X-Git-Url: https://git.openstreetmap.org./nominatim.git/blobdiff_plain/832547f192904a9ec92e173c27a91e0874fcc757..179542f9388d235a2ced6676dfa95b6c45984f62:/wikidata/pull_wikipedia_logs.php

diff --git a/wikidata/pull_wikipedia_logs.php b/wikidata/pull_wikipedia_logs.php
index 3c5f5dfe..bdaab52a 100644
--- a/wikidata/pull_wikipedia_logs.php
+++ b/wikidata/pull_wikipedia_logs.php
@@ -1,17 +1,14 @@
 <?php
 
-for($iTimestamp = mktime(0, 0, 0, 5, 1, 2013); $iTimestamp < mktime(0, 0, 0, 6, 15, 2013); $iTimestamp += 24*60*60)
-{
+for ($iTimestamp = mktime(0, 0, 0, 5, 1, 2013); $iTimestamp < mktime(0, 0, 0, 6, 15, 2013); $iTimestamp += 24*60*60) {
     $sYear = date("Y", $iTimestamp);
     $sMonth = date("Y-m", $iTimestamp);
     $sDay = date("Ymd", $iTimestamp);
 
-    for($iHour = 0; $iHour < 24; $iHour++)
-    {
+    for ($iHour = 0; $iHour < 24; $iHour++) {
         $sFilename = sprintf("pagecounts-".$sDay."-%02d0000", $iHour);
         echo $sFilename."\n";
-        if (!file_exists($sFilename.'.gz'))
-        {
+        if (!file_exists($sFilename.'.gz')) {
             exec('wget http://dumps.wikimedia.org/other/pagecounts-raw/'.$sYear.'/'.$sMonth.'/'.$sFilename.'.gz');
         }
 
@@ -25,57 +22,42 @@ for($iTimestamp = mktime(0, 0, 0, 5, 1, 2013); $iTimestamp < mktime(0, 0, 0, 6,
         $sPrevLine = true;
         $sDayLine = true;
 
-        do
-        {
-            if ($sPrevKey === $sDayKey)
-            {
+        do {
+            if ($sPrevKey === $sDayKey) {
                 if ($sPrevLine !== true) fputs($hNewTotals, "$sPrevKey ".($iPrevValue+$iDayValue)."\n");
                 $sPrevLine = true;
                 $sDayLine = true;
-            }
-            else if ($sDayKey !== false && ($sPrevKey > $sDayKey || $sPrevKey === false))
-            {
+            } elseif ($sDayKey !== false && ($sPrevKey > $sDayKey || $sPrevKey === false)) {
                 fputs($hNewTotals, "$sDayKey ".($iDayValue)."\n");
                 $sDayLine = true;
-            }
-            else if ($sPrevKey !== false && ($sDayKey > $sPrevKey || $sDayKey === false))
-            {
+            } elseif ($sPrevKey !== false && ($sDayKey > $sPrevKey || $sDayKey === false)) {
                 fputs($hNewTotals, "$sPrevKey ".($iPrevValue)."\n");
                 $sPrevLine = true;
             }
 
-            if ($sPrevLine === true)
-            {
+            if ($sPrevLine === true) {
                 $sPrevLine = $hPrevTotals?fgets($hPrevTotals, 4096):false;
-                if ($sPrevLine !== false)
-                {
+                if ($sPrevLine !== false) {
                     $aPrevLine = explode(' ', $sPrevLine);
                     $sPrevKey = $aPrevLine[0].' '.$aPrevLine[1];
                     $iPrevValue = (int)$aPrevLine[2];
-                }
-                else
-                {
+                } else {
                     $sPrevKey = false;
                     $iPrevValue =  0;
                 }
             }
 
-            if ($sDayLine === true)
-            {
+            if ($sDayLine === true) {
                 $sDayLine = $hDayTotals?fgets($hDayTotals, 4096):false;
-                if ($sDayLine !== false)
-                {
+                if ($sDayLine !== false) {
                     preg_match('#^([a-z]{2}) ([^ :]+) ([0-9]+) [0-9]+$#', $sDayLine, $aMatch);
                     $sDayKey = $aMatch[1].' '.$aMatch[2];
                     $iDayValue = (int)$aMatch[3];
-                }
-                else
-                {
+                } else {
                     $sDayKey = false;
                     $iDayValue = 0;
                 }
             }
-
         } while ($sPrevLine !== false || $sDayLine !== false);
 
         @fclose($hPrevTotals);
@@ -89,7 +71,9 @@ for($iTimestamp = mktime(0, 0, 0, 5, 1, 2013); $iTimestamp < mktime(0, 0, 0, 6,
 
 // Notes:
 /*
- gzip -dc $FILE.gz | grep -e "^en [^ :]\+ [0-9]\+" | sed "s#\(^[a-z]\{2\}\) \([^ :]\+\) \([0-9]\+\) [0-9]\+#update wikipedia_article set hit_count = coalesce(hit_count,0) + \3 where language = '\1' and title = catch_decode_url_part('\2');#g" | /opt/mapquest/stdbase-dev$
+    gzip -dc $FILE.gz | grep -e "^en [^ :]\+ [0-9]\+" |
+    sed "s#\(^[a-z]\{2\}\) \([^ :]\+\) \([0-9]\+\) [0-9]\+#update wikipedia_article set hit_count = coalesce(hit_count,0) + \3 where language = '\1'
+      and title = catch_decode_url_part('\2');#g" | /opt/mapquest/stdbase-dev$
  cat totals.txt | sed "s#\(^[a-z]\{2\}\) \([^ ]\+\) \([0-9]\+\)\$#update entity_link set hits = s,0) + \3 where target = '\1wiki' and value = catch_decode_url_part('\2');#g"
  cat totals.txt | sed "s#\(^[a-z]\{2\}\) \([^ ]\+\) \([0-9]\+\)\$#update entity_link set hits = coalesce(hits,0) + \3 where target = '\1wiki' and value = catch_decode_url_part('\2');#g"
 */