boilerplate for PHP code of tokenizer

[nominatim.git] / nominatim / tokenizer / legacy_tokenizer.py

diff --git a/nominatim/tokenizer/legacy_tokenizer.py b/nominatim/tokenizer/legacy_tokenizer.py
index ac26c5cd5d160257ad66f30fe289d6bd9b791e12..d6755835cae40fb496c287b44f15166917040bf5 100644 (file)
--- a/nominatim/tokenizer/legacy_tokenizer.py
+++ b/nominatim/tokenizer/legacy_tokenizer.py
@@ -5,6 +5,7 @@ from collections import OrderedDict
 import logging
 import re
 import shutil
 import logging
 import re
 import shutil

+from textwrap import dedent

 
 from icu import Transliterator
 import psycopg2
 
 from icu import Transliterator
 import psycopg2


@@ -87,7 +88,7 @@ class LegacyTokenizer:
         self.normalization = None
 
 
         self.normalization = None
 
 

-    def init_new_db(self, config):
+    def init_new_db(self, config, init_db=True):

         """ Set up a new tokenizer for the database.
 
             This copies all necessary data in the project directory to make
         """ Set up a new tokenizer for the database.
 
             This copies all necessary data in the project directory to make


@@ -99,13 +100,16 @@ class LegacyTokenizer:
 
         self.normalization = config.TERM_NORMALIZATION
 
 
         self.normalization = config.TERM_NORMALIZATION
 

+        self._install_php(config)
+

         with connect(self.dsn) as conn:
             _check_module(module_dir, conn)
             self._save_config(conn, config)
             conn.commit()
 
         with connect(self.dsn) as conn:
             _check_module(module_dir, conn)
             self._save_config(conn, config)
             conn.commit()
 

-        self.update_sql_functions(config)
-        self._init_db_tables(config)
+        if init_db:
+            self.update_sql_functions(config)
+            self._init_db_tables(config)

 
 
     def init_from_project(self):
 
 
     def init_from_project(self):


@@ -165,6 +169,18 @@ class LegacyTokenizer:
         return LegacyNameAnalyzer(self.dsn, normalizer)
 
 
         return LegacyNameAnalyzer(self.dsn, normalizer)
 
 

+    def _install_php(self, config):
+        """ Install the php script for the tokenizer.
+        """
+        php_file = self.data_dir / "tokenizer.php"
+        php_file.write_text(dedent("""\
+            <?php
+            @define('CONST_Max_Word_Frequency', {0.MAX_WORD_FREQUENCY});
+            @define('CONST_Term_Normalization_Rules', "{0.TERM_NORMALIZATION}");
+            require_once('{0.lib_dir.php}/tokenizer/legacy_tokenizer.php');
+            """.format(config)))
+
+

     def _init_db_tables(self, config):
         """ Set up the word table and fill it with pre-computed word
             frequencies.
     def _init_db_tables(self, config):
         """ Set up the word table and fill it with pre-computed word
             frequencies.