def parseOpenSearch(self,xmldata,format="atom"):
"""
OpenSearchの出力XMLをパースします。
xmldata=OpenSearch("keyword")[1]
format=atomまたはrss
"""
root=XML(xmldata)#xmldata
if format == "rss":
entrys=[elm for elm in root.getchildren()[0].getchildren() if "item" == elm.tag]
elif format == "atom":
entrys=[elm for elm in root if elm.tag.endswith("entry")]
else:
raise Exception("Unknown format : %s" % format)#xmlのフォーマットはatom/rssのみ対応
hits=len(entrys)
entrys=[entry.getiterator for entry in entrys]
entrys=[j for j in [i() for i in entrys]]
ret=[]
h="content"
for entry in entrys:
buf={}
for val in entry:
dual=1
if val != None and val.text != None:
if not val.text.startswith("\n"):
if buf.get(val.tag[val.tag.find("}")+1:]) == None:
buf[val.tag[val.tag.find("}")+1:]] = val.text
else:
dual= dual+1
buf[val.tag[val.tag.find("}")+1:] + str(dual)] = val.text
if h in buf.keys():buf[h]=self.stripHTMLTags(buf[h])
ret.append(buf)
return hits,ret#data[0]["name"] で著者
def get_status_code(output):
try:
tree = XML(output)
child = tree.getchildren()[0]
return int(child.get('status_code'))
except:
return None
def get_property(output, key):
try:
tree = XML(output)
child = tree.getchildren()[0] # first tag
return child.get(key)
except IndexError:
return None
def queryUniprot(id, expand=[], regex=True):
"""Query Uniprot with *id* and return a `dict` containing the raw results.
Regular users should use :func:`searchUniprot` instead.
:arg expand: entries through which you want to loop dictElements
until there aren't any elements left
:type expand: list
"""
if not isinstance(id, str):
raise TypeError('id should be a string')
try:
record_file = openURL(
'http://www.uniprot.org/uniprot/{0}.xml'.format(id))
except:
raise ValueError('No Uniprot record found with that id')
data = record_file.read()
record_file.close()
data = XML(data)
data = dictElement(data.getchildren()[0],
'{http://uniprot.org/uniprot}',
number_multiples=True)
for key in data:
value = data[key]
if not key.startswith('dbReference'):
continue
try:
if value.get('type') != 'PDB':
continue
except AttributeError:
continue
pdbid = value.get('id')
refdata = {'PDB': pdbid}
for prop in value:
prop_key = prop.get('type')
prop_val = prop.get('value')
refdata[prop_key] = prop_val
data[key] = refdata
if expand:
keys = []
if regex:
for lt in expand:
lt_re = re.compile(lt)
for key in data:
if lt_re.match(key):
keys.append(key)
else:
keys = expand
data = dictElementLoop(data, keys, '{http://uniprot.org/uniprot}')
return data
def queryUniprot(id, expand=[], regex=True):
"""Query Uniprot with *id* and return a `dictionary` containing the results
:arg expand: entries through which you want to loop dictElements
until there aren't any elements left
:type expand: list
"""
if not isinstance(id, str):
raise TypeError('id should be a string')
try:
record_file = openURL('http://www.uniprot.org/uniprot/{0}.xml'.format(id))
except:
raise ValueError('No Uniprot record found with that id')
data = record_file.read()
record_file.close()
data = XML(data)
data = dictElement(data.getchildren()[0], '{http://uniprot.org/uniprot}', number_multiples=True)
for key in data:
value = data[key]
if not key.startswith('dbReference'):
continue
try:
if value.get('type') != 'PDB':
continue
except AttributeError:
continue
pdbid = value.get('id')
refdata = {'PDB': pdbid}
for prop in value:
prop_key = prop.get('type')
prop_val = prop.get('value')
refdata[prop_key] = prop_val
data[key] = refdata
if expand:
keys = []
if regex:
for lt in expand:
lt_re = re.compile(lt)
for key in data.keys():
if lt_re.match(key):
keys.append(key)
else:
keys = expand
data = dictElementLoop(data, keys, '{http://uniprot.org/uniprot}')
return data
def get_property(output, key):
tree = XML(output)
child = tree.getchildren()[0]
return child.get(key)
