Merge remote-tracking branch 'upstream/master'

[nominatim.git] / utils / update.php

diff --git a/utils/update.php b/utils/update.php
index 6965cd57e2dbceed20ee00a8b9edcd1f5d2076c3..87a5a294221e5fed07f83e42af1e602fb8e1c49f 100644 (file)
--- a/utils/update.php
+++ b/utils/update.php
@@ -39,22 +39,22 @@ $aCMDOptions
    array('index-rank', '', 0, 1, 1, 1, 'int', 'Rank to start indexing from'),
    array('index-instances', '', 0, 1, 1, 1, 'int', 'Number of indexing instances (threads)'),
 
    array('index-rank', '', 0, 1, 1, 1, 'int', 'Rank to start indexing from'),
    array('index-instances', '', 0, 1, 1, 1, 'int', 'Number of indexing instances (threads)'),
 

-   array('deduplicate', '', 0, 1, 0, 0, 'bool', 'Deduplicate tokens'),

    array('recompute-word-counts', '', 0, 1, 0, 0, 'bool', 'Compute frequency of full-word search terms'),
    array('update-address-levels', '', 0, 1, 0, 0, 'bool', 'Reimport address level configuration (EXPERT)'),
    array('recompute-word-counts', '', 0, 1, 0, 0, 'bool', 'Compute frequency of full-word search terms'),
    array('update-address-levels', '', 0, 1, 0, 0, 'bool', 'Reimport address level configuration (EXPERT)'),

-   array('recompute-importance', '', 0, 1, 0, 0, 'bool', 'Recompute place importances'),
-   array('no-npi', '', 0, 1, 0, 0, 'bool', '(obsolete)'),
+   array('recompute-importance', '', 0, 1, 0, 0, 'bool', 'Recompute place importances')

   );
 
 getCmdOpt($_SERVER['argv'], $aCMDOptions, $aResult, true, true);
 
 if (!isset($aResult['index-instances'])) $aResult['index-instances'] = 1;
   );
 
 getCmdOpt($_SERVER['argv'], $aCMDOptions, $aResult, true, true);
 
 if (!isset($aResult['index-instances'])) $aResult['index-instances'] = 1;

+

 if (!isset($aResult['index-rank'])) $aResult['index-rank'] = 0;
 
 date_default_timezone_set('Etc/UTC');
 
 $oDB = new Nominatim\DB();
 $oDB->connect();
 if (!isset($aResult['index-rank'])) $aResult['index-rank'] = 0;
 
 date_default_timezone_set('Etc/UTC');
 
 $oDB = new Nominatim\DB();
 $oDB->connect();

+$fPostgresVersion = $oDB->getPostgresVersion();

 
 $aDSNInfo = Nominatim\DB::parseDSN(CONST_Database_DSN);
 if (!isset($aDSNInfo['port']) || !$aDSNInfo['port']) $aDSNInfo['port'] = 5432;
 
 $aDSNInfo = Nominatim\DB::parseDSN(CONST_Database_DSN);
 if (!isset($aDSNInfo['port']) || !$aDSNInfo['port']) $aDSNInfo['port'] = 5432;


@@ -65,30 +65,60 @@ if ($iCacheMemory + 500 > getTotalMemoryMB()) {
     $iCacheMemory = getCacheMemoryMB();
     echo "WARNING: resetting cache memory to $iCacheMemory\n";
 }
     $iCacheMemory = getCacheMemoryMB();
     echo "WARNING: resetting cache memory to $iCacheMemory\n";
 }

-$sOsm2pgsqlCmd = CONST_Osm2pgsql_Binary.' -klas --number-processes 1 -C '.$iCacheMemory.' -O gazetteer -S '.CONST_Import_Style.' -d '.$aDSNInfo['database'].' -P '.$aDSNInfo['port'];
-if (isset($aDSNInfo['username']) && $aDSNInfo['username']) {
-    $sOsm2pgsqlCmd .= ' -U ' . $aDSNInfo['username'];
-}
+
+$oOsm2pgsqlCmd = (new \Nominatim\Shell(CONST_Osm2pgsql_Binary))
+                 ->addParams('--hstore')
+                 ->addParams('--latlong')
+                 ->addParams('--append')
+                 ->addParams('--slim')
+                 ->addParams('--number-processes', 1)
+                 ->addParams('--cache', $iCacheMemory)
+                 ->addParams('--output', 'gazetteer')
+                 ->addParams('--style', CONST_Import_Style)
+                 ->addParams('--database', $aDSNInfo['database'])
+                 ->addParams('--port', $aDSNInfo['port']);
+

 if (isset($aDSNInfo['hostspec']) && $aDSNInfo['hostspec']) {
 if (isset($aDSNInfo['hostspec']) && $aDSNInfo['hostspec']) {

-    $sOsm2pgsqlCmd .= ' -H ' . $aDSNInfo['hostspec'];
+    $oOsm2pgsqlCmd->addParams('--host', $aDSNInfo['hostspec']);
+}
+if (isset($aDSNInfo['username']) && $aDSNInfo['username']) {
+    $oOsm2pgsqlCmd->addParams('--user', $aDSNInfo['username']);

 }
 }

-$aProcEnv = null;

 if (isset($aDSNInfo['password']) && $aDSNInfo['password']) {
 if (isset($aDSNInfo['password']) && $aDSNInfo['password']) {

-    $aProcEnv = array_merge(array('PGPASSWORD' => $aDSNInfo['password']), $_ENV);
+    $oOsm2pgsqlCmd->addEnvPair('PGPASSWORD', $aDSNInfo['password']);

 }
 }

-

 if (!is_null(CONST_Osm2pgsql_Flatnode_File) && CONST_Osm2pgsql_Flatnode_File) {
 if (!is_null(CONST_Osm2pgsql_Flatnode_File) && CONST_Osm2pgsql_Flatnode_File) {

-    $sOsm2pgsqlCmd .= ' --flat-nodes '.CONST_Osm2pgsql_Flatnode_File;
+    $oOsm2pgsqlCmd->addParams('--flat-nodes', CONST_Osm2pgsql_Flatnode_File);
+}
+if ($fPostgresVersion >= 11.0) {
+    $oOsm2pgsqlCmd->addEnvPair(
+        'PGOPTIONS',
+        '-c jit=off -c max_parallel_workers_per_gather=0'
+    );

 }
 
 }
 

-$sIndexCmd = CONST_BasePath.'/nominatim/nominatim.py';
+
+$oIndexCmd = (new \Nominatim\Shell(CONST_BasePath.'/nominatim/nominatim.py'))
+             ->addParams('--database', $aDSNInfo['database'])
+             ->addParams('--port', $aDSNInfo['port'])
+             ->addParams('--threads', $aResult['index-instances']);

 if (!$aResult['quiet']) {
 if (!$aResult['quiet']) {

-    $sIndexCmd .= ' -v';
+    $oIndexCmd->addParams('--verbose');

 }
 if ($aResult['verbose']) {
 }
 if ($aResult['verbose']) {

-    $sIndexCmd .= ' -v';
+    $oIndexCmd->addParams('--verbose');
+}
+if (isset($aDSNInfo['hostspec']) && $aDSNInfo['hostspec']) {
+    $oIndexCmd->addParams('--host', $aDSNInfo['hostspec']);
+}
+if (isset($aDSNInfo['username']) && $aDSNInfo['username']) {
+    $oIndexCmd->addParams('--username', $aDSNInfo['username']);
+}
+if (isset($aDSNInfo['password']) && $aDSNInfo['password']) {
+    $oIndexCmd->addEnvPair('PGPASSWORD', $aDSNInfo['password']);

 }
 
 }
 

+

 if ($aResult['init-updates']) {
     // sanity check that the replication URL is correct
     $sBaseState = file_get_contents(CONST_Replication_Url.'/state.txt');
 if ($aResult['init-updates']) {
     // sanity check that the replication URL is correct
     $sBaseState = file_get_contents(CONST_Replication_Url.'/state.txt');


@@ -104,9 +134,11 @@ if ($aResult['init-updates']) {
         echo "in your local settings file.\n\n";
         fail('CONST_Pyosmium_Binary not configured');
     }
         echo "in your local settings file.\n\n";
         fail('CONST_Pyosmium_Binary not configured');
     }

+

     $aOutput = 0;
     $aOutput = 0;

-    $sCmd = CONST_Pyosmium_Binary.' --help';
-    exec($sCmd, $aOutput, $iRet);
+    $oCMD = new \Nominatim\Shell(CONST_Pyosmium_Binary, '--help');
+    exec($oCMD->escapedCmd(), $aOutput, $iRet);
+

     if ($iRet != 0) {
         echo "Cannot execute pyosmium-get-changes.\n";
         echo "Make sure you have pyosmium installed correctly\n";
     if ($iRet != 0) {
         echo "Cannot execute pyosmium-get-changes.\n";
         echo "Make sure you have pyosmium installed correctly\n";


@@ -120,7 +152,6 @@ if ($aResult['init-updates']) {
                                       'enable-diff-updates' => true,
                                       'verbose' => $aResult['verbose']
                                      ));
                                       'enable-diff-updates' => true,
                                       'verbose' => $aResult['verbose']
                                      ));

-        $cSetup->connect();

         $cSetup->createFunctions();
     }
 
         $cSetup->createFunctions();
     }
 


@@ -132,8 +163,11 @@ if ($aResult['init-updates']) {
 
     // get the appropriate state id
     $aOutput = 0;
 
     // get the appropriate state id
     $aOutput = 0;

-    $sCmd = CONST_Pyosmium_Binary.' -D '.$sWindBack.' --server '.CONST_Replication_Url;
-    exec($sCmd, $aOutput, $iRet);
+    $oCMD = (new \Nominatim\Shell(CONST_Pyosmium_Binary))
+            ->addParams('--start-date', $sWindBack)
+            ->addParams('--server', CONST_Replication_Url);
+
+    exec($oCMD->escapedCmd(), $aOutput, $iRet);

     if ($iRet != 0 || $aOutput[0] == 'None') {
         fail('Error running pyosmium tools');
     }
     if ($iRet != 0 || $aOutput[0] == 'None') {
         fail('Error running pyosmium tools');
     }


@@ -158,7 +192,11 @@ if ($aResult['check-for-updates']) {
         fail('Updates not set up. Please run ./utils/update.php --init-updates.');
     }
 
         fail('Updates not set up. Please run ./utils/update.php --init-updates.');
     }
 

-    system(CONST_BasePath.'/utils/check_server_for_updates.py '.CONST_Replication_Url.' '.$aLastState['sequence_id'], $iRet);
+    $oCmd = (new \Nominatim\Shell(CONST_BasePath.'/utils/check_server_for_updates.py'))
+            ->addParams(CONST_Replication_Url)
+            ->addParams($aLastState['sequence_id']);
+    $iRet = $oCmd->run();
+

     exit($iRet);
 }
 
     exit($iRet);
 }
 


@@ -171,12 +209,12 @@ if (isset($aResult['import-diff']) || isset($aResult['import-file'])) {
     }
 
     // Import the file
     }
 
     // Import the file

-    $sCMD = $sOsm2pgsqlCmd.' '.$sNextFile;
-    echo $sCMD."\n";
-    $iErrorLevel = runWithEnv($sCMD, $aProcEnv);
+    $oCMD = (clone $oOsm2pgsqlCmd)->addParams($sNextFile);
+    echo $oCMD->escapedCmd()."\n";
+    $iRet = $oCMD->run();

 
 

-    if ($iErrorLevel) {
-        fail("Error from osm2pgsql, $iErrorLevel\n");
+    if ($iRet) {
+        fail("Error from osm2pgsql, $iRet\n");

     }
 
     // Don't update the import status - we don't know what this file contains
     }
 
     // Don't update the import status - we don't know what this file contains


@@ -204,15 +242,15 @@ if (isset($aResult['import-way']) && $aResult['import-way']) {
     if ($bUseOSMApi) {
         $sContentURL = 'https://www.openstreetmap.org/api/0.6/way/'.$aResult['import-way'].'/full';
     } else {
     if ($bUseOSMApi) {
         $sContentURL = 'https://www.openstreetmap.org/api/0.6/way/'.$aResult['import-way'].'/full';
     } else {

-        $sContentURL = 'https://overpass-api.de/api/interpreter?data=(way('.$aResult['import-way'].');node(w););out%20meta;';
+        $sContentURL = 'https://overpass-api.de/api/interpreter?data=(way('.$aResult['import-way'].');%3E;);out%20meta;';

     }
 }
 
 if (isset($aResult['import-relation']) && $aResult['import-relation']) {
     if ($bUseOSMApi) {
     }
 }
 
 if (isset($aResult['import-relation']) && $aResult['import-relation']) {
     if ($bUseOSMApi) {

-        $sContentURLsModifyXMLstr = 'https://www.openstreetmap.org/api/0.6/relation/'.$aResult['import-relation'].'/full';
+        $sContentURL = 'https://www.openstreetmap.org/api/0.6/relation/'.$aResult['import-relation'].'/full';

     } else {
     } else {

-        $sContentURL = 'https://overpass-api.de/api/interpreter?data=((rel('.$aResult['import-relation'].');way(r);node(w));node(r));out%20meta;';
+        $sContentURL = 'https://overpass-api.de/api/interpreter?data=(rel(id:'.$aResult['import-relation'].');%3E;);out%20meta;';

     }
 }
 
     }
 }
 


@@ -223,83 +261,13 @@ if ($sContentURL) {
 
 if ($bHaveDiff) {
     // import generated change file
 
 if ($bHaveDiff) {
     // import generated change file

-    $sCMD = $sOsm2pgsqlCmd.' '.$sTemporaryFile;
-    echo $sCMD."\n";
-    $iErrorLevel = runWithEnv($sCMD, $aProcEnv);
-    if ($iErrorLevel) {
-        fail("osm2pgsql exited with error level $iErrorLevel\n");
-    }
-}
-
-if ($aResult['deduplicate']) {
-    $oDB = new Nominatim\DB();
-    $oDB->connect();
-
-    if ($oDB->getPostgresVersion() < 9.3) {
-        fail('ERROR: deduplicate is only currently supported in postgresql 9.3');
-    }

 
 

-    $sSQL = 'select partition from country_name order by country_code';
-    $aPartitions = $oDB->getCol($sSQL);
-    $aPartitions[] = 0;
+    $oCMD = (clone $oOsm2pgsqlCmd)->addParams($sTemporaryFile);
+    echo $oCMD->escapedCmd()."\n";

 
 

-    // we don't care about empty search_name_* partitions, they can't contain mentions of duplicates
-    foreach ($aPartitions as $i => $sPartition) {
-        $sSQL = 'select count(*) from search_name_'.$sPartition;
-        $nEntries = $oDB->getOne($sSQL);
-        if ($nEntries == 0) {
-            unset($aPartitions[$i]);
-        }
-    }
-
-    $sSQL = "select word_token,count(*) from word where substr(word_token, 1, 1) = ' '";
-    $sSQL .= ' and class is null and type is null and country_code is null';
-    $sSQL .= ' group by word_token having count(*) > 1 order by word_token';
-    $aDuplicateTokens = $oDB->getAll($sSQL);
-    foreach ($aDuplicateTokens as $aToken) {
-        if (trim($aToken['word_token']) == '' || trim($aToken['word_token']) == '-') continue;
-        echo 'Deduping '.$aToken['word_token']."\n";
-        $sSQL = 'select word_id,';
-        $sSQL .= ' (select count(*) from search_name where nameaddress_vector @> ARRAY[word_id]) as num';
-        $sSQL .= " from word where word_token = '".$aToken['word_token'];
-        $sSQL .= "' and class is null and type is null and country_code is null order by num desc";
-        $aTokenSet = $oDB->getAll($sSQL);
-
-        $aKeep = array_shift($aTokenSet);
-        $iKeepID = $aKeep['word_id'];
-
-        foreach ($aTokenSet as $aRemove) {
-            $sSQL = 'update search_name set';
-            $sSQL .= ' name_vector = array_replace(name_vector,'.$aRemove['word_id'].','.$iKeepID.'),';
-            $sSQL .= ' nameaddress_vector = array_replace(nameaddress_vector,'.$aRemove['word_id'].','.$iKeepID.')';
-            $sSQL .= ' where name_vector @> ARRAY['.$aRemove['word_id'].']';
-            $oDB->exec($sSQL);
-
-            $sSQL = 'update search_name set';
-            $sSQL .= ' nameaddress_vector = array_replace(nameaddress_vector,'.$aRemove['word_id'].','.$iKeepID.')';
-            $sSQL .= ' where nameaddress_vector @> ARRAY['.$aRemove['word_id'].']';
-            $oDB->exec($sSQL);
-
-            $sSQL = 'update location_area_country set';
-            $sSQL .= ' keywords = array_replace(keywords,'.$aRemove['word_id'].','.$iKeepID.')';
-            $sSQL .= ' where keywords @> ARRAY['.$aRemove['word_id'].']';
-            $oDB->exec($sSQL);
-
-            foreach ($aPartitions as $sPartition) {
-                $sSQL = 'update search_name_'.$sPartition.' set';
-                $sSQL .= ' name_vector = array_replace(name_vector,'.$aRemove['word_id'].','.$iKeepID.')';
-                $sSQL .= ' where name_vector @> ARRAY['.$aRemove['word_id'].']';
-                $oDB->exec($sSQL);
-
-                $sSQL = 'update location_area_country set';
-                $sSQL .= ' keywords = array_replace(keywords,'.$aRemove['word_id'].','.$iKeepID.')';
-                $sSQL .= ' where keywords @> ARRAY['.$aRemove['word_id'].']';
-                $oDB->exec($sSQL);
-            }
-
-            $sSQL = 'delete from word where word_id = '.$aRemove['word_id'];
-            $oDB->exec($sSQL);
-        }
+    $iRet = $oCMD->run();
+    if ($iRet) {
+        fail("osm2pgsql exited with error level $iRet\n");

     }
 }
 
     }
 }
 


@@ -310,19 +278,13 @@ if ($aResult['recompute-word-counts']) {
 }
 
 if ($aResult['index']) {
 }
 
 if ($aResult['index']) {

-    $sCmd = $sIndexCmd
-            .' -d '.$aDSNInfo['database']
-            .' -P '.$aDSNInfo['port']
-            .' -t '.$aResult['index-instances']
-            .' -r '.$aResult['index-rank'];
-    if (isset($aDSNInfo['hostspec']) && $aDSNInfo['hostspec']) {
-        $sCmd .= ' -H ' . $aDSNInfo['hostspec'];
-    }
-    if (isset($aDSNInfo['username']) && $aDSNInfo['username']) {
-        $sCmd .= ' -U ' . $aDSNInfo['username'];
-    }
+    $oCmd = (clone $oIndexCmd)
+            ->addParams('--minrank', $aResult['index-rank'], '-b');
+    $oCmd->run();

 
 

-    runWithEnv($sCmd, $aProcEnv);
+    $oCmd = (clone $oIndexCmd)
+            ->addParams('--minrank', $aResult['index-rank']);
+    $oCmd->run();

 
     $oDB->exec('update import_status set indexed = true');
 }
 
     $oDB->exec('update import_status set indexed = true');
 }


@@ -358,18 +320,13 @@ if ($aResult['import-osmosis'] || $aResult['import-osmosis-all']) {
     }
 
     $sImportFile = CONST_InstallPath.'/osmosischange.osc';
     }
 
     $sImportFile = CONST_InstallPath.'/osmosischange.osc';

-    $sCMDDownload = CONST_Pyosmium_Binary.' --server '.CONST_Replication_Url.' -o '.$sImportFile.' -s '.CONST_Replication_Max_Diff_size;
-    $sCMDImport = $sOsm2pgsqlCmd.' '.$sImportFile;
-    $sCMDIndex = $sIndexCmd
-                 .' -d '.$aDSNInfo['database']
-                 .' -P '.$aDSNInfo['port']
-                 .' -t '.$aResult['index-instances'];
-    if (isset($aDSNInfo['hostspec']) && $aDSNInfo['hostspec']) {
-        $sCMDIndex .= ' -H ' . $aDSNInfo['hostspec'];
-    }
-    if (isset($aDSNInfo['username']) && $aDSNInfo['username']) {
-        $sCMDIndex .= ' -U ' . $aDSNInfo['username'];
-    }
+
+    $oCMDDownload = (new \Nominatim\Shell(CONST_Pyosmium_Binary))
+                    ->addParams('--server', CONST_Replication_Url)
+                    ->addParams('--outfile', $sImportFile)
+                    ->addParams('--size', CONST_Replication_Max_Diff_size);
+
+    $oCMDImport = (clone $oOsm2pgsqlCmd)->addParams($sImportFile);

 
     while (true) {
         $fStartTime = time();
 
     while (true) {
         $fStartTime = time();


@@ -399,11 +356,13 @@ if ($aResult['import-osmosis'] || $aResult['import-osmosis-all']) {
                 $fCMDStartTime = time();
                 $iNextSeq = (int) $aLastState['sequence_id'];
                 unset($aOutput);
                 $fCMDStartTime = time();
                 $iNextSeq = (int) $aLastState['sequence_id'];
                 unset($aOutput);

-                echo "$sCMDDownload -I $iNextSeq\n";
+
+                $oCMD = (clone $oCMDDownload)->addParams('--start-id', $iNextSeq);
+                echo $oCMD->escapedCmd()."\n";

                 if (file_exists($sImportFile)) {
                     unlink($sImportFile);
                 }
                 if (file_exists($sImportFile)) {
                     unlink($sImportFile);
                 }

-                exec($sCMDDownload.' -I '.$iNextSeq, $aOutput, $iResult);
+                exec($oCMD->escapedCmd(), $aOutput, $iResult);

 
                 if ($iResult == 3) {
                     echo 'No new updates. Sleeping for '.CONST_Replication_Recheck_Interval." sec.\n";
 
                 if ($iResult == 3) {
                     echo 'No new updates. Sleeping for '.CONST_Replication_Recheck_Interval." sec.\n";


@@ -419,7 +378,8 @@ if ($aResult['import-osmosis'] || $aResult['import-osmosis-all']) {
             // get the newest object from the diff file
             $sBatchEnd = 0;
             $iRet = 0;
             // get the newest object from the diff file
             $sBatchEnd = 0;
             $iRet = 0;

-            exec(CONST_BasePath.'/utils/osm_file_date.py '.$sImportFile, $sBatchEnd, $iRet);
+            $oCMD = new \Nominatim\Shell(CONST_BasePath.'/utils/osm_file_date.py', $sImportFile);
+            exec($oCMD->escapedCmd(), $sBatchEnd, $iRet);

             if ($iRet == 5) {
                 echo "Diff file is empty. skipping import.\n";
                 if (!$aResult['import-osmosis-all']) {
             if ($iRet == 5) {
                 echo "Diff file is empty. skipping import.\n";
                 if (!$aResult['import-osmosis-all']) {


@@ -435,9 +395,11 @@ if ($aResult['import-osmosis'] || $aResult['import-osmosis-all']) {
 
             // Import the file
             $fCMDStartTime = time();
 
             // Import the file
             $fCMDStartTime = time();

-            echo $sCMDImport."\n";
+
+
+            echo $oCMDImport->escapedCmd()."\n";

             unset($sJunk);
             unset($sJunk);

-            $iErrorLevel = runWithEnv($sCMDImport, $aProcEnv);
+            $iErrorLevel = $oCMDImport->run();

             if ($iErrorLevel) {
                 echo "Error executing osm2pgsql: $iErrorLevel\n";
                 exit($iErrorLevel);
             if ($iErrorLevel) {
                 echo "Error executing osm2pgsql: $iErrorLevel\n";
                 exit($iErrorLevel);


@@ -462,11 +424,20 @@ if ($aResult['import-osmosis'] || $aResult['import-osmosis-all']) {
 
         // Index file
         if (!$aResult['no-index']) {
 
         // Index file
         if (!$aResult['no-index']) {

-            $sThisIndexCmd = $sCMDIndex;

             $fCMDStartTime = time();
 
             $fCMDStartTime = time();
 

-            echo "$sThisIndexCmd\n";
-            $iErrorLevel = runWithEnv($sThisIndexCmd, $aProcEnv);
+            $oThisIndexCmd = clone($oIndexCmd);
+            $oThisIndexCmd->addParams('-b');
+            echo $oThisIndexCmd->escapedCmd()."\n";
+            $iErrorLevel = $oThisIndexCmd->run();
+            if ($iErrorLevel) {
+                echo "Error: $iErrorLevel\n";
+                exit($iErrorLevel);
+            }
+
+            $oThisIndexCmd = clone($oIndexCmd);
+            echo $oThisIndexCmd->escapedCmd()."\n";
+            $iErrorLevel = $oThisIndexCmd->run();

             if ($iErrorLevel) {
                 echo "Error: $iErrorLevel\n";
                 exit($iErrorLevel);
             if ($iErrorLevel) {
                 echo "Error: $iErrorLevel\n";
                 exit($iErrorLevel);