AI/ragflow
1
0
Fork 0
mirror of https://git.mirrors.martin98.com/https://github.com/infiniflow/ragflow.git synced 2025-08-13 22:46:07 +08:00
Code Issues Packages Projects Releases Wiki Activity

Set wikipedia lang (#1515)

Browse Source 
### What problem does this PR solve?



### Type of change

- [x] New Feature (non-breaking change which adds functionality)
This commit is contained in:
H 2024-07-15 17:45:40 +08:00 committed by GitHub
parent 1a6bd437f5
commit 60d406acaa
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 11 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
graph/component/wikipedia.py
View File

@ -30,9 +30,16 @@ class WikipediaParam(ComponentParamBase):
def __init__(self): def __init__(self):
super().__init__() super().__init__()
self.top_n = 10 self.top_n = 10
self.lang = 'en'
def check(self): def check(self):
self.check_positive_integer(self.top_n, "Top N") self.check_positive_integer(self.top_n, "Top N")
self.check_valid_value(self.lang, "Wikipedia languages",
['af', 'pl', 'ar', 'ast', 'az', 'bg', 'nan', 'bn', 'be', 'ca', 'cs', 'cy', 'da', 'de',
'et', 'el', 'en', 'es', 'eo', 'eu', 'fa', 'fr', 'gl', 'ko', 'hy', 'hi', 'hr', 'id',
'it', 'he', 'ka', 'lld', 'la', 'lv', 'lt', 'hu', 'mk', 'arz', 'ms', 'min', 'my', 'nl',
'ja', 'nb', 'nn', 'ce', 'uz', 'pt', 'kk', 'ro', 'ru', 'ceb', 'sk', 'sl', 'sr', 'sh',
'fi', 'sv', 'ta', 'tt', 'th', 'tg', 'azb', 'tr', 'uk', 'ur', 'vi', 'war', 'zh', 'yue'])
class Wikipedia(ComponentBase, ABC): class Wikipedia(ComponentBase, ABC):
@ -45,9 +52,11 @@ class Wikipedia(ComponentBase, ABC):
return Wikipedia.be_output(self._param.no) return Wikipedia.be_output(self._param.no)
wiki_res = [] wiki_res = []
for wiki_key in wikipedia.search(ans, results=self._param.top_n): wikipedia.set_lang(self._param.lang)
wiki_engine = wikipedia
for wiki_key in wiki_engine.search(ans, results=self._param.top_n):
try: try:
page = wikipedia.page(title=wiki_key, auto_suggest=False) page = wiki_engine.page(title=wiki_key, auto_suggest=False)
wiki_res.append({"content": '<a href="' + page.url + '">' + page.title + '</a> ' + page.summary}) wiki_res.append({"content": '<a href="' + page.url + '">' + page.title + '</a> ' + page.summary})
except Exception as e: except Exception as e:
print(e) print(e)