Merge branch 'fix-grep-forbidden-words.sh' into 'preview'

Added tab-separated ruleset to grep-forbidden-words.sh to ease adding more rules.

See merge request !273

doc.zih.tu-dresden.de/util/grep-forbidden-words.sh

@@ -2,58 +2,107 @@
 
 set -euo pipefail
 
-branch="preview"
-if [ -n "$CI_MERGE_REQUEST_TARGET_BRANCH_NAME" ]; then
-    branch="origin/$CI_MERGE_REQUEST_TARGET_BRANCH_NAME"
-fi
+scriptpath=${BASH_SOURCE[0]}
+basedir=`dirname "$scriptpath"`
+basedir=`dirname "$basedir"`
+
+#This is the ruleset. Each line represents a rule of tab-separated fields.
+#The first field represents whether the match should be case-sensitive (s) or insensitive (i).
+#The second field represents the pattern that should not be contained in any file that is checked.
+#Further fields represent patterns with exceptions.
+#For example, the first rule says:
+# The pattern \<io\> should not be present in any file (case-insensitive match), except when it appears as ".io".
+ruleset="i	\<io\>	\.io
+s	\<SLURM\>
+i	file \+system
+i	\<taurus\>	taurus\.hrsk	/taurus
+i	\<hrskii\>
+i	hpc \+system
+i	hpc[ -]\+da\>"
 
-any_fails=false
+function grepExceptions () {
+  if [ $# -gt 0 ]; then
+    firstPattern=$1
+    shift
+    grep -v "$firstPattern" | grepExceptions "$@"
+  else
+    cat -
+  fi
+}
 
-files=$(git diff --name-only "$(git merge-base HEAD "$branch")")
+function usage () {
+  echo "$0 [options]"
+  echo "Search forbidden patterns in markdown files."
+  echo ""
+  echo "Options:"
+  echo "  -a     Search in all markdown files (default: git-changed files)" 
+  echo "  -s     Silent mode"
+  echo "  -h     Show help message"
+}
+
+# Options
+all_files=false
+silent=false
+while getopts ":ahs" option; do
+ case $option in
+   a)
+     all_files=true
+     ;;
+   s)
+     silent=true
+     ;;
+   h)
+     usage
+     exit;;
+   \?) # Invalid option
+     echo "Error: Invalid option."
+     usage
+     exit;;
+ esac
+done
+
+branch="origin/${CI_MERGE_REQUEST_TARGET_BRANCH_NAME:-preview}"
+
+if [ $all_files = true ]; then
+  echo "Search in all markdown files."
+  files=$(git ls-tree --full-tree -r --name-only HEAD $basedir/docs/ | grep .md)
+else
+  echo "Search in git-changed files."
+  files=`git diff --name-only "$(git merge-base HEAD "$branch")"`
+fi
+
+cnt=0
 for f in $files; do
-    if [ "$f" != doc.zih.tu-dresden.de/README.md -a "${f: -3}" == ".md" ]; then
-        #The following checks assume that grep signals success when it finds something,
-        #while it signals failure if it doesn't find something.
-        #We assume that we are successful if we DON'T find the pattern,
-        #which is the other way around, hence the "!".
-
-        echo "Checking wording of $f: IO"
-        #io must be the whole word
-        if ! grep -n -i '\<io\>' "$f" | grep -v '\.io'; then
-            any_fails=true
-        fi
-        echo "Checking wording of $f: SLURM"
-        #SLURM must be the whole word, otherwise it might match script variables
-        #such as SLURM_JOB_ID
-        if ! grep -n '\<SLURM\>' "$f"; then
-            any_fails=true
-        fi
-        echo "Checking wording of $f: file system"
-        #arbitrary white space in between
-        if ! grep -n -i 'file \+system' "$f"; then
-            any_fails=true
-        fi
-        #check for word taurus, except when used in conjunction with .hrsk or /taurus,
-        #which might appear in code snippets
-        echo "Checking wording of $f: taurus"
-        if ! grep -n -i '\<taurus\>' "$f" | grep -v 'taurus\.hrsk' | grep -v '/taurus'; then
-            any_fails=true
-        fi
-        echo "Checking wording of $f: hrskii"
-        if ! grep -n -i '\<hrskii\>' "$f"; then
-            any_fails=true
-        fi
-        echo "Checking wording of $f: hpc system"
-        if ! grep -n -i 'hpc \+system' "$f"; then
-            any_fails=true
+  if [ "$f" != doc.zih.tu-dresden.de/README.md -a "${f: -3}" == ".md" ]; then
+    echo "Check wording in file $f"
+    while IFS=$'\t' read -r flags pattern exceptionPatterns; do
+      while IFS=$'\t' read -r -a exceptionPatternsArray; do
+        if [ $silent = false ]; then
+          echo "  Pattern: $pattern"
         fi
-        echo "Checking wording of $f: hpc-da"
-        if ! grep -n -i 'hpc[ -]\+da\>' "$f"; then
-            any_fails=true
+        grepflag=
+        case "$flags" in
+          "i")
+            grepflag=-i
+          ;;
+        esac
+        if grep -n $grepflag "$pattern" "$f" | grepExceptions "${exceptionPatternsArray[@]}" ; then
+          ((cnt=cnt+1))
         fi
-    fi
+      done <<< $exceptionPatterns
+    done <<< $ruleset
+  fi
 done
 
-if [ "$any_fails" == true ]; then
-    exit 1
+echo "" 
+case $cnt in
+  1)
+    echo "Forbidden Patterns: 1 match found"
+  ;;
+  *)
+    echo "Forbidden Patterns: $cnt matches found"
+  ;;
+esac
+if [ $cnt -gt 0 ]; then
+  exit 1
 fi