Localization: Clean up Microsoft workflow (#29675)

* remove comments that no longer apply

* remove unused matrix variables

* remove CrowdIn-specific step of resetting broken files

* udpate link to Microsoft-specific log

* bring back reset script still in use

* add acceptable ja translation of TOS to rendering test

* add ability to remove files that have been translated but don't exist (wip)

* document what `languageFiles()` returns

* fix path issues for removed translations

* cleaner script execution

* add removed files to CSV report

* add workflow link to PR body for better troubleshooting

* catch missing regex for reporting file removals

* fix capturing groups in regex for csv reporting

* fix link to CSV file for PR

* fix formatting for PR

* pass in branch name from workflow

* put back helper scripts, make msft-specific copies

.github/actions-scripts/msft-create-translation-batch-pr.js

@@ -1,5 +1,6 @@
 #!/usr/bin/env node
 
+import fs from 'fs'
 import github from '@actions/github'
 
 const OPTIONS = Object.fromEntries(
@@ -32,6 +33,7 @@ const {
   BASE,
   HEAD,
   LANGUAGE,
+  BODY_FILE,
   GITHUB_TOKEN,
 } = OPTIONS
 const [OWNER, REPO] = GITHUB_REPOSITORY.split('/')
@@ -119,7 +121,7 @@ async function main() {
     title: TITLE,
     base: BASE,
     head: HEAD,
-    body: `New translation batch for ${LANGUAGE}. You can see the log in [\`translations/log/${LANGUAGE}-resets.csv\`](https://github.com/${OWNER}/${REPO}/tree/${HEAD}/translations/log/msft-${LANGUAGE}-resets.csv).`,
+    body: fs.readFileSync(BODY_FILE, 'utf8'),
     labels: ['translation-batch', `translation-batch-${LANGUAGE}`],
     owner: OWNER,
     repo: REPO,

.github/workflows/msft-create-translation-batch-pr.yml

@@ -1,4 +1,4 @@
-name: Create translation Batch Pull Request
+name: Create translation Batch Pull Request (Microsoft)
 
 # **What it does**:
 #  - Creates one pull request per language after running a series of automated checks,
@@ -31,48 +31,39 @@ jobs:
       matrix:
         include:
           - language: es
-            crowdin_language: es-ES
             language_dir: translations/es-ES
             language_repo: github/docs-internal.es-es
 
           - language: ja
-            crowdin_language: ja-JP
             language_dir: translations/ja-JP
             language_repo: github/docs-internal.ja-jp
 
           - language: pt
-            crowdin_language: pt-BR
             language_dir: translations/pt-BR
             language_repo: github/docs-internal.pt-br
 
           - language: cn
-            crowdin_language: zh-CN
             language_dir: translations/zh-CN
             language_repo: github/docs-internal.zh-cn
 
           # We'll be ready to add the following languages in a future effort.
 
           # - language: ru
-          #   crowdin_language: ru-RU
           #   language_dir: translations/ru-RU
           #   language_repo: github/docs-internal.ru-ru
 
           # - language: ko
-          #   crowdin_language: ko-KR
           #   language_dir: translations/ko-KR
           #   language_repo: github/docs-internal.ko-kr
 
           # - language: fr
-          #   crowdin_language: fr-FR
           #   language_dir: translations/fr-FR
           #   language_repo: github/docs-internal.fr-fr
 
           # - language: de
-          #   crowdin_language: de-DE
           #   language_dir: translations/de-DE
           #   language_repo: github/docs-internal.de-de
 
-    # TODO: replace the branch name
     steps:
       - name: Set branch name
         id: set-branch
@@ -109,11 +100,10 @@ jobs:
       - name: Remove .git from the language-specific repo
         run: rm -rf ${{ matrix.language_dir }}/.git
 
-      # TODO: Rename this step
-      - name: Commit crowdin sync
+      - name: Commit translated files
         run: |
           git add ${{ matrix.language_dir }}
-          git commit -m "Add crowdin translations" || echo "Nothing to commit"
+          git commit -m "Add translations" || echo "Nothing to commit"
 
       - name: 'Setup node'
         uses: actions/setup-node@17f8bd926464a1afa4c6a11669539e9c1ba77048
@@ -122,19 +112,16 @@ jobs:
 
       - run: npm ci
 
-      # step 6 in docs-engineering/crowdin.md
       - name: Homogenize frontmatter
         run: |
           node script/i18n/homogenize-frontmatter.js
           git add ${{ matrix.language_dir }} && git commit -m "Run script/i18n/homogenize-frontmatter.js" || echo "Nothing to commit"
 
-      # step 7 in docs-engineering/crowdin.md
       - name: Fix translation errors
         run: |
           node script/i18n/fix-translation-errors.js
           git add ${{ matrix.language_dir }} && git commit -m "Run script/i18n/fix-translation-errors.js" || echo "Nothing to commit"
 
-      # step 8b in docs-engineering/crowdin.md
       - name: Check rendering
         run: |
           node script/i18n/lint-translation-files.js --check rendering | tee -a /tmp/batch.log | cat
@@ -142,26 +129,18 @@ jobs:
 
       - name: Reset files with broken liquid tags
         run: |
-          node script/i18n/reset-files-with-broken-liquid-tags.js --language=${{ matrix.language }} | tee -a /tmp/batch.log | cat
-          git add ${{ matrix.language_dir }} && git commit -m "run script/i18n/reset-files-with-broken-liquid-tags.js --language=${{ matrix.language }}" || echo "Nothing to commit"
-
-      # step 5 in docs-engineering/crowdin.md using script from docs-internal#22709
-      - name: Reset known broken files
-        run: |
-          node script/i18n/reset-known-broken-translation-files.js | tee -a /tmp/batch.log | cat
-          git add ${{ matrix.language_dir }} && git commit -m "run script/i18n/reset-known-broken-translation-files.js" || echo "Nothing to commit"
-        env:
-          GITHUB_TOKEN: ${{ secrets.DOCUBOT_REPO_PAT }}
+          node script/i18n/msft-reset-files-with-broken-liquid-tags.js --language=${{ matrix.language }} | tee -a /tmp/batch.log | cat
+          git add ${{ matrix.language_dir }} && git commit -m "run script/i18n/msft-reset-files-with-broken-liquid-tags.js --language=${{ matrix.language }}" || echo "Nothing to commit"
 
       - name: Check in CSV report
         run: |
           mkdir -p translations/log
           csvFile=translations/log/msft-${{ matrix.language }}-resets.csv
-          script/i18n/report-reset-files.js --report-type=csv --language=${{ matrix.language }} --log-file=/tmp/batch.log > $csvFile
+          script/i18n/msft-report-reset-files.js --report-type=csv --language=${{ matrix.language }} --log-file=/tmp/batch.log > $csvFile
           git add -f $csvFile && git commit -m "Check in ${{ matrix.language }} CSV report" || echo "Nothing to commit"
 
       - name: Write the reported files that were reset to /tmp/pr-body.txt
-        run: script/i18n/report-reset-files.js --report-type=pull-request-body --language=${{ matrix.language }} --log-file=/tmp/batch.log > /tmp/pr-body.txt
+        run: script/i18n/msft-report-reset-files.js --report-type=pull-request-body --language=${{ matrix.language }} --log-file=/tmp/batch.log --csv-path=${{ steps.set-branch.outputs.BRANCH_NAME }}/translations/log/msft-${{ matrix.language }}-resets.csv > /tmp/pr-body.txt
 
       - name: Push filtered translations
         run: git push origin ${{ steps.set-branch.outputs.BRANCH_NAME }}