-#!@PHP_BIN@ -Cq
<?php
-require_once(dirname(dirname(__FILE__)).'/settings/settings.php');
require_once(CONST_BasePath.'/lib/init-cmd.php');
require_once(CONST_BasePath.'/lib/setup_functions.php');
require_once(CONST_BasePath.'/lib/setup/SetupClass.php');
+require_once(CONST_BasePath.'/lib/setup/AddressLevelParser.php');
ini_set('memory_limit', '800M');
// (long-opt, short-opt, min-occurs, max-occurs, num-arguments, num-arguments, type, help)
$aCMDOptions
= array(
- 'Import / update / index osm data',
- array('help', 'h', 0, 1, 0, 0, false, 'Show Help'),
- array('quiet', 'q', 0, 1, 0, 0, 'bool', 'Quiet output'),
- array('verbose', 'v', 0, 1, 0, 0, 'bool', 'Verbose output'),
-
- array('init-updates', '', 0, 1, 0, 0, 'bool', 'Set up database for updating'),
- array('check-for-updates', '', 0, 1, 0, 0, 'bool', 'Check if new updates are available'),
- array('no-update-functions', '', 0, 1, 0, 0, 'bool', 'Do not update trigger functions to support differential updates (assuming the diff update logic is already present)'),
- array('import-osmosis', '', 0, 1, 0, 0, 'bool', 'Import updates once'),
- array('import-osmosis-all', '', 0, 1, 0, 0, 'bool', 'Import updates forever'),
- array('no-index', '', 0, 1, 0, 0, 'bool', 'Do not index the new data'),
-
- array('calculate-postcodes', '', 0, 1, 0, 0, 'bool', 'Update postcode centroid table'),
-
- array('import-file', '', 0, 1, 1, 1, 'realpath', 'Re-import data from an OSM file'),
- array('import-diff', '', 0, 1, 1, 1, 'realpath', 'Import a diff (osc) file from local file system'),
- array('osm2pgsql-cache', '', 0, 1, 1, 1, 'int', 'Cache size used by osm2pgsql'),
-
- array('import-node', '', 0, 1, 1, 1, 'int', 'Re-import node'),
- array('import-way', '', 0, 1, 1, 1, 'int', 'Re-import way'),
- array('import-relation', '', 0, 1, 1, 1, 'int', 'Re-import relation'),
- array('import-from-main-api', '', 0, 1, 0, 0, 'bool', 'Use OSM API instead of Overpass to download objects'),
-
- array('index', '', 0, 1, 0, 0, 'bool', 'Index'),
- 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('no-npi', '', 0, 1, 0, 0, 'bool', '(obsolete)'),
-);
+ 'Import / update / index osm data',
+ array('help', 'h', 0, 1, 0, 0, false, 'Show Help'),
+ array('quiet', 'q', 0, 1, 0, 0, 'bool', 'Quiet output'),
+ array('verbose', 'v', 0, 1, 0, 0, 'bool', 'Verbose output'),
+
+ array('init-updates', '', 0, 1, 0, 0, 'bool', 'Set up database for updating'),
+ array('check-for-updates', '', 0, 1, 0, 0, 'bool', 'Check if new updates are available'),
+ array('no-update-functions', '', 0, 1, 0, 0, 'bool', 'Do not update trigger functions to support differential updates (assuming the diff update logic is already present)'),
+ array('import-osmosis', '', 0, 1, 0, 0, 'bool', 'Import updates once'),
+ array('import-osmosis-all', '', 0, 1, 0, 0, 'bool', 'Import updates forever'),
+ array('no-index', '', 0, 1, 0, 0, 'bool', 'Do not index the new data'),
+
+ array('calculate-postcodes', '', 0, 1, 0, 0, 'bool', 'Update postcode centroid table'),
+
+ array('import-file', '', 0, 1, 1, 1, 'realpath', 'Re-import data from an OSM file'),
+ array('import-diff', '', 0, 1, 1, 1, 'realpath', 'Import a diff (osc) file from local file system'),
+ array('osm2pgsql-cache', '', 0, 1, 1, 1, 'int', 'Cache size used by osm2pgsql'),
+
+ array('import-node', '', 0, 1, 1, 1, 'int', 'Re-import node'),
+ array('import-way', '', 0, 1, 1, 1, 'int', 'Re-import way'),
+ array('import-relation', '', 0, 1, 1, 1, 'int', 'Re-import relation'),
+ array('import-from-main-api', '', 0, 1, 0, 0, 'bool', 'Use OSM API instead of Overpass to download objects'),
+
+ array('index', '', 0, 1, 0, 0, 'bool', 'Index'),
+ 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-importance', '', 0, 1, 0, 0, 'bool', 'Recompute place importances'),
+ array('no-npi', '', 0, 1, 0, 0, 'bool', '(obsolete)'),
+ );
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 =& getDB();
+$oDB = new Nominatim\DB();
+$oDB->connect();
-$aDSNInfo = DB::parseDSN(CONST_Database_DSN);
+$aDSNInfo = Nominatim\DB::parseDSN(CONST_Database_DSN);
if (!isset($aDSNInfo['port']) || !$aDSNInfo['port']) $aDSNInfo['port'] = 5432;
// cache memory to be used by osm2pgsql, should not be more than the available memory
$iCacheMemory = getCacheMemoryMB();
echo "WARNING: resetting cache memory to $iCacheMemory\n";
}
-$sOsm2pgsqlCmd = CONST_Osm2pgsql_Binary.' -klas --number-processes 1 -C '.$iCacheMemory.' -O gazetteer -d '.$aDSNInfo['database'].' -P '.$aDSNInfo['port'];
+$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'];
}
$sOsm2pgsqlCmd .= ' --flat-nodes '.CONST_Osm2pgsql_Flatnode_File;
}
+$sIndexCmd = CONST_BasePath.'/nominatim/nominatim.py';
+if (!$aResult['quiet']) {
+ $sIndexCmd .= ' -v';
+}
+if ($aResult['verbose']) {
+ $sIndexCmd .= ' -v';
+}
+
if ($aResult['init-updates']) {
// sanity check that the replication URL is correct
$sBaseState = file_get_contents(CONST_Replication_Url.'/state.txt');
if (!$aResult['no-update-functions']) {
// instantiate setupClass to use the function therein
- $cSetup = new SetupFunctions('update');
+ $cSetup = new SetupFunctions(array(
+ 'enable-diff-updates' => true,
+ 'verbose' => $aResult['verbose']
+ ));
+ $cSetup->connect();
$cSetup->createFunctions();
}
$sDatabaseDate = getDatabaseDate($oDB);
- if ($sDatabaseDate === false) {
+ if (!$sDatabaseDate) {
fail('Cannot determine date of database.');
}
$sWindBack = strftime('%Y-%m-%dT%H:%M:%SZ', strtotime($sDatabaseDate) - (3*60*60));
fail('Error running pyosmium tools');
}
- pg_query($oDB->connection, 'TRUNCATE import_status');
+ $oDB->exec('TRUNCATE import_status');
$sSQL = "INSERT INTO import_status (lastimportdate, sequence_id, indexed) VALUES('";
$sSQL .= $sDatabaseDate."',".$aOutput[0].', true)';
- if (!pg_query($oDB->connection, $sSQL)) {
+
+ try {
+ $oDB->exec($sSQL);
+ } catch (\Nominatim\DatabaseError $e) {
fail('Could not enter sequence into database.');
}
}
if ($aResult['check-for-updates']) {
- $aLastState = chksql($oDB->getRow('SELECT sequence_id FROM import_status'));
+ $aLastState = $oDB->getRow('SELECT sequence_id FROM import_status');
if (!$aLastState['sequence_id']) {
fail('Updates not set up. Please run ./utils/update.php --init-updates.');
}
if ($aResult['deduplicate']) {
- $oDB =& getDB();
+ $oDB = new Nominatim\DB();
+ $oDB->connect();
- if (getPostgresVersion($oDB) < 9.3) {
+ 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';
- $aPartitions = chksql($oDB->getCol($sSQL));
+ $aPartitions = $oDB->getCol($sSQL);
$aPartitions[] = 0;
// we don't care about empty search_name_* partitions, they can't contain mentions of duplicates
foreach ($aPartitions as $i => $sPartition) {
$sSQL = 'select count(*) from search_name_'.$sPartition;
- $nEntries = chksql($oDB->getOne($sSQL));
+ $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';
- $aDuplicateTokens = chksql($oDB->getAll($sSQL));
+ $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 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";
- $aTokenSet = chksql($oDB->getAll($sSQL));
+ $aTokenSet = $oDB->getAll($sSQL);
$aKeep = array_shift($aTokenSet);
$iKeepID = $aKeep['word_id'];
$sSQL .= ' name_vector = array_replace(name_vector,'.$aRemove['word_id'].','.$iKeepID.'),';
$sSQL .= ' nameaddress_vector = array_replace(nameaddress_vector,'.$aRemove['word_id'].','.$iKeepID.')';
$sSQL .= ' where name_vector @> ARRAY['.$aRemove['word_id'].']';
- chksql($oDB->query($sSQL));
+ $oDB->exec($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'].']';
- chksql($oDB->query($sSQL));
+ $oDB->exec($sSQL);
$sSQL = 'update location_area_country set';
$sSQL .= ' keywords = array_replace(keywords,'.$aRemove['word_id'].','.$iKeepID.')';
$sSQL .= ' where keywords @> ARRAY['.$aRemove['word_id'].']';
- chksql($oDB->query($sSQL));
+ $oDB->exec($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'].']';
- chksql($oDB->query($sSQL));
+ $oDB->exec($sSQL);
$sSQL = 'update location_area_country set';
$sSQL .= ' keywords = array_replace(keywords,'.$aRemove['word_id'].','.$iKeepID.')';
$sSQL .= ' where keywords @> ARRAY['.$aRemove['word_id'].']';
- chksql($oDB->query($sSQL));
+ $oDB->exec($sSQL);
}
$sSQL = 'delete from word where word_id = '.$aRemove['word_id'];
- chksql($oDB->query($sSQL));
+ $oDB->exec($sSQL);
}
}
}
}
if ($aResult['index']) {
- $sCmd = CONST_InstallPath.'/nominatim/nominatim -i -d '.$aDSNInfo['database'].' -P '.$aDSNInfo['port'].' -t '.$aResult['index-instances'].' -r '.$aResult['index-rank'];
+ $sCmd = $sIndexCmd
+ .' -d '.$aDSNInfo['database']
+ .' -P '.$aDSNInfo['port']
+ .' -t '.$aResult['index-instances']
+ .' -r '.$aResult['index-rank'];
if (isset($aDSNInfo['hostspec']) && $aDSNInfo['hostspec']) {
$sCmd .= ' -H ' . $aDSNInfo['hostspec'];
}
}
runWithEnv($sCmd, $aProcEnv);
+
+ $oDB->exec('update import_status set indexed = true');
+}
+
+if ($aResult['update-address-levels']) {
+ echo 'Updating address levels from '.CONST_Address_Level_Config.".\n";
+ $oAlParser = new \Nominatim\Setup\AddressLevelParser(CONST_Address_Level_Config);
+ $oAlParser->createTable($oDB, 'address_levels');
+}
+
+if ($aResult['recompute-importance']) {
+ echo "Updating importance values for database.\n";
+ $oDB = new Nominatim\DB();
+ $oDB->connect();
+
+ $sSQL = 'ALTER TABLE placex DISABLE TRIGGER ALL;';
+ $sSQL .= 'UPDATE placex SET (wikipedia, importance) =';
+ $sSQL .= ' (SELECT wikipedia, importance';
+ $sSQL .= ' FROM compute_importance(extratags, country_code, osm_type, osm_id));';
+ $sSQL .= 'UPDATE placex s SET wikipedia = d.wikipedia, importance = d.importance';
+ $sSQL .= ' FROM placex d';
+ $sSQL .= ' WHERE s.place_id = d.linked_place_id and d.wikipedia is not null';
+ $sSQL .= ' and (s.wikipedia is null or s.importance < d.importance);';
+ $sSQL .= 'ALTER TABLE placex ENABLE TRIGGER ALL;';
+ $oDB->exec($sSQL);
}
if ($aResult['import-osmosis'] || $aResult['import-osmosis-all']) {
$sImportFile = CONST_InstallPath.'/osmosischange.osc';
$sCMDDownload = CONST_Pyosmium_Binary.' --server '.CONST_Replication_Url.' -o '.$sImportFile.' -s '.CONST_Replication_Max_Diff_size;
$sCMDImport = $sOsm2pgsqlCmd.' '.$sImportFile;
- $sCMDIndex = CONST_InstallPath.'/nominatim/nominatim -i -d '.$aDSNInfo['database'].' -P '.$aDSNInfo['port'].' -t '.$aResult['index-instances'];
+ $sCMDIndex = $sIndexCmd
+ .' -d '.$aDSNInfo['database']
+ .' -P '.$aDSNInfo['port']
+ .' -t '.$aResult['index-instances'];
if (isset($aDSNInfo['hostspec']) && $aDSNInfo['hostspec']) {
$sCMDIndex .= ' -H ' . $aDSNInfo['hostspec'];
}
while (true) {
$fStartTime = time();
- $aLastState = chksql($oDB->getRow('SELECT *, EXTRACT (EPOCH FROM lastimportdate) as unix_ts FROM import_status'));
+ $aLastState = $oDB->getRow('SELECT *, EXTRACT (EPOCH FROM lastimportdate) as unix_ts FROM import_status');
if (!$aLastState['sequence_id']) {
echo "Updates not set up. Please run ./utils/update.php --init-updates.\n";
$sBatchEnd = $aLastState['lastimportdate'];
$iEndSequence = $aLastState['sequence_id'];
- if ($aLastState['indexed'] == 't') {
+ if ($aLastState['indexed']) {
// Sleep if the update interval has not yet been reached.
$fNextUpdate = $aLastState['unix_ts'] + CONST_Replication_Update_Interval;
if ($fNextUpdate > $fStartTime) {
$sSQL .= date('Y-m-d H:i:s', $fCMDStartTime)."','";
$sSQL .= date('Y-m-d H:i:s')."','import')";
var_Dump($sSQL);
- chksql($oDB->query($sSQL));
+ $oDB->exec($sSQL);
// update the status
$sSQL = "UPDATE import_status SET lastimportdate = '$sBatchEnd', indexed=false, sequence_id = $iEndSequence";
var_Dump($sSQL);
- chksql($oDB->query($sSQL));
+ $oDB->exec($sSQL);
echo date('Y-m-d H:i:s')." Completed download step for $sBatchEnd in ".round((time()-$fCMDStartTime)/60, 2)." minutes\n";
}
$sSQL = 'INSERT INTO import_osmosis_log';
$sSQL .= '(batchend, batchseq, batchsize, starttime, endtime, event)';
- $sSQL .= " values ('$sBatchEnd',$iEndSequence,$iFileSize,'";
+ $sSQL .= " values ('$sBatchEnd',$iEndSequence,NULL,'";
$sSQL .= date('Y-m-d H:i:s', $fCMDStartTime)."','";
$sSQL .= date('Y-m-d H:i:s')."','index')";
var_Dump($sSQL);
- $oDB->query($sSQL);
+ $oDB->exec($sSQL);
echo date('Y-m-d H:i:s')." Completed index step for $sBatchEnd in ".round((time()-$fCMDStartTime)/60, 2)." minutes\n";
$sSQL = 'update import_status set indexed = true';
- $oDB->query($sSQL);
+ $oDB->exec($sSQL);
+ } else {
+ if ($aResult['import-osmosis-all']) {
+ echo "Error: --no-index cannot be used with continuous imports (--import-osmosis-all).\n";
+ exit(1);
+ }
}
$fDuration = time() - $fStartTime;
projects / nominatim.git / blobdiff