def get_imdbID(self, mop):
"""Return the imdbID for the given Movie, Person, Character or Company
object."""
imdbID = None
if mop.accessSystem == self.accessSystem:
aSystem = self
else:
aSystem = IMDb(mop.accessSystem)
if isinstance(mop, Movie.Movie):
if mop.movieID is not None:
imdbID = aSystem.get_imdbMovieID(mop.movieID)
else:
imdbID = aSystem.title2imdbID(build_title(mop, canonical=0,
ptdf=0, appendKind=False),
mop['kind'])
elif isinstance(mop, Person.Person):
if mop.personID is not None:
imdbID = aSystem.get_imdbPersonID(mop.personID)
else:
imdbID = aSystem.name2imdbID(build_name(mop, canonical=1))
elif isinstance(mop, Character.Character):
if mop.characterID is not None:
imdbID = aSystem.get_imdbCharacterID(mop.characterID)
else:
# canonical=0 ?
imdbID = aSystem.character2imdbID(build_name(mop, canonical=1))
elif isinstance(mop, Company.Company):
if mop.companyID is not None:
imdbID = aSystem.get_imdbCompanyID(mop.companyID)
else:
imdbID = aSystem.company2imdbID(build_company_name(mop))
else:
raise IMDbError('object ' + repr(mop) + \
' is not a Movie, Person or Character instance')
return imdbID
def get_imdbID(self, mop):
"""Return the imdbID for the given Movie, Person, Character or Company
object."""
imdbID = None
if mop.accessSystem == self.accessSystem:
aSystem = self
else:
aSystem = IMDb(mop.accessSystem)
if isinstance(mop, Movie.Movie):
if mop.movieID is not None:
imdbID = aSystem.get_imdbMovieID(mop.movieID)
else:
imdbID = aSystem.title2imdbID(
build_title(mop, canonical=0, ptdf=0, appendKind=False),
mop['kind'])
elif isinstance(mop, Person.Person):
if mop.personID is not None:
imdbID = aSystem.get_imdbPersonID(mop.personID)
else:
imdbID = aSystem.name2imdbID(build_name(mop, canonical=1))
elif isinstance(mop, Character.Character):
if mop.characterID is not None:
imdbID = aSystem.get_imdbCharacterID(mop.characterID)
else:
# canonical=0 ?
imdbID = aSystem.character2imdbID(build_name(mop, canonical=1))
elif isinstance(mop, Company.Company):
if mop.companyID is not None:
imdbID = aSystem.get_imdbCompanyID(mop.companyID)
else:
imdbID = aSystem.company2imdbID(build_company_name(mop))
else:
raise IMDbError('object ' + repr(mop) + \
' is not a Movie, Person or Character instance')
return imdbID
def _getitem(self, key):
"""Handle special keys."""
if self.data.has_key('name'):
if key == 'name':
return normalizeName(self.data['name'])
elif key == 'canonical name':
return self.data['name']
elif key == 'long imdb name':
return build_name(self.data, canonical=0)
elif key == 'long imdb canonical name':
return build_name(self.data)
return None
def _getitem(self, key):
"""Handle special keys."""
if self.data.has_key('name'):
if key == 'name':
return normalizeName(self.data['name'])
elif key == 'canonical name':
return self.data['name']
elif key == 'long imdb name':
return build_name(self.data, canonical=0)
elif key == 'long imdb canonical name':
return build_name(self.data)
return None
def isSameName(self, other):
"""Return true if two character have the same name
and/or characterID."""
if not isinstance(other, self.__class__):
return False
if 'name' in self.data and 'name' in other.data and \
build_name(self.data, canonical=False) == build_name(other.data, canonical=False):
return True
if self.accessSystem == other.accessSystem and \
self.characterID is not None and \
self.characterID == other.characterID:
return True
return False
def isSameName(self, other):
"""Return true if two character have the same name
and/or characterID."""
if not isinstance(other, self.__class__):
return False
if 'name' in self.data and 'name' in other.data and \
build_name(self.data, canonical=False) == build_name(other.data, canonical=False):
return True
if self.accessSystem == other.accessSystem and \
self.characterID is not None and \
self.characterID == other.characterID:
return True
return False
def _getitem(self, key):
"""Handle special keys."""
if 'name' in self.data:
if key == 'name':
return normalizeName(self.data['name'])
elif key == 'canonical name':
return canonicalName(self.data['name'])
elif key == 'long imdb name':
return build_name(self.data, canonical=False)
elif key == 'long imdb canonical name':
return build_name(self.data, canonical=True)
if key == 'full-size headshot':
return self.get_fullsizeURL()
return None
def _getitem(self, key):
"""Handle special keys."""
if self.data.has_key('name'):
if key == 'name':
return normalizeName(self.data['name'])
elif key == 'canonical name':
return self.data['name']
elif key == 'long imdb name':
return build_name(self.data, canonical=0)
elif key == 'long imdb canonical name':
return build_name(self.data)
if key == 'full-size headshot' and self.data.has_key('headshot'):
return self._re_fullsizeURL.sub('', self.data.get('headshot', ''))
return None
def _getitem(self, key):
"""Handle special keys."""
if 'name' in self.data:
if key == 'name':
return normalizeName(self.data['name'])
elif key == 'canonical name':
return self.data['name']
elif key == 'long imdb name':
return build_name(self.data, canonical=False)
elif key == 'long imdb canonical name':
return build_name(self.data)
if key == 'full-size headshot' and 'headshot' in self.data:
return self._re_fullsizeURL.sub('', self.data.get('headshot', ''))
return None
def _getitem(self, key):
"""Handle special keys."""
if self.data.has_key("name"):
if key == "name":
return normalizeName(self.data["name"])
elif key == "canonical name":
return self.data["name"]
elif key == "long imdb name":
return build_name(self.data, canonical=0)
elif key == "long imdb canonical name":
return build_name(self.data)
if key == "full-size headshot" and self.data.has_key("headshot"):
return self._re_fullsizeURL.sub("", self.data.get("headshot", ""))
return None
def _getitem(self, key):
"""Handle special keys."""
if 'name' in self.data:
if key == 'name':
return normalizeName(self.data['name'])
elif key == 'canonical name':
return self.data['name']
elif key == 'long imdb name':
return build_name(self.data, canonical=False)
elif key == 'long imdb canonical name':
return build_name(self.data)
if key == 'full-size headshot':
return self.get_fullsizeURL()
return None
def isSameName(self, other):
"""Return true if two persons have the same name and imdbIndex
and/or personID.
"""
if not isinstance(other, self.__class__):
return 0
if self.data.has_key('name') and \
other.data.has_key('name') and \
build_name(self.data, canonical=1) == \
build_name(other.data, canonical=1):
return 1
if self.accessSystem == other.accessSystem and \
self.personID and self.personID == other.personID:
return 1
return 0
def isSameName(self, other):
"""Return true if two character have the same name
and/or characterID."""
if not isinstance(other, self.__class__):
return 0
if self.data.has_key('name') and \
other.data.has_key('name') and \
build_name(self.data, canonical=0) == \
build_name(other.data, canonical=0):
return 1
if self.accessSystem == other.accessSystem and \
self.characterID is not None and \
self.characterID == other.characterID:
return 1
return 0
def isSameName(self, other):
"""Return true if two persons have the same name and imdbIndex
and/or personID.
"""
if not isinstance(other, self.__class__):
return False
if 'name' in self.data and \
'name' in other.data and \
build_name(self.data, canonical=True) == \
build_name(other.data, canonical=True):
return True
if self.accessSystem == other.accessSystem and \
self.personID and self.personID == other.personID:
return True
return False
def isSameName(self, other):
"""Return true if two persons have the same name and imdbIndex
and/or personID.
"""
if not isinstance(other, self.__class__):
return False
if 'name' in self.data and \
'name' in other.data and \
build_name(self.data, canonical=True) == \
build_name(other.data, canonical=True):
return True
if self.accessSystem == other.accessSystem and \
self.personID and self.personID == other.personID:
return True
return False
def isSameName(self, other):
"""Return true if two character have the same name
and/or characterID."""
if not isinstance(other, self.__class__):
return 0
if self.data.has_key('name') and \
other.data.has_key('name') and \
build_name(self.data, canonical=0) == \
build_name(other.data, canonical=0):
return 1
if self.accessSystem == other.accessSystem and \
self.characterID is not None and \
self.characterID == other.characterID:
return 1
return 0
def isSameName(self, other):
"""Return true if two persons have the same name and imdbIndex
and/or personID.
"""
if not isinstance(other, self.__class__):
return 0
if self.data.has_key('name') and \
other.data.has_key('name') and \
build_name(self.data, canonical=1) == \
build_name(other.data, canonical=1):
return 1
if self.accessSystem == other.accessSystem and \
self.personID and self.personID == other.personID:
return 1
return 0
class DOMHTMLSearchCharacterParser(DOMHTMLSearchMovieParser):
_BaseParser = DOMBasicCharacterParser
_notDirectHitTitle = '<title>find - imdb'
_titleBuilder = lambda self, x: build_name(x, canonical=False)
_linkPrefix = '/character/ch'
_attrs = [
Attribute(
key='data',
multi=True,
path={
'link': "./a[1]/@href",
'name': "./a[1]/text()"
},
postprocess=lambda x: (
analyze_imdbid(x.get('link') or ''),
{'name': x.get('name')}
)
)
]
extractors = [
Extractor(
label='search',
path="//td[@class='result_text']/a[starts-with(@href, '/character/ch')]/..",
attrs=_attrs
)
]
class DOMHTMLSearchPersonParser(DOMHTMLSearchMovieParser):
"""Parse the html page that the IMDb web server shows when the
"new search system" is used, for persons."""
_titleBuilder = lambda self, x: build_name(x, canonical=True)
_linkPrefix = '/name/nm'
_attrs = [
Attribute(
key='data',
multi=True,
path={
'link': "./a[1]/@href",
'name': "./a[1]/text()",
'index': "./text()[1]",
'akas': ".//div[@class='_imdbpyAKA']/text()"
},
postprocess=lambda x: (
analyze_imdbid(x.get('link') or ''),
analyze_name((x.get('name') or '') + (x.get('index') or ''),
canonical=1), x.get('akas')
)
)
]
extractors = [
Extractor(
label='search',
path="//td[@class='result_text']/a[starts-with(@href, '/name/nm')]/..",
attrs=_attrs
)
]
def _getitem(self, key):
"""Handle special keys."""
# XXX: can a character have an imdbIndex?
if 'name' in self.data:
if key == 'long imdb name':
return build_name(self.data)
return None
def _getitem(self, key):
"""Handle special keys."""
# XXX: can a character have an imdbIndex?
if 'name' in self.data:
if key == 'long imdb name':
return build_name(self.data)
return None
class DOMHTMLSearchPersonParser(DOMHTMLSearchMovieParser):
"""Parse the html page that the IMDb web server shows when the
"new search system" is used, for persons."""
_BaseParser = DOMBasicPersonParser
_notDirectHitTitle = '<title>imdb name'
_titleBuilder = lambda self, x: build_name(x, canonical=True)
_linkPrefix = '/name/nm'
_attrs = [Attribute(key='data',
multi=True,
path={
'link': "./a[1]/@href",
'name': "./a[1]/text()",
'index': "./text()[1]",
'akas': ".//div[@class='_imdbpyAKA']/text()"
},
postprocess=lambda x: (
analyze_imdbid(x.get('link') or u''),
analyze_name((x.get('name') or u'') + \
(x.get('index') or u''),
canonical=1), x.get('akas')
))]
extractors = [
Extractor(label='search',
path="//td[3]/a[starts-with(@href, '/name/nm')]/..",
attrs=_attrs)
]
def preprocess_string(self, html_string):
if self._notDirectHitTitle in html_string[:1024].lower():
html_string = _reAKASp.sub(
r'\1<div class="_imdbpyAKA">\2::</div>\3', html_string)
return DOMHTMLSearchMovieParser.preprocess_string(self, html_string)
def scan_names(name_list,
name1,
name2,
name3,
results=0,
ro_thresold=None,
_scan_character=False):
"""Scan a list of names, searching for best matches against
the given variations."""
if ro_thresold is not None: RO_THRESHOLD = ro_thresold
else: RO_THRESHOLD = 0.6
sm1 = SequenceMatcher()
sm2 = SequenceMatcher()
sm3 = SequenceMatcher()
sm1.set_seq1(name1.lower())
if name2: sm2.set_seq1(name2.lower())
if name3: sm3.set_seq1(name3.lower())
resd = {}
for i, n_data in name_list:
nil = n_data['name']
# XXX: on Symbian, here we get a str; not sure this is the
# right place to fix it.
if isinstance(nil, str):
nil = unicode(nil, 'latin1', 'ignore')
# Distance with the canonical name.
ratios = [ratcliff(name1, nil, sm1) + 0.05]
namesurname = u''
if not _scan_character:
nils = nil.split(', ', 1)
surname = nils[0]
if len(nils) == 2: namesurname = '%s %s' % (nils[1], surname)
else:
nils = nil.split(' ', 1)
surname = nils[-1]
namesurname = nil
if surname != nil:
# Distance with the "Surname" in the database.
ratios.append(ratcliff(name1, surname, sm1))
if not _scan_character:
ratios.append(ratcliff(name1, namesurname, sm1))
if name2:
ratios.append(ratcliff(name2, surname, sm2))
# Distance with the "Name Surname" in the database.
if namesurname:
ratios.append(ratcliff(name2, namesurname, sm2))
if name3:
# Distance with the long imdb canonical name.
ratios.append(
ratcliff(name3, build_name(n_data, canonical=1), sm3) + 0.1)
ratio = max(ratios)
if ratio >= RO_THRESHOLD:
if resd.has_key(i):
if ratio > resd[i][0]: resd[i] = (ratio, (i, n_data))
else: resd[i] = (ratio, (i, n_data))
res = resd.values()
res.sort()
res.reverse()
if results > 0: res[:] = res[:results]
return res
def _getitem(self, key):
"""Handle special keys."""
## XXX: can a character have an imdbIndex?
if self.data.has_key('name'):
if key == 'long imdb name':
return build_name(self.data)
if key == 'full-size headshot' and self.data.has_key('headshot'):
return self._re_fullsizeURL.sub('', self.data.get('headshot', ''))
return None
def _getitem(self, key):
"""Handle special keys."""
## XXX: can a character have an imdbIndex?
if self.data.has_key('name'):
if key == 'long imdb name':
return build_name(self.data)
if key == 'full-size headshot' and self.data.has_key('headshot'):
return self._re_fullsizeURL.sub('', self.data.get('headshot', ''))
return None
def scan_names(name_list, name1, name2, name3, results=0, ro_thresold=None,
_scan_character=False):
"""Scan a list of names, searching for best matches against
the given variations."""
if ro_thresold is not None: RO_THRESHOLD = ro_thresold
else: RO_THRESHOLD = 0.6
sm1 = SequenceMatcher()
sm2 = SequenceMatcher()
sm3 = SequenceMatcher()
sm1.set_seq1(name1.lower())
if name2: sm2.set_seq1(name2.lower())
if name3: sm3.set_seq1(name3.lower())
resd = {}
for i, n_data in name_list:
nil = n_data['name']
# XXX: on Symbian, here we get a str; not sure this is the
# right place to fix it.
if isinstance(nil, str):
nil = unicode(nil, 'latin1', 'ignore')
# Distance with the canonical name.
ratios = [ratcliff(name1, nil, sm1) + 0.05]
namesurname = u''
if not _scan_character:
nils = nil.split(', ', 1)
surname = nils[0]
if len(nils) == 2: namesurname = '%s %s' % (nils[1], surname)
else:
nils = nil.split(' ', 1)
surname = nils[-1]
namesurname = nil
if surname != nil:
# Distance with the "Surname" in the database.
ratios.append(ratcliff(name1, surname, sm1))
if not _scan_character:
ratios.append(ratcliff(name1, namesurname, sm1))
if name2:
ratios.append(ratcliff(name2, surname, sm2))
# Distance with the "Name Surname" in the database.
if namesurname:
ratios.append(ratcliff(name2, namesurname, sm2))
if name3:
# Distance with the long imdb canonical name.
ratios.append(ratcliff(name3,
build_name(n_data, canonical=1), sm3) + 0.1)
ratio = max(ratios)
if ratio >= RO_THRESHOLD:
if resd.has_key(i):
if ratio > resd[i][0]: resd[i] = (ratio, (i, n_data))
else: resd[i] = (ratio, (i, n_data))
res = resd.values()
res.sort()
res.reverse()
if results > 0: res[:] = res[:results]
return res
def _findRefs(self, o, trefs, nrefs):
"""Find titles or names references in strings."""
if isinstance(o, (unicode, str)):
for title in re_titleRef.findall(o):
a_title = analyze_title(title, canonical=0)
rtitle = build_title(a_title, ptdf=1)
if trefs.has_key(rtitle): continue
movieID = self._getTitleID(rtitle)
if movieID is None:
movieID = self._getTitleID(title)
if movieID is None:
continue
m = Movie(title=rtitle,
movieID=movieID,
accessSystem=self.accessSystem)
trefs[rtitle] = m
rtitle2 = canonicalTitle(a_title.get('title', u''))
if rtitle2 and rtitle2 != rtitle and rtitle2 != title:
trefs[rtitle2] = m
if title != rtitle:
trefs[title] = m
for name in re_nameRef.findall(o):
a_name = analyze_name(name, canonical=1)
rname = build_name(a_name, canonical=1)
if nrefs.has_key(rname): continue
personID = self._getNameID(rname)
if personID is None:
personID = self._getNameID(name)
if personID is None: continue
p = Person(name=rname,
personID=personID,
accessSystem=self.accessSystem)
nrefs[rname] = p
rname2 = normalizeName(a_name.get('name', u''))
if rname2 and rname2 != rname:
nrefs[rname2] = p
if name != rname and name != rname2:
nrefs[name] = p
elif isinstance(o, (list, tuple)):
for item in o:
self._findRefs(item, trefs, nrefs)
elif isinstance(o, dict):
for value in o.values():
self._findRefs(value, trefs, nrefs)
return (trefs, nrefs)
def _findRefs(self, o, trefs, nrefs):
"""Find titles or names references in strings."""
if isinstance(o, (UnicodeType, StringType)):
for title in re_titleRef.findall(o):
a_title = analyze_title(title, canonical=1)
rtitle = build_title(a_title, canonical=1, ptdf=1)
if trefs.has_key(rtitle): continue
movieID = self._getTitleID(rtitle)
if movieID is None:
movieID = self._getTitleID(title)
if movieID is None:
continue
m = Movie(title=rtitle, movieID=movieID,
accessSystem=self.accessSystem)
trefs[rtitle] = m
rtitle2 = canonicalTitle(a_title.get('title', u''))
if rtitle2 and rtitle2 != rtitle and rtitle2 != title:
trefs[rtitle2] = m
if title != rtitle:
trefs[title] = m
for name in re_nameRef.findall(o):
a_name = analyze_name(name, canonical=1)
rname = build_name(a_name, canonical=1)
if nrefs.has_key(rname): continue
personID = self._getNameID(rname)
if personID is None:
personID = self._getNameID(name)
if personID is None: continue
p = Person(name=rname, personID=personID,
accessSystem=self.accessSystem)
nrefs[rname] = p
rname2 = normalizeName(a_name.get('name', u''))
if rname2 and rname2 != rname:
nrefs[rname2] = p
if name != rname and name != rname2:
nrefs[name] = p
elif isinstance(o, (ListType, TupleType)):
for item in o:
self._findRefs(item, trefs, nrefs)
elif isinstance(o, DictType):
for value in o.values():
self._findRefs(value, trefs, nrefs)
return (trefs, nrefs)
def get_imdbCharacterID(self, characterID):
"""Translate a characterID in an imdbID.
If not in the database, try an Exact Primary Name search on IMDb;
return None if it's unable to get the imdbID.
"""
try:
character = CharName.get(characterID)
except NotFoundError:
return None
imdbID = character.imdbID
if imdbID is not None: return '%07d' % imdbID
n_dict = {'name': character.name, 'imdbIndex': character.imdbIndex}
namline = build_name(n_dict, canonical=1)
imdbID = self.character2imdbID(namline)
if imdbID is not None:
try:
character.imdbID = int(imdbID)
except:
pass
return imdbID
def get_imdbPersonID(self, personID):
"""Translate a personID in an imdbID.
If not in the database, try an Exact Primary Name search on IMDb;
return None if it's unable to get the imdbID.
"""
try:
person = Name.get(personID)
except SQLObjectNotFound:
return None
imdbID = person.imdbID
if imdbID is not None: return '%07d' % imdbID
n_dict = {'name': person.name, 'imdbIndex': person.imdbIndex}
namline = build_name(n_dict, canonical=1)
imdbID = self.name2imdbID(namline)
if imdbID is not None:
try:
person.imdbID = int(imdbID)
except:
pass
return imdbID
def nameVariations(name, fromPtdf=0):
"""Build name variations useful for searches; if fromPtdf is true,
the input is assumed to be in the plain text data files format."""
name1 = name2 = name3 = u''
if fromPtdf or re_nameIndex.search(name):
# We've a name with an (imdbIndex)
namedict = analyze_name(name, canonical=1)
# name1 is the name in the canonical format.
name1 = namedict['name']
# name3 is the canonical name with the imdbIndex.
if fromPtdf:
if namedict.has_key('imdbIndex'):
name3 = name
else:
name3 = build_name(namedict, canonical=1)
else:
# name1 is the name in the canonical format.
name1 = canonicalName(name)
name3 = u''
# name2 is the name in the normal format, if it differs from name1.
name2 = normalizeName(name1)
if name1 == name2: name2 = u''
return name1, name2, name3
def nameVariations(name, fromPtdf=0):
"""Build name variations useful for searches; if fromPtdf is true,
the input is assumed to be in the plain text data files format."""
name1 = name2 = name3 = u''
if fromPtdf or re_nameIndex.search(name):
# We've a name with an (imdbIndex)
namedict = analyze_name(name, canonical=1)
# name1 is the name in the canonical format.
name1 = namedict['name']
# name3 is the canonical name with the imdbIndex.
if fromPtdf:
if namedict.has_key('imdbIndex'):
name3 = name
else:
name3 = build_name(namedict, canonical=1)
else:
# name1 is the name in the canonical format.
name1 = canonicalName(name)
name3 = u''
# name2 is the name in the normal format, if it differs from name1.
name2 = normalizeName(name1)
if name1 == name2: name2 = u''
return name1, name2, name3
def _search_person(self, name, results):
name = name.strip()
if not name: return []
name1, name2, name3 = nameVariations(name)
res = _scan_names('%snames.key' % self.__db,
name1, name2, name3, results)
res[:] = [x[1] for x in res]
new_res = []
seen_PID = []
for idx, (personID, r) in enumerate(res):
# Remove duplicates.
# XXX: find a way to prefer names with an AKA? Or prefer
# the original name?
if personID in seen_PID:
continue
else:
seen_PID.append(personID)
realPID = self._get_real_personID(personID)
if personID == realPID:
new_res.append((personID, r))
continue
if realPID in seen_PID:
continue
else:
seen_PID.append(realPID)
aka_name = build_name(r, canonical=1)
real_name = getLabel(realPID, '%snames.index' % self.__db,
'%snames.key' % self.__db)
if aka_name == real_name:
new_res.append((realPID, r))
continue
new_r = analyze_name(real_name, canonical=1)
new_r['akas'] = [aka_name]
new_res.append((realPID, new_r))
if results > 0: new_res[:] = new_res[:results]
return new_res
def _parseBiography(biol):
"""Parse the biographies.data file."""
res = {}
bio = ' '.join(_parseList(biol, 'BG', mline=0))
bio = _parseBioBy(biol)
if bio: res['mini biography'] = bio
for x in biol:
x4 = x[:4]
x6 = x[:6]
if x4 == 'DB: ':
date, notes = date_and_notes(x[4:])
if date:
res['birth date'] = date
if notes:
res['birth notes'] = notes
elif x4 == 'DD: ':
date, notes = date_and_notes(x[4:])
if date:
res['death date'] = date
if notes:
res['death notes'] = notes
elif x6 == 'SP: * ':
res.setdefault('spouse', []).append(x[6:].strip())
elif x4 == 'RN: ':
n = x[4:].strip()
if not n: continue
rn = build_name(analyze_name(n, canonical=1), canonical=1)
res['birth name'] = rn
elif x6 == 'AT: * ':
res.setdefault('articles', []).append(x[6:].strip())
elif x4 == 'HT: ':
res['height'] = x[4:].strip()
elif x6 == 'PT: * ':
res.setdefault('pictorials', []).append(x[6:].strip())
elif x6 == 'CV: * ':
res.setdefault('magazine covers', []).append(x[6:].strip())
elif x4 == 'NK: ':
res.setdefault('nick names', []).append(normalizeName(x[4:]))
elif x6 == 'PI: * ':
res.setdefault('portrayed', []).append(x[6:].strip())
elif x6 == 'SA: * ':
sal = x[6:].strip().replace(' -> ', '::')
res.setdefault('salary history', []).append(sal)
trl = _parseList(biol, 'TR')
if trl: res['trivia'] = trl
quotes = _parseList(biol, 'QU')
if quotes: res['quotes'] = quotes
otherworks = _parseList(biol, 'OW')
if otherworks: res['other works'] = otherworks
books = _parseList(biol, 'BO')
if books: res['books'] = books
agent = _parseList(biol, 'AG')
if agent: res['agent address'] = agent
wherenow = _parseList(biol, 'WN')
if wherenow: res['where now'] = wherenow[0]
biomovies = _parseList(biol, 'BT')
if biomovies: res['biographical movies'] = biomovies
guestapp = _buildGuests([x[6:].strip() for x in biol if x[:6] == 'GA: * '])
if guestapp: res['notable tv guest appearances'] = guestapp
tm = _parseList(biol, 'TM')
if tm: res['trademarks'] = tm
interv = _parseList(biol, 'IT')
if interv: res['interviews'] = interv
return res
pinfo = _groupListBy(pinfo, 0)
for group in pinfo:
sect = group[0][0]
for pdata in group:
data = pdata[1]
if pdata[2]: data += '::%s' % pdata[2]
res.setdefault(sect, []).append(data)
# AKA names.
akan = [(an.name, an.imdbIndex)
for an in AkaName.select(AkaName.q.personID == personID)]
if akan:
res['akas'] = []
for n in akan:
nd = {'name': n[0]}
if n[1]: nd['imdbIndex'] = n[1]
nt = build_name(nd, canonical=1)
res['akas'].append(nt)
# Do some transformation to preserve consistency with other
# data access systems.
for key in ('birth date', 'birth notes', 'death date', 'death notes',
'birth name', 'height'):
if res.has_key(key):
res[key] = res[key][0]
#if res.has_key('mini biography'):
# nl = []
# for i in res['mini biography']:
# if i[-1] == ')':
# sauth = i.rfind('::(author: ')
# if sauth != -1:
# nl.append(i[sauth+11:-1] + '::' + i[:sauth])
# else: nl.append(i)
def _parseBiography(biol):
"""Parse the biographies.data file."""
res = {}
bio = ' '.join(_parseList(biol, 'BG', mline=0))
bio = _parseBioBy(biol)
if bio: res['mini biography'] = bio
for x in biol:
x4 = x[:4]
x6 = x[:6]
if x4 == 'DB: ':
date, notes = date_and_notes(x[4:])
if date:
res['birth date'] = date
if notes:
res['birth notes'] = notes
#bdate = x.strip()
#i = bdate.find(',')
#if i != -1:
# res['birth notes'] = bdate[i+1:].strip()
# bdate = bdate[:i]
#res['birth date'] = bdate[4:]
elif x4 == 'DD: ':
date, notes = date_and_notes(x[4:])
if date:
res['death date'] = date
if notes:
res['death notes'] = notes
#ddate = x.strip()
#i = ddate.find(',')
#if i != -1:
# res['death notes'] = ddate[i+1:].strip()
# ddate = ddate[:i]
#res['death date'] = ddate[4:]
elif x6 == 'SP: * ':
res.setdefault('spouse', []).append(x[6:].strip())
elif x4 == 'RN: ':
n = x[4:].strip()
if not n: continue
rn = build_name(analyze_name(n, canonical=1), canonical=1)
res['birth name'] = rn
elif x6 == 'AT: * ':
res.setdefault('articles', []).append(x[6:].strip())
elif x4 == 'HT: ':
res['height'] = x[4:].strip()
elif x6 == 'PT: * ':
res.setdefault('pictorials', []).append(x[6:].strip())
elif x6 == 'CV: * ':
res.setdefault('magazine covers', []).append(x[6:].strip())
elif x4 == 'NK: ':
res.setdefault('nick names', []).append(normalizeName(x[4:]))
elif x6 == 'PI: * ':
res.setdefault('portrayed', []).append(x[6:].strip())
elif x6 == 'SA: * ':
sal = x[6:].strip().replace(' -> ', '::')
res.setdefault('salary history', []).append(sal)
trl = _parseList(biol, 'TR')
if trl: res['trivia'] = trl
quotes = _parseList(biol, 'QU')
if quotes: res['quotes'] = quotes
otherworks = _parseList(biol, 'OW')
if otherworks: res['other works'] = otherworks
books = _parseList(biol, 'BO')
if books: res['books'] = books
agent = _parseList(biol, 'AG')
if agent: res['agent address'] = agent
wherenow = _parseList(biol, 'WN')
if wherenow: res['where now'] = wherenow[0]
biomovies = _parseList(biol, 'BT')
if biomovies: res['biographical movies'] = biomovies
guestapp = _buildGuests([x[6:].strip() for x in biol if x[:6] == 'GA: * '])
if guestapp: res['notable tv guest appearances'] = guestapp
tm = _parseList(biol, 'TM')
if tm: res['trademarks'] = tm
interv = _parseList(biol, 'IT')
if interv: res['interviews'] = interv
return res
