hans/Nominatim
1
0
Fork 1
mirror of https://github.com/osm-search/Nominatim.git synced 2026-02-26 11:08:13 +00:00
Code Issues Packages Projects Releases Wiki Activity

add PHP part for new ICU-base tokenizer

Browse Source
This commit is contained in:
Sarah Hoffmann
2021-05-02 21:21:41 +02:00
parent f44af49df9
commit ba8ed7967d
2 changed files with 244 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

9
nominatim/tokenizer/legacy_icu_tokenizer.py
View File

@@ -133,15 +133,18 @@ class LegacyICUTokenizer:
def _install_php(self, config):
""" Install the php script for the tokenizer.
"""
abbr_inverse = list(zip(*self.abbreviations))
php_file = self.data_dir / "tokenizer.php"
php_file.write_text(dedent("""\
<?php
@define('CONST_Max_Word_Frequency', {1.MAX_WORD_FREQUENCY});
@define('CONST_Term_Normalization_Rules', "{0.normalization}");
@define('CONST_Transliteration'. "{0.transliteration}");
# XXX abreviations
@define('CONST_Transliteration', "{0.transliteration}");
@define('CONST_Abbreviations', array(array('{2}'), array('{3}')));
require_once('{1.lib_dir.php}/tokenizer/legacy_icu_tokenizer.php');
""".format(self, config)))
""".format(self, config,
"','".join(abbr_inverse[0]),
"','".join(abbr_inverse[1]))))
def _save_config(self, config):