Refactored decoder responsibilities

- improved comment handling
- yaml decoder now responsible for leading content work around

README.md

@@ -7,27 +7,6 @@ a lightweight and portable command-line YAML, JSON and XML processor. `yq` uses
 
 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.
 
-## Notice for v4.x versions prior to 4.18.1
-Since 4.18.1, yq's 'eval/e' command is the _default_ command and no longer needs to be specified.
-
-Older versions will still need to specify 'eval/e'.
-
-Similarly, '-' is no longer required as a filename to read from STDIN (unless reading from one or more files).
-
-TLDR:
-
-Prior to 4.18.1 
-```bash
-yq e '.cool' - < file.yaml
-```
-
-4.18+ 
-```bash
-yq '.cool' < file.yaml
-```
-
-When merging multiple files together, `eval-all/ea` is still required to tell `yq` to run the expression against all the document at once.
-
 ## Quick Usage Guide
 
 Read a value:
@@ -80,7 +59,7 @@ Take a look at the discussions for [common questions](https://github.com/mikefar
 ### [Download the latest binary](https://github.com/mikefarah/yq/releases/latest)
 
 ### wget
-Use wget to download the pre-compiled binaries:
+Use wget to download, gzipped pre-compiled binaries:
 
 #### Compressed via tar.gz
 ```bash
@@ -97,6 +76,13 @@ wget https://github.com/mikefarah/yq/releases/download/${VERSION}/${BINARY} -O /
 
 For instance, VERSION=v4.2.0 and BINARY=yq_linux_amd64
 
+Use VERSION=latest to always get the latest version:
+
+```bash
+wget https://github.com/mikefarah/yq/releases/latest/download/yq_linux_amd64 -O /usr/bin/yq &&\
+    chmod +x /usr/bin/yq
+```
+
 ### MacOS / Linux via Homebrew:
 Using [Homebrew](https://brew.sh/)
 ```

acceptance_tests/empty.sh

@@ -9,7 +9,7 @@ EOL
 
 testEmptyEval() {
   X=$(./yq e test.yml)
-  expected=$(cat test.yml)
+  expected="# comment"
   assertEquals 0 $?
   assertEquals "$expected" "$X"
 }

acceptance_tests/inputs-format.sh

@@ -127,6 +127,7 @@ testInputXmlNamespaces() {
 EOL
 
   read -r -d '' expected << EOM
++p_xml: version="1.0"
 map:
   +xmlns: some-namespace
   +xmlns:xsi: some-instance
@@ -140,6 +141,26 @@ EOM
   assertEquals "$expected" "$X"
 }
 
+testInputXmlRoundtrip() {
+  cat >test.yml <<EOL
+<?xml version="1.0"?>
+<!DOCTYPE config SYSTEM "/etc/iwatch/iwatch.dtd" >
+<map xmlns="some-namespace" xmlns:xsi="some-instance" xsi:schemaLocation="some-url">Meow</map>
+EOL
+
+  read -r -d '' expected << EOM
+<?xml version="1.0"?>
+<!DOCTYPE config SYSTEM "/etc/iwatch/iwatch.dtd" >
+<map xmlns="some-namespace" xmlns:xsi="some-instance" xsi:schemaLocation="some-url">Meow</map>
+EOM
+
+  X=$(./yq -p=xml -o=xml test.yml)
+  assertEquals "$expected" "$X"
+
+  X=$(./yq ea -p=xml -o=xml test.yml)
+  assertEquals "$expected" "$X"
+}
+
 
 testInputXmlStrict() {
   cat >test.yml <<EOL
@@ -153,11 +174,11 @@ testInputXmlStrict() {
 </root>
 EOL
 
-  X=$(./yq -p=xml --xml-strict-mode test.yml 2>&1)
+  X=$(./yq -p=xml --xml-strict-mode test.yml -o=xml 2>&1)
   assertEquals 1 $?
   assertEquals "Error: bad file 'test.yml': XML syntax error on line 7: invalid character entity &writer;" "$X"
 
-  X=$(./yq ea -p=xml --xml-strict-mode test.yml 2>&1)
+  X=$(./yq ea -p=xml --xml-strict-mode test.yml -o=xml 2>&1)
   assertEquals "Error: bad file 'test.yml': XML syntax error on line 7: invalid character entity &writer;" "$X"
 }
 

cmd/constant.go

@@ -1,6 +1,5 @@
 package cmd
 
-var leadingContentPreProcessing = true
 var unwrapScalar = true
 
 var writeInplace = false
@@ -8,12 +7,6 @@ var outputToJSON = false
 var outputFormat = "yaml"
 var inputFormat = "yaml"
 
-var xmlAttributePrefix = "+"
-var xmlContentName = "+content"
-var xmlStrictMode = false
-var xmlKeepNamespace = true
-var xmlUseRawToken = true
-
 var exitStatus = false
 var forceColor = false
 var forceNoColor = false

cmd/evaluate_all_command.go

@@ -75,7 +75,7 @@ func evaluateAll(cmd *cobra.Command, args []string) (cmdError error) {
 		return err
 	}
 
-	decoder, err := configureDecoder()
+	decoder, err := configureDecoder(true)
 	if err != nil {
 		return err
 	}
@@ -109,13 +109,13 @@ func evaluateAll(cmd *cobra.Command, args []string) (cmdError error) {
 	switch len(args) {
 	case 0:
 		if nullInput {
-			err = yqlib.NewStreamEvaluator().EvaluateNew(processExpression(expression), printer, "")
+			err = yqlib.NewStreamEvaluator().EvaluateNew(processExpression(expression), printer)
 		} else {
 			cmd.Println(cmd.UsageString())
 			return nil
 		}
 	default:
-		err = allAtOnceEvaluator.EvaluateFiles(processExpression(expression), args, printer, leadingContentPreProcessing, decoder)
+		err = allAtOnceEvaluator.EvaluateFiles(processExpression(expression), args, printer, decoder)
 	}
 
 	completedSuccessfully = err == nil

cmd/evalute_sequence_command.go

@@ -97,7 +97,7 @@ func evaluateSequence(cmd *cobra.Command, args []string) (cmdError error) {
 
 	printer := yqlib.NewPrinter(encoder, printerWriter)
 
-	decoder, err := configureDecoder()
+	decoder, err := configureDecoder(false)
 	if err != nil {
 		return err
 	}
@@ -123,13 +123,13 @@ func evaluateSequence(cmd *cobra.Command, args []string) (cmdError error) {
 	switch len(args) {
 	case 0:
 		if nullInput {
-			err = streamEvaluator.EvaluateNew(processExpression(expression), printer, "")
+			err = streamEvaluator.EvaluateNew(processExpression(expression), printer)
 		} else {
 			cmd.Println(cmd.UsageString())
 			return nil
 		}
 	default:
-		err = streamEvaluator.EvaluateFiles(processExpression(expression), args, printer, leadingContentPreProcessing, decoder)
+		err = streamEvaluator.EvaluateFiles(processExpression(expression), args, printer, decoder)
 	}
 	completedSuccessfully = err == nil
 

cmd/root.go

@@ -47,14 +47,23 @@ yq -P sample.json
 			if verbose {
 				backend.SetLevel(logging.DEBUG, "")
 			} else {
-				backend.SetLevel(logging.ERROR, "")
+				backend.SetLevel(logging.WARNING, "")
 			}
 
 			logging.SetBackend(backend)
 			yqlib.InitExpressionParser()
-			yqlib.XMLPreferences.AttributePrefix = xmlAttributePrefix
-			yqlib.XMLPreferences.ContentName = xmlContentName
-			yqlib.XMLPreferences.StrictMode = xmlStrictMode
+			if (inputFormat == "x" || inputFormat == "xml") &&
+				outputFormat != "x" && outputFormat != "xml" &&
+				yqlib.ConfiguredXMLPreferences.AttributePrefix == "+" {
+				yqlib.GetLogger().Warning("The default xml-attribute-prefix will change in the v4.30 to `+@` to avoid " +
+					"naming conflicts with the default content name, directive name and proc inst prefix. If you need to keep " +
+					"`+` please set that value explicityly with --xml-attribute-prefix.")
+			}
+
+			//copy preference form global setting
+			yqlib.ConfiguredYamlPreferences.UnwrapScalar = unwrapScalar
+
+			yqlib.ConfiguredYamlPreferences.PrintDocSeparators = !noDocSeparators
 		},
 	}
 
@@ -69,11 +78,15 @@ yq -P sample.json
 	rootCmd.PersistentFlags().StringVarP(&outputFormat, "output-format", "o", "yaml", "[yaml|y|json|j|props|p|xml|x] output format type.")
 	rootCmd.PersistentFlags().StringVarP(&inputFormat, "input-format", "p", "yaml", "[yaml|y|props|p|xml|x] parse format for input. Note that json is a subset of yaml.")
 
-	rootCmd.PersistentFlags().StringVar(&xmlAttributePrefix, "xml-attribute-prefix", "+", "prefix for xml attributes")
-	rootCmd.PersistentFlags().StringVar(&xmlContentName, "xml-content-name", "+content", "name for xml content (if no attribute name is present).")
-	rootCmd.PersistentFlags().BoolVar(&xmlStrictMode, "xml-strict-mode", false, "enables strict parsing of XML. See https://pkg.go.dev/encoding/xml for more details.")
-	rootCmd.PersistentFlags().BoolVar(&xmlKeepNamespace, "xml-keep-namespace", true, "enables keeping namespace after parsing attributes")
-	rootCmd.PersistentFlags().BoolVar(&xmlUseRawToken, "xml-raw-token", true, "enables using RawToken method instead Token. Commonly disables namespace translations. See https://pkg.go.dev/encoding/xml#Decoder.RawToken for details.")
+	rootCmd.PersistentFlags().StringVar(&yqlib.ConfiguredXMLPreferences.AttributePrefix, "xml-attribute-prefix", "+", "prefix for xml attributes")
+	rootCmd.PersistentFlags().StringVar(&yqlib.ConfiguredXMLPreferences.ContentName, "xml-content-name", "+content", "name for xml content (if no attribute name is present).")
+	rootCmd.PersistentFlags().BoolVar(&yqlib.ConfiguredXMLPreferences.StrictMode, "xml-strict-mode", false, "enables strict parsing of XML. See https://pkg.go.dev/encoding/xml for more details.")
+	rootCmd.PersistentFlags().BoolVar(&yqlib.ConfiguredXMLPreferences.KeepNamespace, "xml-keep-namespace", true, "enables keeping namespace after parsing attributes")
+	rootCmd.PersistentFlags().BoolVar(&yqlib.ConfiguredXMLPreferences.UseRawToken, "xml-raw-token", true, "enables using RawToken method instead Token. Commonly disables namespace translations. See https://pkg.go.dev/encoding/xml#Decoder.RawToken for details.")
+	rootCmd.PersistentFlags().StringVar(&yqlib.ConfiguredXMLPreferences.ProcInstPrefix, "xml-proc-inst-prefix", "+p_", "prefix for xml processing instructions (e.g. <?xml version=\"1\"?>)")
+	rootCmd.PersistentFlags().StringVar(&yqlib.ConfiguredXMLPreferences.DirectiveName, "xml-directive-name", "+directive", "name for xml directives (e.g. <!DOCTYPE thing cat>)")
+	rootCmd.PersistentFlags().BoolVar(&yqlib.ConfiguredXMLPreferences.SkipProcInst, "xml-skip-proc-inst", false, "skip over process instructions (e.g. <?xml version=\"1\"?>)")
+	rootCmd.PersistentFlags().BoolVar(&yqlib.ConfiguredXMLPreferences.SkipDirectives, "xml-skip-directives", false, "skip over directives (e.g. <!DOCTYPE thing cat>)")
 
 	rootCmd.PersistentFlags().BoolVarP(&nullInput, "null-input", "n", false, "Don't read input, simply evaluate the expression given. Useful for creating docs from scratch.")
 	rootCmd.PersistentFlags().BoolVarP(&noDocSeparators, "no-doc", "N", false, "Don't print document separators (---)")
@@ -89,7 +102,7 @@ yq -P sample.json
 	rootCmd.PersistentFlags().BoolVarP(&forceNoColor, "no-colors", "M", false, "force print with no colors")
 	rootCmd.PersistentFlags().StringVarP(&frontMatter, "front-matter", "f", "", "(extract|process) first input as yaml front-matter. Extract will pull out the yaml content, process will run the expression against the yaml content, leaving the remaining data intact")
 	rootCmd.PersistentFlags().StringVarP(&forceExpression, "expression", "", "", "forcibly set the expression argument. Useful when yq argument detection thinks your expression is a file.")
-	rootCmd.PersistentFlags().BoolVarP(&leadingContentPreProcessing, "header-preprocess", "", true, "Slurp any header comments and separators before processing expression.")
+	rootCmd.PersistentFlags().BoolVarP(&yqlib.ConfiguredYamlPreferences.LeadingContentPreProcessing, "header-preprocess", "", true, "Slurp any header comments and separators before processing expression.")
 
 	rootCmd.PersistentFlags().StringVarP(&splitFileExp, "split-exp", "s", "", "print each result (or doc) into a file named (exp). [exp] argument must return a string. You can use $index in the expression as the result counter.")
 	rootCmd.PersistentFlags().StringVarP(&splitFileExpFile, "split-exp-file", "", "", "Use a file to specify the split-exp expression.")

cmd/utils.go

@@ -56,14 +56,14 @@ func initCommand(cmd *cobra.Command, args []string) (string, []string, error) {
 	return expression, args, nil
 }
 
-func configureDecoder() (yqlib.Decoder, error) {
+func configureDecoder(evaluateTogether bool) (yqlib.Decoder, error) {
 	yqlibInputFormat, err := yqlib.InputFormatFromString(inputFormat)
 	if err != nil {
 		return nil, err
 	}
 	switch yqlibInputFormat {
 	case yqlib.XMLInputFormat:
-		return yqlib.NewXMLDecoder(xmlAttributePrefix, xmlContentName, xmlStrictMode, xmlKeepNamespace, xmlUseRawToken), nil
+		return yqlib.NewXMLDecoder(yqlib.ConfiguredXMLPreferences), nil
 	case yqlib.PropertiesInputFormat:
 		return yqlib.NewPropertiesDecoder(), nil
 	case yqlib.JsonInputFormat:
@@ -73,8 +73,9 @@ func configureDecoder() (yqlib.Decoder, error) {
 	case yqlib.TSVObjectInputFormat:
 		return yqlib.NewCSVObjectDecoder('\t'), nil
 	}
-
-	return yqlib.NewYamlDecoder(), nil
+	prefs := yqlib.ConfiguredYamlPreferences
+	prefs.EvaluateTogether = evaluateTogether
+	return yqlib.NewYamlDecoder(prefs), nil
 }
 
 func configurePrinterWriter(format yqlib.PrinterOutputFormat, out io.Writer) (yqlib.PrinterWriter, error) {
@@ -105,9 +106,9 @@ func configureEncoder(format yqlib.PrinterOutputFormat) yqlib.Encoder {
 	case yqlib.TSVOutputFormat:
 		return yqlib.NewCsvEncoder('\t')
 	case yqlib.YamlOutputFormat:
-		return yqlib.NewYamlEncoder(indent, colorsEnabled, !noDocSeparators, unwrapScalar)
+		return yqlib.NewYamlEncoder(indent, colorsEnabled, yqlib.ConfiguredYamlPreferences)
 	case yqlib.XMLOutputFormat:
-		return yqlib.NewXMLEncoder(indent, xmlAttributePrefix, xmlContentName)
+		return yqlib.NewXMLEncoder(indent, yqlib.ConfiguredXMLPreferences)
 	}
 	panic("invalid encoder")
 }

examples/mike.xml

@@ -1,14 +1,7 @@
-<!-- before cat -->
-<cat>
-	<!-- in cat before -->
-	<x>3<!-- multi
-line comment
-for x --></x>
-	<y>
-		<!-- in y before -->
-		<d><!-- in d before -->4<!-- in d after --></d>
-		<!-- in y after -->
-	</y>
-	<!-- in_cat_after -->
-</cat>
-<!-- after cat -->
+<?xml version="1.0"?>
+<!DOCTYPE config SYSTEM "/etc/iwatch/iwatch.dtd" >
+<apple>
+<?coolioo version="1.0"?>
+<!DOCTYPE config SYSTEM "/etc/iwatch/iwatch.dtd" >
+<b>things</b>
+</apple>

go.mod

@@ -8,11 +8,11 @@ require (
 	github.com/elliotchance/orderedmap v1.5.0
 	github.com/fatih/color v1.13.0
 	github.com/goccy/go-json v0.9.11
-	github.com/goccy/go-yaml v1.9.5
+	github.com/goccy/go-yaml v1.9.6
 	github.com/jinzhu/copier v0.3.5
 	github.com/magiconair/properties v1.8.6
 	github.com/pkg/diff v0.0.0-20210226163009-20ebb0f2a09e
-	github.com/spf13/cobra v1.6.0
+	github.com/spf13/cobra v1.6.1
 	golang.org/x/net v0.0.0-20220708220712-1185a9018129
 	gopkg.in/op/go-logging.v1 v1.0.0-20160211212156-b2cb9fa56473
 	gopkg.in/yaml.v3 v3.0.1

go.sum

@@ -21,8 +21,8 @@ github.com/go-playground/universal-translator v0.17.0/go.mod h1:UkSxE5sNxxRwHyU+
 github.com/go-playground/validator/v10 v10.4.1/go.mod h1:nlOn6nFhuKACm19sB/8EGNn9GlaMV7XkbRSipzJ0Ii4=
 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/goccy/go-yaml v1.9.6 h1:KhAu1zf9JXnm3vbG49aDE0E5uEBUsM4uwD31/58ZWyI=
+github.com/goccy/go-yaml v1.9.6/go.mod h1:JubOolP3gh0HpiBc4BLRD4YmjEjHAmIIB2aaXKkTfoE=
 github.com/hexops/gotextdiff v1.0.3 h1:gitA9+qJrrTCsiCl7+kh75nPqQt1cx4ZkudSTLoUqJM=
 github.com/inconshreveable/mousetrap v1.0.1 h1:U3uMjPSQEBMNp1lFxmllqCPM6P5u/Xq7Pgzkat/bFNc=
 github.com/inconshreveable/mousetrap v1.0.1/go.mod h1:vpF70FUmC8bwa3OWnCshd2FqLfsEA9PFc4w1p2J65bw=
@@ -43,8 +43,8 @@ github.com/pkg/diff v0.0.0-20210226163009-20ebb0f2a09e/go.mod h1:pJLUxLENpZxwdsK
 github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM=
 github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
 github.com/russross/blackfriday/v2 v2.1.0/go.mod h1:+Rmxgy9KzJVeS9/2gXHxylqXiyQDYRxCVz55jmeOWTM=
-github.com/spf13/cobra v1.6.0 h1:42a0n6jwCot1pUmomAp4T7DeMD+20LFv4Q54pxLf2LI=
-github.com/spf13/cobra v1.6.0/go.mod h1:IOw/AERYS7UzyrGinqmz6HLUo219MORXGxhbaJUqzrY=
+github.com/spf13/cobra v1.6.1 h1:o94oiPyS4KD1mPy2fmcYYHHfCxLqYjJOhGsCHFZtEzA=
+github.com/spf13/cobra v1.6.1/go.mod h1:IOw/AERYS7UzyrGinqmz6HLUo219MORXGxhbaJUqzrY=
 github.com/spf13/pflag v1.0.5 h1:iy+VFUOCP1a+8yFto/drg2CJ5u0yRoB7fZw3DKv/JXA=
 github.com/spf13/pflag v1.0.5/go.mod h1:McXfInJRrz4CZXVZOBLb0bTZqETkiAhM9Iw0y3An2Bg=
 github.com/stretchr/objx v0.1.0/go.mod h1:HFkY916IF+rwdDfMAkV7OtwuqBVzrE8GR6GFx+wExME=
@@ -62,6 +62,7 @@ golang.org/x/sys v0.0.0-20200116001909-b77594299b42/go.mod h1:h1NjWce9XRLGQEsW7w
 golang.org/x/sys v0.0.0-20200223170610-d5e6a3e2c0ae/go.mod h1:h1NjWce9XRLGQEsW7wpKNCjG9DtNlClVuFLEZdDNbEs=
 golang.org/x/sys v0.0.0-20210630005230-0f9fa26af87c/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
 golang.org/x/sys v0.0.0-20210927094055-39ccf1dd6fa6/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
+golang.org/x/sys v0.0.0-20220406163625-3f8b81556e12/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
 golang.org/x/sys v0.0.0-20220712014510-0a85c31ab51e h1:NHvCuwuS43lGnYhten69ZWqi2QOj/CiDNcKbVqwVoew=
 golang.org/x/sys v0.0.0-20220712014510-0a85c31ab51e/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
 golang.org/x/text v0.3.0/go.mod h1:NqM8EUOU14njkJ3fqMW+pc6Ldnwhi/IjpwHt7yyuwOQ=

pkg/yqlib/all_at_once_evaluator.go

@@ -8,7 +8,7 @@ import (
 
 // A yaml expression evaluator that runs the expression once against all files/nodes in memory.
 type Evaluator interface {
-	EvaluateFiles(expression string, filenames []string, printer Printer, leadingContentPreProcessing bool, decoder Decoder) error
+	EvaluateFiles(expression string, filenames []string, printer Printer, decoder Decoder) error
 
 	// EvaluateNodes takes an expression and one or more yaml nodes, returning a list of matching candidate nodes
 	EvaluateNodes(expression string, nodes ...*yaml.Node) (*list.List, error)
@@ -46,21 +46,16 @@ func (e *allAtOnceEvaluator) EvaluateCandidateNodes(expression string, inputCand
 	return context.MatchingNodes, nil
 }
 
-func (e *allAtOnceEvaluator) EvaluateFiles(expression string, filenames []string, printer Printer, leadingContentPreProcessing bool, decoder Decoder) error {
+func (e *allAtOnceEvaluator) EvaluateFiles(expression string, filenames []string, printer Printer, decoder Decoder) error {
 	fileIndex := 0
-	firstFileLeadingContent := ""
 
 	var allDocuments = list.New()
 	for _, filename := range filenames {
-		reader, leadingContent, err := readStream(filename, fileIndex == 0 && leadingContentPreProcessing)
+		reader, err := readStream(filename)
 		if err != nil {
 			return err
 		}
 
-		if fileIndex == 0 {
-			firstFileLeadingContent = leadingContent
-		}
-
 		fileDocuments, err := readDocuments(reader, filename, fileIndex, decoder)
 		if err != nil {
 			return err
@@ -75,11 +70,9 @@ func (e *allAtOnceEvaluator) EvaluateFiles(expression string, filenames []string
 			Filename:       "",
 			Node:           &yaml.Node{Kind: yaml.DocumentNode, Content: []*yaml.Node{{Tag: "!!null", Kind: yaml.ScalarNode}}},
 			FileIndex:      0,
-			LeadingContent: firstFileLeadingContent,
+			LeadingContent: "",
 		}
 		allDocuments.PushBack(candidateNode)
-	} else {
-		allDocuments.Front().Value.(*CandidateNode).LeadingContent = firstFileLeadingContent
 	}
 
 	matches, err := e.EvaluateCandidateNodes(expression, allDocuments)

pkg/yqlib/csv_test.go

@@ -136,13 +136,13 @@ var csvScenarios = []formatScenario{
 func testCSVScenario(t *testing.T, s formatScenario) {
 	switch s.scenarioType {
 	case "encode-csv":
-		test.AssertResultWithContext(t, s.expected, processFormatScenario(s, NewYamlDecoder(), NewCsvEncoder(',')), s.description)
+		test.AssertResultWithContext(t, s.expected, processFormatScenario(s, NewYamlDecoder(ConfiguredYamlPreferences), NewCsvEncoder(',')), s.description)
 	case "encode-tsv":
-		test.AssertResultWithContext(t, s.expected, processFormatScenario(s, NewYamlDecoder(), NewCsvEncoder('\t')), s.description)
+		test.AssertResultWithContext(t, s.expected, processFormatScenario(s, NewYamlDecoder(ConfiguredYamlPreferences), NewCsvEncoder('\t')), s.description)
 	case "decode-csv-object":
-		test.AssertResultWithContext(t, s.expected, processFormatScenario(s, NewCSVObjectDecoder(','), NewYamlEncoder(2, false, true, true)), s.description)
+		test.AssertResultWithContext(t, s.expected, processFormatScenario(s, NewCSVObjectDecoder(','), NewYamlEncoder(2, false, ConfiguredYamlPreferences)), s.description)
 	case "decode-tsv-object":
-		test.AssertResultWithContext(t, s.expected, processFormatScenario(s, NewCSVObjectDecoder('\t'), NewYamlEncoder(2, false, true, true)), s.description)
+		test.AssertResultWithContext(t, s.expected, processFormatScenario(s, NewCSVObjectDecoder('\t'), NewYamlEncoder(2, false, ConfiguredYamlPreferences)), s.description)
 	case "roundtrip-csv":
 		test.AssertResultWithContext(t, s.expected, processFormatScenario(s, NewCSVObjectDecoder(','), NewCsvEncoder(',')), s.description)
 	default:
@@ -171,7 +171,7 @@ func documentCSVDecodeObjectScenario(w *bufio.Writer, s formatScenario, formatTy
 	}
 
 	writeOrPanic(w, fmt.Sprintf("```yaml\n%v```\n\n",
-		processFormatScenario(s, NewCSVObjectDecoder(separator), NewYamlEncoder(s.indent, false, true, true))),
+		processFormatScenario(s, NewCSVObjectDecoder(separator), NewYamlEncoder(s.indent, false, ConfiguredYamlPreferences))),
 	)
 }
 
@@ -203,7 +203,7 @@ func documentCSVEncodeScenario(w *bufio.Writer, s formatScenario, formatType str
 	}
 
 	writeOrPanic(w, fmt.Sprintf("```%v\n%v```\n\n", formatType,
-		processFormatScenario(s, NewYamlDecoder(), NewCsvEncoder(separator))),
+		processFormatScenario(s, NewYamlDecoder(ConfiguredYamlPreferences), NewCsvEncoder(separator))),
 	)
 }
 

pkg/yqlib/decoder.go

@@ -3,8 +3,6 @@ package yqlib
 import (
 	"fmt"
 	"io"
-
-	yaml "gopkg.in/yaml.v3"
 )
 
 type InputFormat uint
@@ -20,8 +18,8 @@ const (
 )
 
 type Decoder interface {
-	Init(reader io.Reader)
-	Decode(node *yaml.Node) error
+	Init(reader io.Reader) error
+	Decode() (*CandidateNode, error)
 }
 
 func InputFormatFromString(format string) (InputFormat, error) {

pkg/yqlib/decoder_base64.go

@@ -19,21 +19,22 @@ func NewBase64Decoder() Decoder {
 	return &base64Decoder{finished: false, encoding: *base64.StdEncoding}
 }
 
-func (dec *base64Decoder) Init(reader io.Reader) {
+func (dec *base64Decoder) Init(reader io.Reader) error {
 	dec.reader = reader
 	dec.readAnything = false
 	dec.finished = false
+	return nil
 }
 
-func (dec *base64Decoder) Decode(rootYamlNode *yaml.Node) error {
+func (dec *base64Decoder) Decode() (*CandidateNode, error) {
 	if dec.finished {
-		return io.EOF
+		return nil, io.EOF
 	}
 	base64Reader := base64.NewDecoder(&dec.encoding, dec.reader)
 	buf := new(bytes.Buffer)
 
 	if _, err := buf.ReadFrom(base64Reader); err != nil {
-		return err
+		return nil, err
 	}
 	if buf.Len() == 0 {
 		dec.finished = true
@@ -42,12 +43,15 @@ func (dec *base64Decoder) Decode(rootYamlNode *yaml.Node) error {
 		// otherwise if we've already read some bytes, and now we get
 		// an empty string, then we are done.
 		if dec.readAnything {
-			return io.EOF
+			return nil, io.EOF
 		}
 	}
 	dec.readAnything = true
-	rootYamlNode.Kind = yaml.ScalarNode
-	rootYamlNode.Tag = "!!str"
-	rootYamlNode.Value = buf.String()
-	return nil
+	return &CandidateNode{
+		Node: &yaml.Node{
+			Kind:  yaml.ScalarNode,
+			Tag:   "!!str",
+			Value: buf.String(),
+		},
+	}, nil
 }

pkg/yqlib/decoder_csv_object.go

@@ -19,12 +19,13 @@ func NewCSVObjectDecoder(separator rune) Decoder {
 	return &csvObjectDecoder{separator: separator}
 }
 
-func (dec *csvObjectDecoder) Init(reader io.Reader) {
+func (dec *csvObjectDecoder) Init(reader io.Reader) error {
 	cleanReader, enc := utfbom.Skip(reader)
 	log.Debugf("Detected encoding: %s\n", enc)
 	dec.reader = *csv.NewReader(cleanReader)
 	dec.reader.Comma = dec.separator
 	dec.finished = false
+	return nil
 }
 
 func (dec *csvObjectDecoder) convertToYamlNode(content string) *yaml.Node {
@@ -47,14 +48,14 @@ func (dec *csvObjectDecoder) createObject(headerRow []string, contentRow []strin
 	return objectNode
 }
 
-func (dec *csvObjectDecoder) Decode(rootYamlNode *yaml.Node) error {
+func (dec *csvObjectDecoder) Decode() (*CandidateNode, error) {
 	if dec.finished {
-		return io.EOF
+		return nil, io.EOF
 	}
 	headerRow, err := dec.reader.Read()
 	log.Debugf(": headerRow%v", headerRow)
 	if err != nil {
-		return err
+		return nil, err
 	}
 
 	rootArray := &yaml.Node{Kind: yaml.SequenceNode, Tag: "!!seq"}
@@ -68,13 +69,13 @@ func (dec *csvObjectDecoder) Decode(rootYamlNode *yaml.Node) error {
 		log.Debugf("Read next contentRow: %v, %v", contentRow, err)
 	}
 	if !errors.Is(err, io.EOF) {
-		return err
+		return nil, err
 	}
 
-	log.Debugf("finished, contentRow%v", contentRow)
-	log.Debugf("err: %v", err)
-
-	rootYamlNode.Kind = yaml.DocumentNode
-	rootYamlNode.Content = []*yaml.Node{rootArray}
-	return nil
+	return &CandidateNode{
+		Node: &yaml.Node{
+			Kind:    yaml.DocumentNode,
+			Content: []*yaml.Node{rootArray},
+		},
+	}, nil
 }

pkg/yqlib/decoder_json.go

@@ -16,26 +16,31 @@ func NewJSONDecoder() Decoder {
 	return &jsonDecoder{}
 }
 
-func (dec *jsonDecoder) Init(reader io.Reader) {
+func (dec *jsonDecoder) Init(reader io.Reader) error {
 	dec.decoder = *json.NewDecoder(reader)
+	return nil
 }
 
-func (dec *jsonDecoder) Decode(rootYamlNode *yaml.Node) error {
+func (dec *jsonDecoder) Decode() (*CandidateNode, error) {
 
 	var dataBucket orderedMap
 	log.Debug("going to decode")
 	err := dec.decoder.Decode(&dataBucket)
 	if err != nil {
-		return err
+		return nil, err
 	}
 	node, err := dec.convertToYamlNode(&dataBucket)
 
 	if err != nil {
-		return err
+		return nil, err
 	}
-	rootYamlNode.Kind = yaml.DocumentNode
-	rootYamlNode.Content = []*yaml.Node{node}
-	return nil
+
+	return &CandidateNode{
+		Node: &yaml.Node{
+			Kind:    yaml.DocumentNode,
+			Content: []*yaml.Node{node},
+		},
+	}, nil
 }
 
 func (dec *jsonDecoder) convertToYamlNode(data *orderedMap) (*yaml.Node, error) {

pkg/yqlib/decoder_properties.go

@@ -2,6 +2,7 @@ package yqlib
 
 import (
 	"bytes"
+	"fmt"
 	"io"
 	"strconv"
 	"strings"
@@ -20,9 +21,10 @@ func NewPropertiesDecoder() Decoder {
 	return &propertiesDecoder{d: NewDataTreeNavigator(), finished: false}
 }
 
-func (dec *propertiesDecoder) Init(reader io.Reader) {
+func (dec *propertiesDecoder) Init(reader io.Reader) error {
 	dec.reader = reader
 	dec.finished = false
+	return nil
 }
 
 func parsePropKey(key string) []interface{} {
@@ -46,15 +48,49 @@ func (dec *propertiesDecoder) processComment(c string) string {
 	return "# " + c
 }
 
-func (dec *propertiesDecoder) applyProperty(properties *properties.Properties, context Context, key string) error {
+func (dec *propertiesDecoder) applyPropertyComments(context Context, path []interface{}, comments []string) error {
+	assignmentOp := &Operation{OperationType: assignOpType, Preferences: assignPreferences{}}
+
+	rhsCandidateNode := &CandidateNode{
+		Path: path,
+		Node: &yaml.Node{
+			Tag:         "!!str",
+			Value:       fmt.Sprintf("%v", path[len(path)-1]),
+			HeadComment: dec.processComment(strings.Join(comments, "\n")),
+			Kind:        yaml.ScalarNode,
+		},
+	}
+
+	rhsCandidateNode.Node.Tag = guessTagFromCustomType(rhsCandidateNode.Node)
+
+	rhsOp := &Operation{OperationType: valueOpType, CandidateNode: rhsCandidateNode}
+
+	assignmentOpNode := &ExpressionNode{
+		Operation: assignmentOp,
+		LHS:       createTraversalTree(path, traversePreferences{}, true),
+		RHS:       &ExpressionNode{Operation: rhsOp},
+	}
+
+	_, err := dec.d.GetMatchingNodes(context, assignmentOpNode)
+	return err
+}
+
+func (dec *propertiesDecoder) applyProperty(context Context, properties *properties.Properties, key string) error {
 	value, _ := properties.Get(key)
 	path := parsePropKey(key)
 
+	propertyComments := properties.GetComments(key)
+	if len(propertyComments) > 0 {
+		err := dec.applyPropertyComments(context, path, propertyComments)
+		if err != nil {
+			return nil
+		}
+	}
+
 	rhsNode := &yaml.Node{
-		Value:       value,
-		Tag:         "!!str",
-		Kind:        yaml.ScalarNode,
-		LineComment: dec.processComment(properties.GetComment(key)),
+		Value: value,
+		Tag:   "!!str",
+		Kind:  yaml.ScalarNode,
 	}
 
 	rhsNode.Tag = guessTagFromCustomType(rhsNode)
@@ -78,22 +114,22 @@ func (dec *propertiesDecoder) applyProperty(properties *properties.Properties, c
 	return err
 }
 
-func (dec *propertiesDecoder) Decode(rootYamlNode *yaml.Node) error {
+func (dec *propertiesDecoder) Decode() (*CandidateNode, error) {
 	if dec.finished {
-		return io.EOF
+		return nil, io.EOF
 	}
 	buf := new(bytes.Buffer)
 
 	if _, err := buf.ReadFrom(dec.reader); err != nil {
-		return err
+		return nil, err
 	}
 	if buf.Len() == 0 {
 		dec.finished = true
-		return io.EOF
+		return nil, io.EOF
 	}
 	properties, err := properties.LoadString(buf.String())
 	if err != nil {
-		return err
+		return nil, err
 	}
 	properties.DisableExpansion = true
 
@@ -108,15 +144,18 @@ func (dec *propertiesDecoder) Decode(rootYamlNode *yaml.Node) error {
 	context = context.SingleChildContext(rootMap)
 
 	for _, key := range properties.Keys() {
-		if err := dec.applyProperty(properties, context, key); err != nil {
-			return err
+		if err := dec.applyProperty(context, properties, key); err != nil {
+			return nil, err
 		}
 
 	}
-
-	rootYamlNode.Kind = yaml.DocumentNode
-	rootYamlNode.Content = []*yaml.Node{rootMap.Node}
 	dec.finished = true
-	return nil
+
+	return &CandidateNode{
+		Node: &yaml.Node{
+			Kind:    yaml.DocumentNode,
+			Content: []*yaml.Node{rootMap.Node},
+		},
+	}, nil
 
 }

pkg/yqlib/decoder_test.go

@@ -23,7 +23,7 @@ func processFormatScenario(s formatScenario, decoder Decoder, encoder Encoder) s
 	writer := bufio.NewWriter(&output)
 
 	if decoder == nil {
-		decoder = NewYamlDecoder()
+		decoder = NewYamlDecoder(ConfiguredYamlPreferences)
 	}
 
 	inputs, err := readDocuments(strings.NewReader(s.input), "sample.yml", 0, decoder)

pkg/yqlib/decoder_xml.go

@@ -12,34 +12,24 @@ import (
 )
 
 type xmlDecoder struct {
-	reader          io.Reader
-	readAnything    bool
-	attributePrefix string
-	contentName     string
-	strictMode      bool
-	keepNamespace   bool
-	useRawToken     bool
-	finished        bool
+	reader       io.Reader
+	readAnything bool
+	finished     bool
+	prefs        XmlPreferences
 }
 
-func NewXMLDecoder(attributePrefix string, contentName string, strictMode bool, keepNamespace bool, useRawToken bool) Decoder {
-	if contentName == "" {
-		contentName = "content"
-	}
+func NewXMLDecoder(prefs XmlPreferences) Decoder {
 	return &xmlDecoder{
-		attributePrefix: attributePrefix,
-		contentName:     contentName,
-		finished:        false,
-		strictMode:      strictMode,
-		keepNamespace:   keepNamespace,
-		useRawToken:     useRawToken,
+		finished: false,
+		prefs:    prefs,
 	}
 }
 
-func (dec *xmlDecoder) Init(reader io.Reader) {
+func (dec *xmlDecoder) Init(reader io.Reader) error {
 	dec.reader = reader
 	dec.readAnything = false
 	dec.finished = false
+	return nil
 }
 
 func (dec *xmlDecoder) createSequence(nodes []*xmlNode) (*yaml.Node, error) {
@@ -67,7 +57,7 @@ func (dec *xmlDecoder) createMap(n *xmlNode) (*yaml.Node, error) {
 	yamlNode := &yaml.Node{Kind: yaml.MappingNode, Tag: "!!map"}
 
 	if len(n.Data) > 0 {
-		label := dec.contentName
+		label := dec.prefs.ContentName
 		labelNode := createScalarNode(label, label)
 		labelNode.HeadComment = dec.processComment(n.HeadComment)
 		labelNode.FootComment = dec.processComment(n.FootComment)
@@ -86,9 +76,7 @@ func (dec *xmlDecoder) createMap(n *xmlNode) (*yaml.Node, error) {
 
 		}
 
-		// if i == len(n.Children)-1 {
 		labelNode.FootComment = dec.processComment(keyValuePair.FootComment)
-		// }
 
 		log.Debug("len of children in %v is %v", label, len(children))
 		if len(children) > 1 {
@@ -131,32 +119,36 @@ func (dec *xmlDecoder) convertToYamlNode(n *xmlNode) (*yaml.Node, error) {
 	return scalar, nil
 }
 
-func (dec *xmlDecoder) Decode(rootYamlNode *yaml.Node) error {
+func (dec *xmlDecoder) Decode() (*CandidateNode, error) {
 	if dec.finished {
-		return io.EOF
+		return nil, io.EOF
 	}
 	root := &xmlNode{}
 	// cant use xj - it doesn't keep map order.
 	err := dec.decodeXML(root)
 
 	if err != nil {
-		return err
+		return nil, err
 	}
 	firstNode, err := dec.convertToYamlNode(root)
 
 	if err != nil {
-		return err
+		return nil, err
 	} else if firstNode.Tag == "!!null" {
 		dec.finished = true
 		if dec.readAnything {
-			return io.EOF
+			return nil, io.EOF
 		}
 	}
 	dec.readAnything = true
-	rootYamlNode.Kind = yaml.DocumentNode
-	rootYamlNode.Content = []*yaml.Node{firstNode}
 	dec.finished = true
-	return nil
+
+	return &CandidateNode{
+		Node: &yaml.Node{
+			Kind:    yaml.DocumentNode,
+			Content: []*yaml.Node{firstNode},
+		},
+	}, nil
 }
 
 type xmlNode struct {
@@ -205,7 +197,7 @@ type element struct {
 // of the map keys.
 func (dec *xmlDecoder) decodeXML(root *xmlNode) error {
 	xmlDec := xml.NewDecoder(dec.reader)
-	xmlDec.Strict = dec.strictMode
+	xmlDec.Strict = dec.prefs.StrictMode
 	// That will convert the charset if the provided XML is non-UTF-8
 	xmlDec.CharsetReader = charset.NewReaderLabel
 
@@ -216,7 +208,7 @@ func (dec *xmlDecoder) decodeXML(root *xmlNode) error {
 	}
 
 	getToken := func() (xml.Token, error) {
-		if dec.useRawToken {
+		if dec.prefs.UseRawToken {
 			return xmlDec.RawToken()
 		}
 		return xmlDec.Token()
@@ -244,12 +236,12 @@ func (dec *xmlDecoder) decodeXML(root *xmlNode) error {
 
 			// Extract attributes as children
 			for _, a := range se.Attr {
-				if dec.keepNamespace {
+				if dec.prefs.KeepNamespace {
 					if a.Name.Space != "" {
 						a.Name.Local = a.Name.Space + ":" + a.Name.Local
 					}
 				}
-				elem.n.AddChild(dec.attributePrefix+a.Name.Local, &xmlNode{Data: a.Value})
+				elem.n.AddChild(dec.prefs.AttributePrefix+a.Name.Local, &xmlNode{Data: a.Value})
 			}
 		case xml.CharData:
 			// Extract XML data (if any)
@@ -282,6 +274,14 @@ func (dec *xmlDecoder) decodeXML(root *xmlNode) error {
 				elem.n.HeadComment = joinFilter([]string{elem.n.HeadComment, commentStr})
 			}
 
+		case xml.ProcInst:
+			if !dec.prefs.SkipProcInst {
+				elem.n.AddChild(dec.prefs.ProcInstPrefix+se.Target, &xmlNode{Data: string(se.Inst)})
+			}
+		case xml.Directive:
+			if !dec.prefs.SkipDirectives {
+				elem.n.AddChild(dec.prefs.DirectiveName, &xmlNode{Data: string(se)})
+			}
 		}
 	}
 

pkg/yqlib/decoder_yaml.go

@@ -1,23 +1,114 @@
 package yqlib
 
 import (
+	"bufio"
+	"errors"
 	"io"
+	"regexp"
+	"strings"
 
 	yaml "gopkg.in/yaml.v3"
 )
 
 type yamlDecoder struct {
 	decoder yaml.Decoder
+	// work around of various parsing issues by yaml.v3 with document headers
+	prefs          YamlPreferences
+	leadingContent string
+	readAnything   bool
+	firstFile      bool
 }
 
-func NewYamlDecoder() Decoder {
-	return &yamlDecoder{}
+func NewYamlDecoder(prefs YamlPreferences) Decoder {
+	return &yamlDecoder{prefs: prefs, firstFile: true}
 }
 
-func (dec *yamlDecoder) Init(reader io.Reader) {
-	dec.decoder = *yaml.NewDecoder(reader)
+func (dec *yamlDecoder) processReadStream(reader *bufio.Reader) (io.Reader, string, error) {
+	var commentLineRegEx = regexp.MustCompile(`^\s*#`)
+	var sb strings.Builder
+	for {
+		peekBytes, err := reader.Peek(3)
+		if errors.Is(err, io.EOF) {
+			// EOF are handled else where..
+			return reader, sb.String(), nil
+		} else if err != nil {
+			return reader, sb.String(), err
+		} else if string(peekBytes) == "---" {
+			_, err := reader.ReadString('\n')
+			sb.WriteString("$yqDocSeperator$\n")
+			if errors.Is(err, io.EOF) {
+				return reader, sb.String(), nil
+			} else if err != nil {
+				return reader, sb.String(), err
+			}
+		} else if commentLineRegEx.MatchString(string(peekBytes)) {
+			line, err := reader.ReadString('\n')
+			sb.WriteString(line)
+			if errors.Is(err, io.EOF) {
+				return reader, sb.String(), nil
+			} else if err != nil {
+				return reader, sb.String(), err
+			}
+		} else {
+			return reader, sb.String(), nil
+		}
+	}
 }
 
-func (dec *yamlDecoder) Decode(rootYamlNode *yaml.Node) error {
-	return dec.decoder.Decode(rootYamlNode)
+func (dec *yamlDecoder) Init(reader io.Reader) error {
+	readerToUse := reader
+	leadingContent := ""
+	var err error
+	// if we 'evaluating together' - we only process the leading content
+	// of the first file - this ensures comments from subsequent files are
+	// merged together correctly.
+	if dec.prefs.LeadingContentPreProcessing && (!dec.prefs.EvaluateTogether || dec.firstFile) {
+		readerToUse, leadingContent, err = dec.processReadStream(bufio.NewReader(reader))
+		if err != nil {
+			return err
+		}
+	}
+	dec.leadingContent = leadingContent
+	dec.readAnything = false
+	dec.decoder = *yaml.NewDecoder(readerToUse)
+	dec.firstFile = false
+	return nil
+}
+
+func (dec *yamlDecoder) Decode() (*CandidateNode, error) {
+	var dataBucket yaml.Node
+
+	err := dec.decoder.Decode(&dataBucket)
+	if errors.Is(err, io.EOF) && dec.leadingContent != "" && !dec.readAnything {
+		// force returning an empty node with a comment.
+		dec.readAnything = true
+		return dec.blankNodeWithComment(), nil
+
+	} else if err != nil {
+		return nil, err
+	}
+
+	candidateNode := &CandidateNode{
+		Node: &dataBucket,
+	}
+
+	if dec.leadingContent != "" {
+		candidateNode.LeadingContent = dec.leadingContent
+		dec.leadingContent = ""
+	}
+	// move document comments into candidate node
+	// otherwise unwrap drops them.
+	candidateNode.TrailingContent = dataBucket.FootComment
+	dataBucket.FootComment = ""
+	return candidateNode, nil
+}
+
+func (dec *yamlDecoder) blankNodeWithComment() *CandidateNode {
+	return &CandidateNode{
+		Document:       0,
+		Filename:       "",
+		Node:           &yaml.Node{Kind: yaml.DocumentNode, Content: []*yaml.Node{{Tag: "!!null", Kind: yaml.ScalarNode}}},
+		FileIndex:      0,
+		LeadingContent: dec.leadingContent,
+	}
 }

pkg/yqlib/doc/notification-snippet.md

@@ -1,6 +0,0 @@
-
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}

pkg/yqlib/doc/operators/add.md

@@ -9,12 +9,6 @@ Add behaves differently according to the type of the LHS:
 Use `+=` as a relative append assign for things like increment. Note that `.a += .x` is equivalent to running `.a = .a + .x`.
 
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Concatenate arrays
 Given a sample.yml file of:
 ```yaml

pkg/yqlib/doc/operators/alternative-default-value.md

@@ -2,12 +2,6 @@
 
 This operator is used to provide alternative (or default) values when a particular expression is either null or false.
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## LHS is defined
 Given a sample.yml file of:
 ```yaml

pkg/yqlib/doc/operators/anchor-and-alias-operators.md

@@ -5,12 +5,6 @@ Use the `alias` and `anchor` operators to read and write yaml aliases and anchor
 `yq` supports merge aliases (like `<<: *blah`) however this is no longer in the standard yaml spec (1.2) and so `yq` will automatically add the `!!merge` tag to these nodes as it is effectively a custom tag.
 
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified.&#x20;
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Merge one map
 see https://yaml.org/type/merge.html
 

pkg/yqlib/doc/operators/assign-update.md

@@ -12,12 +12,6 @@ This will do a similar thing to the plain form, however, the RHS expression is r
 ### Flags
 - `c` clobber custom tags
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Create yaml file
 Running
 ```bash

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

@@ -16,12 +16,6 @@ These are most commonly used with the `select` operator to filter particular nod
 - comparison (`>=`, `<` etc) operators [here](https://mikefarah.gitbook.io/yq/operators/compare)
 - select operator [here](https://mikefarah.gitbook.io/yq/operators/select)
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified.&#x20;
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## `or` example
 Running
 ```bash

pkg/yqlib/doc/operators/collect-into-array.md

@@ -3,12 +3,6 @@
 This creates an array using the expression between the square brackets.
 
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Collect empty
 Running
 ```bash

pkg/yqlib/doc/operators/column.md

@@ -2,12 +2,6 @@
 
 Returns the column of the matching node. Starts from 1, 0 indicates there was no column data.
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Returns column of _value_ node
 Given a sample.yml file of:
 ```yaml

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

@@ -10,12 +10,6 @@ This will assign the LHS nodes comments to the expression on the RHS. The RHS is
 ### relative form: `|=` 
 Similar to the plain form, however the RHS evaluates against each matching LHS node! This is useful if you want to set the comments as a relative expression of the node, for instance its value or path.
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Set line comment
 Set the comment on the key node for more reliability (see below).
 
@@ -252,6 +246,7 @@ Note the use of `...` to ensure key nodes are included.
 
 Given a sample.yml file of:
 ```yaml
+# hi
 a: cat # comment
 # great
 b: # key comment
@@ -269,6 +264,7 @@ b:
 ## Get line comment
 Given a sample.yml file of:
 ```yaml
+# welcome!
 a: cat # meow
 # have a great day
 ```

pkg/yqlib/doc/operators/compare.md

@@ -14,12 +14,6 @@ The following types are currently supported:
 - boolean operators (`and`, `or`, `any` etc) [here](https://mikefarah.gitbook.io/yq/operators/boolean-operators)
 - select operator [here](https://mikefarah.gitbook.io/yq/operators/select)
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Compare numbers (>)
 Given a sample.yml file of:
 ```yaml

pkg/yqlib/doc/operators/contains.md

@@ -2,12 +2,6 @@
 
 This returns `true` if the context contains the passed in parameter, and false otherwise.
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Array contains array
 Array is equal or subset of
 

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

@@ -2,12 +2,6 @@
 
 This is used to construct objects (or maps). This can be used against existing yaml, or to create fresh yaml documents.
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Collect empty object
 Running
 ```bash

pkg/yqlib/doc/operators/datetime.md

@@ -25,12 +25,6 @@ Durations are parsed using golangs built in [ParseDuration](https://pkg.go.dev/t
 
 You can durations to time using the `+` operator.
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Format: from standard RFC3339 format
 Providing a single parameter assumes a standard RFC3339 datetime format. If the target format is not a valid yaml datetime format, the result will be a string tagged node.
 

pkg/yqlib/doc/operators/delete.md

@@ -2,12 +2,6 @@
 
 Deletes matching entries in maps or arrays.
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Delete entry in map
 Given a sample.yml file of:
 ```yaml

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

@@ -2,12 +2,6 @@
 
 Use the `documentIndex` operator (or the `di` shorthand) to select nodes of a particular document.
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Retrieve a document index
 Given a sample.yml file of:
 ```yaml

pkg/yqlib/doc/operators/encode-decode.md

@@ -25,12 +25,6 @@ XML uses the `--xml-attribute-prefix` and `xml-content-name` flags to identify a
 
 Base64 assumes [rfc4648](https://rfc-editor.org/rfc/rfc4648.html) encoding. Encoding and decoding both assume that the content is a string.
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Encode value as json string
 Given a sample.yml file of:
 ```yaml

pkg/yqlib/doc/operators/entries.md

@@ -2,12 +2,6 @@
 
 Similar to the same named functions in `jq` these functions convert to/from an object and an array of key-value pairs. This is most useful for performing operations on keys of maps.
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## to_entries Map
 Given a sample.yml file of:
 ```yaml

pkg/yqlib/doc/operators/env-variable-operators.md

@@ -30,12 +30,6 @@ yq '(.. | select(tag == "!!str")) |= envsubst' file.yaml
 ```
 
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Read string environment variable
 Running
 ```bash

pkg/yqlib/doc/operators/equals.md

@@ -21,12 +21,6 @@ The not equals `!=` operator returns `false` if the LHS is equal to the RHS.
 - select operator [here](https://mikefarah.gitbook.io/yq/operators/select)
 
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Match string
 Given a sample.yml file of:
 ```yaml

pkg/yqlib/doc/operators/error.md

@@ -2,12 +2,6 @@
 
 Use this operation to short-circuit expressions. Useful for validation.
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Validate a particular value
 Given a sample.yml file of:
 ```yaml

pkg/yqlib/doc/operators/eval.md

@@ -6,12 +6,6 @@ Use `eval` to dynamically process an expression - for instance from an environme
 
 Tip: This can be useful way parameterise complex scripts.
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Dynamically evaluate a path
 Given a sample.yml file of:
 ```yaml

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

@@ -10,12 +10,6 @@ Note the use of eval-all to ensure all documents are loaded into memory.
 yq eval-all 'select(fi == 0) * select(filename == "file2.yaml")' file1.yaml file2.yaml
 ```
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Get filename
 Given a sample.yml file of:
 ```yaml

pkg/yqlib/doc/operators/flatten.md

@@ -1,12 +1,6 @@
 # Flatten
 This recursively flattens arrays.
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Flatten
 Recursively flattens all arrays
 

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

@@ -2,12 +2,6 @@
 
 This is used to group items in an array by an expression.
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Group by field
 Given a sample.yml file of:
 ```yaml

pkg/yqlib/doc/operators/has.md

@@ -2,12 +2,6 @@
 
 This is operation that returns true if the key exists in a map (or index in an array), false otherwise.
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Has map key
 Given a sample.yml file of:
 ```yaml

pkg/yqlib/doc/operators/keys.md

@@ -2,12 +2,6 @@
 
 Use the `keys` operator to return map keys or array indices. 
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Map keys
 Given a sample.yml file of:
 ```yaml

pkg/yqlib/doc/operators/length.md

@@ -2,12 +2,6 @@
 
 Returns the lengths of the nodes. Length is defined according to the type of the node.
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## String length
 returns length of string
 

pkg/yqlib/doc/operators/line.md

@@ -2,12 +2,6 @@
 
 Returns the line of the matching node. Starts from 1, 0 indicates there was no line data.
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Returns line of _value_ node
 Given a sample.yml file of:
 ```yaml

pkg/yqlib/doc/operators/load.md

@@ -45,12 +45,6 @@ this.is = a properties file
 bXkgc2VjcmV0IGNoaWxsaSByZWNpcGUgaXMuLi4u
 ```
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Simple example
 Given a sample.yml file of:
 ```yaml

pkg/yqlib/doc/operators/map.md

@@ -2,12 +2,6 @@
 
 Maps values of an array. Use `map_values` to map values of an object.
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Map array
 Given a sample.yml file of:
 ```yaml

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

@@ -36,12 +36,6 @@ By default - `yq` merge is naive. It merges maps when they match the key name, a
 For more complex array merging (e.g. merging items that match on a certain key) please see the example [here](https://mikefarah.gitbook.io/yq/operators/multiply-merge#merge-arrays-of-objects-together-matching-on-a-key)
 
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Multiply integers
 Given a sample.yml file of:
 ```yaml

pkg/yqlib/doc/operators/parent.md

@@ -2,12 +2,6 @@
 
 Parent simply returns the parent nodes of the matching nodes.
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Simple example
 Given a sample.yml file of:
 ```yaml

pkg/yqlib/doc/operators/path.md

@@ -7,12 +7,6 @@ You can get the key/index of matching nodes by using the `path` operator to retu
 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. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Map path
 Given a sample.yml file of:
 ```yaml

pkg/yqlib/doc/operators/pick.md

@@ -4,12 +4,6 @@ Filter a map by the specified list of keys. Map is returned with the key in the
 
 Similarly, filter an array by the specified list of indices.
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Pick keys from map
 Note that the order of the keys matches the pick order and non existent keys are skipped.
 

pkg/yqlib/doc/operators/pipe.md

@@ -2,12 +2,6 @@
 
 Pipe the results of an expression into another. Like the bash operator.
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Simple Pipe
 Given a sample.yml file of:
 ```yaml

pkg/yqlib/doc/operators/recursive-descent-glob.md

@@ -19,12 +19,6 @@ For instance to set the `style` of all nodes in a yaml doc, including the map ke
 ```bash
 yq '... style= "flow"' file.yaml
 ```
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Recurse map (values only)
 Given a sample.yml file of:
 ```yaml

pkg/yqlib/doc/operators/reduce.md

@@ -21,12 +21,6 @@ Reduce syntax in `yq` is a little different from `jq` - as `yq` (currently) isn'
 
 To that end, the reduce operator is called `ireduce` for backwards compatibility if a `jq` like prefix version of `reduce` is ever added.
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Sum numbers
 Given a sample.yml file of:
 ```yaml

pkg/yqlib/doc/operators/reverse.md

@@ -2,12 +2,6 @@
 
 Reverses the order of the items in an array 
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Reverse
 Given a sample.yml file of:
 ```yaml

pkg/yqlib/doc/operators/select.md

@@ -8,12 +8,6 @@ Select is used to filter arrays and maps by a boolean expression.
 - comparison (`>=`, `<` etc) operators [here](https://mikefarah.gitbook.io/yq/operators/compare)
 - boolean operators (`and`, `or`, `any` etc) [here](https://mikefarah.gitbook.io/yq/operators/boolean-operators)
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified.&#x20;
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Select elements from array using wildcard prefix
 Given a sample.yml file of:
 ```yaml

pkg/yqlib/doc/operators/sort-keys.md

@@ -15,12 +15,6 @@ Note that `yq` does not yet consider anchors when sorting by keys - this may res
 For more advanced sorting, using `to_entries` to convert the map to an array, then sort/process the array as you like (e.g. using `sort_by`) and convert back to a map using `from_entries`.
 See [here](https://mikefarah.gitbook.io/yq/operators/entries#custom-sort-map-keys) for an example. 
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Sort keys of map
 Given a sample.yml file of:
 ```yaml

pkg/yqlib/doc/operators/sort.md

@@ -7,12 +7,6 @@ To sort by descending order, pipe the results through the `reverse` operator aft
 Note that at this stage, `yq` only sorts scalar fields.
 
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Sort by string field
 Given a sample.yml file of:
 ```yaml

pkg/yqlib/doc/operators/split-into-documents.md

@@ -2,12 +2,6 @@
 
 This operator splits all matches into separate documents
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Split empty
 Running
 ```bash

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

@@ -56,12 +56,6 @@ IFS= read -rd '' output < <(cat my_file)
 output=$output ./yq '.data.values = strenv(output)' first.yml
 ```
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified.&#x20;
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## To up (upper) case
 Works with unicode characters
 

pkg/yqlib/doc/operators/style.md

@@ -2,12 +2,6 @@
 
 The style operator can be used to get or set the style of nodes (e.g. string style, yaml style)
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Update and set style of a particular node (simple)
 Given a sample.yml file of:
 ```yaml

pkg/yqlib/doc/operators/subtract.md

@@ -2,12 +2,6 @@
 
 You can use subtract to subtract numbers, as well as removing elements from an array.
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Array subtraction
 Running
 ```bash

pkg/yqlib/doc/operators/tag.md

@@ -2,12 +2,6 @@
 
 The tag operator can be used to get or set the tag of nodes (e.g. `!!str`, `!!int`, `!!bool`).
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Get tag
 Given a sample.yml file of:
 ```yaml

pkg/yqlib/doc/operators/traverse-read.md

@@ -2,12 +2,6 @@
 
 This is the simplest (and perhaps most used) operator, it is used to navigate deeply into yaml structures.
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Simple map navigation
 Given a sample.yml file of:
 ```yaml

pkg/yqlib/doc/operators/union.md

@@ -2,12 +2,6 @@
 
 This operator is used to combine different results together.
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Combine scalars
 Running
 ```bash

pkg/yqlib/doc/operators/unique.md

@@ -3,12 +3,6 @@
 This is used to filter out duplicated items in an array. Note that the original order of the array is maintained.
 
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Unique array of scalars (string/numbers)
 Note that unique maintains the original order of the array.
 

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

@@ -4,12 +4,6 @@ Like the `jq` equivalents, variables are sometimes required for the more complex
 
 Note that there is also an additional `ref` operator that holds a reference (instead of a copy) of the path, allowing you to make multiple changes to the same path.
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Single value variable
 Given a sample.yml file of:
 ```yaml

pkg/yqlib/doc/operators/with.md

@@ -2,12 +2,6 @@
 
 Use the `with` operator to conveniently make multiple updates to a deeply nested path, or to update array elements relatively to each other. The first argument expression sets the root context, and the second expression runs against that root context.
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Update and style
 Given a sample.yml file of:
 ```yaml

pkg/yqlib/doc/usage/convert.md

@@ -5,12 +5,6 @@ Encode and decode to and from JSON. Supports multiple JSON documents in a single
 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.
 
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Parse json: simple
 JSON is a subset of yaml, so all you need to do is prettify the output
 

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

@@ -29,12 +29,6 @@ Fifi,cat
 ```
 
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Encode CSV simple
 Given a sample.yml file of:
 ```yaml

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

@@ -4,4 +4,37 @@ Encode and decode to and from XML. Whitespace is not conserved for round trips -
 
 Consecutive xml nodes with the same name are assumed to be arrays.
 
-XML content data and attributes are created as fields. This can be controlled by the `'--xml-attribute-prefix` and `--xml-content-name` flags - see below for examples.
+XML content data, attributes processing instructions and directives are all created as plain fields. 
+
+This can be controlled by:
+
+| Flag | Default |Sample XML | 
+| -- | -- |  -- |
+ | `--xml-attribute-prefix` | `+` (changing to `+@` soon) | Legs in ```<cat legs="4"/>``` |  
+ |  `--xml-content-name` | `+content` | Meow in ```<cat>Meow <fur>true</true></cat>``` |
+ | `--xml-directive-name` | `+directive` | ```<!DOCTYPE config system "blah">``` |
+ | `--xml-proc-inst-prefix` | `+p_` |  ```<?xml version="1"?>``` |
+
+
+{% hint style="warning" %}
+Default Attribute Prefix will be changing in v4.30!
+In order to avoid name conflicts (e.g. having an attribute named "content" will create a field that clashes with the default content name of "+content") the attribute prefix will be changing to "+@".
+
+This will affect users that have not set their own prefix and are not roundtripping XML changes.
+
+{% endhint %}
+
+## Encoder / Decoder flag options
+
+In addition to the above flags, there are the following xml encoder/decoder options controlled by flags:
+
+| Flag | Default | Description |
+| -- | -- | -- |
+| `--xml-strict-mode` | false | Strict mode enforces the requirements of the XML specification. When switched off the parser allows input containing common mistakes. See [the Golang xml decoder ](https://pkg.go.dev/encoding/xml#Decoder) for more details.| 
+| `--xml-keep-namespace` | true | Keeps the namespace of attributes |
+| `--xml-raw-token` | true |  Does not verify that start and end elements match and does not translate name space prefixes to their corresponding URLs. |
+| `--xml-skip-proc-inst` | false | Skips over processing instructions, e.g. `<?xml version="1"?>` |
+| `--xml-skip-directives` | false | Skips over directives, e.g. ```<!DOCTYPE config system "blah">``` |
+
+
+See below for examples

pkg/yqlib/doc/usage/properties.md

@@ -4,18 +4,12 @@ Encode/Decode/Roundtrip to/from a property file. Line comments on value nodes wi
 
 By default, empty maps and arrays are not encoded - see below for an example on how to encode a value for these.
 
-{% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified. 
-
-`yq e <exp> <file>`
-{% endhint %}
-
 ## Encode properties
 Note that empty arrays and maps are not encoded by default.
 
 Given a sample.yml file of:
 ```yaml
-# block comments don't come through
+# block comments come through
 person: # neither do comments on maps
     name: Mike Wazowski # comments on values appear
     pets: 
@@ -31,6 +25,7 @@ yq -o=props sample.yml
 ```
 will output
 ```properties
+# block comments come through
 # comments on values appear
 person.name = Mike Wazowski
 
@@ -44,7 +39,7 @@ Note that string values with blank characters in them are encapsulated with doub
 
 Given a sample.yml file of:
 ```yaml
-# block comments don't come through
+# block comments come through
 person: # neither do comments on maps
     name: Mike Wazowski # comments on values appear
     pets: 
@@ -60,6 +55,7 @@ yq -o=props --unwrapScalar=false sample.yml
 ```
 will output
 ```properties
+# block comments come through
 # comments on values appear
 person.name = "Mike Wazowski"
 
@@ -71,7 +67,7 @@ person.food.0 = pizza
 ## Encode properties: no comments
 Given a sample.yml file of:
 ```yaml
-# block comments don't come through
+# block comments come through
 person: # neither do comments on maps
     name: Mike Wazowski # comments on values appear
     pets: 
@@ -97,7 +93,7 @@ Use a yq expression to set the empty maps and sequences to your desired value.
 
 Given a sample.yml file of:
 ```yaml
-# block comments don't come through
+# block comments come through
 person: # neither do comments on maps
     name: Mike Wazowski # comments on values appear
     pets: 
@@ -113,6 +109,7 @@ yq -o=props '(.. | select( (tag == "!!map" or tag =="!!seq") and length == 0)) =
 ```
 will output
 ```properties
+# block comments come through
 # comments on values appear
 person.name = Mike Wazowski
 
@@ -126,6 +123,7 @@ emptyMap =
 ## Decode properties
 Given a sample.properties file of:
 ```properties
+# block comments come through
 # comments on values appear
 person.name = Mike Wazowski
 
@@ -141,9 +139,12 @@ yq -p=props sample.properties
 will output
 ```yaml
 person:
-    name: Mike Wazowski # comments on values appear
+    # block comments come through
+    # comments on values appear
+    name: Mike Wazowski
     pets:
-        - cat # comments on array values appear
+        # comments on array values appear
+        - cat
     food:
         - pizza
 ```
@@ -151,6 +152,7 @@ person:
 ## Roundtrip
 Given a sample.properties file of:
 ```properties
+# block comments come through
 # comments on values appear
 person.name = Mike Wazowski
 
@@ -165,6 +167,7 @@ yq -p=props -o=props '.person.pets.0 = "dog"' sample.properties
 ```
 will output
 ```properties
+# block comments come through
 # comments on values appear
 person.name = Mike Wazowski
 

pkg/yqlib/doc/usage/xml.md

@@ -4,14 +4,41 @@ Encode and decode to and from XML. Whitespace is not conserved for round trips -
 
 Consecutive xml nodes with the same name are assumed to be arrays.
 
-XML content data and attributes are created as fields. This can be controlled by the `'--xml-attribute-prefix` and `--xml-content-name` flags - see below for examples.
+XML content data, attributes processing instructions and directives are all created as plain fields. 
+
+This can be controlled by:
+
+| Flag | Default |Sample XML | 
+| -- | -- |  -- |
+ | `--xml-attribute-prefix` | `+` (changing to `+@` soon) | Legs in ```<cat legs="4"/>``` |  
+ |  `--xml-content-name` | `+content` | Meow in ```<cat>Meow <fur>true</true></cat>``` |
+ | `--xml-directive-name` | `+directive` | ```<!DOCTYPE config system "blah">``` |
+ | `--xml-proc-inst-prefix` | `+p_` |  ```<?xml version="1"?>``` |
+
 
 {% hint style="warning" %}
-Note that versions prior to 4.18 require the 'eval/e' command to be specified.&#x20;
+Default Attribute Prefix will be changing in v4.30!
+In order to avoid name conflicts (e.g. having an attribute named "content" will create a field that clashes with the default content name of "+content") the attribute prefix will be changing to "+@".
+
+This will affect users that have not set their own prefix and are not roundtripping XML changes.
 
-`yq e <exp> <file>`
 {% endhint %}
 
+## Encoder / Decoder flag options
+
+In addition to the above flags, there are the following xml encoder/decoder options controlled by flags:
+
+| Flag | Default | Description |
+| -- | -- | -- |
+| `--xml-strict-mode` | false | Strict mode enforces the requirements of the XML specification. When switched off the parser allows input containing common mistakes. See [the Golang xml decoder ](https://pkg.go.dev/encoding/xml#Decoder) for more details.| 
+| `--xml-keep-namespace` | true | Keeps the namespace of attributes |
+| `--xml-raw-token` | true |  Does not verify that start and end elements match and does not translate name space prefixes to their corresponding URLs. |
+| `--xml-skip-proc-inst` | false | Skips over processing instructions, e.g. `<?xml version="1"?>` |
+| `--xml-skip-directives` | false | Skips over directives, e.g. ```<!DOCTYPE config system "blah">``` |
+
+
+See below for examples
+
 ## Parse xml: simple
 Notice how all the values are strings, see the next example on how you can fix that.
 
@@ -30,6 +57,7 @@ yq -p=xml '.' sample.xml
 ```
 will output
 ```yaml
++p_xml: version="1.0" encoding="UTF-8"
 cat:
   says: meow
   legs: "4"
@@ -54,6 +82,7 @@ yq -p=xml ' (.. | select(tag == "!!str")) |= from_yaml' sample.xml
 ```
 will output
 ```yaml
++p_xml: version="1.0" encoding="UTF-8"
 cat:
   says: meow
   legs: 4
@@ -75,6 +104,7 @@ yq -p=xml '.' sample.xml
 ```
 will output
 ```yaml
++p_xml: version="1.0" encoding="UTF-8"
 animal:
   - cat
   - goat
@@ -96,6 +126,7 @@ yq -p=xml '.' sample.xml
 ```
 will output
 ```yaml
++p_xml: version="1.0" encoding="UTF-8"
 cat:
   +legs: "4"
   legs: "7"
@@ -115,13 +146,14 @@ yq -p=xml '.' sample.xml
 ```
 will output
 ```yaml
++p_xml: version="1.0" encoding="UTF-8"
 cat:
   +content: meow
   +legs: "4"
 ```
 
 ## Parse xml: custom dtd
-DTD entities are ignored.
+DTD entities are processed as directives.
 
 Given a sample.xml file of:
 ```xml
@@ -137,12 +169,45 @@ Given a sample.xml file of:
 ```
 then
 ```bash
-yq -p=xml '.' sample.xml
+yq -p=xml -o=xml '.' sample.xml
 ```
 will output
-```yaml
-root:
-  item: '&writer;&copyright;'
+```xml
+<?xml version="1.0"?>
+<!DOCTYPE root [
+<!ENTITY writer "Blah.">
+<!ENTITY copyright "Blah">
+]>
+<root>
+  <item>&amp;writer;&amp;copyright;</item>
+</root>
+```
+
+## Parse xml: skip custom dtd
+DTDs are directives, skip over directives to skip DTDs.
+
+Given a sample.xml file of:
+```xml
+
+<?xml version="1.0"?>
+<!DOCTYPE root [
+<!ENTITY writer "Blah.">
+<!ENTITY copyright "Blah">
+]>
+<root>
+    <item>&writer;&copyright;</item>
+</root>
+```
+then
+```bash
+yq -p=xml -o=xml --xml-skip-directives '.' sample.xml
+```
+will output
+```xml
+<?xml version="1.0"?>
+<root>
+  <item>&amp;writer;&amp;copyright;</item>
+</root>
 ```
 
 ## Parse xml: with comments
@@ -207,12 +272,14 @@ yq -p=xml -o=xml --xml-keep-namespace '.' sample.xml
 ```
 will output
 ```xml
+<?xml version="1.0"?>
 <map xmlns="some-namespace" xmlns:xsi="some-instance" some-instance:schemaLocation="some-url"></map>
 ```
 
 instead of
 ```xml
-<map xmlns="some-namespace" xsi="some-instance" schemaLocation="some-url"></map>
+<?xml version="1.0"?>
+<map xmlns="some-namespace" xmlns:xsi="some-instance" some-instance:schemaLocation="some-url"></map>
 ```
 
 ## Parse xml: keep raw attribute namespace
@@ -230,11 +297,13 @@ yq -p=xml -o=xml --xml-keep-namespace --xml-raw-token '.' sample.xml
 ```
 will output
 ```xml
-<map xmlns="some-namespace" xmlns:xsi="some-instance" xsi:schemaLocation="some-url"></map>
+<?xml version="1.0"?>
+<map xmlns="some-namespace" xmlns:xsi="some-instance" some-instance:schemaLocation="some-url"></map>
 ```
 
 instead of
 ```xml
+<?xml version="1.0"?>
 <map xmlns="some-namespace" xsi="some-instance" schemaLocation="some-url"></map>
 ```
 
@@ -317,6 +386,7 @@ A best attempt is made to copy comments to xml.
 
 Given a sample.yml file of:
 ```yaml
+# header comment
 # above_cat
 cat: # inline_cat
   # above_array
@@ -333,12 +403,41 @@ yq -o=xml '.' sample.yml
 ```
 will output
 ```xml
-<!-- above_cat inline_cat --><cat><!-- above_array inline_array -->
+<!--
+ header comment
+ above_cat
+ --><!-- inline_cat --><cat><!-- above_array inline_array -->
   <array>val1<!-- inline_val1 --></array>
   <array><!-- above_val2 -->val2<!-- inline_val2 --></array>
 </cat><!-- below_cat -->
 ```
 
+## Encode: doctype and xml declaration
+Use the special xml names to add/modify proc instructions and directives.
+
+Given a sample.yml file of:
+```yaml
++p_xml: version="1.0"
++directive: 'DOCTYPE config SYSTEM "/etc/iwatch/iwatch.dtd" '
+apple:
+  +p_coolioo: version="1.0"
+  +directive: 'CATYPE meow purr puss '
+  b: things
+
+```
+then
+```bash
+yq -o=xml '.' sample.yml
+```
+will output
+```xml
+<?xml version="1.0"?>
+<!DOCTYPE config SYSTEM "/etc/iwatch/iwatch.dtd" >
+<apple><?coolioo version="1.0"?><!CATYPE meow purr puss >
+  <b>things</b>
+</apple>
+```
+
 ## Round trip: with comments
 A best effort is made, but comment positions and white space are not preserved perfectly.
 
@@ -380,3 +479,31 @@ in d before -->
 </cat><!-- after cat -->
 ```
 
+## Roundtrip: with doctype and declaration
+yq parses XML proc instructions and directives into nodes.
+Unfortunately the underlying XML parser loses whitespace information.
+
+Given a sample.xml file of:
+```xml
+<?xml version="1.0"?>
+<!DOCTYPE config SYSTEM "/etc/iwatch/iwatch.dtd" >
+<apple>
+  <?coolioo version="1.0"?>
+  <!CATYPE meow purr puss >
+  <b>things</b>
+</apple>
+
+```
+then
+```bash
+yq -p=xml -o=xml '.' sample.xml
+```
+will output
+```xml
+<?xml version="1.0"?>
+<!DOCTYPE config SYSTEM "/etc/iwatch/iwatch.dtd" >
+<apple><?coolioo version="1.0"?><!CATYPE meow purr puss >
+  <b>things</b>
+</apple>
+```
+

pkg/yqlib/encoder_properties.go

@@ -65,7 +65,7 @@ func (pe *propertiesEncoder) PrintLeadingContent(writer io.Writer, content strin
 func (pe *propertiesEncoder) Encode(writer io.Writer, node *yaml.Node) error {
 	mapKeysToStrings(node)
 	p := properties.NewProperties()
-	err := pe.doEncode(p, node, "")
+	err := pe.doEncode(p, node, "", nil)
 	if err != nil {
 		return err
 	}
@@ -74,8 +74,17 @@ func (pe *propertiesEncoder) Encode(writer io.Writer, node *yaml.Node) error {
 	return err
 }
 
-func (pe *propertiesEncoder) doEncode(p *properties.Properties, node *yaml.Node, path string) error {
-	p.SetComment(path, headAndLineComment(node))
+func (pe *propertiesEncoder) doEncode(p *properties.Properties, node *yaml.Node, path string, keyNode *yaml.Node) error {
+
+	comments := ""
+	if keyNode != nil {
+		// include the key node comments if present
+		comments = headAndLineComment(keyNode)
+	}
+	comments = comments + headAndLineComment(node)
+	commentsWithSpaces := strings.ReplaceAll(comments, "\n", "\n ")
+	p.SetComments(path, strings.Split(commentsWithSpaces, "\n"))
+
 	switch node.Kind {
 	case yaml.ScalarNode:
 		var nodeValue string
@@ -87,13 +96,13 @@ func (pe *propertiesEncoder) doEncode(p *properties.Properties, node *yaml.Node,
 		_, _, err := p.Set(path, nodeValue)
 		return err
 	case yaml.DocumentNode:
-		return pe.doEncode(p, node.Content[0], path)
+		return pe.doEncode(p, node.Content[0], path, node)
 	case yaml.SequenceNode:
 		return pe.encodeArray(p, node.Content, path)
 	case yaml.MappingNode:
 		return pe.encodeMap(p, node.Content, path)
 	case yaml.AliasNode:
-		return pe.doEncode(p, node.Alias, path)
+		return pe.doEncode(p, node.Alias, path, nil)
 	default:
 		return fmt.Errorf("Unsupported node %v", node.Tag)
 	}
@@ -108,7 +117,7 @@ func (pe *propertiesEncoder) appendPath(path string, key interface{}) string {
 
 func (pe *propertiesEncoder) encodeArray(p *properties.Properties, kids []*yaml.Node, path string) error {
 	for index, child := range kids {
-		err := pe.doEncode(p, child, pe.appendPath(path, index))
+		err := pe.doEncode(p, child, pe.appendPath(path, index), nil)
 		if err != nil {
 			return err
 		}
@@ -120,7 +129,7 @@ func (pe *propertiesEncoder) encodeMap(p *properties.Properties, kids []*yaml.No
 	for index := 0; index < len(kids); index = index + 2 {
 		key := kids[index]
 		value := kids[index+1]
-		err := pe.doEncode(p, value, pe.appendPath(path, key.Value))
+		err := pe.doEncode(p, value, pe.appendPath(path, key.Value), key)
 		if err != nil {
 			return err
 		}

pkg/yqlib/encoder_properties_test.go

@@ -14,7 +14,7 @@ func yamlToProps(sampleYaml string, unwrapScalar bool) string {
 	writer := bufio.NewWriter(&output)
 
 	var propsEncoder = NewPropertiesEncoder(unwrapScalar)
-	inputs, err := readDocuments(strings.NewReader(sampleYaml), "sample.yml", 0, NewYamlDecoder())
+	inputs, err := readDocuments(strings.NewReader(sampleYaml), "sample.yml", 0, NewYamlDecoder(ConfiguredYamlPreferences))
 	if err != nil {
 		panic(err)
 	}

pkg/yqlib/encoder_test.go

@@ -14,7 +14,7 @@ func yamlToJSON(sampleYaml string, indent int) string {
 	writer := bufio.NewWriter(&output)
 
 	var jsonEncoder = NewJSONEncoder(indent, false)
-	inputs, err := readDocuments(strings.NewReader(sampleYaml), "sample.yml", 0, NewYamlDecoder())
+	inputs, err := readDocuments(strings.NewReader(sampleYaml), "sample.yml", 0, NewYamlDecoder(ConfiguredYamlPreferences))
 	if err != nil {
 		panic(err)
 	}

pkg/yqlib/encoder_xml.go

@@ -4,26 +4,28 @@ import (
 	"encoding/xml"
 	"fmt"
 	"io"
+	"regexp"
 	"strings"
 
 	yaml "gopkg.in/yaml.v3"
 )
 
-var XMLPreferences = xmlPreferences{AttributePrefix: "+", ContentName: "+content", StrictMode: false, UseRawToken: false}
-
 type xmlEncoder struct {
-	attributePrefix string
-	contentName     string
-	indentString    string
+	indentString   string
+	writer         io.Writer
+	prefs          XmlPreferences
+	leadingContent string
 }
 
-func NewXMLEncoder(indent int, attributePrefix string, contentName string) Encoder {
+var commentPrefix = regexp.MustCompile(`(^|\n)\s*#`)
+
+func NewXMLEncoder(indent int, prefs XmlPreferences) Encoder {
 	var indentString = ""
 
 	for index := 0; index < indent; index++ {
 		indentString = indentString + " "
 	}
-	return &xmlEncoder{attributePrefix, contentName, indentString}
+	return &xmlEncoder{indentString, nil, prefs, ""}
 }
 
 func (e *xmlEncoder) CanHandleAliases() bool {
@@ -35,13 +37,23 @@ func (e *xmlEncoder) PrintDocumentSeparator(writer io.Writer) error {
 }
 
 func (e *xmlEncoder) PrintLeadingContent(writer io.Writer, content string) error {
+	e.leadingContent = commentPrefix.ReplaceAllString(content, "\n")
 	return nil
 }
 
 func (e *xmlEncoder) Encode(writer io.Writer, node *yaml.Node) error {
 	encoder := xml.NewEncoder(writer)
+	// hack so we can manually add newlines to procInst and directives
+	e.writer = writer
 	encoder.Indent("", e.indentString)
 
+	if e.leadingContent != "" {
+		err := e.encodeComment(encoder, e.leadingContent)
+		if err != nil {
+			return err
+		}
+	}
+
 	switch node.Kind {
 	case yaml.MappingNode:
 		err := e.encodeTopLevelMap(encoder, node)
@@ -77,6 +89,23 @@ func (e *xmlEncoder) Encode(writer io.Writer, node *yaml.Node) error {
 }
 
 func (e *xmlEncoder) encodeTopLevelMap(encoder *xml.Encoder, node *yaml.Node) error {
+	// make sure <?xml .. ?> processing instructions are encoded first
+	for i := 0; i < len(node.Content); i += 2 {
+		key := node.Content[i]
+		value := node.Content[i+1]
+
+		if key.Value == (e.prefs.ProcInstPrefix + "xml") {
+			name := strings.Replace(key.Value, e.prefs.ProcInstPrefix, "", 1)
+			procInst := xml.ProcInst{Target: name, Inst: []byte(value.Value)}
+			if err := encoder.EncodeToken(procInst); err != nil {
+				return err
+			}
+			if _, err := e.writer.Write([]byte("\n")); err != nil {
+				log.Warning("Unable to write newline, skipping: %w", err)
+			}
+		}
+	}
+
 	err := e.encodeComment(encoder, headAndLineComment(node))
 	if err != nil {
 		return err
@@ -92,11 +121,33 @@ func (e *xmlEncoder) encodeTopLevelMap(encoder *xml.Encoder, node *yaml.Node) er
 			return err
 		}
 
-		log.Debugf("recursing")
+		if key.Value == (e.prefs.ProcInstPrefix + "xml") {
+			// dont double process these.
+		} else if strings.HasPrefix(key.Value, e.prefs.ProcInstPrefix) {
+			name := strings.Replace(key.Value, e.prefs.ProcInstPrefix, "", 1)
+			procInst := xml.ProcInst{Target: name, Inst: []byte(value.Value)}
+			if err := encoder.EncodeToken(procInst); err != nil {
+				return err
+			}
+			if _, err := e.writer.Write([]byte("\n")); err != nil {
+				log.Warning("Unable to write newline, skipping: %w", err)
+			}
+		} else if key.Value == e.prefs.DirectiveName {
+			var directive xml.Directive = []byte(value.Value)
+			if err := encoder.EncodeToken(directive); err != nil {
+				return err
+			}
+			if _, err := e.writer.Write([]byte("\n")); err != nil {
+				log.Warning("Unable to write newline, skipping: %w", err)
+			}
+		} else {
 
-		err = e.doEncode(encoder, value, start)
-		if err != nil {
-			return err
+			log.Debugf("recursing")
+
+			err = e.doEncode(encoder, value, start)
+			if err != nil {
+				return err
+			}
 		}
 		err = e.encodeComment(encoder, footComment(key))
 		if err != nil {
@@ -180,6 +231,13 @@ func (e *xmlEncoder) encodeArray(encoder *xml.Encoder, node *yaml.Node, start xm
 	return e.encodeComment(encoder, footComment(node))
 }
 
+func (e *xmlEncoder) isAttribute(name string) bool {
+	return strings.HasPrefix(name, e.prefs.AttributePrefix) &&
+		name != e.prefs.ContentName &&
+		name != e.prefs.DirectiveName &&
+		!strings.HasPrefix(name, e.prefs.ProcInstPrefix)
+}
+
 func (e *xmlEncoder) encodeMap(encoder *xml.Encoder, node *yaml.Node, start xml.StartElement) error {
 	log.Debug("its a map")
 
@@ -188,9 +246,9 @@ func (e *xmlEncoder) encodeMap(encoder *xml.Encoder, node *yaml.Node, start xml.
 		key := node.Content[i]
 		value := node.Content[i+1]
 
-		if strings.HasPrefix(key.Value, e.attributePrefix) && key.Value != e.contentName {
+		if e.isAttribute(key.Value) {
 			if value.Kind == yaml.ScalarNode {
-				attributeName := strings.Replace(key.Value, e.attributePrefix, "", 1)
+				attributeName := strings.Replace(key.Value, e.prefs.AttributePrefix, "", 1)
 				start.Attr = append(start.Attr, xml.Attr{Name: xml.Name{Local: attributeName}, Value: value.Value})
 			} else {
 				return fmt.Errorf("cannot use %v as attribute, only scalars are supported", value.Tag)
@@ -212,14 +270,18 @@ func (e *xmlEncoder) encodeMap(encoder *xml.Encoder, node *yaml.Node, start xml.
 		if err != nil {
 			return err
 		}
-
-		if !strings.HasPrefix(key.Value, e.attributePrefix) && key.Value != e.contentName {
-			start := xml.StartElement{Name: xml.Name{Local: key.Value}}
-			err := e.doEncode(encoder, value, start)
-			if err != nil {
+		if strings.HasPrefix(key.Value, e.prefs.ProcInstPrefix) {
+			name := strings.Replace(key.Value, e.prefs.ProcInstPrefix, "", 1)
+			procInst := xml.ProcInst{Target: name, Inst: []byte(value.Value)}
+			if err := encoder.EncodeToken(procInst); err != nil {
 				return err
 			}
-		} else if key.Value == e.contentName {
+		} else if key.Value == e.prefs.DirectiveName {
+			var directive xml.Directive = []byte(value.Value)
+			if err := encoder.EncodeToken(directive); err != nil {
+				return err
+			}
+		} else if key.Value == e.prefs.ContentName {
 			// directly encode the contents
 			err = e.encodeComment(encoder, headAndLineComment(value))
 			if err != nil {
@@ -234,6 +296,12 @@ func (e *xmlEncoder) encodeMap(encoder *xml.Encoder, node *yaml.Node, start xml.
 			if err != nil {
 				return err
 			}
+		} else if !e.isAttribute(key.Value) {
+			start := xml.StartElement{Name: xml.Name{Local: key.Value}}
+			err := e.doEncode(encoder, value, start)
+			if err != nil {
+				return err
+			}
 		}
 		err = e.encodeComment(encoder, footComment(key))
 		if err != nil {

pkg/yqlib/encoder_yaml.go

@@ -11,17 +11,16 @@ import (
 )
 
 type yamlEncoder struct {
-	indent             int
-	colorise           bool
-	printDocSeparators bool
-	unwrapScalar       bool
+	indent   int
+	colorise bool
+	prefs    YamlPreferences
 }
 
-func NewYamlEncoder(indent int, colorise bool, printDocSeparators bool, unwrapScalar bool) Encoder {
+func NewYamlEncoder(indent int, colorise bool, prefs YamlPreferences) Encoder {
 	if indent < 0 {
 		indent = 0
 	}
-	return &yamlEncoder{indent, colorise, printDocSeparators, unwrapScalar}
+	return &yamlEncoder{indent, colorise, prefs}
 }
 
 func (ye *yamlEncoder) CanHandleAliases() bool {
@@ -29,7 +28,7 @@ func (ye *yamlEncoder) CanHandleAliases() bool {
 }
 
 func (ye *yamlEncoder) PrintDocumentSeparator(writer io.Writer) error {
-	if ye.printDocSeparators {
+	if ye.prefs.PrintDocSeparators {
 		log.Debug("-- writing doc sep")
 		if err := writeString(writer, "---\n"); err != nil {
 			return err
@@ -76,7 +75,7 @@ func (ye *yamlEncoder) PrintLeadingContent(writer io.Writer, content string) err
 
 func (ye *yamlEncoder) Encode(writer io.Writer, node *yaml.Node) error {
 
-	if node.Kind == yaml.ScalarNode && ye.unwrapScalar {
+	if node.Kind == yaml.ScalarNode && ye.prefs.UnwrapScalar {
 		return writeString(writer, node.Value+"\n")
 	}
 

pkg/yqlib/json_test.go

@@ -262,11 +262,11 @@ func documentDecodeNdJsonScenario(w *bufio.Writer, s formatScenario) {
 
 	writeOrPanic(w, "will output\n")
 
-	writeOrPanic(w, fmt.Sprintf("```yaml\n%v```\n\n", processFormatScenario(s, NewJSONDecoder(), NewYamlEncoder(s.indent, false, true, true))))
+	writeOrPanic(w, fmt.Sprintf("```yaml\n%v```\n\n", processFormatScenario(s, NewJSONDecoder(), NewYamlEncoder(s.indent, false, ConfiguredYamlPreferences))))
 }
 
 func decodeJSON(t *testing.T, jsonString string) *CandidateNode {
-	docs, err := readDocumentWithLeadingContent(jsonString, "sample.json", 0)
+	docs, err := readDocument(jsonString, "sample.json", 0)
 
 	if err != nil {
 		t.Error(err)
@@ -293,12 +293,12 @@ func decodeJSON(t *testing.T, jsonString string) *CandidateNode {
 func testJSONScenario(t *testing.T, s formatScenario) {
 	switch s.scenarioType {
 	case "encode", "decode":
-		test.AssertResultWithContext(t, s.expected, processFormatScenario(s, NewYamlDecoder(), NewJSONEncoder(s.indent, false)), s.description)
+		test.AssertResultWithContext(t, s.expected, processFormatScenario(s, NewYamlDecoder(ConfiguredYamlPreferences), NewJSONEncoder(s.indent, false)), s.description)
 	case "":
 		var actual = resultToString(t, decodeJSON(t, s.input))
 		test.AssertResultWithContext(t, s.expected, actual, s.description)
 	case "decode-ndjson":
-		test.AssertResultWithContext(t, s.expected, processFormatScenario(s, NewJSONDecoder(), NewYamlEncoder(2, false, true, true)), s.description)
+		test.AssertResultWithContext(t, s.expected, processFormatScenario(s, NewJSONDecoder(), NewYamlEncoder(2, false, ConfiguredYamlPreferences)), s.description)
 	case "roundtrip-ndjson":
 		test.AssertResultWithContext(t, s.expected, processFormatScenario(s, NewJSONDecoder(), NewJSONEncoder(0, false)), s.description)
 	case "roundtrip-multi":
@@ -385,7 +385,7 @@ func documentJSONEncodeScenario(w *bufio.Writer, s formatScenario) {
 	}
 	writeOrPanic(w, "will output\n")
 
-	writeOrPanic(w, fmt.Sprintf("```json\n%v```\n\n", processFormatScenario(s, NewYamlDecoder(), NewJSONEncoder(s.indent, false))))
+	writeOrPanic(w, fmt.Sprintf("```json\n%v```\n\n", processFormatScenario(s, NewYamlDecoder(ConfiguredYamlPreferences), NewJSONEncoder(s.indent, false))))
 }
 
 func TestJSONScenarios(t *testing.T) {

pkg/yqlib/lexer_participle.go

@@ -76,7 +76,7 @@ var participleYqRules = []*participleYqRule{
 	{"Base64d", `@base64d`, decodeOp(Base64InputFormat), 0},
 	{"Base64", `@base64`, encodeWithIndent(Base64OutputFormat, 0), 0},
 
-	{"LoadXML", `load_?xml|xml_?load`, loadOp(NewXMLDecoder(XMLPreferences.AttributePrefix, XMLPreferences.ContentName, XMLPreferences.StrictMode, XMLPreferences.KeepNamespace, XMLPreferences.UseRawToken), false), 0},
+	{"LoadXML", `load_?xml|xml_?load`, loadOp(NewXMLDecoder(ConfiguredXMLPreferences), false), 0},
 
 	{"LoadBase64", `load_?base64`, loadOp(NewBase64Decoder(), false), 0},
 
@@ -84,7 +84,7 @@ var participleYqRules = []*participleYqRule{
 
 	{"LoadString", `load_?str|str_?load`, loadOp(nil, true), 0},
 
-	{"LoadYaml", `load`, loadOp(NewYamlDecoder(), false), 0},
+	{"LoadYaml", `load`, loadOp(NewYamlDecoder(ConfiguredYamlPreferences), false), 0},
 
 	{"SplitDocument", `splitDoc|split_?doc`, opToken(splitDocumentOpType), 0},
 

pkg/yqlib/lib.go

@@ -21,14 +21,6 @@ func InitExpressionParser() {
 	}
 }
 
-type xmlPreferences struct {
-	AttributePrefix string
-	ContentName     string
-	StrictMode      bool
-	KeepNamespace   bool
-	UseRawToken     bool
-}
-
 var log = logging.MustGetLogger("yq-lib")
 
 var PrettyPrintExp = `(... | (select(tag != "!!str"), select(tag == "!!str") | select(test("(?i)^(y|yes|n|no|on|off)$") | not))  ) style=""`
@@ -256,14 +248,16 @@ func guessTagFromCustomType(node *yaml.Node) string {
 }
 
 func parseSnippet(value string) (*yaml.Node, error) {
-	decoder := NewYamlDecoder()
-	decoder.Init(strings.NewReader(value))
-	var dataBucket yaml.Node
-	err := decoder.Decode(&dataBucket)
-	if len(dataBucket.Content) == 0 {
+	decoder := NewYamlDecoder(ConfiguredYamlPreferences)
+	err := decoder.Init(strings.NewReader(value))
+	if err != nil {
+		return nil, err
+	}
+	parsedNode, err := decoder.Decode()
+	if len(parsedNode.Node.Content) == 0 {
 		return nil, fmt.Errorf("bad data")
 	}
-	return dataBucket.Content[0], err
+	return unwrapDoc(parsedNode.Node), err
 }
 
 func recursiveNodeEqual(lhs *yaml.Node, rhs *yaml.Node) bool {

pkg/yqlib/operator_collect.go

@@ -56,7 +56,7 @@ func collectOperator(d *dataTreeNavigator, context Context, expressionNode *Expr
 		collectedNode := &yaml.Node{Kind: yaml.SequenceNode, Tag: "!!seq"}
 		collectCandidate := candidate.CreateReplacement(collectedNode)
 
-		collectExpResults, err := d.GetMatchingNodes(context.SingleReadonlyChildContext(candidate), expressionNode.RHS)
+		collectExpResults, err := d.GetMatchingNodes(context.SingleChildContext(candidate), expressionNode.RHS)
 		if err != nil {
 			return Context{}, err
 		}

pkg/yqlib/operator_collect_test.go

@@ -20,6 +20,14 @@ var collectOperatorScenarios = []expressionScenario{
 			"D0, P[], (!!seq)::- 1\n- 2\n- 3\n",
 		},
 	},
+	{
+		skipDoc:     true,
+		description: "update in collect",
+		expression:  `[.a = "cat"]`,
+		expected: []string{
+			"D0, P[], (!!seq)::- a: cat\n",
+		},
+	},
 	{
 		description: "Collect empty",
 		document:    ``,

pkg/yqlib/operator_comments.go

@@ -83,6 +83,10 @@ func getCommentsOperator(d *dataTreeNavigator, context Context, expressionNode *
 	log.Debugf("GetComments operator!")
 	var results = list.New()
 
+	yamlPrefs := NewDefaultYamlPreferences()
+	yamlPrefs.PrintDocSeparators = false
+	yamlPrefs.UnwrapScalar = false
+
 	for el := context.MatchingNodes.Front(); el != nil; el = el.Next() {
 		candidate := el.Value.(*CandidateNode)
 		comment := ""
@@ -92,7 +96,7 @@ func getCommentsOperator(d *dataTreeNavigator, context Context, expressionNode *
 			var chompRegexp = regexp.MustCompile(`\n$`)
 			var output bytes.Buffer
 			var writer = bufio.NewWriter(&output)
-			var encoder = NewYamlEncoder(2, false, false, false)
+			var encoder = NewYamlEncoder(2, false, yamlPrefs)
 			if err := encoder.PrintLeadingContent(writer, candidate.LeadingContent); err != nil {
 				return Context{}, err
 			}

pkg/yqlib/operator_datetime.go

@@ -4,7 +4,6 @@ import (
 	"container/list"
 	"errors"
 	"fmt"
-	"strings"
 	"time"
 
 	"gopkg.in/yaml.v3"
@@ -54,7 +53,6 @@ func nowOp(d *dataTreeNavigator, context Context, expressionNode *ExpressionNode
 func formatDateTime(d *dataTreeNavigator, context Context, expressionNode *ExpressionNode) (Context, error) {
 	format, err := getStringParamter("format", d, context, expressionNode.RHS)
 	layout := context.GetDateTimeLayout()
-	decoder := NewYamlDecoder()
 
 	if err != nil {
 		return Context{}, err
@@ -69,19 +67,15 @@ func formatDateTime(d *dataTreeNavigator, context Context, expressionNode *Expre
 			return Context{}, fmt.Errorf("could not parse datetime of [%v]: %w", candidate.GetNicePath(), err)
 		}
 		formattedTimeStr := parsedTime.Format(format)
-		decoder.Init(strings.NewReader(formattedTimeStr))
-		var dataBucket yaml.Node
-		errorReading := decoder.Decode(&dataBucket)
-		var node *yaml.Node
+
+		node, errorReading := parseSnippet(formattedTimeStr)
 		if errorReading != nil {
-			log.Debugf("could not parse %v - lets just leave it as a string", formattedTimeStr)
+			log.Debugf("could not parse %v - lets just leave it as a string: %w", formattedTimeStr, errorReading)
 			node = &yaml.Node{
 				Kind:  yaml.ScalarNode,
 				Tag:   "!!str",
 				Value: formattedTimeStr,
 			}
-		} else {
-			node = unwrapDoc(&dataBucket)
 		}
 
 		results.PushBack(candidate.CreateReplacement(node))

pkg/yqlib/operator_encoder_decoder.go

@@ -21,9 +21,9 @@ func configureEncoder(format PrinterOutputFormat, indent int) Encoder {
 	case TSVOutputFormat:
 		return NewCsvEncoder('\t')
 	case YamlOutputFormat:
-		return NewYamlEncoder(indent, false, true, true)
+		return NewYamlEncoder(indent, false, ConfiguredYamlPreferences)
 	case XMLOutputFormat:
-		return NewXMLEncoder(indent, XMLPreferences.AttributePrefix, XMLPreferences.ContentName)
+		return NewXMLEncoder(indent, ConfiguredXMLPreferences)
 	case Base64OutputFormat:
 		return NewBase64Encoder()
 	}
@@ -102,14 +102,9 @@ func decodeOperator(d *dataTreeNavigator, context Context, expressionNode *Expre
 	var decoder Decoder
 	switch preferences.format {
 	case YamlInputFormat:
-		decoder = NewYamlDecoder()
+		decoder = NewYamlDecoder(ConfiguredYamlPreferences)
 	case XMLInputFormat:
-		decoder = NewXMLDecoder(
-			XMLPreferences.AttributePrefix,
-			XMLPreferences.ContentName,
-			XMLPreferences.StrictMode,
-			XMLPreferences.KeepNamespace,
-			XMLPreferences.UseRawToken)
+		decoder = NewXMLDecoder(ConfiguredXMLPreferences)
 	case Base64InputFormat:
 		decoder = NewBase64Decoder()
 	case PropertiesInputFormat:
@@ -126,17 +121,19 @@ func decodeOperator(d *dataTreeNavigator, context Context, expressionNode *Expre
 
 		context.SetVariable("decoded: "+candidate.GetKey(), candidate.AsList())
 
-		var dataBucket yaml.Node
 		log.Debugf("got: [%v]", candidate.Node.Value)
 
-		decoder.Init(strings.NewReader(unwrapDoc(candidate.Node).Value))
+		err := decoder.Init(strings.NewReader(unwrapDoc(candidate.Node).Value))
+		if err != nil {
+			return Context{}, err
+		}
 
-		errorReading := decoder.Decode(&dataBucket)
+		decodedNode, errorReading := decoder.Decode()
 		if errorReading != nil {
 			return Context{}, errorReading
 		}
 		//first node is a doc
-		node := unwrapDoc(&dataBucket)
+		node := unwrapDoc(decodedNode.Node)
 
 		results.PushBack(candidate.CreateReplacement(node))
 	}

pkg/yqlib/operator_load.go

@@ -48,7 +48,7 @@ func loadYaml(filename string, decoder Decoder) (*CandidateNode, error) {
 	} else {
 		sequenceNode := &CandidateNode{Node: &yaml.Node{Kind: yaml.SequenceNode}}
 		for doc := documents.Front(); doc != nil; doc = doc.Next() {
-			sequenceNode.Node.Content = append(sequenceNode.Node.Content, doc.Value.(*CandidateNode).Node)
+			sequenceNode.Node.Content = append(sequenceNode.Node.Content, unwrapDoc(doc.Value.(*CandidateNode).Node))
 		}
 		return sequenceNode, nil
 	}

pkg/yqlib/operator_load_test.go

@@ -5,6 +5,30 @@ import (
 )
 
 var loadScenarios = []expressionScenario{
+	{
+		skipDoc:     true,
+		description: "Load empty file with a comment",
+		expression:  `load("../../examples/empty.yaml")`,
+		expected: []string{
+			"D0, P[], (doc)::# comment\n\n",
+		},
+	},
+	{
+		skipDoc:     true,
+		description: "Load empty file with no comment",
+		expression:  `load("../../examples/empty-no-comment.yaml")`,
+		expected: []string{
+			"D0, P[], (!!null)::\n",
+		},
+	},
+	{
+		skipDoc:     true,
+		description: "Load multiple documents",
+		expression:  `load("../../examples/multiple_docs_small.yaml")`,
+		expected: []string{
+			"D0, P[], ()::- a: Easy! as one two three\n- another:\n    document: here\n- - 1\n  - 2\n",
+		},
+	},
 	{
 		description: "Simple example",
 		document:    `{myFile: "../../examples/thing.yml"}`,

pkg/yqlib/operators_test.go

@@ -40,21 +40,16 @@ func TestMain(m *testing.M) {
 }
 
 func NewSimpleYamlPrinter(writer io.Writer, outputFormat PrinterOutputFormat, unwrapScalar bool, colorsEnabled bool, indent int, printDocSeparators bool) Printer {
-	return NewPrinter(NewYamlEncoder(indent, colorsEnabled, printDocSeparators, unwrapScalar), NewSinglePrinterWriter(writer))
+	prefs := NewDefaultYamlPreferences()
+	prefs.PrintDocSeparators = printDocSeparators
+	prefs.UnwrapScalar = unwrapScalar
+	return NewPrinter(NewYamlEncoder(indent, colorsEnabled, prefs), NewSinglePrinterWriter(writer))
 }
 
-func readDocumentWithLeadingContent(content string, fakefilename string, fakeFileIndex int) (*list.List, error) {
-	reader, firstFileLeadingContent, err := processReadStream(bufio.NewReader(strings.NewReader(content)))
-	if err != nil {
-		return nil, err
-	}
+func readDocument(content string, fakefilename string, fakeFileIndex int) (*list.List, error) {
+	reader := bufio.NewReader(strings.NewReader(content))
 
-	inputs, err := readDocuments(reader, fakefilename, fakeFileIndex, NewYamlDecoder())
-	if err != nil {
-		return nil, err
-	}
-	inputs.Front().Value.(*CandidateNode).LeadingContent = firstFileLeadingContent
-	return inputs, nil
+	return readDocuments(reader, fakefilename, fakeFileIndex, NewYamlDecoder(ConfiguredYamlPreferences))
 }
 
 func testScenario(t *testing.T, s *expressionScenario) {
@@ -67,7 +62,7 @@ func testScenario(t *testing.T, s *expressionScenario) {
 	inputs := list.New()
 
 	if s.document != "" {
-		inputs, err = readDocumentWithLeadingContent(s.document, "sample.yml", 0)
+		inputs, err = readDocument(s.document, "sample.yml", 0)
 
 		if err != nil {
 			t.Error(err, s.document, s.expression)
@@ -75,7 +70,7 @@ func testScenario(t *testing.T, s *expressionScenario) {
 		}
 
 		if s.document2 != "" {
-			moreInputs, err := readDocumentWithLeadingContent(s.document2, "another.yml", 1)
+			moreInputs, err := readDocument(s.document2, "another.yml", 1)
 			if err != nil {
 				t.Error(err, s.document2, s.expression)
 				return
@@ -176,7 +171,7 @@ func formatYaml(yaml string, filename string) string {
 		panic(err)
 	}
 	streamEvaluator := NewStreamEvaluator()
-	_, err = streamEvaluator.Evaluate(filename, strings.NewReader(yaml), node, printer, "", NewYamlDecoder())
+	_, err = streamEvaluator.Evaluate(filename, strings.NewReader(yaml), node, printer, NewYamlDecoder(ConfiguredYamlPreferences))
 	if err != nil {
 		panic(err)
 	}
@@ -322,13 +317,13 @@ func documentOutput(t *testing.T, w *bufio.Writer, s expressionScenario, formatt
 
 	if s.document != "" {
 
-		inputs, err = readDocumentWithLeadingContent(formattedDoc, "sample.yml", 0)
+		inputs, err = readDocument(formattedDoc, "sample.yml", 0)
 		if err != nil {
 			t.Error(err, s.document, s.expression)
 			return
 		}
 		if s.document2 != "" {
-			moreInputs, err := readDocumentWithLeadingContent(formattedDoc2, "another.yml", 1)
+			moreInputs, err := readDocument(formattedDoc2, "another.yml", 1)
 			if err != nil {
 				t.Error(err, s.document, s.expression)
 				return

pkg/yqlib/printer.go

@@ -111,7 +111,7 @@ func (p *resultsPrinter) PrintResults(matchingNodes *list.List) error {
 
 		mappedDoc := el.Value.(*CandidateNode)
 		log.Debug("-- print sep logic: p.firstTimePrinting: %v, previousDocIndex: %v, mappedDoc.Document: %v", p.firstTimePrinting, p.previousDocIndex, mappedDoc.Document)
-
+		log.Debug("%v", NodeToString(mappedDoc))
 		writer, errorWriting := p.printerWriter.GetWriter(mappedDoc)
 		if errorWriting != nil {
 			return errorWriting

pkg/yqlib/printer_test.go

@@ -38,7 +38,7 @@ func TestPrinterMultipleDocsInSequenceOnly(t *testing.T) {
 	var writer = bufio.NewWriter(&output)
 	printer := NewSimpleYamlPrinter(writer, YamlOutputFormat, true, false, 2, true)
 
-	inputs, err := readDocuments(strings.NewReader(multiDocSample), "sample.yml", 0, NewYamlDecoder())
+	inputs, err := readDocuments(strings.NewReader(multiDocSample), "sample.yml", 0, NewYamlDecoder(ConfiguredYamlPreferences))
 	if err != nil {
 		panic(err)
 	}
@@ -76,7 +76,7 @@ func TestPrinterMultipleDocsInSequenceWithLeadingContent(t *testing.T) {
 	var writer = bufio.NewWriter(&output)
 	printer := NewSimpleYamlPrinter(writer, YamlOutputFormat, true, false, 2, true)
 
-	inputs, err := readDocuments(strings.NewReader(multiDocSample), "sample.yml", 0, NewYamlDecoder())
+	inputs, err := readDocuments(strings.NewReader(multiDocSample), "sample.yml", 0, NewYamlDecoder(ConfiguredYamlPreferences))
 	if err != nil {
 		panic(err)
 	}
@@ -118,7 +118,7 @@ func TestPrinterMultipleFilesInSequence(t *testing.T) {
 	var writer = bufio.NewWriter(&output)
 	printer := NewSimpleYamlPrinter(writer, YamlOutputFormat, true, false, 2, true)
 
-	inputs, err := readDocuments(strings.NewReader(multiDocSample), "sample.yml", 0, NewYamlDecoder())
+	inputs, err := readDocuments(strings.NewReader(multiDocSample), "sample.yml", 0, NewYamlDecoder(ConfiguredYamlPreferences))
 	if err != nil {
 		panic(err)
 	}
@@ -165,7 +165,7 @@ func TestPrinterMultipleFilesInSequenceWithLeadingContent(t *testing.T) {
 	var writer = bufio.NewWriter(&output)
 	printer := NewSimpleYamlPrinter(writer, YamlOutputFormat, true, false, 2, true)
 
-	inputs, err := readDocuments(strings.NewReader(multiDocSample), "sample.yml", 0, NewYamlDecoder())
+	inputs, err := readDocuments(strings.NewReader(multiDocSample), "sample.yml", 0, NewYamlDecoder(ConfiguredYamlPreferences))
 	if err != nil {
 		panic(err)
 	}
@@ -215,7 +215,7 @@ func TestPrinterMultipleDocsInSinglePrint(t *testing.T) {
 	var writer = bufio.NewWriter(&output)
 	printer := NewSimpleYamlPrinter(writer, YamlOutputFormat, true, false, 2, true)
 
-	inputs, err := readDocuments(strings.NewReader(multiDocSample), "sample.yml", 0, NewYamlDecoder())
+	inputs, err := readDocuments(strings.NewReader(multiDocSample), "sample.yml", 0, NewYamlDecoder(ConfiguredYamlPreferences))
 	if err != nil {
 		panic(err)
 	}
@@ -234,7 +234,7 @@ func TestPrinterMultipleDocsInSinglePrintWithLeadingDoc(t *testing.T) {
 	var writer = bufio.NewWriter(&output)
 	printer := NewSimpleYamlPrinter(writer, YamlOutputFormat, true, false, 2, true)
 
-	inputs, err := readDocuments(strings.NewReader(multiDocSample), "sample.yml", 0, NewYamlDecoder())
+	inputs, err := readDocuments(strings.NewReader(multiDocSample), "sample.yml", 0, NewYamlDecoder(ConfiguredYamlPreferences))
 	if err != nil {
 		panic(err)
 	}
@@ -263,7 +263,7 @@ func TestPrinterMultipleDocsInSinglePrintWithLeadingDocTrailing(t *testing.T) {
 	var writer = bufio.NewWriter(&output)
 	printer := NewSimpleYamlPrinter(writer, YamlOutputFormat, true, false, 2, true)
 
-	inputs, err := readDocuments(strings.NewReader(multiDocSample), "sample.yml", 0, NewYamlDecoder())
+	inputs, err := readDocuments(strings.NewReader(multiDocSample), "sample.yml", 0, NewYamlDecoder(ConfiguredYamlPreferences))
 	if err != nil {
 		panic(err)
 	}
@@ -294,7 +294,7 @@ func TestPrinterScalarWithLeadingCont(t *testing.T) {
 		panic(err)
 	}
 	streamEvaluator := NewStreamEvaluator()
-	_, err = streamEvaluator.Evaluate("sample", strings.NewReader(multiDocSample), node, printer, "# blah\n", NewYamlDecoder())
+	_, err = streamEvaluator.Evaluate("sample", strings.NewReader(multiDocSample), node, printer, NewYamlDecoder(ConfiguredYamlPreferences))
 	if err != nil {
 		panic(err)
 	}
@@ -316,7 +316,7 @@ func TestPrinterMultipleDocsJson(t *testing.T) {
 	// when outputing JSON.
 	printer := NewPrinter(NewJSONEncoder(0, false), NewSinglePrinterWriter(writer))
 
-	inputs, err := readDocuments(strings.NewReader(multiDocSample), "sample.yml", 0, NewYamlDecoder())
+	inputs, err := readDocuments(strings.NewReader(multiDocSample), "sample.yml", 0, NewYamlDecoder(ConfiguredYamlPreferences))
 	if err != nil {
 		panic(err)
 	}

pkg/yqlib/properties_test.go

@@ -8,7 +8,49 @@ import (
 	"github.com/mikefarah/yq/v4/test"
 )
 
-const samplePropertiesYaml = `# block comments don't come through
+const propertiesWithCommentsOnMap = `this.thing = hi hi
+# important notes
+# about this value
+this.value = cool
+`
+
+const expectedPropertiesWithCommentsOnMapProps = `this.thing = hi hi
+
+# important notes
+# about this value
+this.value = cool
+`
+
+const expectedPropertiesWithCommentsOnMapYaml = `this:
+  thing: hi hi
+  # important notes
+  # about this value
+  value: cool
+`
+
+const propertiesWithCommentInArray = `
+this.array.0 = cat
+# important notes
+# about dogs
+this.array.1 = dog
+`
+
+const expectedPropertiesWithCommentInArrayProps = `this.array.0 = cat
+
+# important notes
+# about dogs
+this.array.1 = dog
+`
+
+const expectedPropertiesWithCommentInArrayYaml = `this:
+  array:
+    - cat
+    # important notes
+    # about dogs
+    - dog
+`
+
+const samplePropertiesYaml = `# block comments come through
 person: # neither do comments on maps
     name: Mike Wazowski # comments on values appear
     pets: 
@@ -18,7 +60,8 @@ emptyArray: []
 emptyMap: []
 `
 
-const expectedPropertiesUnwrapped = `# comments on values appear
+const expectedPropertiesUnwrapped = `# block comments come through
+# comments on values appear
 person.name = Mike Wazowski
 
 # comments on array values appear
@@ -26,7 +69,8 @@ person.pets.0 = cat
 person.food.0 = pizza
 `
 
-const expectedPropertiesWrapped = `# comments on values appear
+const expectedPropertiesWrapped = `# block comments come through
+# comments on values appear
 person.name = "Mike Wazowski"
 
 # comments on array values appear
@@ -34,7 +78,8 @@ person.pets.0 = cat
 person.food.0 = pizza
 `
 
-const expectedUpdatedProperties = `# comments on values appear
+const expectedUpdatedProperties = `# block comments come through
+# comments on values appear
 person.name = Mike Wazowski
 
 # comments on array values appear
@@ -43,9 +88,12 @@ person.food.0 = pizza
 `
 
 const expectedDecodedYaml = `person:
-  name: Mike Wazowski # comments on values appear
+  # block comments come through
+  # comments on values appear
+  name: Mike Wazowski
   pets:
-    - cat # comments on array values appear
+    # comments on array values appear
+    - cat
   food:
     - pizza
 `
@@ -55,7 +103,8 @@ person.pets.0 = cat
 person.food.0 = pizza
 `
 
-const expectedPropertiesWithEmptyMapsAndArrays = `# comments on values appear
+const expectedPropertiesWithEmptyMapsAndArrays = `# block comments come through
+# comments on values appear
 person.name = Mike Wazowski
 
 # comments on array values appear
@@ -112,6 +161,34 @@ var propertyScenarios = []formatScenario{
 		expected:     expectedUpdatedProperties,
 		scenarioType: "roundtrip",
 	},
+	{
+		skipDoc:      true,
+		description:  "comments on arrays roundtrip",
+		input:        propertiesWithCommentInArray,
+		expected:     expectedPropertiesWithCommentInArrayProps,
+		scenarioType: "roundtrip",
+	},
+	{
+		skipDoc:      true,
+		description:  "comments on arrays decode",
+		input:        propertiesWithCommentInArray,
+		expected:     expectedPropertiesWithCommentInArrayYaml,
+		scenarioType: "decode",
+	},
+	{
+		skipDoc:      true,
+		description:  "comments on map roundtrip",
+		input:        propertiesWithCommentsOnMap,
+		expected:     expectedPropertiesWithCommentsOnMapProps,
+		scenarioType: "roundtrip",
+	},
+	{
+		skipDoc:      true,
+		description:  "comments on map decode",
+		input:        propertiesWithCommentsOnMap,
+		expected:     expectedPropertiesWithCommentsOnMapYaml,
+		scenarioType: "decode",
+	},
 	{
 		description:  "Empty doc",
 		skipDoc:      true,
@@ -143,7 +220,7 @@ func documentUnwrappedEncodePropertyScenario(w *bufio.Writer, s formatScenario)
 	}
 	writeOrPanic(w, "will output\n")
 
-	writeOrPanic(w, fmt.Sprintf("```properties\n%v```\n\n", processFormatScenario(s, NewYamlDecoder(), NewPropertiesEncoder(true))))
+	writeOrPanic(w, fmt.Sprintf("```properties\n%v```\n\n", processFormatScenario(s, NewYamlDecoder(ConfiguredYamlPreferences), NewPropertiesEncoder(true))))
 }
 
 func documentWrappedEncodePropertyScenario(w *bufio.Writer, s formatScenario) {
@@ -168,7 +245,7 @@ func documentWrappedEncodePropertyScenario(w *bufio.Writer, s formatScenario) {
 	}
 	writeOrPanic(w, "will output\n")
 
-	writeOrPanic(w, fmt.Sprintf("```properties\n%v```\n\n", processFormatScenario(s, NewYamlDecoder(), NewPropertiesEncoder(false))))
+	writeOrPanic(w, fmt.Sprintf("```properties\n%v```\n\n", processFormatScenario(s, NewYamlDecoder(ConfiguredYamlPreferences), NewPropertiesEncoder(false))))
 }
 
 func documentDecodePropertyScenario(w *bufio.Writer, s formatScenario) {
@@ -193,7 +270,7 @@ func documentDecodePropertyScenario(w *bufio.Writer, s formatScenario) {
 
 	writeOrPanic(w, "will output\n")
 
-	writeOrPanic(w, fmt.Sprintf("```yaml\n%v```\n\n", processFormatScenario(s, NewPropertiesDecoder(), NewYamlEncoder(s.indent, false, true, true))))
+	writeOrPanic(w, fmt.Sprintf("```yaml\n%v```\n\n", processFormatScenario(s, NewPropertiesDecoder(), NewYamlEncoder(s.indent, false, ConfiguredYamlPreferences))))
 }
 
 func documentRoundTripPropertyScenario(w *bufio.Writer, s formatScenario) {
@@ -245,11 +322,11 @@ func TestPropertyScenarios(t *testing.T) {
 	for _, s := range propertyScenarios {
 		switch s.scenarioType {
 		case "":
-			test.AssertResultWithContext(t, s.expected, processFormatScenario(s, NewYamlDecoder(), NewPropertiesEncoder(true)), s.description)
+			test.AssertResultWithContext(t, s.expected, processFormatScenario(s, NewYamlDecoder(ConfiguredYamlPreferences), NewPropertiesEncoder(true)), s.description)
 		case "decode":
-			test.AssertResultWithContext(t, s.expected, processFormatScenario(s, NewPropertiesDecoder(), NewYamlEncoder(2, false, true, true)), s.description)
+			test.AssertResultWithContext(t, s.expected, processFormatScenario(s, NewPropertiesDecoder(), NewYamlEncoder(2, false, ConfiguredYamlPreferences)), s.description)
 		case "encode-wrapped":
-			test.AssertResultWithContext(t, s.expected, processFormatScenario(s, NewYamlDecoder(), NewPropertiesEncoder(false)), s.description)
+			test.AssertResultWithContext(t, s.expected, processFormatScenario(s, NewYamlDecoder(ConfiguredYamlPreferences), NewPropertiesEncoder(false)), s.description)
 		case "roundtrip":
 			test.AssertResultWithContext(t, s.expected, processFormatScenario(s, NewPropertiesDecoder(), NewPropertiesEncoder(true)), s.description)
 

pkg/yqlib/stream_evaluator.go

@@ -14,9 +14,9 @@ import (
 // Uses less memory than loading all documents and running the expression once, but this cannot process
 // cross document expressions.
 type StreamEvaluator interface {
-	Evaluate(filename string, reader io.Reader, node *ExpressionNode, printer Printer, leadingContent string, decoder Decoder) (uint, error)
-	EvaluateFiles(expression string, filenames []string, printer Printer, leadingContentPreProcessing bool, decoder Decoder) error
-	EvaluateNew(expression string, printer Printer, leadingContent string) error
+	Evaluate(filename string, reader io.Reader, node *ExpressionNode, printer Printer, decoder Decoder) (uint, error)
+	EvaluateFiles(expression string, filenames []string, printer Printer, decoder Decoder) error
+	EvaluateNew(expression string, printer Printer) error
 }
 
 type streamEvaluator struct {
@@ -28,17 +28,16 @@ func NewStreamEvaluator() StreamEvaluator {
 	return &streamEvaluator{treeNavigator: NewDataTreeNavigator()}
 }
 
-func (s *streamEvaluator) EvaluateNew(expression string, printer Printer, leadingContent string) error {
+func (s *streamEvaluator) EvaluateNew(expression string, printer Printer) error {
 	node, err := ExpressionParser.ParseExpression(expression)
 	if err != nil {
 		return err
 	}
 	candidateNode := &CandidateNode{
-		Document:       0,
-		Filename:       "",
-		Node:           &yaml.Node{Kind: yaml.DocumentNode, Content: []*yaml.Node{{Tag: "!!null", Kind: yaml.ScalarNode}}},
-		FileIndex:      0,
-		LeadingContent: leadingContent,
+		Document:  0,
+		Filename:  "",
+		Node:      &yaml.Node{Kind: yaml.DocumentNode, Content: []*yaml.Node{{Tag: "!!null", Kind: yaml.ScalarNode}}},
+		FileIndex: 0,
 	}
 	inputList := list.New()
 	inputList.PushBack(candidateNode)
@@ -50,27 +49,20 @@ func (s *streamEvaluator) EvaluateNew(expression string, printer Printer, leadin
 	return printer.PrintResults(result.MatchingNodes)
 }
 
-func (s *streamEvaluator) EvaluateFiles(expression string, filenames []string, printer Printer, leadingContentPreProcessing bool, decoder Decoder) error {
+func (s *streamEvaluator) EvaluateFiles(expression string, filenames []string, printer Printer, decoder Decoder) error {
 	var totalProcessDocs uint
 	node, err := ExpressionParser.ParseExpression(expression)
 	if err != nil {
 		return err
 	}
 
-	var firstFileLeadingContent string
-
-	for index, filename := range filenames {
-		reader, leadingContent, err := readStream(filename, leadingContentPreProcessing)
-		log.Debug("leadingContent: %v", leadingContent)
-
-		if index == 0 {
-			firstFileLeadingContent = leadingContent
-		}
+	for _, filename := range filenames {
+		reader, err := readStream(filename)
 
 		if err != nil {
 			return err
 		}
-		processedDocs, err := s.Evaluate(filename, reader, node, printer, leadingContent, decoder)
+		processedDocs, err := s.Evaluate(filename, reader, node, printer, decoder)
 		if err != nil {
 			return err
 		}
@@ -83,19 +75,22 @@ func (s *streamEvaluator) EvaluateFiles(expression string, filenames []string, p
 	}
 
 	if totalProcessDocs == 0 {
-		return s.EvaluateNew(expression, printer, firstFileLeadingContent)
+		// problem is I've already slurped the leading content sadface
+		return s.EvaluateNew(expression, printer)
 	}
 
 	return nil
 }
 
-func (s *streamEvaluator) Evaluate(filename string, reader io.Reader, node *ExpressionNode, printer Printer, leadingContent string, decoder Decoder) (uint, error) {
+func (s *streamEvaluator) Evaluate(filename string, reader io.Reader, node *ExpressionNode, printer Printer, decoder Decoder) (uint, error) {
 
 	var currentIndex uint
-	decoder.Init(reader)
+	err := decoder.Init(reader)
+	if err != nil {
+		return 0, err
+	}
 	for {
-		var dataBucket yaml.Node
-		errorReading := decoder.Decode(&dataBucket)
+		candidateNode, errorReading := decoder.Decode()
 
 		if errors.Is(errorReading, io.EOF) {
 			s.fileIndex = s.fileIndex + 1
@@ -103,21 +98,10 @@ func (s *streamEvaluator) Evaluate(filename string, reader io.Reader, node *Expr
 		} else if errorReading != nil {
 			return currentIndex, fmt.Errorf("bad file '%v': %w", filename, errorReading)
 		}
+		candidateNode.Document = currentIndex
+		candidateNode.Filename = filename
+		candidateNode.FileIndex = s.fileIndex
 
-		candidateNode := &CandidateNode{
-			Document:  currentIndex,
-			Filename:  filename,
-			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)
 

pkg/yqlib/string_evaluator.go

@@ -1,17 +1,17 @@
 package yqlib
 
 import (
+	"bufio"
 	"bytes"
 	"container/list"
 	"errors"
 	"fmt"
 	"io"
-
-	yaml "gopkg.in/yaml.v3"
+	"strings"
 )
 
 type StringEvaluator interface {
-	Evaluate(expression string, input string, encoder Encoder, leadingContentPreProcessing bool, decoder Decoder) (string, error)
+	Evaluate(expression string, input string, encoder Encoder, decoder Decoder) (string, error)
 }
 
 type stringEvaluator struct {
@@ -25,7 +25,7 @@ func NewStringEvaluator() StringEvaluator {
 	}
 }
 
-func (s *stringEvaluator) Evaluate(expression string, input string, encoder Encoder, leadingContentPreProcessing bool, decoder Decoder) (string, error) {
+func (s *stringEvaluator) Evaluate(expression string, input string, encoder Encoder, decoder Decoder) (string, error) {
 
 	// Use bytes.Buffer for output of string
 	out := new(bytes.Buffer)
@@ -37,16 +37,15 @@ func (s *stringEvaluator) Evaluate(expression string, input string, encoder Enco
 		return "", err
 	}
 
-	reader, leadingContent, err := readString(input, leadingContentPreProcessing)
+	reader := bufio.NewReader(strings.NewReader(input))
+
+	var currentIndex uint
+	err = decoder.Init(reader)
 	if err != nil {
 		return "", err
 	}
-
-	var currentIndex uint
-	decoder.Init(reader)
 	for {
-		var dataBucket yaml.Node
-		errorReading := decoder.Decode(&dataBucket)
+		candidateNode, errorReading := decoder.Decode()
 
 		if errors.Is(errorReading, io.EOF) {
 			s.fileIndex = s.fileIndex + 1
@@ -54,20 +53,9 @@ func (s *stringEvaluator) Evaluate(expression string, input string, encoder Enco
 		} else if errorReading != nil {
 			return "", fmt.Errorf("bad input '%v': %w", input, errorReading)
 		}
+		candidateNode.Document = currentIndex
+		candidateNode.FileIndex = s.fileIndex
 
-		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)
 

pkg/yqlib/string_evaluator_test.go

@@ -18,10 +18,10 @@ func TestStringEvaluator_Evaluate_Nominal(t *testing.T) {
 		`---` + "\n" +
 		` - name: jq` + "\n" +
 		`   description: Command-line JSON processor` + "\n"
-	encoder := NewYamlEncoder(2, true, true, true)
-	decoder := NewYamlDecoder()
+	encoder := NewYamlEncoder(2, true, ConfiguredYamlPreferences)
+	decoder := NewYamlDecoder(ConfiguredYamlPreferences)
 
-	result, err := NewStringEvaluator().Evaluate(expression, input, encoder, true, decoder)
+	result, err := NewStringEvaluator().Evaluate(expression, input, encoder, decoder)
 	if err != nil {
 		t.Error(err)
 	}