hans/Nominatim
1
0
Fork 1
mirror of https://github.com/osm-search/Nominatim.git synced 2026-02-14 10:27:57 +00:00
Code Issues Packages Projects Releases Wiki Activity

move parsing of mutation config to setup phase

Browse Source
This commit is contained in:
Sarah Hoffmann
2022-01-12 19:41:16 +01:00
parent 4a41bff3ab
commit adbaf700cd
1 changed files with 17 additions and 16 deletions
Download Patch File Download Diff File Expand all files Collapse all files

33
nominatim/tokenizer/token_analysis/generic.py
View File

@@ -25,7 +25,22 @@ def configure(rules, normalization_rules):
config['replacements'], config['chars'] = get_variant_config(rules.get('variants'),
normalization_rules)
config['variant_only'] = rules.get('mode', '') == 'variant-only'
config['mutations'] = rules.get('mutations', [])
# parse mutation rules
config['mutations'] = []
for rule in rules.get('mutations', []):
if 'pattern' not in rule:
raise UsageError("Missing field 'pattern' in mutation configuration.")
if not isinstance(rule['pattern'], str):
raise UsageError("Field 'pattern' in mutation configuration "
"must be a simple text field.")
if 'replacements' not in rule:
raise UsageError("Missing field 'replacements' in mutation configuration.")
if not isinstance(rule['replacements'], list):
raise UsageError("Field 'replacements' in mutation configuration "
"must be a list of texts.")
config['mutations'].append((rule['pattern'], rule['replacements']))
return config
@@ -56,21 +71,7 @@ class GenericTokenAnalysis:
self.replacements = None
# set up mutation rules
self.mutations = []
for cfg in config['mutations']:
if 'pattern' not in cfg:
raise UsageError("Missing field 'pattern' in mutation configuration.")
if not isinstance(cfg['pattern'], str):
raise UsageError("Field 'pattern' in mutation configuration "
"must be a simple text field.")
if 'replacements' not in cfg:
raise UsageError("Missing field 'replacements' in mutation configuration.")
if not isinstance(cfg['replacements'], list):
raise UsageError("Field 'replacements' in mutation configuration "
"must be a list of texts.")
self.mutations.append(MutationVariantGenerator(cfg['pattern'],
cfg['replacements']))
self.mutations = [MutationVariantGenerator(*cfg) for cfg in config['mutations']]
def get_variants_ascii(self, norm_name):