Source Code for Module rdkit.Dbase.Pubmed.Records

  1  # $Id: Records.py 997 2009-02-25 06:12:43Z glandrum $ 
  2  # 
  3  # Copyright (C) 2003-2006 Rational Discovery LLC 
  4  # 
  5  #   @@ All Rights Reserved  @@ 
  6  # 
  7  from xml.etree import ElementTree 
  8  # check the version of ElementTree.  We need at least version 1.2 
  9  # in order for the XPath-style parsing stuff to work 
 10  import re 
 11  vers = re.split("[a-zA-Z]",ElementTree.VERSION)[0] 
 12  if vers < '1.2': 
 13    raise ImportError,'The PubMed record interface requires a version of ElementTree >= 1.2' 
 14   
 15   
 16 -class Record(object): 
 17 -  def __init__(self,element): 
 18      for field in self._fieldsOfInterest: 
 19          setattr(self,field,'') 
 20      self._element = element 
 21 -  def toXML(self): 
 22      from cStringIO import StringIO 
 23      sio = StringIO() 
 24      ElementTree.ElementTree(self._element).write(sio) 
 25      return sio.getvalue() 
 26     
 27 -class SummaryRecord(Record): 
 28    _fieldsOfInterest=['PubMedId','PubDate','Source','Authors', 
 29                      'Title','Volume','Issue','Pages','Lang', 
 30                      'HasAbstract','RecordStatus'] 
 31 -  def __init__(self,element): 
 32      Record.__init__(self,element) 
 33      for item in element.getiterator('Item'): 
 34        if item.attrib['Name'] in self._fieldsOfInterest: 
 35          setattr(self,item.attrib['Name'],item.text) 
 36      if self.PubDate: 
 37        self.PubYear = str(self.PubDate).split(' ')[0] 
 38   
 39 -class JournalArticleRecord(Record): 
 40    _fieldsOfInterest=['PubMedId','PubYear','Source','Authors', 
 41                      'Title','Volume','Issue','Pages','Lang', 
 42                      'Abstract'] 
 43 -  def __init__(self,element): 
 44      Record.__init__(self,element) 
 45   
 46      cite = self._element.find('MedlineCitation') 
 47      self.PubMedId = cite.findtext('PMID') 
 48      article = cite.find('Article') 
 49      issue = article.find('Journal/JournalIssue') 
 50      self.Volume = issue.findtext('Volume') 
 51      self.Issue = issue.findtext('Issue') 
 52      self.PubYear = issue.findtext('PubDate/Year') 
 53      if not self.PubYear: 
 54        txt = issue.findtext('PubDate/MedlineDate') 
 55        self.PubYear = txt.split(' ')[0] 
 56      self.Title = unicode(article.findtext('ArticleTitle')) 
 57      self.Pages = article.findtext('Pagination/MedlinePgn') 
 58      abs = article.findtext('Abstract/AbstractText') 
 59      if abs: 
 60        self.Abstract = unicode(abs) 
 61   
 62      self.authors = [] 
 63      tmp = [] 
 64      for author in article.find('AuthorList').getiterator('Author'): 
 65        last = unicode(author.findtext('LastName')) 
 66        first = unicode(author.findtext('ForeName')) 
 67        initials =  unicode(author.findtext('Initials')) 
 68        self.authors.append((last,first,initials)) 
 69        tmp.append('%s %s'%(last,initials)) 
 70      self.Authors=', '.join(tmp) 
 71      journal = cite.findtext('MedlineJournalInfo/MedlineTA') 
 72      if journal: 
 73        self.Source = unicode(journal) 
 74   
 75      self.ParseKeywords() 
 76      self.ParseChemicals() 
 77   
 78 -  def ParseKeywords(self): 
 79      self.keywords = [] 
 80      headings = self.find('MedlineCitation/MeshHeadingList') 
 81      if headings: 
 82        for heading in headings.getiterator('MeshHeading'): 
 83          kw = unicode(heading.findtext('DescriptorName')) 
 84          for qualifier in heading.getiterator('QualifierName'): 
 85            kw += ' / %s'%(unicode(qualifier.text)) 
 86          self.keywords.append(kw) 
 87       
 88 -  def ParseChemicals(self): 
 89      self.chemicals = [] 
 90      chemicals = self.find('MedlineCitation/ChemicalList') 
 91      if chemicals: 
 92        for chemical in chemicals.getiterator('Chemical'): 
 93          name = chemical.findtext('NameOfSubstance').encode('utf-8') 
 94          rn = chemical.findtext('RegistryNumber').encode('utf-8') 
 95          if rn != '0': 
 96            self.chemicals.append('%s <%s>'%(name,rn)) 
 97          else: 
 98            self.chemicals.append('%s'%(name)) 
 99       
100   
101    # -------------------------------------------- 
102    # 
103    #  We'll expose these ElementTree methods in case 
104    #  client code wants to pull extra info 
105    # 
106 -  def getiterator(self,key=None): 
107      if key is not None: 
108        return self._element.getiterator(key) 
109      else: 
110        return self._element.getiterator() 
111 -  def find(self,key): 
112      return self._element.find(key) 
113 -  def findtext(self,key): 
114      return self._element.findtext(key) 
115 -  def findall(self,key): 
116      return self._element.findall(key) 
117     
118 -class LinkRecord(Record): 
119    _fieldsOfInterest=[] 
120 -  def __init__(self,element): 
121      Record.__init__(self,element) 
122      self.PubMedId = self._element.text 
123      nbr = self._element.get('HasNeighbor','N') 
124      if nbr == 'Y': 
125        self.HasNeighbor = 1 
126      else: 
127        self.HasNeighbor = 0 
128