From db7e32441d2ae823d8c73506f4f13fad55ba7164 Mon Sep 17 00:00:00 2001
From: MusikAnimal <musikanimal@gmail.com>
Date: Mon, 18 Sep 2017 16:57:26 -0400
Subject: [PATCH] Attempt to fix populateIpChanges maintenance script

With the previous edition, some IP revisions failed to copy,
e.g. rev_id 1507693 on enwikivoyage. This change first fixes to go
by mBatchSize, and removes other redundant clauses in the query.

Bug: T175962
Change-Id: Ia5f3e275d8bef120090b4c60e6aec4c204d55d53
---
 maintenance/populateIpChanges.php | 19 +++++++++----------
 1 file changed, 9 insertions(+), 10 deletions(-)

diff --git a/maintenance/populateIpChanges.php b/maintenance/populateIpChanges.php
index c173270d21..17f49ee358 100644
--- a/maintenance/populateIpChanges.php
+++ b/maintenance/populateIpChanges.php
@@ -77,25 +77,26 @@ TEXT
 		$this->output( "Copying IP revisions to ip_changes, from rev_id $start to rev_id $end\n" );
 
 		while ( $blockStart <= $end ) {
-			$blockEnd = min( $blockStart + 200, $end );
+			$blockEnd = min( $blockStart + $this->mBatchSize, $end );
 			$rows = $dbr->select(
 				'revision',
 				[ 'rev_id', 'rev_timestamp', 'rev_user_text' ],
 				[ "rev_id BETWEEN $blockStart AND $blockEnd", 'rev_user' => 0 ],
-				__METHOD__,
-				[ 'ORDER BY' => 'rev_id ASC', 'LIMIT' => $this->mBatchSize ]
+				__METHOD__
 			);
 
-			if ( !$rows || $rows->numRows() === 0 ) {
+			$numRows = $rows->numRows();
+
+			if ( !$rows || $numRows === 0 ) {
 				break;
 			}
 
-			$this->output( "...checking $this->mBatchSize revisions for IP edits that need copying, " .
+			$this->output( "...checking $numRows revisions for IP edits that need copying, " .
 				"between rev_ids $blockStart and $blockEnd\n" );
 
 			$insertRows = [];
 			foreach ( $rows as $row ) {
-				// Double-check to make sure this is an IP, e.g. not maintenance user or imported revision.
+				// Make sure this is really an IP, e.g. not maintenance user or imported revision.
 				if ( IP::isValid( $row->rev_user_text ) ) {
 					$insertRows[] = [
 						'ipc_rev_id' => $row->rev_id,
@@ -105,12 +106,8 @@ TEXT
 
 					$revCount++;
 				}
-
-				$blockStart = (int)$row->rev_id;
 			}
 
-			$blockStart++;
-
 			$dbw->insert(
 				'ip_changes',
 				$insertRows,
@@ -120,6 +117,8 @@ TEXT
 
 			$lbFactory->waitForReplication();
 			usleep( $throttle * 1000 );
+
+			$blockStart = $blockEnd + 1;
 		}
 
 		$this->output( "$revCount IP revisions copied.\n" );
-- 
2.20.1