Revcheck deduplication: ignore, regex, xml.

php · Nov 6, 2024 · 60b5f1e · 60b5f1e
1 parent 324f48c
commit 60b5f1e
Show file tree

Hide file tree

Showing 4 changed files with 36 additions and 17 deletions.
diff --git a/scripts/translation/lib/RevcheckIgnore.php b/scripts/translation/lib/RevcheckIgnore.php
@@ -42,26 +42,27 @@ public static function ignore( $filename ) : bool
         if ( str_contains( $filename , "/versions.xml" ) )
             return true;
 
-        // Only in English
+        // Only in English, autogenerated, marked not translatable
 
         if ( $filename == "contributors.ent" )
             return true;
         if ( $filename == "contributors.xml" )
             return true;
         if ( $filename == "appendices/license.xml" )
             return true;
-        if ( $filename == "appendices/license.xml" )
-            return true;
         if ( $filename == "appendices/extensions.xml" )
             return true;
         if ( $filename == "appendices/reserved.constants.xml" )
             return true;
         if ( $filename == "reference/datetime/timezones.xml" )
             return true;
-        if ( str_starts_with( $filename , 'chmonly/') )
-            return true;
-        if ( str_ends_with( $filename , '/book.developer.xml') )
-            return true;
+
+        if ( IGNORE_EXTENSIONS_XML )
+            if ( str_ends_with( $filename , '/extension.xml') ) // wrong
+                return true;
+        if ( IGNORE_CHMONLY_DIR )
+            if ( str_starts_with( $filename , 'chmonly/') )     // possibly wrong
+                return true;
 
         // Only in translations
 

diff --git a/scripts/translation/lib/RevcheckRun.php b/scripts/translation/lib/RevcheckRun.php
@@ -217,13 +217,13 @@ private function parseTranslationXml() : void
         $this->revData->lang = $this->targetDir;
         $this->revData->date = date("r");
 
-        $xml = XmlUtil::loadFile( $this->targetDir . '/translation.xml' );
+        $dom = XmlUtil::loadFile( $this->targetDir . '/translation.xml' );
 
         $this->revData->intro =
-            $xml->getElementsByTagName( 'intro' )[0]->textContent
+            $dom->saveXML( $dom->getElementsByTagName( 'intro' )[0] )
             ?? "No intro available for the {$lang} translation of the manual.";
 
-        $persons = $xml->getElementsByTagName( 'person' );
+        $persons = $dom->getElementsByTagName( 'person' );
         foreach( $persons as $person )
         {
             $nick = $person->getAttribute( 'nick' );

diff --git a/scripts/translation/lib/RevtagParser.php b/scripts/translation/lib/RevtagParser.php
@@ -67,11 +67,11 @@ public static function parseComment( DOMNode $node , RevtagInfo $ret , $filename
 
         if ( str_starts_with( $text , "EN-" ) )
         {
-            // /EN-Revision:\s*(\S+)\s*Maintainer:\s*(\S+)\s*Status:\s*(\S+)/ // restrict maintainer without spaces
-            // /EN-Revision:\s*(\S+)\s*Maintainer:\s(.*?)\sStatus:\s*(\S+)/   // accepts maintainer with spaces
+            // /EN-Revision:\s*(\S+)\s*Maintainer:\s*(\S+)\s*Status:\s*(\S+)/       // restrict maintainer without spaces
+            // /EN-Revision:\s*(\S+)\s*Maintainer:\s(.*?)\sStatus:\s*(\S+)/         // accepts maintainer with spaces
 
             $match = array();
-            $regex = "/EN-Revision:\s*(\S+)\s*Maintainer:\s*(\S+)\s*Status:\s*(\S+)/";
+            $regex = "/EN-Revision:\s*(\S+)\s*Maintainer:\s(.*?)\sStatus:\s*(\S+)/";
             if ( preg_match( $regex , $text , $match ) )
             {
                 $ret->revision = trim( $match[1] );

diff --git a/scripts/translation/lib/backport.php b/scripts/translation/lib/backport.php
@@ -9,12 +9,30 @@
 
 // Constants
 
+const IGNORE_CHMONLY_DIR = true;
+const IGNORE_EXTENSIONS_XML = true;
 const STATUS_COUNT_MISMATCH = true;
 const LOOSE_SKIP_REVCHECK = true;
 const FIXED_SKIP_REVCHECK = true;
 
 // Documentation
 
+/* # IGNORE_CHMONLY_DIR
+
+The chmonly/ dir contains files that appears to be
+translatable. See recent efforts to re-enabling
+CHM build: https://github.com/php/doc-base/pull/163
+*/
+
+/* # IGNORE_EXTENSIONS_XML
+
+The actual revcheck ignores any files called extensions.xml,
+but are at least two of translated files with this name.
+
+- appendices/migration56/extensions.xml
+- install/windows/legacy/extensions.xml
+*/
+
 /* # STATUS_COUNT_MISMATCH
 
 The actual revcheck counts "outdated" files in two different ways;
@@ -31,14 +49,14 @@
 was added here.
 
 Also, NotInEnTree is missing on first case, and files
-in this situation goes uncounted...
+in this situation goes uncounted.
 
-Also, RevTagProblem is counted towards as Old, but it
-is impossible to generate diffs with invalid hashes... */
+Also, RevTagProblem is counted towards as Old, but files
+are show in revtag missing/problem list, and is
+impossible to generate diffs with invalid hashes... */
 
 assert( STATUS_COUNT_MISMATCH || ! STATUS_COUNT_MISMATCH );
 
-
 /* # LOOSE_SKIP_REVCHECK
 
 Consider the output of: git show f80105b4fc1196bd8d5fecb98d686b580b1ff65d