Added new CSV option to turn off auto-parsing #1947

acceptance_tests/inputs-format.sh

@@ -63,6 +63,40 @@ EOM
   assertEquals "$expected" "$X"
 }
 
+testInputCSVNoAuto() {
+  cat >test.csv <<EOL
+thing1
+name: cat
+EOL
+
+  read -r -d '' expected << EOM
+- thing1: 'name: cat'
+EOM
+
+  X=$(./yq --csv-auto-parse=f test.csv -oy)
+  assertEquals "$expected" "$X"
+
+  X=$(./yq ea --csv-auto-parse=f test.csv -oy)
+  assertEquals "$expected" "$X"
+}
+
+testInputTSVNoAuto() {
+  cat >test.tsv <<EOL
+thing1
+name: cat
+EOL
+
+  read -r -d '' expected << EOM
+- thing1: 'name: cat'
+EOM
+
+  X=$(./yq --tsv-auto-parse=f test.tsv -oy)
+  assertEquals "$expected" "$X"
+
+  X=$(./yq ea --tsv-auto-parse=f test.tsv -oy)
+  assertEquals "$expected" "$X"
+}
+
 testInputCSVUTF8() {
   read -r -d '' expected << EOM
 - id: 1

cmd/root.go

@@ -80,6 +80,9 @@ yq -P -oy sample.json
 	rootCmd.PersistentFlags().BoolVar(&yqlib.ConfiguredXMLPreferences.SkipProcInst, "xml-skip-proc-inst", yqlib.ConfiguredXMLPreferences.SkipProcInst, "skip over process instructions (e.g. <?xml version=\"1\"?>)")
 	rootCmd.PersistentFlags().BoolVar(&yqlib.ConfiguredXMLPreferences.SkipDirectives, "xml-skip-directives", yqlib.ConfiguredXMLPreferences.SkipDirectives, "skip over directives (e.g. <!DOCTYPE thing cat>)")
 
+	rootCmd.PersistentFlags().BoolVar(&yqlib.ConfiguredCsvPreferences.AutoParse, "csv-auto-parse", yqlib.ConfiguredCsvPreferences.AutoParse, "parse CSV YAML/JSON values")
+	rootCmd.PersistentFlags().BoolVar(&yqlib.ConfiguredTsvPreferences.AutoParse, "tsv-auto-parse", yqlib.ConfiguredTsvPreferences.AutoParse, "parse TSV YAML/JSON values")
+
 	rootCmd.PersistentFlags().StringVar(&yqlib.ConfiguredLuaPreferences.DocPrefix, "lua-prefix", yqlib.ConfiguredLuaPreferences.DocPrefix, "prefix")
 	rootCmd.PersistentFlags().StringVar(&yqlib.ConfiguredLuaPreferences.DocSuffix, "lua-suffix", yqlib.ConfiguredLuaPreferences.DocSuffix, "suffix")
 	rootCmd.PersistentFlags().BoolVar(&yqlib.ConfiguredLuaPreferences.UnquotedKeys, "lua-unquoted", yqlib.ConfiguredLuaPreferences.UnquotedKeys, "output unquoted string keys (e.g. {foo=\"bar\"})")

cmd/utils.go

@@ -140,9 +140,9 @@ func createDecoder(format yqlib.InputFormat, evaluateTogether bool) (yqlib.Decod
 	case yqlib.JsonInputFormat:
 		return yqlib.NewJSONDecoder(), nil
 	case yqlib.CSVObjectInputFormat:
-		return yqlib.NewCSVObjectDecoder(','), nil
+		return yqlib.NewCSVObjectDecoder(yqlib.ConfiguredCsvPreferences), nil
 	case yqlib.TSVObjectInputFormat:
-		return yqlib.NewCSVObjectDecoder('\t'), nil
+		return yqlib.NewCSVObjectDecoder(yqlib.ConfiguredTsvPreferences), nil
 	case yqlib.TomlInputFormat:
 		return yqlib.NewTomlDecoder(), nil
 	case yqlib.YamlInputFormat:

pkg/yqlib/csv.go

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

pkg/yqlib/csv_test.go

@@ -12,6 +12,11 @@ const csvSimple = `name,numberOfCats,likesApples,height
 Gary,1,true,168.8
 Samantha's Rabbit,2,false,-188.8
 `
+
+const csvSimpleWithObject = `name,numberOfCats,likesApples,height,facts
+Gary,1,true,168.8,cool: true
+Samantha's Rabbit,2,false,-188.8,tall: indeed
+`
 const csvMissing = `name,numberOfCats,likesApples,height
 ,null,,168.8
 `
@@ -39,6 +44,31 @@ const expectedYamlFromCSV = `- name: Gary
   likesApples: false
   height: -188.8
 `
+const expectedYamlFromCSVWithObject = `- name: Gary
+  numberOfCats: 1
+  likesApples: true
+  height: 168.8
+  facts:
+    cool: true
+- name: Samantha's Rabbit
+  numberOfCats: 2
+  likesApples: false
+  height: -188.8
+  facts:
+    tall: indeed
+`
+
+const expectedYamlFromCSVNoParsing = `- name: Gary
+  numberOfCats: 1
+  likesApples: true
+  height: 168.8
+  facts: 'cool: true'
+- name: Samantha's Rabbit
+  numberOfCats: 2
+  likesApples: false
+  height: -188.8
+  facts: 'tall: indeed'
+`
 
 const expectedYamlFromCSVMissingData = `- name: Gary
   numberOfCats: 1
@@ -125,22 +155,29 @@ var csvScenarios = []formatScenario{
 		input:        csvSimple,
 		expression:   ".[0].name | key",
 		expected:     "name\n",
-		scenarioType: "decode-csv-object",
+		scenarioType: "decode-csv",
 	},
 	{
 		description:  "decode csv parent",
 		skipDoc:      true,
 		input:        csvSimple,
 		expression:   ".[0].name | parent | .height",
 		expected:     "168.8\n",
-		scenarioType: "decode-csv-object",
+		scenarioType: "decode-csv",
 	},
 	{
 		description:    "Parse CSV into an array of objects",
-		subdescription: "First row is assumed to be the header row.",
-		input:          csvSimple,
-		expected:       expectedYamlFromCSV,
-		scenarioType:   "decode-csv-object",
+		subdescription: "First row is assumed to be the header row. By default, entries with YAML/JSON formatting will be parsed!",
+		input:          csvSimpleWithObject,
+		expected:       expectedYamlFromCSVWithObject,
+		scenarioType:   "decode-csv",
+	},
+	{
+		description:    "Parse CSV into an array of objects, no auto-parsing",
+		subdescription: "First row is assumed to be the header row. Entries with YAML/JSON will be left as strings.",
+		input:          csvSimpleWithObject,
+		expected:       expectedYamlFromCSVNoParsing,
+		scenarioType:   "decode-csv-no-auto",
 	},
 	{
 		description:  "Scalar roundtrip",
@@ -172,12 +209,14 @@ func testCSVScenario(t *testing.T, s formatScenario) {
 		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewYamlDecoder(ConfiguredYamlPreferences), NewCsvEncoder(',')), s.description)
 	case "encode-tsv":
 		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewYamlDecoder(ConfiguredYamlPreferences), NewCsvEncoder('\t')), s.description)
-	case "decode-csv-object":
-		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewCSVObjectDecoder(','), NewYamlEncoder(2, false, ConfiguredYamlPreferences)), s.description)
+	case "decode-csv":
+		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewCSVObjectDecoder(ConfiguredCsvPreferences), NewYamlEncoder(2, false, ConfiguredYamlPreferences)), s.description)
+	case "decode-csv-no-auto":
+		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewCSVObjectDecoder(CsvPreferences{Separator: ',', AutoParse: false}), NewYamlEncoder(2, false, ConfiguredYamlPreferences)), s.description)
 	case "decode-tsv-object":
-		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewCSVObjectDecoder('\t'), NewYamlEncoder(2, false, ConfiguredYamlPreferences)), s.description)
+		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewCSVObjectDecoder(ConfiguredTsvPreferences), NewYamlEncoder(2, false, ConfiguredYamlPreferences)), s.description)
 	case "roundtrip-csv":
-		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewCSVObjectDecoder(','), NewCsvEncoder(',')), s.description)
+		test.AssertResultWithContext(t, s.expected, mustProcessFormatScenario(s, NewCSVObjectDecoder(ConfiguredCsvPreferences), NewCsvEncoder(',')), s.description)
 	default:
 		panic(fmt.Sprintf("unhandled scenario type %q", s.scenarioType))
 	}
@@ -204,7 +243,32 @@ func documentCSVDecodeObjectScenario(w *bufio.Writer, s formatScenario, formatTy
 	}
 
 	writeOrPanic(w, fmt.Sprintf("```yaml\n%v```\n\n",
-		mustProcessFormatScenario(s, NewCSVObjectDecoder(separator), NewYamlEncoder(s.indent, false, ConfiguredYamlPreferences))),
+		mustProcessFormatScenario(s, NewCSVObjectDecoder(CsvPreferences{Separator: separator, AutoParse: true}), NewYamlEncoder(s.indent, false, ConfiguredYamlPreferences))),
+	)
+}
+
+func documentCSVDecodeObjectNoAutoScenario(w *bufio.Writer, s formatScenario, formatType string) {
+	writeOrPanic(w, fmt.Sprintf("## %v\n", s.description))
+
+	if s.subdescription != "" {
+		writeOrPanic(w, s.subdescription)
+		writeOrPanic(w, "\n\n")
+	}
+
+	writeOrPanic(w, fmt.Sprintf("Given a sample.%v file of:\n", formatType))
+	writeOrPanic(w, fmt.Sprintf("```%v\n%v\n```\n", formatType, s.input))
+
+	writeOrPanic(w, "then\n")
+	writeOrPanic(w, fmt.Sprintf("```bash\nyq -p=%v --csv-auto-parse=f sample.%v\n```\n", formatType, formatType))
+	writeOrPanic(w, "will output\n")
+
+	separator := ','
+	if formatType == "tsv" {
+		separator = '\t'
+	}
+
+	writeOrPanic(w, fmt.Sprintf("```yaml\n%v```\n\n",
+		mustProcessFormatScenario(s, NewCSVObjectDecoder(CsvPreferences{Separator: separator, AutoParse: false}), NewYamlEncoder(s.indent, false, ConfiguredYamlPreferences))),
 	)
 }
 
@@ -268,7 +332,7 @@ func documentCSVRoundTripScenario(w *bufio.Writer, s formatScenario, formatType
 	}
 
 	writeOrPanic(w, fmt.Sprintf("```%v\n%v```\n\n", formatType,
-		mustProcessFormatScenario(s, NewCSVObjectDecoder(separator), NewCsvEncoder(separator))),
+		mustProcessFormatScenario(s, NewCSVObjectDecoder(CsvPreferences{Separator: separator, AutoParse: true}), NewCsvEncoder(separator))),
 	)
 }
 
@@ -282,8 +346,10 @@ func documentCSVScenario(_ *testing.T, w *bufio.Writer, i interface{}) {
 		documentCSVEncodeScenario(w, s, "csv")
 	case "encode-tsv":
 		documentCSVEncodeScenario(w, s, "tsv")
-	case "decode-csv-object":
+	case "decode-csv":
 		documentCSVDecodeObjectScenario(w, s, "csv")
+	case "decode-csv-no-auto":
+		documentCSVDecodeObjectNoAutoScenario(w, s, "csv")
 	case "decode-tsv-object":
 		documentCSVDecodeObjectScenario(w, s, "tsv")
 	case "roundtrip-csv":

pkg/yqlib/decoder_csv_object.go

@@ -9,27 +9,29 @@ import (
 )
 
 type csvObjectDecoder struct {
-	separator rune
-	reader    csv.Reader
-	finished  bool
+	prefs    CsvPreferences
+	reader   csv.Reader
+	finished bool
 }
 
-func NewCSVObjectDecoder(separator rune) Decoder {
-	return &csvObjectDecoder{separator: separator}
+func NewCSVObjectDecoder(prefs CsvPreferences) Decoder {
+	return &csvObjectDecoder{prefs: prefs}
 }
 
 func (dec *csvObjectDecoder) Init(reader io.Reader) error {
 	cleanReader, enc := utfbom.Skip(reader)
 	log.Debugf("Detected encoding: %s\n", enc)
 	dec.reader = *csv.NewReader(cleanReader)
-	dec.reader.Comma = dec.separator
+	dec.reader.Comma = dec.prefs.Separator
 	dec.finished = false
 	return nil
 }
 
 func (dec *csvObjectDecoder) convertToNode(content string) *CandidateNode {
 	node, err := parseSnippet(content)
-	if err != nil {
+	// if we're not auto-parsing, then we wont put in parsed objects or arrays
+	// but we still parse scalars
+	if err != nil || (!dec.prefs.AutoParse && node.Kind != ScalarNode) {
 		return createScalarNode(content, content)
 	}
 	return node

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

@@ -136,13 +136,13 @@ Samantha's Rabbit,,-188.8
 ```
 
 ## Parse CSV into an array of objects
-First row is assumed to be the header row.
+First row is assumed to be the header row. By default, entries with YAML/JSON formatting will be parsed!
 
 Given a sample.csv file of:
 ```csv
-name,numberOfCats,likesApples,height
-Gary,1,true,168.8
-Samantha's Rabbit,2,false,-188.8
+name,numberOfCats,likesApples,height,facts
+Gary,1,true,168.8,cool: true
+Samantha's Rabbit,2,false,-188.8,tall: indeed
 
 ```
 then
@@ -155,10 +155,42 @@ will output
   numberOfCats: 1
   likesApples: true
   height: 168.8
+  facts:
+    cool: true
+- name: Samantha's Rabbit
+  numberOfCats: 2
+  likesApples: false
+  height: -188.8
+  facts:
+    tall: indeed
+```
+
+## Parse CSV into an array of objects, no auto-parsing
+First row is assumed to be the header row. Entries with YAML/JSON will be left as strings.
+
+Given a sample.csv file of:
+```csv
+name,numberOfCats,likesApples,height,facts
+Gary,1,true,168.8,cool: true
+Samantha's Rabbit,2,false,-188.8,tall: indeed
+
+```
+then
+```bash
+yq -p=csv --csv-auto-parse=f sample.csv
+```
+will output
+```yaml
+- name: Gary
+  numberOfCats: 1
+  likesApples: true
+  height: 168.8
+  facts: 'cool: true'
 - name: Samantha's Rabbit
   numberOfCats: 2
   likesApples: false
   height: -188.8
+  facts: 'tall: indeed'
 ```
 
 ## Parse TSV into an array of objects

pkg/yqlib/operator_encoder_decoder.go

@@ -112,9 +112,9 @@ func createDecoder(format InputFormat) Decoder {
 	case PropertiesInputFormat:
 		decoder = NewPropertiesDecoder()
 	case CSVObjectInputFormat:
-		decoder = NewCSVObjectDecoder(',')
+		decoder = NewCSVObjectDecoder(ConfiguredCsvPreferences)
 	case TSVObjectInputFormat:
-		decoder = NewCSVObjectDecoder('\t')
+		decoder = NewCSVObjectDecoder(ConfiguredTsvPreferences)
 	case UriInputFormat:
 		decoder = NewUriDecoder()
 	}