Spaces:
Sleeping
Sleeping
| # genesis/api_clients/pubmed_api.py | |
| import requests | |
| import xml.etree.ElementTree as ET | |
| from typing import List, Dict, Optional | |
| from datetime import datetime | |
| PUBMED_SEARCH_URL = "https://eutils.ncbi.nlm.nih.gov/entrez/eutils/esearch.fcgi" | |
| PUBMED_FETCH_URL = "https://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi" | |
| NCBI_API_KEY = None # Optional: Set in Hugging Face secrets for higher rate limits | |
| def search_pubmed(query: str, max_results: int = 20, start_date: Optional[str] = None, end_date: Optional[str] = None) -> List[str]: | |
| """ | |
| Search PubMed for a given query and return a list of PubMed IDs. | |
| Optionally filter by start_date and end_date (YYYY/MM/DD format). | |
| """ | |
| params = { | |
| "db": "pubmed", | |
| "term": query, | |
| "retmax": max_results, | |
| "retmode": "json", | |
| "api_key": NCBI_API_KEY | |
| } | |
| if start_date and end_date: | |
| params["mindate"] = start_date | |
| params["maxdate"] = end_date | |
| params["datetype"] = "pdat" | |
| r = requests.get(PUBMED_SEARCH_URL, params=params) | |
| r.raise_for_status() | |
| data = r.json() | |
| return data.get("esearchresult", {}).get("idlist", []) | |
| def fetch_pubmed_details(pmid_list: List[str]) -> List[Dict]: | |
| """ | |
| Fetch detailed metadata for a list of PubMed IDs. | |
| Returns title, abstract, authors, journal, and publication date. | |
| """ | |
| if not pmid_list: | |
| return [] | |
| params = { | |
| "db": "pubmed", | |
| "id": ",".join(pmid_list), | |
| "retmode": "xml", | |
| "api_key": NCBI_API_KEY | |
| } | |
| r = requests.get(PUBMED_FETCH_URL, params=params) | |
| r.raise_for_status() | |
| root = ET.fromstring(r.text) | |
| results = [] | |
| for article in root.findall(".//PubmedArticle"): | |
| try: | |
| title = article.find(".//ArticleTitle").text or "No title" | |
| abstract = " ".join([t.text for t in article.findall(".//AbstractText") if t.text]) or "No abstract" | |
| authors = [] | |
| for a in article.findall(".//Author"): | |
| last = a.findtext("LastName", "") | |
| first = a.findtext("ForeName", "") | |
| if last or first: | |
| authors.append(f"{first} {last}".strip()) | |
| journal = article.findtext(".//Journal/Title", "Unknown Journal") | |
| pub_date = article.find(".//PubDate") | |
| if pub_date is not None: | |
| year = pub_date.findtext("Year", "") | |
| month = pub_date.findtext("Month", "") | |
| day = pub_date.findtext("Day", "") | |
| date_str = f"{year}-{month}-{day}" if year else "Unknown" | |
| else: | |
| date_str = "Unknown" | |
| results.append({ | |
| "title": title, | |
| "abstract": abstract, | |
| "authors": authors, | |
| "journal": journal, | |
| "publication_date": date_str, | |
| "pubmed_link": f"https://pubmed.ncbi.nlm.nih.gov/{article.findtext('.//PMID')}/" | |
| }) | |
| except Exception: | |
| continue | |
| return results | |
| def search_and_fetch_pubmed(query: str, max_results: int = 20, start_date: Optional[str] = None, end_date: Optional[str] = None) -> List[Dict]: | |
| """ | |
| Search and fetch PubMed results in one call. | |
| """ | |
| pmids = search_pubmed(query, max_results, start_date, end_date) | |
| return fetch_pubmed_details(pmids) | |