HBASE-29878 Improve the Documentation UI (addendum) - fix author check in nightly (#7952)

Ignore files under hbase-website in author checks.
For this we monkey-patch YETUS-1266 fix for author_postcompile function in our hbase-personality.sh.

YETUS-1266 Escape regex metacharacters in author plugin ignore list

The author_postcompile function writes --author-ignore-list entries directly into a grep -E filter file without escaping. Filenames containing regex metacharacters (parentheses, brackets, etc.) are silently misinterpreted as pattern syntax, causing those files to pass through unfiltered.

Signed-off-by: Nick Dimiduk <ndimiduk@apache.org>

Author: PDavid

.github/workflows/yetus-general-check.yml

@@ -115,7 +115,7 @@ jobs:
           SOURCEDIR: "${{ github.workspace }}/src"
           TESTS_FILTER: "checkstyle,javac,pylint,shellcheck,shelldocs,blanks,perlcritic,ruby-lint,rubocop"
           YETUSDIR: "${{ github.workspace }}/yetus"
-          AUTHOR_IGNORE_LIST: "src/main/asciidoc/_chapters/developer.adoc"
+          AUTHOR_IGNORE_LIST: "hbase-website/app/pages/_docs/docs/_mdx/(multi-page)/building-and-developing/developer-guidelines.mdx,hbase-website/public/book.html"
           BLANKS_EOL_IGNORE_FILE: "dev-support/blanks-eol-ignore.txt"
           BLANKS_TABS_IGNORE_FILE: "dev-support/blanks-tabs-ignore.txt"
           EXCLUDE_TESTS_URL: "https://ci-hbase.apache.org/job/HBase-Find-Flaky-Tests/job/${{ github.base_ref }}/lastSuccessfulBuild/artifact/output/excludes"

.github/workflows/yetus-jdk17-hadoop3-compile-check.yml

@@ -114,7 +114,7 @@ jobs:
           SOURCEDIR: "${{ github.workspace }}/src"
           TESTS_FILTER: "javac,javadoc"
           YETUSDIR: "${{ github.workspace }}/yetus"
-          AUTHOR_IGNORE_LIST: "src/main/asciidoc/_chapters/developer.adoc"
+          AUTHOR_IGNORE_LIST: "hbase-website/app/pages/_docs/docs/_mdx/(multi-page)/building-and-developing/developer-guidelines.mdx,hbase-website/public/book.html"
           BLANKS_EOL_IGNORE_FILE: "dev-support/blanks-eol-ignore.txt"
           BLANKS_TABS_IGNORE_FILE: "dev-support/blanks-tabs-ignore.txt"
           BUILD_THREAD: "4"

.github/workflows/yetus-jdk17-hadoop3-unit-check.yml

@@ -132,7 +132,7 @@ jobs:
           SET_JAVA_HOME: "/usr/lib/jvm/java-17"
           SOURCEDIR: "${{ github.workspace }}/src"
           YETUSDIR: "${{ github.workspace }}/yetus"
-          AUTHOR_IGNORE_LIST: "src/main/asciidoc/_chapters/developer.adoc"
+          AUTHOR_IGNORE_LIST: "hbase-website/app/pages/_docs/docs/_mdx/(multi-page)/building-and-developing/developer-guidelines.mdx,hbase-website/public/book.html"
           BLANKS_EOL_IGNORE_FILE: "dev-support/blanks-eol-ignore.txt"
           BLANKS_TABS_IGNORE_FILE: "dev-support/blanks-tabs-ignore.txt"
           EXCLUDE_TESTS_URL: "https://ci-hbase.apache.org/job/HBase-Find-Flaky-Tests/job/${{ github.base_ref }}/lastSuccessfulBuild/artifact/output/excludes"

dev-support/Jenkinsfile

@@ -43,7 +43,7 @@ pipeline {
     PROJECT_PERSONALITY = 'https://raw.githubusercontent.com/apache/hbase/master/dev-support/hbase-personality.sh'
     PERSONALITY_FILE = 'tools/personality.sh'
     // This section of the docs tells folks not to use the javadoc tag. older branches have our old version of the check for said tag.
-    AUTHOR_IGNORE_LIST = 'src/main/asciidoc/_chapters/developer.adoc'
+    AUTHOR_IGNORE_LIST = 'hbase-website/app/pages/_docs/docs/_mdx/(multi-page)/building-and-developing/developer-guidelines.mdx,hbase-website/public/book.html'
     BLANKS_EOL_IGNORE_FILE = 'dev-support/blanks-eol-ignore.txt'
     BLANKS_TABS_IGNORE_FILE = 'dev-support/blanks-tabs-ignore.txt'
     // output from surefire; sadly the archive function in yetus only works on file names.

dev-support/Jenkinsfile_GitHub

@@ -41,7 +41,7 @@ pipeline {
         GENERAL_CHECK_PLUGINS = 'all,-javadoc,-jira,-shadedjars,-unit'
         JDK_SPECIFIC_PLUGINS = 'compile,github,htmlout,javac,javadoc,maven,mvninstall,shadedjars,unit'
         // This section of the docs tells folks not to use the javadoc tag. older branches have our old version of the check for said tag.
-        AUTHOR_IGNORE_LIST = 'src/main/asciidoc/_chapters/developer.adoc'
+        AUTHOR_IGNORE_LIST = 'hbase-website/app/pages/_docs/docs/_mdx/(multi-page)/building-and-developing/developer-guidelines.mdx,hbase-website/public/book.html'
         BLANKS_EOL_IGNORE_FILE = 'dev-support/blanks-eol-ignore.txt'
         BLANKS_TABS_IGNORE_FILE = 'dev-support/blanks-tabs-ignore.txt'
         // output from surefire; sadly the archive function in yetus only works on file names.

dev-support/hadoop3-backwards-compatibility-check.Jenkinsfile

@@ -135,7 +135,7 @@ pipeline {
           SKIP_ERRORPRONE = true
           OUTPUT_DIR_RELATIVE = "output-jdk17-hadoop3-backwards-${HADOOP3_VERSION}"
           OUTPUT_DIR = "${WORKSPACE}/${OUTPUT_DIR_RELATIVE}"
-          AUTHOR_IGNORE_LIST = 'src/main/asciidoc/_chapters/developer.adoc'
+          AUTHOR_IGNORE_LIST = 'hbase-website/app/pages/_docs/docs/_mdx/(multi-page)/building-and-developing/developer-guidelines.mdx,hbase-website/public/book.html'
           BLANKS_EOL_IGNORE_FILE = 'dev-support/blanks-eol-ignore.txt'
           BLANKS_TABS_IGNORE_FILE = 'dev-support/blanks-tabs-ignore.txt'
           // output from surefire; sadly the archive function in yetus only works on file names.

dev-support/hbase-personality.sh

@@ -290,6 +290,57 @@ function personality_modules
   done
 }
 
+# This is a workaround to fix the author check until YETUS-1266 is released.
+# TODO: Remove this when we upgraded to Yetus having YETUS-1266!
+## @description  Check the current directory for author tags
+## @audience     private
+## @stability    evolving
+## @replaceable  no
+## @return       0 on success
+## @return       1 on failure
+function author_postcompile
+{
+  # shellcheck disable=SC2155
+  declare -r appname=$(basename "${BASH_SOURCE-$0}")
+  declare -a globalignore
+
+  if [[ "${BUILDMODE}" != full ]]; then
+    return
+  fi
+
+  big_console_header "Checking for @author tags: ${BUILDMODE}"
+
+  start_clock
+
+  if [[ -f "${PATCH_DIR}/excluded.txt" ]]; then
+    globalignore=("${GREP}" "-v" "-f" "${PATCH_DIR}/excluded.txt")
+  else
+    globalignore=("cat")
+  fi
+
+  "${GIT}" grep -n -I --extended-regexp -i -e '^[^-].*@author' \
+    | "${GREP}" -v "${appname}" \
+    | "${globalignore[@]}" \
+    >> "${PATCH_DIR}/author-tags-git.txt"
+
+  if [[ -z "${AUTHOR_IGNORE_LIST[0]}" ]]; then
+    cp -p "${PATCH_DIR}/author-tags-git.txt" "${PATCH_DIR}/${AUTHOR_LOGNAME}"
+  else
+    for i in "${AUTHOR_IGNORE_LIST[@]}"; do
+      printf "%s\n" "${i}"
+    done \
+      | "${SED}" 's/[][\\.^$*+?{}()|]/\\&/g' \
+      | "${SED}" 's/^/^/' \
+      > "${PATCH_DIR}/author-tags-filter.txt"
+    "${GREP}" -v -E \
+      -f "${PATCH_DIR}/author-tags-filter.txt" \
+      "${PATCH_DIR}/author-tags-git.txt" \
+      > "${PATCH_DIR}/${AUTHOR_LOGNAME}"
+  fi
+
+  author_generic
+}
+
 ## @description places where we override the built in assumptions about what tests to run
 ## @audience    private
 ## @stability   evolving