AI/ragflow
1
0
Fork 0
mirror of https://git.mirrors.martin98.com/https://github.com/infiniflow/ragflow.git synced 2025-08-12 02:19:05 +08:00
Code Issues Packages Projects Releases Wiki Activity

Fix component PubMed (#2192)

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


### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)
This commit is contained in:
H 2024-09-02 16:19:41 +08:00 committed by GitHub
parent 922f79e757
commit ab9efb3c23
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 13 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

18
agent/component/pubmed.py
View File

@ -49,11 +49,19 @@ class PubMed(ComponentBase, ABC):
pubmedids = Entrez.read(Entrez.esearch(db='pubmed', retmax=self._param.top_n, term=ans))['IdList']
pubmedcnt = ET.fromstring(
Entrez.efetch(db='pubmed', id=",".join(pubmedids), retmode="xml").read().decode("utf-8"))
pubmed_res = [{"content": 'Title:' + child.find("MedlineCitation").find("Article").find(
"ArticleTitle").text + '\nUrl:<a href=" https://pubmed.ncbi.nlm.nih.gov/' + child.find(
"MedlineCitation").find("PMID").text + '">' + '</a>\n' + 'Abstract:' + child.find(
"MedlineCitation").find("Article").find("Abstract").find("AbstractText").text} for child in
pubmedcnt.findall("PubmedArticle")]
pubmed_res = []
for child in pubmedcnt.findall("PubmedArticle"):
if child.find("MedlineCitation").find("Article").find("ArticleTitle").text:
title_tmp = 'Title:' + child.find("MedlineCitation").find("Article").find("ArticleTitle").text
else:
title_tmp = 'Title:' + "".join(
[childtitle.text for childtitle in
child.find("MedlineCitation").find("Article").find("ArticleTitle")])
url_tmp = '\nUrl:<a href=" https://pubmed.ncbi.nlm.nih.gov/' + child.find("MedlineCitation").find(
"PMID").text + '">' + '</a>'
abstract_tmp = '\nAbstract:' + child.find("MedlineCitation").find("Article").find("Abstract").find(
"AbstractText").text
pubmed_res.append({"content": title_tmp + url_tmp + abstract_tmp})
except Exception as e:
return PubMed.be_output("**ERROR**: " + str(e))