forked from ostis-apps/data2sc-parser
-
Notifications
You must be signed in to change notification settings - Fork 0
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
perf(loader): add loading in threads
- Loading branch information
1 parent
75b3c46
commit f99fc2e
Showing
1 changed file
with
68 additions
and
46 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -1,62 +1,84 @@ | ||
from .WikiDataLoader import WikiDataLoader | ||
import wptools | ||
import re | ||
import threading | ||
from queue import Queue | ||
|
||
|
||
class WikiDataWithContextLoader(WikiDataLoader): | ||
def __init__(self): | ||
super().__init__() | ||
self._lock = threading.Lock() | ||
|
||
def load_object(self, obj, obj_type): | ||
page = wptools.page(wikibase=obj, skip=['labels'], silent=True) | ||
de_page = wptools.page(wikibase=obj, lang='de', skip=[ | ||
'labels', 'imageinfo'], silent=True) | ||
ru_page = wptools.page(wikibase=obj, lang='ru', skip=[ | ||
'labels', 'imageinfo'], silent=True) | ||
|
||
page.get_wikidata() | ||
de_page.get_wikidata() | ||
ru_page.get_wikidata() | ||
|
||
try: | ||
obj_id = re.sub(r"\+|-|–|\/|:|\s", '_', | ||
re.sub(r"'s?|\(|\)|,", '', page.data['title'])) | ||
except KeyError: | ||
obj_id = obj | ||
|
||
with self._lock: | ||
self._info[obj_type][obj] = { | ||
'identifier': obj_id, | ||
'label': {'en': page.data['label'], 'de': de_page.data['label'], 'ru': ru_page.data['label']}, | ||
'description': {'en': page.data['description'], 'de': de_page.data['description'], 'ru': ru_page.data['description']} | ||
} | ||
self.add_image_url(page, obj_type, obj) | ||
|
||
def thread_fun(self, queue): | ||
while True: | ||
obj, obj_type = queue.get() | ||
self.load_object(obj, obj_type) | ||
queue.task_done() | ||
|
||
def resolve_ids(self): | ||
for triplet in self._info['triplets']: | ||
_, rlt, ent = triplet | ||
triplet[1] = self._info['relations'][rlt]['identifier'] | ||
triplet[2] = self._info['entities'][ent]['identifier'] | ||
|
||
temp_ent = self._info['entities'] | ||
self._info['entities'] = {} | ||
for ent in temp_ent: | ||
self._info['entities'][temp_ent[ent]['identifier']] = temp_ent[ent] | ||
|
||
temp_rlt = self._info['relations'] | ||
self._info['relations'] = {} | ||
for rlt in temp_rlt: | ||
self._info['relations'][temp_rlt[rlt] | ||
['identifier']] = temp_rlt[rlt] | ||
|
||
def getEntity(self, entity, lang='en'): | ||
super().getEntity(entity, lang=lang) | ||
self._page = wptools.page(wikibase=self._page.data['wikibase'], lang='en', skip=[ | ||
'labels', 'imageinfo'], silent=True) | ||
self._page.get_wikidata() | ||
page_title = re.sub(r"\+|-|–|\/|:|\s", '_', | ||
re.sub(r"'s?|\(|\)|,", '', self._page.data['title'])) | ||
loading_queue = Queue() | ||
|
||
for _ in range(16): | ||
threading.Thread(target=self.thread_fun, args=[ | ||
loading_queue, ], daemon=True).start() | ||
|
||
context = self._page.data['claims'] | ||
for rlt, ents in context.items(): | ||
rlt_page = wptools.page( | ||
wikibase=rlt, skip=['labels', 'imageinfo'], silent=True) | ||
rlt_page.get_wikidata() | ||
rlt_de_page = wptools.page(wikibase=rlt, lang='de', skip=[ | ||
'labels', 'imageinfo'], silent=True) | ||
rlt_de_page.get_wikidata() | ||
rlt_ru_page = wptools.page(wikibase=rlt, lang='ru', skip=[ | ||
'labels', 'imageinfo'], silent=True) | ||
rlt_ru_page.get_wikidata() | ||
if 'title' in rlt_page.data: | ||
rlt_id = re.sub( | ||
r"\+|-|–|\/|:|\s", '_', re.sub(r"'s?|\(|\)|,", '', rlt_page.data['title'])) | ||
else: | ||
continue | ||
|
||
self._info['relations'][rlt_id] = { | ||
'identifier': rlt_id, | ||
'label': {'en': rlt_page.data['label'], 'de': rlt_de_page.data['label'], 'ru': rlt_ru_page.data['label']}, | ||
'description': {'en': rlt_page.data['description'], 'de': rlt_de_page.data['description'], 'ru': rlt_ru_page.data['description']} | ||
} | ||
|
||
loading_queue.put([rlt, 'relations']) | ||
for ent in ents: | ||
if type(ent) is str: | ||
if ent.startswith('Q'): | ||
ent_page = wptools.page( | ||
wikibase=ent, skip=['labels'], silent=True) | ||
ent_page.get_wikidata() | ||
ent_de_page = wptools.page(wikibase=ent, lang='de', skip=[ | ||
'labels', 'imageinfo'], silent=True) | ||
ent_de_page.get_wikidata() | ||
ent_ru_page = wptools.page(wikibase=ent, lang='ru', skip=[ | ||
'labels', 'imageinfo'], silent=True) | ||
ent_ru_page.get_wikidata() | ||
if 'title' in ent_page.data: | ||
ent_id = re.sub( | ||
r"\+|-|–|\/|:|\s", '_', re.sub(r"'s?|\(|\)|,", '', ent_page.data['title'])) | ||
else: | ||
continue | ||
|
||
self._info['entities'][ent_id] = { | ||
'identifier': ent_id, | ||
'label': {'en': ent_page.data['label'], 'de': ent_de_page.data['label'], 'ru': ent_ru_page.data['label']}, | ||
'description': {'en': ent_page.data['description'], 'de': ent_de_page.data['description'], 'ru': ent_ru_page.data['description']} | ||
} | ||
|
||
self.add_image_url(ent_page, 'entities', ent_id) | ||
|
||
self._info['triplets'].append( | ||
[self._page.data['title'], rlt_id, ent_id]) | ||
loading_queue.put([ent, 'entities']) | ||
self._info['triplets'].append([page_title, rlt, ent]) | ||
|
||
loading_queue.join() | ||
self.resolve_ids() |