Replaces importDump.php with PagePort import script (#212)

* Replaces importDump.php with PagePort import script
* Introduces new `MW_AUTO_IMPORT` env variable which is `1` by default
* Always loads the `PagePort` extension of the `MW_AUTO_IMPORT` is set
* Performs automatic PagePort import from the `MW_IMPORT_VOLUME` if the volume is present and the ENV is set
* Runs import as maintenance user
* Updates PagePort to `65cb6908c852c5483efa5f8022b8bb97b5adb208`

Dockerfile

@@ -546,7 +546,7 @@ RUN set -x; \
 	# PagePort
 	&& git clone --single-branch -b master https://github.com/WikiTeq/PagePort.git $MW_HOME/extensions/PagePort \
 	&& cd $MW_HOME/extensions/PagePort \
-	&& git checkout -q a6b800c9b3f58c151cdda4ec2f1aa396536c3a7d
+	&& git checkout -q 65cb6908c852c5483efa5f8022b8bb97b5adb208
 
 # R
 RUN set -x; \
@@ -1059,7 +1059,8 @@ ENV MW_AUTOUPDATE=true \
 	MEDIAWIKI_MAINTENANCE_AUTO_ENABLED=false \
 	MW_SENTRY_DSN="" \
 	MW_USE_CACHE_DIRECTORY=1 \
-	APACHE_REMOTE_IP_HEADER=X-Forwarded-For
+	APACHE_REMOTE_IP_HEADER=X-Forwarded-For \
+	MW_AUTO_IMPORT=1
 
 COPY _sources/configs/msmtprc /etc/
 COPY _sources/configs/mediawiki.conf /etc/apache2/sites-enabled/

_sources/canasta/DockerSettings.php

@@ -606,6 +606,10 @@ function isEnvTrue( $name ): bool {
 	$wgUseCdn = false; // make sure profile is not cached
 }
 
+if ( getenv( 'MW_AUTO_IMPORT' ) ) {
+	wfLoadExtension( 'PagePort' );
+}
+
 # Include all php files in config/settings directory
 foreach ( glob( getenv( 'MW_CONFIG_DIR' ) . '/settings/*.php' ) as $filename ) {
 	if ( is_readable( $filename ) ) {

_sources/scripts/run-maintenance-scripts.sh

@@ -361,25 +361,16 @@ run_autoupdate () {
 }
 
 run_import () {
-    # Import dumps if any
+    # Import PagePort dumps if any
     if mountpoint -q -- "$MW_IMPORT_VOLUME"; then
-        echo "Found $MW_IMPORT_VOLUME, checking for XML dump presence.."
-        local FILES_IMPORTED=0
-        for filename in $MW_IMPORT_VOLUME/*.xml; do
-            echo "Found $filename file, importing.."
-            php maintenance/importDump.php \
-                --username-prefix="" \
-                --memory-limit=max \
-                --report 1 \
-                "$filename"
-            FILES_IMPORTED=$((FILES_IMPORTED + 1))
-        done
-        if [ "$FILES_IMPORTED" -gt "0" ]; then
-            echo "Running post-import maintenance scripts.."
-            php maintenance/rebuildrecentchanges.php
-            php maintenance/initSiteStats.php
+        echo "Found $MW_IMPORT_VOLUME, running PagePort import.."
+        XML_TEST=($(find $MW_IMPORT_VOLUME -maxdepth 1 -name "*.xml"))
+        if [ ${#XML_TEST[@]} -gt 0 ]; then
+        	echo "WARNING! The directory contains XML files, did you forget to migrate the dump to PagePort?"
+        	return
         fi
-        echo "Imported $FILES_IMPORTED files!"
+        php extensions/PagePort/maintenance/importPages.php --source "$MW_IMPORT_VOLUME" --user 'Maintenance script'
+        echo "Imported completed!"
     fi
 }
 
@@ -391,7 +382,9 @@ else
 fi
 
 # Run import after install and update are completed
-run_import
+if isTrue "$MW_AUTO_IMPORT"; then
+	run_import
+fi
 
 jobrunner &
 transcoder &