given a sample.yml file of:
+```yaml
+- ' cat'
+- 'dog '
+- ' cow cow '
+- horse
+```
+then
+```bash
+yq '.[] | trim' sample.yml
+```
+will output
+```yaml
+cat
+dog
+cow cow
+horse
+```
+
 ## Match string
 Given a sample.yml file of:
 ```yaml
@@ -1,10 +1,9 @@
 # JSON

-Encode and decode to and from JSON. Note that, unless you have multiple JSON documents in a single file, YAML is a _superset_ of JSON - so `yq` can read any json file without doing anything special.
+Encode and decode to and from JSON. Supports multiple JSON documents in a single file (e.g. NDJSON).

-If you do have mulitple JSON documents in a single file (e.g. NDJSON) then you will need to explicity use the json parser `-p=json`.
+Note that YAML is a superset of (single document) JSON - so you don't have to use the JSON parser to read JSON when there is only one JSON document in the input. You will probably want to pretty print the result in this case, to get idiomatic YAML styling.

-This means you don't need to 'convert' a JSON file to YAML - however if you want idiomatic YAML styling, then you can use the `-P/--prettyPrint` flag, see examples below.

 {% hint style="warning" %}
 Note that versions prior to 4.18 require the 'eval/e' command to be specified.&#x20;
@@ -0,0 +1,214 @@
+# CSV
+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:
+
+```yaml
+- name: Bobo
+  type: dog
+- name: Fifi
+  type: cat
+```
+
+As well as arrays of arrays of scalars (strings/numbers/booleans):
+
+```yaml
+- [Bobo, dog]
+- [Fifi, cat]
+```
+
+## Decode
+Decode assumes the first CSV/TSV row is the header row, and all rows beneath are the entries.
+The data will be coded into an array of objects, using the header rows as keys.
+
+```csv
+name,type
+Bobo,dog
+Fifi,cat
+```
+
+
+{% hint style="warning" %}
+Note that versions prior to 4.18 require the 'eval/e' command to be specified.&#x20;
+
+`yq e <exp> <file>`
+{% endhint %}
+
+## Encode CSV simple
+Given a sample.yml file of:
+```yaml
+- [i, like, csv]
+- [because, excel, is, cool]
+```
+then
+```bash
+yq -o=csv sample.yml
+```
+will output
+```csv
+i,like,csv
+because,excel,is,cool
+```
+
+## Encode TSV simple
+Given a sample.yml file of:
+```yaml
+- [i, like, csv]
+- [because, excel, is, cool]
+```
+then
+```bash
+yq -o=tsv sample.yml
+```
+will output
+```tsv
+i	like	csv
+because	excel	is	cool
+```
+
+## Encode array of objects to csv
+Given a sample.yml file of:
+```yaml
+- name: Gary
+  numberOfCats: 1
+  likesApples: true
+  height: 168.8
+- name: Samantha's Rabbit
+  numberOfCats: 2
+  likesApples: false
+  height: -188.8
+
+```
+then
+```bash
+yq -o=csv sample.yml
+```
+will output
+```csv
+name,numberOfCats,likesApples,height
+Gary,1,true,168.8
+Samantha's Rabbit,2,false,-188.8
+```
+
+## Encode array of objects to custom csv format
+Add the header row manually, then the we convert each object into an array of values - resulting in an array of arrays. Pick the columns and call the header whatever you like.
+
+Given a sample.yml file of:
+```yaml
+- name: Gary
+  numberOfCats: 1
+  likesApples: true
+  height: 168.8
+- name: Samantha's Rabbit
+  numberOfCats: 2
+  likesApples: false
+  height: -188.8
+
+```
+then
+```bash
+yq -o=csv '[["Name", "Number of Cats"]] +  [.[] | [.name, .numberOfCats ]]' sample.yml
+```
+will output
+```csv
+Name,Number of Cats
+Gary,1
+Samantha's Rabbit,2
+```
+
+## Encode array of objects to csv - missing fields behaviour
+First entry is used to determine the headers, and it is missing 'likesApples', so it is not included in the csv. Second entry does not have 'numberOfCats' so that is blank
+
+Given a sample.yml file of:
+```yaml
+- name: Gary
+  numberOfCats: 1
+  height: 168.8
+- name: Samantha's Rabbit
+  height: -188.8
+  likesApples: false
+
+```
+then
+```bash
+yq -o=csv sample.yml
+```
+will output
+```csv
+name,numberOfCats,height
+Gary,1,168.8
+Samantha's Rabbit,,-188.8
+```
+
+## Parse CSV into an array of objects
+First row is assumed to be the header row.
+
+Given a sample.csv file of:
+```csv
+name,numberOfCats,likesApples,height
+Gary,1,true,168.8
+Samantha's Rabbit,2,false,-188.8
+
+```
+then
+```bash
+yq -p=csv sample.csv
+```
+will output
+```yaml
+- name: Gary
+  numberOfCats: 1
+  likesApples: true
+  height: 168.8
+- name: Samantha's Rabbit
+  numberOfCats: 2
+  likesApples: false
+  height: -188.8
+```
+
+## Parse TSV into an array of objects
+First row is assumed to be the header row.
+
+Given a sample.tsv file of:
+```tsv
+name	numberOfCats	likesApples	height
+Gary	1	true	168.8
+Samantha's Rabbit	2	false	-188.8
+
+```
+then
+```bash
+yq -p=tsv sample.tsv
+```
+will output
+```yaml
+- name: Gary
+  numberOfCats: 1
+  likesApples: true
+  height: 168.8
+- name: Samantha's Rabbit
+  numberOfCats: 2
+  likesApples: false
+  height: -188.8
+```
+
+## Round trip
+Given a sample.csv file of:
+```csv
+name,numberOfCats,likesApples,height
+Gary,1,true,168.8
+Samantha's Rabbit,2,false,-188.8
+
+```
+then
+```bash
+yq -p=csv -o=csv '(.[] | select(.name == "Gary") | .numberOfCats) = 3' sample.csv
+```
+will output
+```csv
+name,numberOfCats,likesApples,height
+Gary,3,true,168.8
+Samantha's Rabbit,2,false,-188.8
+```
+
@@ -1,7 +1,6 @@
 # JSON

-Encode and decode to and from JSON. Note that, unless you have multiple JSON documents in a single file, YAML is a _superset_ of JSON - so `yq` can read any json file without doing anything special.
+Encode and decode to and from JSON. Supports multiple JSON documents in a single file (e.g. NDJSON).

-If you do have mulitple JSON documents in a single file (e.g. NDJSON) then you will need to explicity use the json parser `-p=json`.
+Note that YAML is a superset of (single document) JSON - so you don't have to use the JSON parser to read JSON when there is only one JSON document in the input. You will probably want to pretty print the result in this case, to get idiomatic YAML styling.

-This means you don't need to 'convert' a JSON file to YAML - however if you want idiomatic YAML styling, then you can use the `-P/--prettyPrint` flag, see examples below.
@@ -0,0 +1,30 @@
+# CSV
+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:
+
+```yaml
+- name: Bobo
+  type: dog
+- name: Fifi
+  type: cat
+```
+
+As well as arrays of arrays of scalars (strings/numbers/booleans):
+
+```yaml
+- [Bobo, dog]
+- [Fifi, cat]
+```
+
+## Decode
+Decode assumes the first CSV/TSV row is the header row, and all rows beneath are the entries.
+The data will be coded into an array of objects, using the header rows as keys.
+
+```csv
+name,type
+Bobo,dog
+Fifi,cat
+```
+
@@ -1,5 +1,5 @@
 # Properties

-Encode to a property file (decode not yet supported). Line comments on value nodes will be copied across.
+Encode/Decode/Roundtrip to/from a property file. Line comments on value nodes will be copied across.

 By default, empty maps and arrays are not encoded - see below for an example on how to encode a value for these.
@@ -1,6 +1,6 @@
 # Properties

-Encode to a property file (decode not yet supported). Line comments on value nodes will be copied across.
+Encode/Decode/Roundtrip to/from a property file. Line comments on value nodes will be copied across.

 By default, empty maps and arrays are not encoded - see below for an example on how to encode a value for these.

@@ -66,8 +66,13 @@ func (o *orderedMap) UnmarshalJSON(data []byte) error {
 		}
 		return nil
 	case '[':
-		var arr []orderedMap
-		return json.Unmarshal(data, &arr)
+		var res []*orderedMap
+		if err := json.Unmarshal(data, &res); err != nil {
+			return err
+		}
+		o.altVal = res
+		o.kv = nil
+		return nil
 	}

 	return json.Unmarshal(data, &o.altVal)
@@ -13,7 +13,7 @@ type csvEncoder struct {
 }

 func NewCsvEncoder(separator rune) Encoder {
-	return &csvEncoder{separator}
+	return &csvEncoder{separator: separator}
 }

 func (e *csvEncoder) CanHandleAliases() bool {
@@ -41,6 +41,67 @@ func (e *csvEncoder) encodeRow(csvWriter *csv.Writer, contents []*yaml.Node) err
 	return csvWriter.Write(stringValues)
 }

+func (e *csvEncoder) encodeArrays(csvWriter *csv.Writer, content []*yaml.Node) error {
+	for i, child := range content {
+
+		if child.Kind != yaml.SequenceNode {
+			return fmt.Errorf("csv encoding only works for arrays of scalars (string/numbers/booleans), child[%v] is a %v", i, child.Tag)
+		}
+		err := e.encodeRow(csvWriter, child.Content)
+		if err != nil {
+			return err
+		}
+	}
+	return nil
+}
+
+func (e *csvEncoder) extractHeader(child *yaml.Node) ([]*yaml.Node, error) {
+	if child.Kind != yaml.MappingNode {
+		return nil, fmt.Errorf("csv object encoding only works for arrays of flat objects (string key => string/numbers/boolean value), child[0] is a %v", child.Tag)
+	}
+	mapKeys := getMapKeys(child)
+	return mapKeys.Content, nil
+}
+
+func (e *csvEncoder) createChildRow(child *yaml.Node, headers []*yaml.Node) []*yaml.Node {
+	childRow := make([]*yaml.Node, 0)
+	for _, header := range headers {
+		keyIndex := findKeyInMap(child, header)
+		value := createScalarNode(nil, "")
+		if keyIndex != -1 {
+			value = child.Content[keyIndex+1]
+		}
+		childRow = append(childRow, value)
+	}
+	return childRow
+
+}
+
+func (e *csvEncoder) encodeObjects(csvWriter *csv.Writer, content []*yaml.Node) error {
+	headers, err := e.extractHeader(content[0])
+	if err != nil {
+		return nil
+	}
+
+	err = e.encodeRow(csvWriter, headers)
+	if err != nil {
+		return nil
+	}
+
+	for i, child := range content {
+		if child.Kind != yaml.MappingNode {
+			return fmt.Errorf("csv object encoding only works for arrays of flat objects (string key => string/numbers/boolean value), child[%v] is a %v", i, child.Tag)
+		}
+		row := e.createChildRow(child, headers)
+		err = e.encodeRow(csvWriter, row)
+		if err != nil {
+			return err
+		}
+
+	}
+	return nil
+}
+
 func (e *csvEncoder) Encode(writer io.Writer, originalNode *yaml.Node) error {
 	csvWriter := csv.NewWriter(writer)
 	csvWriter.Comma = e.separator
@@ -56,15 +117,10 @@ func (e *csvEncoder) Encode(writer io.Writer, originalNode *yaml.Node) error {
 		return e.encodeRow(csvWriter, node.Content)
 	}

-	for i, child := range node.Content {
-
-		if child.Kind != yaml.SequenceNode {
-			return fmt.Errorf("csv encoding only works for arrays of scalars (string/numbers/booleans), child[%v] is a %v", i, child.Tag)
-		}
-		err := e.encodeRow(csvWriter, child.Content)
-		if err != nil {
-			return err
-		}
+	if node.Content[0].Kind == yaml.MappingNode {
+		return e.encodeObjects(csvWriter, node.Content)
 	}
-	return nil
+
+	return e.encodeArrays(csvWriter, node.Content)
+
 }
@@ -1,60 +0,0 @@
-package yqlib
-
-import (
-	"bufio"
-	"bytes"
-	"strings"
-	"testing"
-
-	"github.com/mikefarah/yq/v4/test"
-)
-
-func yamlToCsv(sampleYaml string, separator rune) string {
-	var output bytes.Buffer
-	writer := bufio.NewWriter(&output)
-
-	var jsonEncoder = NewCsvEncoder(separator)
-	inputs, err := readDocuments(strings.NewReader(sampleYaml), "sample.yml", 0, NewYamlDecoder())
-	if err != nil {
-		panic(err)
-	}
-	node := inputs.Front().Value.(*CandidateNode).Node
-	err = jsonEncoder.Encode(writer, node)
-	if err != nil {
-		panic(err)
-	}
-	writer.Flush()
-
-	return strings.TrimSuffix(output.String(), "\n")
-}
-
-var sampleYaml = `["apple", apple2, "comma, in, value", "new
-line", 3, 3.40, true, "tab	here"]`
-
-var sampleYamlArray = "[" + sampleYaml + ", [bob, cat, meow, puss]]"
-
-func TestCsvEncoderEmptyArray(t *testing.T) {
-	var actualCsv = yamlToCsv(`[]`, ',')
-	test.AssertResult(t, "", actualCsv)
-}
-
-func TestCsvEncoder(t *testing.T) {
-	var expectedCsv = `apple,apple2,"comma, in, value",new line,3,3.40,true,tab	here`
-
-	var actualCsv = yamlToCsv(sampleYaml, ',')
-	test.AssertResult(t, expectedCsv, actualCsv)
-}
-
-func TestCsvEncoderArrayOfArrays(t *testing.T) {
-	var actualCsv = yamlToCsv(sampleYamlArray, ',')
-	var expectedCsv = "apple,apple2,\"comma, in, value\",new line,3,3.40,true,tab	here\nbob,cat,meow,puss"
-	test.AssertResult(t, expectedCsv, actualCsv)
-}
-
-func TestTsvEncoder(t *testing.T) {
-
-	var expectedCsv = `apple	apple2	comma, in, value	new line	3	3.40	true	"tab	here"`
-
-	var actualCsv = yamlToCsv(sampleYaml, '\t')
-	test.AssertResult(t, expectedCsv, actualCsv)
-}
@@ -22,6 +22,13 @@ const sampleNdJson = `{"this": "is a multidoc json file"}
 {"a number": 4}
 `

+const sampleNdJsonKey = `{"a": "first", "b": "next", "ab": "last"}`
+
+const expectedJsonKeysInOrder = `a: first
+b: next
+ab: last
+`
+
 const expectedNdJsonYaml = `this: is a multidoc json file
 ---
 each:
@@ -157,6 +164,13 @@ var jsonScenarios = []formatScenario{
 		expected:     expectedNdJsonYaml,
 		scenarioType: "decode-ndjson",
 	},
+	{
+		description:  "Decode NDJSON, maintain key order",
+		skipDoc:      true,
+		input:        sampleNdJsonKey,
+		expected:     expectedJsonKeysInOrder,
+		scenarioType: "decode-ndjson",
+	},
 	{
 		description:  "numbers",
 		skipDoc:      true,
@@ -164,6 +178,20 @@ var jsonScenarios = []formatScenario{
 		expected:     "- 3\n- 3\n- 3.1\n- -1\n",
 		scenarioType: "decode-ndjson",
 	},
+	{
+		description:  "number single",
+		skipDoc:      true,
+		input:        "3",
+		expected:     "3\n",
+		scenarioType: "decode-ndjson",
+	},
+	{
+		description:  "empty string",
+		skipDoc:      true,
+		input:        `""`,
+		expected:     "\"\"\n",
+		scenarioType: "decode-ndjson",
+	},
 	{
 		description:  "strings",
 		skipDoc:      true,
@@ -67,7 +67,10 @@ var participleYqRules = []*participleYqRule{
 	{"XMLEncode", `to_?xml`, encodeWithIndent(XMLOutputFormat, 2), 0},
 	{"XMLEncodeNoIndent", `@xml`, encodeWithIndent(XMLOutputFormat, 0), 0},

+	{"CSVDecode", `from_?csv|@csvd`, decodeOp(CSVObjectInputFormat), 0},
 	{"CSVEncode", `to_?csv|@csv`, encodeWithIndent(CSVOutputFormat, 0), 0},
+
+	{"TSVDecode", `from_?tsv|@tsvd`, decodeOp(TSVObjectInputFormat), 0},
 	{"TSVEncode", `to_?tsv|@tsv`, encodeWithIndent(TSVOutputFormat, 0), 0},

 	{"Base64d", `@base64d`, decodeOp(Base64InputFormat), 0},
@@ -120,10 +123,13 @@ var participleYqRules = []*participleYqRule{

 	simpleOp("keys", keysOpType),
 	simpleOp("key", getKeyOpType),
+	simpleOp("is_?key", isKeyOpType),

 	simpleOp("file_?name|fileName", getFilenameOpType),
 	simpleOp("file_?index|fileIndex|fi", getFileIndexOpType),
 	simpleOp("path", getPathOpType),
+	simpleOp("set_?path", setPathOpType),
+	simpleOp("del_?paths", delPathsOpType),

 	simpleOp("to_?entries|toEntries", toEntriesOpType),
 	simpleOp("from_?entries|fromEntries", fromEntriesOpType),
@@ -149,6 +155,7 @@ var participleYqRules = []*participleYqRule{

 	{"Uppercase", `upcase|ascii_?upcase`, opTokenWithPrefs(changeCaseOpType, nil, changeCasePrefs{ToUpperCase: true}), 0},
 	{"Downcase", `downcase|ascii_?downcase`, opTokenWithPrefs(changeCaseOpType, nil, changeCasePrefs{ToUpperCase: false}), 0},
+	simpleOp("trim", trimOpType),

 	{"HexValue", `0[xX][0-9A-Fa-f]+`, hexValue(), 0},
 	{"FloatValueScientific", `-?[1-9](\.\d+)?[Ee][-+]?\d+`, floatValue(), 0},
@@ -178,8 +185,8 @@ var participleYqRules = []*participleYqRule{
 	{"GreaterThan", `\s*>\s*`, opTokenWithPrefs(compareOpType, nil, compareTypePref{OrEqual: false, Greater: true}), 0},
 	{"LessThan", `\s*<\s*`, opTokenWithPrefs(compareOpType, nil, compareTypePref{OrEqual: false, Greater: false}), 0},

-	{"AssignRelative", `\|=`, assignOpToken(true), 0},
-	{"Assign", `=`, assignOpToken(false), 0},
+	{"AssignRelative", `\|=[c]*`, assignOpToken(true), 0},
+	{"Assign", `=[c]*`, assignOpToken(false), 0},

 	{`whitespace`, `[ \t\n]+`, nil, 0},

@@ -190,8 +197,8 @@ var participleYqRules = []*participleYqRule{

 	{"Union", `,`, opToken(unionOpType), 0},

-	{"MultiplyAssign", `\*=[\+|\?dn]*`, multiplyWithPrefs(multiplyAssignOpType), 0},
-	{"Multiply", `\*[\+|\?dn]*`, multiplyWithPrefs(multiplyOpType), 0},
+	{"MultiplyAssign", `\*=[\+|\?cdn]*`, multiplyWithPrefs(multiplyAssignOpType), 0},
+	{"Multiply", `\*[\+|\?cdn]*`, multiplyWithPrefs(multiplyOpType), 0},

 	{"AddAssign", `\+=`, opToken(addAssignOpType), 0},
 	{"Add", `\+`, opToken(addOpType), 0},
@@ -313,6 +320,9 @@ func assignOpToken(updateAssign bool) yqAction {
 		log.Debug("assignOpToken %v", rawToken.Value)
 		value := rawToken.Value
 		prefs := assignPreferences{DontOverWriteAnchor: true}
+		if strings.Contains(value, "c") {
+			prefs.ClobberCustomTags = true
+		}
 		op := &Operation{OperationType: assignOpType, Value: assignOpType.Type, StringValue: value, UpdateAssign: updateAssign, Preferences: prefs}
 		return &token{TokenType: operationToken, Operation: op}, nil
 	}
@@ -383,6 +393,7 @@ func envSubstWithOptions() yqAction {
 func multiplyWithPrefs(op *operationType) yqAction {
 	return func(rawToken lexer.Token) (*token, error) {
 		prefs := multiplyPreferences{}
+		prefs.AssignPrefs = assignPreferences{}
 		options := rawToken.Value
 		if strings.Contains(options, "+") {
 			prefs.AppendArrays = true
@@ -391,11 +402,14 @@ func multiplyWithPrefs(op *operationType) yqAction {
 			prefs.TraversePrefs = traversePreferences{DontAutoCreate: true}
 		}
 		if strings.Contains(options, "n") {
-			prefs.AssignPrefs = assignPreferences{OnlyWriteNull: true}
+			prefs.AssignPrefs.OnlyWriteNull = true
 		}
 		if strings.Contains(options, "d") {
 			prefs.DeepMergeArrays = true
 		}
+		if strings.Contains(options, "c") {
+			prefs.AssignPrefs.ClobberCustomTags = true
+		}
 		prefs.TraversePrefs.DontFollowAlias = true
 		op := &Operation{OperationType: op, Value: multiplyOpType.Type, StringValue: options, Preferences: prefs}
 		return &token{TokenType: operationToken, Operation: op}, nil
@@ -523,7 +523,6 @@ var participleLexerScenarios = []participleLexerScenario{
 }

 func TestParticipleLexer(t *testing.T) {
-	log.Errorf("TestParticiple")
 	lexer := newParticipleLexer()

 	for _, scenario := range participleLexerScenarios {
@@ -1,5 +1,4 @@
 // Use the top level Evaluator or StreamEvaluator to evaluate expressions and return matches.
-//
 package yqlib

 import (
@@ -80,7 +79,7 @@ var notEqualsOpType = &operationType{Type: "NOT_EQUALS", NumArgs: 2, Precedence:

 var compareOpType = &operationType{Type: "COMPARE", NumArgs: 2, Precedence: 40, Handler: compareOperator}

-//createmap needs to be above union, as we use union to build the components of the objects
+// createmap needs to be above union, as we use union to build the components of the objects
 var createMapOpType = &operationType{Type: "CREATE_MAP", NumArgs: 2, Precedence: 15, Handler: createMapOperator}

 var shortPipeOpType = &operationType{Type: "SHORT_PIPE", NumArgs: 2, Precedence: 45, Handler: pipeOperator}
@@ -122,6 +121,7 @@ var getStyleOpType = &operationType{Type: "GET_STYLE", NumArgs: 0, Precedence: 5
 var getTagOpType = &operationType{Type: "GET_TAG", NumArgs: 0, Precedence: 50, Handler: getTagOperator}

 var getKeyOpType = &operationType{Type: "GET_KEY", NumArgs: 0, Precedence: 50, Handler: getKeyOperator}
+var isKeyOpType = &operationType{Type: "IS_KEY", NumArgs: 0, Precedence: 50, Handler: isKeyOperator}
 var getParentOpType = &operationType{Type: "GET_PARENT", NumArgs: 0, Precedence: 50, Handler: getParentOperator}

 var getCommentOpType = &operationType{Type: "GET_COMMENT", NumArgs: 0, Precedence: 50, Handler: getCommentsOperator}
@@ -130,7 +130,10 @@ var getAliasOpType = &operationType{Type: "GET_ALIAS", NumArgs: 0, Precedence: 5
 var getDocumentIndexOpType = &operationType{Type: "GET_DOCUMENT_INDEX", NumArgs: 0, Precedence: 50, Handler: getDocumentIndexOperator}
 var getFilenameOpType = &operationType{Type: "GET_FILENAME", NumArgs: 0, Precedence: 50, Handler: getFilenameOperator}
 var getFileIndexOpType = &operationType{Type: "GET_FILE_INDEX", NumArgs: 0, Precedence: 50, Handler: getFileIndexOperator}
+
 var getPathOpType = &operationType{Type: "GET_PATH", NumArgs: 0, Precedence: 50, Handler: getPathOperator}
+var setPathOpType = &operationType{Type: "SET_PATH", NumArgs: 1, Precedence: 50, Handler: setPathOperator}
+var delPathsOpType = &operationType{Type: "DEL_PATHS", NumArgs: 1, Precedence: 50, Handler: delPathsOperator}

 var explodeOpType = &operationType{Type: "EXPLODE", NumArgs: 1, Precedence: 50, Handler: explodeOperator}
 var sortByOpType = &operationType{Type: "SORT_BY", NumArgs: 1, Precedence: 50, Handler: sortByOperator}
@@ -146,6 +149,7 @@ var captureOpType = &operationType{Type: "CAPTURE", NumArgs: 1, Precedence: 50,
 var testOpType = &operationType{Type: "TEST", NumArgs: 1, Precedence: 50, Handler: testOperator}
 var splitStringOpType = &operationType{Type: "SPLIT", NumArgs: 1, Precedence: 50, Handler: splitStringOperator}
 var changeCaseOpType = &operationType{Type: "CHANGE_CASE", NumArgs: 0, Precedence: 50, Handler: changeCaseOperator}
+var trimOpType = &operationType{Type: "TRIM", NumArgs: 0, Precedence: 50, Handler: trimSpaceOperator}

 var loadOpType = &operationType{Type: "LOAD", NumArgs: 1, Precedence: 50, Handler: loadYamlOperator}

@@ -205,10 +209,10 @@ func findInArray(array *yaml.Node, item *yaml.Node) int {
 	return -1
 }

-func findKeyInMap(array *yaml.Node, item *yaml.Node) int {
+func findKeyInMap(dataMap *yaml.Node, item *yaml.Node) int {

-	for index := 0; index < len(array.Content); index = index + 2 {
-		if recursiveNodeEqual(array.Content[index], item) {
+	for index := 0; index < len(dataMap.Content); index = index + 2 {
+		if recursiveNodeEqual(dataMap.Content[index], item) {
 			return index
 		}
 	}
@@ -430,7 +434,7 @@ func (p *Operation) toString() string {
 	}
 }

-//use for debugging only
+// use for debugging only
 func NodesToString(collection *list.List) string {
 	if !log.IsEnabledFor(logging.DEBUG) {
 		return ""
@@ -62,6 +62,9 @@ func add(d *dataTreeNavigator, context Context, lhs *CandidateNode, rhs *Candida

 	switch lhsNode.Kind {
 	case yaml.MappingNode:
+		if rhs.Node.Kind != yaml.MappingNode {
+			return nil, fmt.Errorf("%v (%v) cannot be added to a %v (%v)", rhs.Node.Tag, rhs.GetNicePath(), lhsNode.Tag, lhs.GetNicePath())
+		}
 		addMaps(target, lhs, rhs)
 	case yaml.SequenceNode:
 		if err := addSequences(target, lhs, rhs); err != nil {
@@ -70,7 +73,7 @@ func add(d *dataTreeNavigator, context Context, lhs *CandidateNode, rhs *Candida

 	case yaml.ScalarNode:
 		if rhs.Node.Kind != yaml.ScalarNode {
-			return nil, fmt.Errorf("%v (%v) cannot be added to a %v", rhs.Node.Tag, rhs.Path, lhsNode.Tag)
+			return nil, fmt.Errorf("%v (%v) cannot be added to a %v (%v)", rhs.Node.Tag, rhs.GetNicePath(), lhsNode.Tag, lhs.GetNicePath())
 		}
 		target.Node.Kind = yaml.ScalarNode
 		target.Node.Style = lhsNode.Style
@@ -350,6 +350,20 @@ var addOperatorScenarios = []expressionScenario{
 			"D0, P[], (doc)::a: &horse [1, 2]\n",
 		},
 	},
+	{
+		skipDoc:       true,
+		description:   "Add sequence to map",
+		document:      "a: {x: cool}",
+		expression:    `.a += [2]`,
+		expectedError: "!!seq () cannot be added to a !!map (a)",
+	},
+	{
+		skipDoc:       true,
+		description:   "Add sequence to scalar",
+		document:      "a: cool",
+		expression:    `.a += [2]`,
+		expectedError: "!!seq () cannot be added to a !!str (a)",
+	},
 }

 func TestAddOperatorScenarios(t *testing.T) {
@@ -2,7 +2,24 @@ package yqlib

 func alternativeOperator(d *dataTreeNavigator, context Context, expressionNode *ExpressionNode) (Context, error) {
 	log.Debugf("-- alternative")
-	return crossFunction(d, context, expressionNode, alternativeFunc, true)
+	prefs := crossFunctionPreferences{
+		CalcWhenEmpty: true,
+		Calculation:   alternativeFunc,
+		LhsResultValue: func(lhs *CandidateNode) (*CandidateNode, error) {
+			if lhs == nil {
+				return nil, nil
+			}
+			truthy, err := isTruthy(lhs)
+			if err != nil {
+				return nil, err
+			}
+			if truthy {
+				return lhs, nil
+			}
+			return nil, nil
+		},
+	}
+	return crossFunctionWithPrefs(d, context, expressionNode, prefs)
 }

 func alternativeFunc(d *dataTreeNavigator, context Context, lhs *CandidateNode, rhs *CandidateNode) (*CandidateNode, error) {
@@ -85,6 +85,24 @@ var alternativeOperatorScenarios = []expressionScenario{
 			"D0, P[], (!!bool)::true\n",
 		},
 	},
+	{
+		description:    "Update or create - entity exists",
+		subdescription: "This initialises `a` if it's not present",
+		expression:     "(.a // (.a = 0)) += 1",
+		document:       `a: 1`,
+		expected: []string{
+			"D0, P[], (doc)::a: 2\n",
+		},
+	},
+	{
+		description:    "Update or create - entity does not exist",
+		subdescription: "This initialises `a` if it's not present",
+		expression:     "(.a // (.a = 0)) += 1",
+		document:       `b: camel`,
+		expected: []string{
+			"D0, P[], (!!map)::b: camel\na: 1\n",
+		},
+	},
 }

 func TestAlternativeOperatorScenarios(t *testing.T) {
@@ -3,6 +3,7 @@ package yqlib
 type assignPreferences struct {
 	DontOverWriteAnchor bool
 	OnlyWriteNull       bool
+	ClobberCustomTags   bool
 }

 func assignUpdateFunc(prefs assignPreferences) crossFunctionCalculation {
@@ -15,18 +16,29 @@ func assignUpdateFunc(prefs assignPreferences) crossFunctionCalculation {
 	}
 }

+// they way *= (multipleAssign) is handled, we set the multiplePrefs
+// on the node, not assignPrefs. Long story.
+func getAssignPreferences(preferences interface{}) assignPreferences {
+	prefs := assignPreferences{}
+
+	switch typedPref := preferences.(type) {
+	case assignPreferences:
+		prefs = typedPref
+	case multiplyPreferences:
+		prefs = typedPref.AssignPrefs
+	}
+	return prefs
+}
+
 func assignUpdateOperator(d *dataTreeNavigator, context Context, expressionNode *ExpressionNode) (Context, error) {
 	lhs, err := d.GetMatchingNodes(context, expressionNode.LHS)
 	if err != nil {
 		return Context{}, err
 	}

-	prefs := assignPreferences{}
-	// they way *= (multipleAssign) is handled, we set the multiplePrefs
-	// on the node, not assignPrefs. Long story.
-	if p, ok := expressionNode.Operation.Preferences.(assignPreferences); ok {
-		prefs = p
-	}
+	prefs := getAssignPreferences(expressionNode.Operation.Preferences)
+
+	log.Debug("assignUpdateOperator prefs: %v", prefs)

 	if !expressionNode.Operation.UpdateAssign {
 		// this works because we already ran against LHS with an editable context.
@@ -208,6 +208,23 @@ var assignOperatorScenarios = []expressionScenario{
 			"D0, P[], (doc)::a:\n    b:\n        - null\n        - c: bogs\n",
 		},
 	},
+	{
+		description: "Custom types are maintained by default",
+		document:    "a: !cat meow\nb: !dog woof",
+		expression:  `.a = .b`,
+		expected: []string{
+			"D0, P[], (doc)::a: !cat woof\nb: !dog woof\n",
+		},
+	},
+	{
+		description:    "Custom types: clovver",
+		subdescription: "Use the `c` option to clobber custom tags",
+		document:       "a: !cat meow\nb: !dog woof",
+		expression:     `.a =c .b`,
+		expected: []string{
+			"D0, P[], (doc)::a: !dog woof\nb: !dog woof\n",
+		},
+	},
 }

 func TestAssignOperatorScenarios(t *testing.T) {
@@ -63,8 +63,12 @@ func assignCommentsOperator(d *dataTreeNavigator, context Context, expressionNod
 		}
 		if preferences.FootComment && candidate.Node.Kind == yaml.DocumentNode && comment != "" {
 			candidate.TrailingContent = "# " + comment
+		} else if preferences.FootComment && candidate.Node.Kind == yaml.DocumentNode {
+			candidate.TrailingContent = comment
+
 		} else if preferences.FootComment && candidate.Node.Kind != yaml.DocumentNode {
 			candidate.Node.FootComment = comment
+			candidate.TrailingContent = ""
 		}

 	}
@@ -4,15 +4,74 @@ import (
 	"testing"
 )

+var expectedWhereIsMyCommentMapKey = `D0, P[], (!!seq)::- p: ""
+  isKey: false
+  hc: ""
+  lc: ""
+  fc: ""
+- p: hello
+  isKey: true
+  hc: ""
+  lc: hello-world-comment
+  fc: ""
+- p: hello
+  isKey: false
+  hc: ""
+  lc: ""
+  fc: ""
+- p: hello.message
+  isKey: true
+  hc: ""
+  lc: ""
+  fc: ""
+- p: hello.message
+  isKey: false
+  hc: ""
+  lc: ""
+  fc: ""
+`
+
+var expectedWhereIsMyCommentArray = `D0, P[], (!!seq)::- p: ""
+  isKey: false
+  hc: ""
+  lc: ""
+  fc: ""
+- p: name
+  isKey: true
+  hc: ""
+  lc: ""
+  fc: ""
+- p: name
+  isKey: false
+  hc: ""
+  lc: ""
+  fc: ""
+- p: name.0
+  isKey: false
+  hc: under-name-comment
+  lc: ""
+  fc: ""
+`
+
 var commentOperatorScenarios = []expressionScenario{
 	{
-		description: "Set line comment",
-		document:    `a: cat`,
-		expression:  `.a line_comment="single"`,
+		description:    "Set line comment",
+		subdescription: "Set the comment on the key node for more reliability (see below).",
+		document:       `a: cat`,
+		expression:     `.a line_comment="single"`,
 		expected: []string{
 			"D0, P[], (doc)::a: cat # single\n",
 		},
 	},
+	{
+		description:    "Set line comment of a maps/arrays",
+		subdescription: "For maps and arrays, you need to set the line comment on the _key_ node. This will also work for scalars.",
+		document:       "a:\n  b: things",
+		expression:     `(.a | key) line_comment="single"`,
+		expected: []string{
+			"D0, P[], (doc)::a: # single\n    b: things\n",
+		},
+	},
 	{
 		skipDoc:    true,
 		document:   "a: cat\nb: dog",
@@ -46,6 +105,42 @@ var commentOperatorScenarios = []expressionScenario{
 			"D0, P[], (!!map)::a: cat # cat\n# cat\n\n# cat\nb: dog # dog\n# dog\n\n# dog\n",
 		},
 	},
+	{
+		description:    "Where is the comment - map key example",
+		subdescription: "The underlying yaml parser can assign comments in a document to surprising nodes. Use an expression like this to find where you comment is. 'p' indicates the path, 'isKey' is if the node is a map key (as opposed to a map value).\nFrom this, you can see the 'hello-world-comment' is actually on the 'hello' key",
+		document:       "hello: # hello-world-comment\n  message: world",
+		expression:     `[... | {"p": path | join("."), "isKey": is_key, "hc": headComment, "lc": lineComment, "fc": footComment}]`,
+		expected: []string{
+			expectedWhereIsMyCommentMapKey,
+		},
+	},
+	{
+		description:    "Retrieve comment - map key example",
+		subdescription: "From the previous example, we know that the comment is on the 'hello' _key_ as a lineComment",
+		document:       "hello: # hello-world-comment\n  message: world",
+		expression:     `.hello | key | line_comment`,
+		expected: []string{
+			"D0, P[hello], (!!str)::hello-world-comment\n",
+		},
+	},
+	{
+		description:    "Where is the comment - array example",
+		subdescription: "The underlying yaml parser can assign comments in a document to surprising nodes. Use an expression like this to find where you comment is. 'p' indicates the path, 'isKey' is if the node is a map key (as opposed to a map value).\nFrom this, you can see the 'under-name-comment' is actually on the first child",
+		document:       "name:\n  # under-name-comment\n  - first-array-child",
+		expression:     `[... | {"p": path | join("."), "isKey": is_key, "hc": headComment, "lc": lineComment, "fc": footComment}]`,
+		expected: []string{
+			expectedWhereIsMyCommentArray,
+		},
+	},
+	{
+		description:    "Retrieve comment - array example",
+		subdescription: "From the previous example, we know that the comment is on the first child as a headComment",
+		document:       "name:\n  # under-name-comment\n  - first-array-child",
+		expression:     `.name[0] | headComment`,
+		expected: []string{
+			"D0, P[name 0], (!!str)::under-name-comment\n",
+		},
+	},
 	{
 		description: "Set head comment",
 		document:    `a: cat`,
@@ -54,6 +149,14 @@ var commentOperatorScenarios = []expressionScenario{
 			"D0, P[], (doc)::# single\n\na: cat\n",
 		},
 	},
+	{
+		description: "Set head comment of a map entry",
+		document:    "f: foo\na:\n  b: cat",
+		expression:  `(.a | key) head_comment="single"`,
+		expected: []string{
+			"D0, P[], (doc)::f: foo\n# single\na:\n    b: cat\n",
+		},
+	},
 	{
 		description: "Set foot comment, using an expression",
 		document:    `a: cat`,
@@ -62,6 +165,15 @@ var commentOperatorScenarios = []expressionScenario{
 			"D0, P[], (doc)::a: cat\n# cat\n",
 		},
 	},
+	{
+		skipDoc:     true,
+		description: "Set foot comment, using an expression",
+		document:    "a: cat\n\n# hi",
+		expression:  `. foot_comment=""`,
+		expected: []string{
+			"D0, P[], (doc)::a: cat\n",
+		},
+	},
 	{
 		skipDoc:    true,
 		document:   `a: cat`,
@@ -112,6 +224,24 @@ var commentOperatorScenarios = []expressionScenario{
 			"D0, P[], (!!str)::welcome!\n",
 		},
 	},
+	{
+		skipDoc:     true,
+		description: "strip trailing comment recurse all",
+		document:    "a: cat\n\n# haha",
+		expression:  `... comments= ""`,
+		expected: []string{
+			"D0, P[], (!!map)::a: cat\n",
+		},
+	},
+	{
+		skipDoc:     true,
+		description: "strip trailing comment recurse values",
+		document:    "a: cat\n\n# haha",
+		expression:  `.. comments= ""`,
+		expected: []string{
+			"D0, P[], (!!map)::a: cat\n",
+		},
+	},
 	{
 		description:           "Head comment with document split",
 		dontFormatInputForDoc: true,
@@ -72,7 +72,7 @@ func sequenceFor(d *dataTreeNavigator, context Context, matchingNode *CandidateN
 	return &CandidateNode{Node: innerList, Document: document, Path: path}, nil
 }

-//NOTE: here the document index gets dropped so we
+// NOTE: here the document index gets dropped so we
 // no longer know where the node originates from.
 func listToNodeSeq(list *list.List) *yaml.Node {
 	node := yaml.Node{Kind: yaml.SequenceNode, Tag: "!!seq"}
@@ -114,6 +114,10 @@ func decodeOperator(d *dataTreeNavigator, context Context, expressionNode *Expre
 		decoder = NewBase64Decoder()
 	case PropertiesInputFormat:
 		decoder = NewPropertiesDecoder()
+	case CSVObjectInputFormat:
+		decoder = NewCSVObjectDecoder(',')
+	case TSVObjectInputFormat:
+		decoder = NewCSVObjectDecoder('\t')
 	}

 	var results = list.New()
@@ -66,11 +66,27 @@ var encoderDecoderOperatorScenarios = []expressionScenario{
 	{
 		description: "Decode props encoded string",
 		document:    `a: "cats=great\ndogs=cool as well"`,
-		expression:  `.a |= from_props`,
+		expression:  `.a |= @propsd`,
 		expected: []string{
 			"D0, P[], (doc)::a:\n    cats: great\n    dogs: cool as well\n",
 		},
 	},
+	{
+		description: "Decode csv encoded string",
+		document:    `a: "cats,dogs\ngreat,cool as well"`,
+		expression:  `.a |= @csvd`,
+		expected: []string{
+			"D0, P[], (doc)::a:\n    - cats: great\n      dogs: cool as well\n",
+		},
+	},
+	{
+		description: "Decode tsv encoded string",
+		document:    `a: "cats	dogs\ngreat	cool as well"`,
+		expression:  `.a |= @tsvd`,
+		expected: []string{
+			"D0, P[], (doc)::a:\n    - cats: great\n      dogs: cool as well\n",
+		},
+	},
 	{
 		skipDoc:    true,
 		document:   "a:\n  cool:\n    bob: dylan",
@@ -7,6 +7,20 @@ import (
 	"gopkg.in/yaml.v3"
 )

+func isKeyOperator(d *dataTreeNavigator, context Context, expressionNode *ExpressionNode) (Context, error) {
+	log.Debugf("-- isKeyOperator")
+
+	var results = list.New()
+
+	for el := context.MatchingNodes.Front(); el != nil; el = el.Next() {
+		candidate := el.Value.(*CandidateNode)
+
+		results.PushBack(createBooleanCandidate(candidate, candidate.IsMapKey))
+	}
+
+	return context.ChildContext(results), nil
+}
+
 func getKeyOperator(d *dataTreeNavigator, context Context, expressionNode *ExpressionNode) (Context, error) {
 	log.Debugf("-- getKeyOperator")

@@ -4,6 +4,32 @@ import (
 	"testing"
 )

+var expectedIsKey = `D0, P[], (!!seq)::- p: ""
+  isKey: false
+  tag: '!!map'
+- p: a
+  isKey: true
+  tag: '!!str'
+- p: a
+  isKey: false
+  tag: '!!map'
+- p: a.b
+  isKey: true
+  tag: '!!str'
+- p: a.b
+  isKey: false
+  tag: '!!seq'
+- p: a.b.0
+  isKey: false
+  tag: '!!str'
+- p: a.c
+  isKey: true
+  tag: '!!str'
+- p: a.c
+  isKey: false
+  tag: '!!str'
+`
+
 var keysOperatorScenarios = []expressionScenario{
 	{
 		description: "Map keys",
@@ -75,6 +101,14 @@ var keysOperatorScenarios = []expressionScenario{
 			"D0, P[a x], (!!str)::comment on key\n",
 		},
 	},
+	{
+		description: "Check node is a key",
+		document:    "a: \n  b: [cat]\n  c: frog\n",
+		expression:  `[... | { "p": path | join("."), "isKey": is_key, "tag": tag }]`,
+		expected: []string{
+			expectedIsKey,
+		},
+	},
 }

 func TestKeysOperatorScenarios(t *testing.T) {
@@ -523,6 +523,35 @@ var multiplyOperatorScenarios = []expressionScenario{
 			"D0, P[], (doc)::a: !horse {cat: meow, dog: woof}\nb: !goat {dog: woof}\n",
 		},
 	},
+	{
+		description:    "Custom types: clobber tags",
+		subdescription: "Use the `c` option to clobber custom tags. Note that the second tag is now used",
+		document:       "a: !horse {cat: meow}\nb: !goat {dog: woof}",
+		expression:     ".a *=c .b",
+		expected: []string{
+			"D0, P[], (doc)::a: !goat {cat: meow, dog: woof}\nb: !goat {dog: woof}\n",
+		},
+	},
+	{
+		skipDoc:        true,
+		description:    "Custom types: clobber tags - *=",
+		subdescription: "Use the `c` option to clobber custom tags - on both the `=` and `*` operator. Note that the second tag is now used",
+		document:       "a: !horse {cat: meow}\nb: !goat {dog: woof}",
+		expression:     ".a =c .a *c .b",
+		expected: []string{
+			"D0, P[], (doc)::a: !goat {cat: meow, dog: woof}\nb: !goat {dog: woof}\n",
+		},
+	},
+	{
+		skipDoc:        true,
+		description:    "Custom types: dont clobber tags - *=",
+		subdescription: "Use the `c` option to clobber custom tags - on both the `=` and `*` operator. Note that the second tag is now used",
+		document:       "a: !horse {cat: meow}\nb: !goat {dog: woof}",
+		expression:     ".a *= .b",
+		expected: []string{
+			"D0, P[], (doc)::a: !horse {cat: meow, dog: woof}\nb: !goat {dog: woof}\n",
+		},
+	},
 	{
 		skipDoc:     true,
 		description: "Custom types: that are really maps",
@@ -532,6 +561,24 @@ var multiplyOperatorScenarios = []expressionScenario{
 			"D0, P[], (doc)::a: {cat: !horse 5}\nb: {cat: 5}\n",
 		},
 	},
+	{
+		skipDoc:     true,
+		description: "Relative merge, new fields only",
+		document:    "a: {a: original}\n",
+		expression:  `.a *=n load("../../examples/thing.yml")`,
+		expected: []string{
+			"D0, P[], (doc)::a: {a: original, b: cool.}\n",
+		},
+	},
+	{
+		skipDoc:     true,
+		description: "Relative merge",
+		document:    "a: {a: original}\n",
+		expression:  `.a *= load("../../examples/thing.yml")`,
+		expected: []string{
+			"D0, P[], (doc)::a: {a: apple is included, b: cool.}\n",
+		},
+	},
 }

 func TestMultiplyOperatorScenarios(t *testing.T) {
@@ -16,6 +16,139 @@ func createPathNodeFor(pathElement interface{}) *yaml.Node {
 	}
 }

+func getPathArrayFromNode(funcName string, node *yaml.Node) ([]interface{}, error) {
+	if node.Kind != yaml.SequenceNode {
+		return nil, fmt.Errorf("%v: expected path array, but got %v instead", funcName, node.Tag)
+	}
+
+	path := make([]interface{}, len(node.Content))
+
+	for i, childNode := range node.Content {
+		if childNode.Tag == "!!str" {
+			path[i] = childNode.Value
+		} else if childNode.Tag == "!!int" {
+			number, err := parseInt(childNode.Value)
+			if err != nil {
+				return nil, fmt.Errorf("%v: could not parse %v as an int: %w", funcName, childNode.Value, err)
+			}
+			path[i] = number
+		} else {
+			return nil, fmt.Errorf("%v: expected either a !!str or !!int in the path, found %v instead", funcName, childNode.Tag)
+		}
+
+	}
+	return path, nil
+}
+
+// SETPATH(pathArray; value)
+func setPathOperator(d *dataTreeNavigator, context Context, expressionNode *ExpressionNode) (Context, error) {
+	log.Debugf("SetPath")
+
+	if expressionNode.RHS.Operation.OperationType != blockOpType {
+		return Context{}, fmt.Errorf("SETPATH must be given a block (;), got %v instead", expressionNode.RHS.Operation.OperationType.Type)
+	}
+
+	lhsPathContext, err := d.GetMatchingNodes(context.ReadOnlyClone(), expressionNode.RHS.LHS)
+
+	if err != nil {
+		return Context{}, err
+	}
+
+	if lhsPathContext.MatchingNodes.Len() != 1 {
+		return Context{}, fmt.Errorf("SETPATH: expected single path but found %v results instead", lhsPathContext.MatchingNodes.Len())
+	}
+	lhsValue := lhsPathContext.MatchingNodes.Front().Value.(*CandidateNode)
+
+	lhsPath, err := getPathArrayFromNode("SETPATH", lhsValue.Node)
+
+	if err != nil {
+		return Context{}, err
+	}
+
+	lhsTraversalTree := createTraversalTree(lhsPath, traversePreferences{}, false)
+
+	assignmentOp := &Operation{OperationType: assignOpType}
+
+	for el := context.MatchingNodes.Front(); el != nil; el = el.Next() {
+		candidate := el.Value.(*CandidateNode)
+
+		targetContextValue, err := d.GetMatchingNodes(context.SingleReadonlyChildContext(candidate), expressionNode.RHS.RHS)
+		if err != nil {
+			return Context{}, err
+		}
+
+		if targetContextValue.MatchingNodes.Len() != 1 {
+			return Context{}, fmt.Errorf("SETPATH: expected single value on RHS but found %v", targetContextValue.MatchingNodes.Len())
+		}
+
+		rhsOp := &Operation{OperationType: valueOpType, CandidateNode: targetContextValue.MatchingNodes.Front().Value.(*CandidateNode)}
+
+		assignmentOpNode := &ExpressionNode{
+			Operation: assignmentOp,
+			LHS:       lhsTraversalTree,
+			RHS:       &ExpressionNode{Operation: rhsOp},
+		}
+
+		_, err = d.GetMatchingNodes(context.SingleChildContext(candidate), assignmentOpNode)
+
+		if err != nil {
+			return Context{}, err
+		}
+
+	}
+	return context, nil
+}
+
+func delPathsOperator(d *dataTreeNavigator, context Context, expressionNode *ExpressionNode) (Context, error) {
+	log.Debugf("delPaths")
+	// single RHS expression that returns an array of paths (array of arrays)
+
+	pathArraysContext, err := d.GetMatchingNodes(context.ReadOnlyClone(), expressionNode.RHS)
+	if err != nil {
+		return Context{}, err
+	}
+	if pathArraysContext.MatchingNodes.Len() != 1 {
+		return Context{}, fmt.Errorf("DELPATHS: expected single value but found %v", pathArraysContext.MatchingNodes.Len())
+	}
+	pathArraysNode := pathArraysContext.MatchingNodes.Front().Value.(*CandidateNode).Node
+
+	if pathArraysNode.Tag != "!!seq" {
+		return Context{}, fmt.Errorf("DELPATHS: expected a sequence of sequences, but found %v", pathArraysNode.Tag)
+	}
+
+	updatedContext := context
+
+	for i, child := range pathArraysNode.Content {
+
+		if child.Tag != "!!seq" {
+			return Context{}, fmt.Errorf("DELPATHS: expected entry [%v] to be a sequence, but its a %v. Note that delpaths takes an array of path arrays, e.g. [[\"a\", \"b\"]]", i, child.Tag)
+		}
+		childPath, err := getPathArrayFromNode("DELPATHS", child)
+
+		if err != nil {
+			return Context{}, err
+		}
+
+		childTraversalExp := createTraversalTree(childPath, traversePreferences{}, false)
+		deleteChildOp := &Operation{OperationType: deleteChildOpType}
+
+		deleteChildOpNode := &ExpressionNode{
+			Operation: deleteChildOp,
+			RHS:       childTraversalExp,
+		}
+
+		updatedContext, err = d.GetMatchingNodes(updatedContext, deleteChildOpNode)
+
+		if err != nil {
+			return Context{}, err
+		}
+
+	}
+
+	return updatedContext, nil
+
+}
+
 func getPathOperator(d *dataTreeNavigator, context Context, expressionNode *ExpressionNode) (Context, error) {
 	log.Debugf("GetPath")

@@ -63,6 +63,60 @@ var pathOperatorScenarios = []expressionScenario{
 			"D0, P[a 2], (!!seq)::- path:\n    - a\n    - 2\n  value: frog\n",
 		},
 	},
+	{
+		description: "Set path",
+		document:    `{a: {b: cat}}`,
+		expression:  `setpath(["a", "b"]; "things")`,
+		expected: []string{
+			"D0, P[], (doc)::{a: {b: things}}\n",
+		},
+	},
+	{
+		description: "Set on empty document",
+		expression:  `setpath(["a", "b"]; "things")`,
+		expected: []string{
+			"D0, P[], ()::a:\n    b: things\n",
+		},
+	},
+	{
+		description: "Set array path",
+		document:    `a: [cat, frog]`,
+		expression:  `setpath(["a", 0]; "things")`,
+		expected: []string{
+			"D0, P[], (doc)::a: [things, frog]\n",
+		},
+	},
+	{
+		description: "Set array path empty",
+		expression:  `setpath(["a", 0]; "things")`,
+		expected: []string{
+			"D0, P[], ()::a:\n    - things\n",
+		},
+	},
+	{
+		description:    "Delete path",
+		subdescription: "Notice delpaths takes an _array_ of paths.",
+		document:       `{a: {b: cat, c: dog, d: frog}}`,
+		expression:     `delpaths([["a", "c"], ["a", "d"]])`,
+		expected: []string{
+			"D0, P[], (doc)::{a: {b: cat}}\n",
+		},
+	},
+	{
+		description: "Delete array path",
+		document:    `a: [cat, frog]`,
+		expression:  `delpaths([["a", 0]])`,
+		expected: []string{
+			"D0, P[], (doc)::a: [frog]\n",
+		},
+	},
+	{
+		description:    "Delete - wrong parameter",
+		subdescription: "delpaths does not work with a single path array",
+		document:       `a: [cat, frog]`,
+		expression:     `delpaths(["a", 0])`,
+		expectedError:  "DELPATHS: expected entry [0] to be a sequence, but its a !!str. Note that delpaths takes an array of path arrays, e.g. [[\"a\", \"b\"]]",
+	},
 }

 func TestPathOperatorsScenarios(t *testing.T) {
@@ -17,7 +17,7 @@ func sortOperator(d *dataTreeNavigator, context Context, expressionNode *Express
 }

 // context represents the current matching nodes in the expression pipeline
-//expressionNode is your current expression (sort_by)
+// expressionNode is your current expression (sort_by)
 func sortByOperator(d *dataTreeNavigator, context Context, expressionNode *ExpressionNode) (Context, error) {

 	results := list.New()
@@ -13,6 +13,26 @@ type changeCasePrefs struct {
 	ToUpperCase bool
 }

+func trimSpaceOperator(d *dataTreeNavigator, context Context, expressionNode *ExpressionNode) (Context, error) {
+	results := list.New()
+	for el := context.MatchingNodes.Front(); el != nil; el = el.Next() {
+		candidate := el.Value.(*CandidateNode)
+
+		node := unwrapDoc(candidate.Node)
+
+		if guessTagFromCustomType(node) != "!!str" {
+			return Context{}, fmt.Errorf("cannot trim %v, can only operate on strings. ", node.Tag)
+		}
+
+		newStringNode := &yaml.Node{Kind: yaml.ScalarNode, Tag: node.Tag, Style: node.Style}
+		newStringNode.Value = strings.TrimSpace(node.Value)
+		results.PushBack(candidate.CreateReplacement(newStringNode))
+
+	}
+
+	return context.ChildContext(results), nil
+}
+
 func changeCaseOperator(d *dataTreeNavigator, context Context, expressionNode *ExpressionNode) (Context, error) {
 	results := list.New()
 	prefs := expressionNode.Operation.Preferences.(changeCasePrefs)
@@ -47,6 +47,17 @@ var stringsOperatorScenarios = []expressionScenario{
 			"D0, P[], (!!str)::cat; meow; 1; ; true\n",
 		},
 	},
+	{
+		description: "Trim strings",
+		document:    `[" cat", "dog ", " cow cow ", horse]`,
+		expression:  `.[] | trim`,
+		expected: []string{
+			"D0, P[0], (!!str)::cat\n",
+			"D0, P[1], (!!str)::dog\n",
+			"D0, P[2], (!!str)::cow cow\n",
+			"D0, P[3], (!!str)::horse\n",
+		},
+	},
 	{
 		skipDoc:    true,
 		document:   `[!horse cat, !goat meow, !frog 1, null, true]`,
@@ -7,7 +7,7 @@ func withOperator(d *dataTreeNavigator, context Context, expressionNode *Express
 	// with(path, exp)

 	if expressionNode.RHS.Operation.OperationType != blockOpType {
-		return Context{}, fmt.Errorf("with must be given a block, got %v instead", expressionNode.RHS.Operation.OperationType.Type)
+		return Context{}, fmt.Errorf("with must be given a block (;), got %v instead", expressionNode.RHS.Operation.OperationType.Type)
 	}

 	pathExp := expressionNode.RHS.LHS
@@ -5,6 +5,7 @@ import (
 	"fmt"

 	"github.com/jinzhu/copier"
+	logging "gopkg.in/op/go-logging.v1"
 	"gopkg.in/yaml.v3"
 )

@@ -18,7 +19,21 @@ func compoundAssignFunction(d *dataTreeNavigator, context Context, expressionNod
 		return Context{}, err
 	}

-	assignmentOp := &Operation{OperationType: assignOpType, Preferences: expressionNode.Operation.Preferences}
+	// tricky logic when we are running *= with flags.
+	// we have an op like: .a *=nc .b
+	// which should roughly translate to .a =c .a *nc .b
+	// note that the 'n' flag only applies to the multiple op, not the assignment
+	// but the clobber flag applies to both!
+
+	prefs := assignPreferences{}
+	switch typedPref := expressionNode.Operation.Preferences.(type) {
+	case assignPreferences:
+		prefs = typedPref
+	case multiplyPreferences:
+		prefs.ClobberCustomTags = typedPref.AssignPrefs.ClobberCustomTags
+	}
+
+	assignmentOp := &Operation{OperationType: assignOpType, Preferences: prefs}

 	for el := lhs.MatchingNodes.Front(); el != nil; el = el.Next() {
 		candidate := el.Value.(*CandidateNode)
@@ -83,8 +98,10 @@ func resultsForRHS(d *dataTreeNavigator, context Context, lhsCandidate *Candidat
 	}

 	for rightEl := rhs.MatchingNodes.Front(); rightEl != nil; rightEl = rightEl.Next() {
-		log.Debugf("Applying calc")
 		rhsCandidate := rightEl.Value.(*CandidateNode)
+		if !log.IsEnabledFor(logging.DEBUG) {
+			log.Debugf("Applying lhs: %v, rhsCandidate, %v", NodeToString(lhsCandidate), NodeToString(rhsCandidate))
+		}
 		resultCandidate, err := prefs.Calculation(d, context, lhsCandidate, rhsCandidate)
 		if err != nil {
 			return err
@@ -100,7 +100,11 @@ func testScenario(t *testing.T, s *expressionScenario) {
 	context, err := NewDataTreeNavigator().GetMatchingNodes(Context{MatchingNodes: inputs}, node)

 	if s.expectedError != "" {
-		test.AssertResultComplexWithContext(t, s.expectedError, err.Error(), fmt.Sprintf("desc: %v\nexp: %v\ndoc: %v", s.description, s.expression, s.document))
+		if err == nil {
+			t.Errorf("Expected error '%v' but it worked!", s.expectedError)
+		} else {
+			test.AssertResultComplexWithContext(t, s.expectedError, err.Error(), fmt.Sprintf("desc: %v\nexp: %v\ndoc: %v", s.description, s.expression, s.document))
+		}
 		return
 	}

@@ -110,7 +110,7 @@ func (s *streamEvaluator) Evaluate(filename string, reader io.Reader, node *Expr
 			Node:      &dataBucket,
 			FileIndex: s.fileIndex,
 		}
-		//move document comments into candidate node
+		// move document comments into candidate node
 		// otherwise unwrap drops them.
 		candidateNode.TrailingContent = dataBucket.FootComment
 		dataBucket.FootComment = ""
@@ -0,0 +1,85 @@
+package yqlib
+
+import (
+	"bytes"
+	"container/list"
+	"errors"
+	"fmt"
+	"io"
+
+	yaml "gopkg.in/yaml.v3"
+)
+
+type StringEvaluator interface {
+	Evaluate(expression string, input string, encoder Encoder, leadingContentPreProcessing bool, decoder Decoder) (string, error)
+}
+
+type stringEvaluator struct {
+	treeNavigator DataTreeNavigator
+	fileIndex     int
+}
+
+func NewStringEvaluator() StringEvaluator {
+	return &stringEvaluator{
+		treeNavigator: NewDataTreeNavigator(),
+	}
+}
+
+func (s *stringEvaluator) Evaluate(expression string, input string, encoder Encoder, leadingContentPreProcessing bool, decoder Decoder) (string, error) {
+
+	// Use bytes.Buffer for output of string
+	out := new(bytes.Buffer)
+	printer := NewPrinter(encoder, NewSinglePrinterWriter(out))
+
+	InitExpressionParser()
+	node, err := ExpressionParser.ParseExpression(expression)
+	if err != nil {
+		return "", err
+	}
+
+	reader, leadingContent, err := readString(input, leadingContentPreProcessing)
+	if err != nil {
+		return "", err
+	}
+
+	var currentIndex uint
+	decoder.Init(reader)
+	for {
+		var dataBucket yaml.Node
+		errorReading := decoder.Decode(&dataBucket)
+
+		if errors.Is(errorReading, io.EOF) {
+			s.fileIndex = s.fileIndex + 1
+			return out.String(), nil
+		} else if errorReading != nil {
+			return "", fmt.Errorf("bad input '%v': %w", input, errorReading)
+		}
+
+		candidateNode := &CandidateNode{
+			Document:  currentIndex,
+			Node:      &dataBucket,
+			FileIndex: s.fileIndex,
+		}
+		// move document comments into candidate node
+		// otherwise unwrap drops them.
+		candidateNode.TrailingContent = dataBucket.FootComment
+		dataBucket.FootComment = ""
+
+		if currentIndex == 0 {
+			candidateNode.LeadingContent = leadingContent
+		}
+		inputList := list.New()
+		inputList.PushBack(candidateNode)
+
+		result, errorParsing := s.treeNavigator.GetMatchingNodes(Context{MatchingNodes: inputList}, node)
+		if errorParsing != nil {
+			return "", errorParsing
+		}
+		err = printer.PrintResults(result.MatchingNodes)
+
+		if err != nil {
+			return "", err
+		}
+		currentIndex = currentIndex + 1
+	}
+}
@@ -0,0 +1,30 @@
+package yqlib
+
+import (
+	"testing"
+
+	"github.com/mikefarah/yq/v4/test"
+)
+
+func TestStringEvaluator_Evaluate_Nominal(t *testing.T) {
+	expected_output := `` +
+		`yq` + "\n" +
+		`---` + "\n" +
+		`jq` + "\n"
+	expression := ".[].name"
+	input := `` +
+		` - name: yq` + "\n" +
+		`   description: yq is a portable command-line YAML, JSON and XML processor` + "\n" +
+		`---` + "\n" +
+		` - name: jq` + "\n" +
+		`   description: Command-line JSON processor` + "\n"
+	encoder := NewYamlEncoder(2, true, true, true)
+	decoder := NewYamlDecoder()
+
+	result, err := NewStringEvaluator().Evaluate(expression, input, encoder, true, decoder)
+	if err != nil {
+		t.Error(err)
+	}
+
+	test.AssertResult(t, expected_output, result)
+}
@@ -33,6 +33,14 @@ func readStream(filename string, leadingContentPreProcessing bool) (io.Reader, s
 	return processReadStream(reader)
 }

+func readString(input string, leadingContentPreProcessing bool) (io.Reader, string, error) {
+	reader := bufio.NewReader(strings.NewReader(input))
+	if !leadingContentPreProcessing {
+		return reader, "", nil
+	}
+	return processReadStream(reader)
+}
+
 func writeString(writer io.Writer, txt string) error {
 	_, errorWriting := writer.Write([]byte(txt))
 	return errorWriting
@@ -2,10 +2,6 @@
 - increment version in version.go
 - increment version in snapcraft.yaml
 - make sure local build passes
- check manpage 
-  - ./scripts/generate-man-page-md.sh
-  - ./scripts/generate-man-page.sh
-  - man ./yq.1
 - run ./scripts/copy-docs.sh (and commit the changed in the yq-book branch)
 - commit version update changes
 - tag git with same version number
@@ -1,5 +1,29 @@
+4.28.1:
+  - Added `setpath` and `delpaths` operators, like jq (#1374)
+  - Added `is_key` operator, to check if a match was a key when recursing
+  - Added validation when attempting to add sequences to maps (#1341)
+
+4.27.5:
+  - Fixed relative merge bug #1333
+
+4.27.4:
+  - Fixed bug in alternative (//) operator, RHS being evaluated when it didn't need to be
+  - Fixed footer comment issue #1231
+  - Github action now runs as root (as recommended by Github Actions doc)
+  - Updated dependencies
+
+4.27.3:
+  - Added new 'c' merge and assign flag that clobbers custom tags
+  - Bumped go dependency to fix CVE (#1316)
+  - Updated dependencies
+
+4.27.2:
+  - Fixed JSON decoder to maintain object key order.
+
 4.27.1:
  - Added 'json' decoder for support for multiple JSON documents in a single file (e.g. NDJSON)
+  - Added 'csv' decoding, array of objects encoding, and round-triping
+  - New StringEvaluator when using yq as a lib (thanks @leviliangtw)
  - Fixed XML decoding issue (#1284)

 4.26.1:
@@ -1,5 +1,5 @@
 #!/bin/sh
 set -ex
 go mod download golang.org/x/tools@latest
-curl -sSfL https://raw.githubusercontent.com/golangci/golangci-lint/master/install.sh | sh -s -- -b $(go env GOPATH)/bin v1.45.2
-wget -O- -nv https://raw.githubusercontent.com/securego/gosec/master/install.sh | sh -s v2.11.0
+curl -sSfL https://raw.githubusercontent.com/golangci/golangci-lint/master/install.sh | sh -s -- -b $(go env GOPATH)/bin v1.49.0
+wget -O- -nv https://raw.githubusercontent.com/securego/gosec/master/install.sh | sh -s v2.13.1
@@ -1,5 +1,5 @@
 name: yq
-version: '4.26.1'
+version: '4.28.1'
 summary: A lightweight and portable command-line YAML processor
 description: |
  The aim of the project is to be the jq or sed of yaml files.
@@ -16,7 +16,7 @@ apps:
 parts:
  yq:
    plugin: go
-    go-channel: 1.18/stable
+    go-channel: 1.19/stable
    source: .
    source-type: git
    go-importpath: github.com/mikefarah/yq
Author	SHA1	Message	Date
Mike Farah	ffe635f943	Bumping version	2022-10-05 20:15:02 +11:00
Mike Farah	b5b81abb90	Added DELPATHS operator	2022-10-05 20:09:53 +11:00
Mike Farah	da3f3b93b4	Added SETPATH operator	2022-10-05 14:12:08 +11:00
Mike Farah	7f4c8e1c78	Added comment operator examples	2022-09-30 11:30:31 +10:00
Mike Farah	60f30f8a48	Added comment operator examples	2022-09-30 11:26:36 +10:00
Mike Farah	2362451fda	Added comment operator examples	2022-09-30 11:22:58 +10:00
Mike Farah	55a7fdfd8a	Added is_key operator	2022-09-30 10:27:35 +10:00
Mike Farah	3dd5f0c80c	Added another comment test	2022-09-30 10:15:41 +10:00
Mike Farah	be304a1102	Added another comment test	2022-09-30 09:46:07 +10:00
Mike Farah	a4b60039ca	Added examples of header-preprocess needing to be turned off	2022-09-30 09:36:38 +10:00
Mike Farah	b20477210f	Added missing validation for adding sequences to maps #1341	2022-09-16 10:04:48 +10:00
Mike Farah	9b6961875e	Bumping version	2022-09-09 12:29:41 +10:00
Mike Farah	51b64e6cf3	Fixing relative merge bug #1333	2022-09-09 12:28:38 +10:00
Mike Farah	851d93e5a8	Bumping version	2022-09-09 11:23:14 +10:00
Mike Farah	9291865eed	Preparing release notes	2022-09-09 11:21:53 +10:00
Mike Farah	fb330ebcb0	Updated github action to run as root	2022-09-09 11:14:25 +10:00
dependabot[bot]GitHubdependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>	ae7a48df2d	Bump github.com/elliotchance/orderedmap from 1.4.0 to 1.5.0 (#1331 ) Bumps [github.com/elliotchance/orderedmap](https://github.com/elliotchance/orderedmap) from 1.4.0 to 1.5.0. - [Release notes](https://github.com/elliotchance/orderedmap/releases) - [Commits](https://github.com/elliotchance/orderedmap/compare/v1.4.0...v1.5.0) --- updated-dependencies: - dependency-name: github.com/elliotchance/orderedmap dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>	2022-09-09 11:05:46 +10:00
dependabot[bot]GitHubdependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>	2e3c50c535	Bump golang from 1.19.0 to 1.19.1 (#1330 ) Bumps golang from 1.19.0 to 1.19.1. --- updated-dependencies: - dependency-name: golang dependency-type: direct:production update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>	2022-09-09 11:05:36 +10:00
Mike Farah	a64eea3b1a	Fixed bug in alternative op, dont evaluate RHS if LHS is truthy	2022-09-09 11:05:12 +10:00
AbirdcflyandGitHub	a12c1a6610	chore: remove duplicate word in doc (#1322 ) Signed-off-by: Abirdcfly <fp544037857@gmail.com> Signed-off-by: Abirdcfly <fp544037857@gmail.com>	2022-08-31 18:43:03 +10:00
Mike Farah	bf6b4d3941	Fixed footer comment issue #1321	2022-08-30 11:24:26 +10:00
Mike Farah	3d11b0545d	Bumping version	2022-08-29 15:41:07 +10:00
Mike Farah	b4ca184108	Added new "c" flag to clobber custom tags when needed	2022-08-29 15:38:21 +10:00
Mike Farah	b1a40a9fb7	Bumped go compiler; bumped quality tools	2022-08-29 14:13:15 +10:00
dependabot[bot]GitHubdependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>	187d9523c6	Bump github.com/goccy/go-json from 0.9.10 to 0.9.11 (#1314 ) Bumps [github.com/goccy/go-json](https://github.com/goccy/go-json) from 0.9.10 to 0.9.11. - [Release notes](https://github.com/goccy/go-json/releases) - [Changelog](https://github.com/goccy/go-json/blob/master/CHANGELOG.md) - [Commits](https://github.com/goccy/go-json/compare/v0.9.10...v0.9.11) --- updated-dependencies: - dependency-name: github.com/goccy/go-json dependency-type: direct:production update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>	2022-08-25 13:41:11 +10:00
Matti SchneiderandGitHub	a44938101d	Document dynamic get and output in GitHub Action (#1303 )	2022-08-14 10:40:41 +10:00
dependabot[bot]GitHubdependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>	a4634cdc86	Bump github.com/alecthomas/participle/v2 (#1302 ) Bumps [github.com/alecthomas/participle/v2](https://github.com/alecthomas/participle) from 2.0.0-beta.4 to 2.0.0-beta.5. - [Release notes](https://github.com/alecthomas/participle/releases) - [Changelog](https://github.com/alecthomas/participle/blob/master/CHANGES.md) - [Commits](https://github.com/alecthomas/participle/compare/v2.0.0-beta.4...v2.0.0-beta.5) --- updated-dependencies: - dependency-name: github.com/alecthomas/participle/v2 dependency-type: direct:production update-type: version-update:semver-patch ... Signed-off-by: dependabot[bot] <support@github.com> Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>	2022-08-11 13:47:22 +10:00
Mike Farah	a696dceea4	Added string trim operator	2022-08-08 13:35:57 +10:00
dependabot[bot]GitHubdependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>	989dabac5a	Bump golang from 1.18.4 to 1.19.0 (#1292 ) Bumps golang from 1.18.4 to 1.19.0. --- updated-dependencies: - dependency-name: golang dependency-type: direct:production update-type: version-update:semver-minor ... Signed-off-by: dependabot[bot] <support@github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>	2022-08-04 14:23:20 +10:00
Mike Farah	8eac186a35	Add link to discussions on readme	2022-08-01 16:55:55 +10:00
Mike Farah	1c7dc0e88a	Bug fix	2022-08-01 16:27:42 +10:00
Mike Farah	2c9b5be408	Fixed json decode to maintain key order	2022-08-01 16:26:43 +10:00
Mike Farah	a91a8ccc66	Improving docs	2022-08-01 14:12:35 +10:00
Mike Farah	6dfef716d7	Bumping version	2022-08-01 14:00:13 +10:00
Mike Farah	06a2b9d426	updating release notes	2022-08-01 13:47:24 +10:00
Mike Farah	6cd0892d59	Added SELinux note #1289	2022-08-01 13:47:16 +10:00
Mike Farah	51b3985aa0	Preparing release notes	2022-08-01 13:40:41 +10:00
Mike FarahandGitHub	c8815f5ab9	Csv decoder (#1290 ) * WIP: adding CSV decoder * Adding CSV decoder * Added CSV roundtrip * Fixing from review	2022-08-01 10:28:34 +10:00
Jih-Wei, LiangandGitHub	3c222d8707	Added StringEvaluator for evaluating string input #1266 (#1278 )	2022-08-01 08:50:56 +10:00