Refactoring: extract text transform library from libs/tags to libs/textutil (#3621)

## Why
<!-- Why are these changes needed? Provide the context that the reviewer
might be missing.
For example, were there any decisions behind the change that are not
reflected in the code itself? -->

- Text transformation logic can be useful outside the domain of working
with tags
- I am currently planning to re-use it for populating a dns-friendly
short user name in a following PR:
#3623

## Tests
<!-- How have you tested the changes? -->
Existing tests

<!-- If your PR needs to be included in the release notes for next
release,
add a separate entry in NEXT_CHANGELOG.md as part of your PR. -->

Author: anton-107

libs/tags/aws.go

@@ -4,6 +4,8 @@ import (
 	"regexp"
 	"unicode"
 
+	"github.com/databricks/cli/libs/textutil"
+
 	"golang.org/x/text/unicode/rangetable"
 )
 
@@ -20,17 +22,17 @@ var awsChars = rangetable.Merge(
 var awsTag = &tag{
 	keyLength:  127,
 	keyPattern: regexp.MustCompile(`^[\d \w\+\-=\.:\/@]*$`),
-	keyNormalize: chain(
-		normalizeMarks(),
-		replaceNotIn(latin1, '_'),
-		replaceNotIn(awsChars, '_'),
+	keyNormalize: textutil.Chain(
+		textutil.NormalizeMarks(),
+		textutil.ReplaceNotIn(textutil.Latin1, '_'),
+		textutil.ReplaceNotIn(awsChars, '_'),
 	),
 
 	valueLength:  255,
 	valuePattern: regexp.MustCompile(`^[\d \w\+\-=\.:/@]*$`),
-	valueNormalize: chain(
-		normalizeMarks(),
-		replaceNotIn(latin1, '_'),
-		replaceNotIn(awsChars, '_'),
+	valueNormalize: textutil.Chain(
+		textutil.NormalizeMarks(),
+		textutil.ReplaceNotIn(textutil.Latin1, '_'),
+		textutil.ReplaceNotIn(awsChars, '_'),
 	),
 }

libs/tags/azure.go

@@ -3,6 +3,8 @@ package tags
 import (
 	"regexp"
 
+	"github.com/databricks/cli/libs/textutil"
+
 	"golang.org/x/text/unicode/rangetable"
 )
 
@@ -12,14 +14,14 @@ var azureForbiddenChars = rangetable.New('<', '>', '*', '&', '%', ';', '\\', '/'
 var azureTag = &tag{
 	keyLength:  512,
 	keyPattern: regexp.MustCompile(`^[^<>\*&%;\\\/\+\?]*$`),
-	keyNormalize: chain(
-		replaceNotIn(latin1, '_'),
-		replaceIn(azureForbiddenChars, '_'),
+	keyNormalize: textutil.Chain(
+		textutil.ReplaceNotIn(textutil.Latin1, '_'),
+		textutil.ReplaceIn(azureForbiddenChars, '_'),
 	),
 
 	valueLength:  256,
 	valuePattern: regexp.MustCompile(`^.*$`),
-	valueNormalize: chain(
-		replaceNotIn(latin1, '_'),
+	valueNormalize: textutil.Chain(
+		textutil.ReplaceNotIn(textutil.Latin1, '_'),
 	),
 }

libs/tags/gcp.go

@@ -3,6 +3,8 @@ package tags
 import (
 	"regexp"
 	"unicode"
+
+	"github.com/databricks/cli/libs/textutil"
 )
 
 // Tag keys and values on GCP are limited to 63 characters and must match the
@@ -45,19 +47,19 @@ var gcpInner = &unicode.RangeTable{
 var gcpTag = &tag{
 	keyLength:  63,
 	keyPattern: regexp.MustCompile(`^([A-Za-z0-9][-A-Za-z0-9_.]*)?[A-Za-z0-9]$`),
-	keyNormalize: chain(
-		normalizeMarks(),
-		replaceNotIn(latin1, '_'),
-		replaceNotIn(gcpInner, '_'),
-		trimIfNotIn(gcpOuter),
+	keyNormalize: textutil.Chain(
+		textutil.NormalizeMarks(),
+		textutil.ReplaceNotIn(textutil.Latin1, '_'),
+		textutil.ReplaceNotIn(gcpInner, '_'),
+		textutil.TrimIfNotIn(gcpOuter),
 	),
 
 	valueLength:  63,
 	valuePattern: regexp.MustCompile(`^(([A-Za-z0-9][-A-Za-z0-9_.]*)?[A-Za-z0-9])?$`),
-	valueNormalize: chain(
-		normalizeMarks(),
-		replaceNotIn(latin1, '_'),
-		replaceNotIn(gcpInner, '_'),
-		trimIfNotIn(gcpOuter),
+	valueNormalize: textutil.Chain(
+		textutil.NormalizeMarks(),
+		textutil.ReplaceNotIn(textutil.Latin1, '_'),
+		textutil.ReplaceNotIn(gcpInner, '_'),
+		textutil.TrimIfNotIn(gcpOuter),
 	),
 }

libs/tags/latin_test.go

@@ -6,18 +6,20 @@ import (
 	"regexp"
 	"strings"
 	"unicode"
+
+	"github.com/databricks/cli/libs/textutil"
 )
 
 // The tag type holds the validation and normalization rules for
 // a cloud provider's resource tags as applied by Databricks.
 type tag struct {
 	keyLength    int
 	keyPattern   *regexp.Regexp
-	keyNormalize transformer
+	keyNormalize textutil.Transformer
 
 	valueLength    int
 	valuePattern   *regexp.Regexp
-	valueNormalize transformer
+	valueNormalize textutil.Transformer
 }
 
 func (t *tag) ValidateKey(s string) error {
@@ -27,7 +29,7 @@ func (t *tag) ValidateKey(s string) error {
 	if len(s) > t.keyLength {
 		return fmt.Errorf("key length %d exceeds maximum of %d", len(s), t.keyLength)
 	}
-	if strings.ContainsFunc(s, func(r rune) bool { return !unicode.Is(latin1, r) }) {
+	if strings.ContainsFunc(s, func(r rune) bool { return !unicode.Is(textutil.Latin1, r) }) {
 		return errors.New("key contains non-latin1 characters")
 	}
 	if !t.keyPattern.MatchString(s) {
@@ -40,7 +42,7 @@ func (t *tag) ValidateValue(s string) error {
 	if len(s) > t.valueLength {
 		return fmt.Errorf("value length %d exceeds maximum of %d", len(s), t.valueLength)
 	}
-	if strings.ContainsFunc(s, func(r rune) bool { return !unicode.Is(latin1, r) }) {
+	if strings.ContainsFunc(s, func(r rune) bool { return !unicode.Is(textutil.Latin1, r) }) {
 		return errors.New("value contains non-latin1 characters")
 	}
 	if !t.valuePattern.MatchString(s) {
@@ -50,9 +52,9 @@ func (t *tag) ValidateValue(s string) error {
 }
 
 func (t *tag) NormalizeKey(s string) string {
-	return t.keyNormalize.transform(s)
+	return t.keyNormalize.TransformString(s)
 }
 
 func (t *tag) NormalizeValue(s string) string {
-	return t.valueNormalize.transform(s)
+	return t.valueNormalize.TransformString(s)
 }

libs/tags/tag.go

@@ -1,9 +1,9 @@
-package tags
+package textutil
 
 import "unicode"
 
 // Range table for all characters in the Latin1 character set.
-var latin1 = &unicode.RangeTable{
+var Latin1 = &unicode.RangeTable{
 	R16: []unicode.Range16{
 		{0x0000, 0x00ff, 1},
 	},

libs/tags/transform.go

@@ -0,0 +1,16 @@
+package textutil
+
+import (
+	"testing"
+	"unicode"
+
+	"github.com/stretchr/testify/assert"
+)
+
+func TestLatinTable(t *testing.T) {
+	assert.True(t, unicode.In('\u0000', Latin1))
+	assert.True(t, unicode.In('A', Latin1))
+	assert.True(t, unicode.In('Z', Latin1))
+	assert.True(t, unicode.In('\u00ff', Latin1))
+	assert.False(t, unicode.In('\u0100', Latin1))
+}