import numpy as np
 from Bio import Medline, Entrez # 一般是通過BioPython的Bio.Entrez模塊訪問Entrez
 from collections import Counter
 
 Entrez.email = "（此處寫你自己在官網(wǎng)注冊(cè)的郵箱賬號(hào)）" # 應(yīng)用自己的賬號(hào)訪問NCBI數(shù)據(jù)庫
 
 # 此處需將服務(wù)器協(xié)議指定為1.0，否則會(huì)出現(xiàn)報(bào)錯(cuò)。http.client.IncompleteRead: IncompleteRead(0 bytes read)
 # 服務(wù)器http協(xié)議1.0，而python的是1.1，解決辦法就是指定客戶端http協(xié)議版本
 import http.client
 http.client.HTTPConnection._http_vsn = 10
 http.client.HTTPConnection._http_vsn_str = 'HTTP/1.0'
 
 """
 Entrez 是一個(gè)檢索系統(tǒng)，可以用其訪問NCBI數(shù)據(jù)庫，比如說PubMed，GenBank，GEO等。
 獲得有關(guān) global PBDE 的所有文獻(xiàn)的PubMed IDs
 """
 # handle_0 = Entrez.esearch(db="pubmed", term="drug therapy[Subheading] AND adverse effects[Subheading] AND humans[MeSH Terms]", retmax=306431)
 handle_0 = Entrez.esearch(db="pubmed", term="drug therapy[MeSH Subheading] AND adverse effects[MeSH Subheading] AND humans[MeSH Terms] AND (2000/01/01[Date - Publication] : 2021/12/31[Date - Publication])",
              ptyp="Review", usehistory="y", retmax=306431)
 record = Entrez.read(handle_0) # 獲取檢索條件的所有文獻(xiàn)
 idlist = record["IdList"] # 提取出文獻(xiàn)id
 print ("Total: ", record["Count"])
 No_Papers = len(idlist)  # 共306431篇文獻(xiàn) 2000-01-01:2021-12-31
 webenv = record['WebEnv']
 query_key = record['QueryKey']
 
 total = No_Papers
 step = 1300
 print("Result items:", total)
 with open("./Data_PubMed/PBDE1.txt", 'w') as f:
   for start in range(0, total, step):
     print("Download record %i to %i" % (start + 1, int(start + step)))
     handle_1 = Entrez.efetch(db="pubmed", retstart=start, rettype="medline", retmode="text",
                 retmax=step, webenv=webenv, query_key=query_key) # 獲取上述所有文獻(xiàn)的PubMed IDs
     records = Medline.parse(handle_1)
     records = list(records) # 將迭代器轉(zhuǎn)換至列表（list）
     for index in np.arange(len(records)):
       id = records[index].get("PMID", "?")
       title = records[index].get("TI", "?")
       title = title.replace('[', '').replace('].', '') # 若提取的標(biāo)題出現(xiàn)[].符號(hào)，則去除
       abstract = records[index].get("AB", "?")
       f.write(id)
       f.write("\n")
       f.write(title)
       f.write("\n")
       f.write(abstract)
       f.write("\n")