[CI] Enforce Vale on Data documentation (ray-project#39029)

Vale is a linter for writing style. This PR enforces Vale on the Data documentation.

Signed-off-by: Balaji Veeramani <[email protected]>

.buildkite/lint.rayci.yml

@@ -67,3 +67,8 @@ steps:
       - RAY_DISABLE_EXTRA_CPP=1 pip install -e python/[all]
       - ./ci/lint/check_api_annotations.py
     depends_on: forge
+
+  - label: ":lint-roller: lint: documentation style"
+    commands:
+      - ./ci/lint/check-documentation-style.sh
+    depends_on: forge

.vale.ini

@@ -1,4 +1,4 @@
-StylesPath = .github/styles
+StylesPath = .vale/styles
 
 Vocab = General, Data
 

.github/styles/Google/Acronyms.yml → .vale/styles/Google/Acronyms.yml

@@ -1,7 +1,7 @@
 extends: conditional
 message: "Spell out '%s', if it's unfamiliar to the audience."
 link: 'https://developers.google.com/style/abbreviations'
-level: warning
+level: error
 ignorecase: false
 # Ensures that the existence of 'first' implies the existence of 'second'.
 first: '\b([A-Z]{3,5})\b'

.github/styles/Google/Contractions.yml → .vale/styles/Google/Contractions.yml

@@ -1,7 +1,7 @@
 extends: substitution
 message: "Use '%s' instead of '%s'."
 link: 'https://developers.google.com/style/contractions'
-level: warning
+level: error
 ignorecase: true
 action:
   name: replace

.github/styles/Google/Ellipses.yml → .vale/styles/Google/Ellipses.yml

@@ -2,7 +2,7 @@ extends: existence
 message: "In general, don't use an ellipsis."
 link: 'https://developers.google.com/style/ellipses'
 nonword: true
-level: warning
+level: error
 action:
   name: remove
 tokens:

.github/styles/Google/FirstPerson.yml → .vale/styles/Google/FirstPerson.yml

@@ -2,7 +2,7 @@ extends: existence
 message: "Avoid first-person pronouns such as '%s'."
 link: 'https://developers.google.com/style/pronouns#personal-pronouns'
 ignorecase: true
-level: warning
+level: error
 nonword: true
 tokens:
   - (?:^|\s)I\s

.github/styles/Google/HeadingPunctuation.yml → .vale/styles/Google/HeadingPunctuation.yml

@@ -2,7 +2,7 @@ extends: existence
 message: "Don't put a period at the end of a heading."
 link: 'https://developers.google.com/style/capitalization#capitalization-in-titles-and-headings'
 nonword: true
-level: warning
+level: error
 scope: heading
 action:
   name: edit

.github/styles/Google/Headings.yml → .vale/styles/Google/Headings.yml

@@ -1,7 +1,7 @@
 extends: capitalization
 message: "'%s' should use sentence-style capitalization."
 link: 'https://developers.google.com/style/capitalization#capitalization-in-titles-and-headings'
-level: warning
+level: error
 scope: heading
 match: $sentence
 indicators:

.github/styles/Google/OxfordComma.yml → .vale/styles/Google/OxfordComma.yml

@@ -2,6 +2,6 @@ extends: existence
 message: "Use the Oxford comma in '%s'."
 link: 'https://developers.google.com/style/commas'
 scope: sentence
-level: warning
+level: error
 tokens:
   - '(?:[^,]+,){1,}\s\w+\s(?:and|or)'

.github/styles/Google/Ranges.yml → .vale/styles/Google/Ranges.yml

@@ -2,6 +2,6 @@ extends: existence
 message: "Don't add words such as 'from' or 'between' to describe a range of numbers."
 link: 'https://developers.google.com/style/hyphens'
 nonword: true
-level: warning
+level: error
 tokens:
   - '(?:from|between)\s\d+\s?-\s?\d+'

.github/styles/Google/Spacing.yml → .vale/styles/Google/Spacing.yml

@@ -1,7 +1,7 @@
 extends: existence
 message: "'%s' should have one space."
 link: 'https://developers.google.com/style/sentence-spacing'
-level: warning
+level: error
 nonword: true
 action:
   name: remove

.github/styles/Google/Spelling.yml → .vale/styles/Google/Spelling.yml

@@ -2,7 +2,7 @@ extends: existence
 message: "In general, use American spelling instead of '%s'."
 link: 'https://developers.google.com/style/spelling'
 ignorecase: true
-level: warning
+level: error
 tokens:
   - '(?:\w+)nised?'
   - 'colour'

.github/styles/Google/We.yml → .vale/styles/Google/We.yml

@@ -1,7 +1,7 @@
 extends: existence
 message: "Try to avoid using first-person plural like '%s'."
 link: 'https://developers.google.com/style/pronouns#personal-pronouns'
-level: warning
+level: error
 ignorecase: true
 tokens:
   - we

.github/styles/Google/Will.yml → .vale/styles/Google/Will.yml

@@ -2,7 +2,7 @@ extends: existence
 message: "Avoid using '%s'."
 link: 'https://developers.google.com/style/tense'
 ignorecase: true
-level: warning
+level: error
 tokens:
   - will
   - "'ll"

.github/styles/Google/WordList.yml → .vale/styles/Google/WordList.yml

@@ -1,7 +1,7 @@
 extends: substitution
 message: "Use '%s' instead of '%s'."
 link: "https://developers.google.com/style/word-list"
-level: warning
+level: error
 ignorecase: false
 action:
   name: replace

.github/styles/Vocab/Data/accept.txt → .vale/styles/Vocab/Data/accept.txt

@@ -6,6 +6,7 @@ Spotify('s)?
 TFRecord(s)?
 UDF(s)?
 [Dd]atasource
+[Dd]iscretizer(s)?
 [Gg]roupby
 [Ii]ndexable
 [Ii]ngest

ci/lint/check-documentation-style.sh

@@ -0,0 +1,9 @@
+#!/usr/bin/env bash
+
+set -euxo pipefail
+
+VALE_BIN=$(mktemp -d)
+wget https://github.com/errata-ai/vale/releases/download/v2.28.0/vale_2.28.0_Linux_64-bit.tar.gz -P "$VALE_BIN"
+tar -xvzf "$VALE_BIN"/vale_2.28.0_Linux_64-bit.tar.gz -C "$VALE_BIN" vale
+"$VALE_BIN"/vale doc/source/data
+rm -rf "$VALE_BIN"