def parse(self, xml_detail, xml_more_info):
title = self.parse_title(xml_detail)
authors = self.parse_authors(xml_detail)
comments = self.parse_comments(xml_detail)
rating = self.parse_rating(xml_detail)
isbn = self.parse_isbn(xml_more_info)
publisher = self.parse_publisher(xml_detail)
tags = self.parse_tags(xml_detail, xml_more_info)
serie, serie_index = self.parse_serie(xml_detail)
pub_year = self.parse_pub_year(xml_detail, xml_more_info)
cover = self.parse_cover(xml_detail)
if title is not None and authors is not None:
mi = Metadata(as_unicode(title), authors)
mi.languages = {'ces'}
mi.comments = as_unicode(comments)
mi.identifiers = {self.plugin.name:self.ident}
mi.rating = rating
mi.tags = tags
mi.publisher = publisher
mi.pubdate = pub_year
mi.isbn = isbn
mi.series = serie
mi.series_index = serie_index
mi.cover_url = cover
if cover:
self.plugin.cache_identifier_to_cover_url(self.ident, cover)
return mi
else:
self.log('Result skipped for because title or authors not found')
return None
def build_meta(log, issue_id):
"""Build metadata record based on comicvine issue_id"""
issue = pycomicvine.Issue(
issue_id,
field_list=[
"id",
"name",
"volume",
"issue_number",
"person_credits",
"description",
"store_date",
"cover_date",
],
)
if not issue or not issue.volume:
log.warn("Unable to load Issue(%d)" % issue_id)
return None
title = "%s #%s" % (issue.volume.name, issue.issue_number)
if issue.name:
title = title + ": %s" % (issue.name)
authors = [p.name for p in issue.person_credits]
meta = Metadata(title, authors)
meta.series = issue.volume.name
meta.series_index = str(issue.issue_number)
meta.set_identifier("comicvine", str(issue.id))
meta.comments = issue.description
meta.has_cover = False
if issue.volume.publisher:
meta.publisher = issue.volume.publisher.name
meta.pubdate = issue.store_date or issue.cover_date
return meta
def parse(self, xml_detail):
title = self.parse_title(xml_detail)
authors = self.parse_authors(xml_detail)
comments = self.parse_comments(xml_detail)
rating = self.parse_rating(xml_detail)
isbn = self.parse_isbn(xml_detail)
publisher = self.parse_publisher(xml_detail)
pub_year = self.parse_pubdate(xml_detail)
tags = self.parse_tags(xml_detail)
serie, serie_index = self.parse_serie(xml_detail)
cover = self.parse_cover(xml_detail)
if title is not None and authors is not None:
mi = Metadata(title, authors)
mi.languages = {'ces'}
mi.comments = as_unicode(comments)
mi.identifiers = {self.plugin.name:str(self.number)}
mi.rating = rating
mi.tags = tags
mi.publisher = publisher
mi.pubdate = pub_year
mi.isbn = isbn
mi.series = serie
mi.series_index = serie_index
mi.cover_url = cover
if cover:
self.plugin.cache_identifier_to_cover_url(str(self.number), cover)
return mi
else:
return None
def convert_markdown_with_metadata(txt, title='', extensions=DEFAULT_MD_EXTENSIONS):
from calibre.ebooks.metadata.book.base import Metadata
from calibre.utils.date import parse_only_date
from calibre.db.write import get_series_values
if 'meta' not in extensions:
extensions.append('meta')
md = create_markdown_object(extensions)
html = md.convert(txt)
mi = Metadata(title or _('Unknown'))
m = md.Meta
for k, v in iteritems({'date':'pubdate', 'summary':'comments'}):
if v not in m and k in m:
m[v] = m.pop(k)
for k in 'title authors series tags pubdate comments publisher rating'.split():
val = m.get(k)
if val:
mf = mi.metadata_for_field(k)
if not mf.get('is_multiple'):
val = val[0]
if k == 'series':
val, si = get_series_values(val)
mi.series_index = 1 if si is None else si
if k == 'rating':
try:
val = max(0, min(int(float(val)), 10))
except Exception:
continue
if mf.get('datatype') == 'datetime':
try:
val = parse_only_date(val, assume_utc=False)
except Exception:
continue
setattr(mi, k, val)
return mi, HTML_TEMPLATE % (mi.title, html)
def test(scale=0.5):
from PyQt5.Qt import QLabel, QApplication, QPixmap, QMainWindow, QWidget, QScrollArea, QGridLayout
app = QApplication([])
mi = Metadata('xxx', ['Kovid Goyal', 'John Q. Doe', 'Author'])
mi.series = 'A series of styles'
m = QMainWindow()
sa = QScrollArea(m)
w = QWidget(m)
sa.setWidget(w)
l = QGridLayout(w)
w.setLayout(l), l.setSpacing(30)
labels = []
for r, color in enumerate(sorted(default_color_themes)):
for c, style in enumerate(sorted(all_styles())):
mi.series_index = c + 1
mi.title = 'An algorithmic cover [%s]' % color
prefs = override_prefs(cprefs, override_color_theme=color, override_style=style)
for x in ('cover_width', 'cover_height', 'title_font_size', 'subtitle_font_size', 'footer_font_size'):
prefs[x] = int(scale * prefs[x])
img = generate_cover(mi, prefs=prefs, as_qimage=True)
la = QLabel()
la.setPixmap(QPixmap.fromImage(img))
l.addWidget(la, r, c)
labels.append(la)
m.setCentralWidget(sa)
w.resize(w.sizeHint())
m.show()
app.exec_()
def test(scale=0.25):
from PyQt5.Qt import QLabel, QPixmap, QMainWindow, QWidget, QScrollArea, QGridLayout
from calibre.gui2 import Application
app = Application([])
mi = Metadata('Unknown', ['Kovid Goyal', 'John & Doe', 'Author'])
mi.series = 'A series & styles'
m = QMainWindow()
sa = QScrollArea(m)
w = QWidget(m)
sa.setWidget(w)
l = QGridLayout(w)
w.setLayout(l), l.setSpacing(30)
scale *= w.devicePixelRatioF()
labels = []
for r, color in enumerate(sorted(default_color_themes)):
for c, style in enumerate(sorted(all_styles())):
mi.series_index = c + 1
mi.title = 'An algorithmic cover [%s]' % color
prefs = override_prefs(cprefs, override_color_theme=color, override_style=style)
scale_cover(prefs, scale)
img = generate_cover(mi, prefs=prefs, as_qimage=True)
img.setDevicePixelRatio(w.devicePixelRatioF())
la = QLabel()
la.setPixmap(QPixmap.fromImage(img))
l.addWidget(la, r, c)
labels.append(la)
m.setCentralWidget(sa)
w.resize(w.sizeHint())
m.show()
app.exec_()
def get_series(title, authors, timeout=60):
mi = Metadata(title, authors)
if title and title[0] in _ignore_starts:
title = title[1:]
title = re.sub(r'^(A|The|An)\s+', '', title).strip()
if not title:
return mi
if isinstance(title, unicode):
title = title.encode('utf-8')
title = urllib.quote_plus(title)
author = authors[0].strip()
if not author:
return mi
if ',' in author:
author = author.split(',')[0]
else:
author = author.split()[-1]
url = URL.format(author, title)
br = browser()
try:
raw = br.open_novisit(url, timeout=timeout).read()
except URLError as e:
if isinstance(e.reason, socket.timeout):
raise Exception('KDL Server busy, try again later')
raise
if 'see the full results' not in raw:
return mi
raw = xml_to_unicode(raw)[0]
soup = BeautifulSoup(raw)
searcharea = soup.find('div', attrs={'class':'searcharea'})
if searcharea is None:
return mi
ss = searcharea.find('div', attrs={'class':'seriessearch'})
if ss is None:
return mi
a = ss.find('a', href=True)
if a is None:
return mi
href = a['href'].partition('?')[-1]
data = urlparse.parse_qs(href)
series = data.get('SeriesName', [])
if not series:
return mi
series = series[0]
series = re.sub(r' series$', '', series).strip()
if series:
mi.series = series
ns = ss.nextSibling
if ns.contents:
raw = unicode(ns.contents[0])
raw = raw.partition('.')[0].strip()
try:
mi.series_index = int(raw)
except:
pass
return mi
def default_mi(self):
from calibre.ebooks.metadata.book.base import Metadata
mi = Metadata(_('A sample book'), [_('Author One'), _('Author Two')])
mi.series = _('A series of samples')
mi.series_index = 4
mi.tags = [_('Tag One'), _('Tag Two')]
mi.publisher = _('Some publisher')
mi.rating = 4
mi.identifiers = {'isbn':'123456789', 'url': 'http://calibre-ebook.com'}
mi.languages = ['eng', 'fra']
mi.pubdate = mi.timestamp = now()
return mi
def data2mi(self, item):
"""Converts a single metadata answer in the form of a dict to a MetadataInformation object"""
mi = Metadata(_('Unknown'))
# Regular metadata
mi.title = item.get('title', None)
mi.authors = item.get('authors', [])
mi.publisher = item.get('publisher', None)
if 'id' in item.keys(): mi.set_identifier(self.idkey, item['id'])
if 'doi' in item.keys(): mi.set_identifier('doi', item['doi'])
if 'isbn' in item.keys(): mi.set_identifier('isbn', item['isbn'])
if 'updated' in item.keys(): mi.pubdate = parse_date(item['updated'], assume_utc=True)
if 'series' in item.keys():
mi.series = item['series']
mi.series_index = self.format_series_index(item.get('series_index'), None)
if 'year' in item.keys(): mi.pubdate = parse_date(item['year'], assume_utc=True)
if 'abstract' in item.keys(): mi.comments = self.format_abstract(item['abstract'])
if 'language' in item.keys(): mi.language = item['language']
if 'journal' in item.keys():
mi.series = item['journal']
mi.series_index = self.format_series_index(item.get('volume'), item.get('number'))
if 'subject' in item.keys():
tags = set([])
for s in item['subject']:
tags.update(msc_tags(s))
tags.update(arxiv_tags(s))
mi.tags = list(sorted(tags))
return mi
def parse(self, xml_detail):
sys_ident = title = isbn = publisher = pub_year = serie = serie_index = cover = None
authors = []
tags = []
xpath = self.XPath('//table[@id="record"]//tr')
for row in xpath(xml_detail):
ch = row.getchildren()
txt = ch[0].text.strip()
data = self.normalize(ch[1].text)
if txt.startswith('245') and title is None:
title = self.parse_title(data)
if txt.startswith('246'):
title = self.parse_title(data)
elif txt.startswith('100') or txt.startswith('700'):
res = self.parse_author(data)
if res is not None:
authors.append(res)
elif txt == 'SYS':
sys_ident = data.strip()
elif txt =='020':
isbn = self.parse_isbn(data)
elif txt == '260':
publisher, pub_year = self.parse_publisher(data)
elif txt.startswith('490') and serie is None:
serie, serie_index = self.parse_serie(data)
elif txt == '655 7':
tags.append(self.parse_tags(data))
if isbn is not None and isbn != '':
cover = self.parse_cover(isbn)
if title is not None and len(authors) > 0 and sys_ident is not None:
mi = Metadata(title, authors)
mi.languages = {'ces'}
mi.identifiers = {self.plugin.name:sys_ident}
mi.tags = tags
mi.publisher = publisher
mi.pubdate = pub_year
mi.isbn = isbn
mi.series = serie
mi.series_index = serie_index
mi.cover_url = cover
if cover:
self.plugin.cache_identifier_to_cover_url(sys_ident, cover)
return mi
else:
self.log('Data not found')
return None
def build_meta(log, issue_id):
"""Build metadata record based on comicvine issue_id."""
issue = PyComicvineWrapper(log).lookup_issue(issue_id)
if issue:
meta = Metadata(issue.get_full_title(), issue.get_authors())
meta.series = issue.volume_name
meta.series_index = issue.issue_number
meta.set_identifier('comicvine', str(issue.id))
meta.set_identifier('comicvine-volume', str(issue.volume_id))
meta.comments = issue.description
meta.has_cover = False
meta.publisher = issue.publisher_name
meta.pubdate = issue.date
return meta
else:
return None
def parse(self, xml_detail):
title = self.parse_title(xml_detail)
authors = self.parse_authors(xml_detail)
comments = self.parse_comments(xml_detail)
rating = self.parse_rating(xml_detail)
tags = self.parse_tags(xml_detail)
serie, serie_index = self.parse_serie(xml_detail)
if title is not None and authors is not None:
mi = Metadata(title, authors)
mi.languages = {'ces'}
mi.comments = as_unicode(comments)
mi.identifiers = {self.plugin.name:self.ident}
mi.rating = rating
mi.tags = tags
mi.series = serie
mi.series_index = serie_index
return mi
else:
return None
def build_meta(log, issue_id):
'''Build metadata record based on comicvine issue_id'''
issue = pycomicvine.Issue(issue_id, field_list=[
'id', 'name', 'volume', 'issue_number', 'person_credits', 'description',
'store_date', 'cover_date'])
if not issue or not issue.volume:
log.warn('Unable to load Issue(%d)' % issue_id)
return None
title = '%s #%s' % (issue.volume.name, issue.issue_number)
if issue.name:
title = title + ': %s' % (issue.name)
authors = [p.name for p in issue.person_credits]
meta = Metadata(title, authors)
meta.series = issue.volume.name
meta.series_index = str(issue.issue_number)
meta.set_identifier('comicvine', str(issue.id))
meta.set_identifier('comicvine-volume', str(issue.volume.id))
meta.comments = issue.description
meta.has_cover = False
if issue.volume.publisher:
meta.publisher = issue.volume.publisher.name
meta.pubdate = issue.store_date or issue.cover_date
return meta
def get_details(self):
self.log.info(" Worker.get_details:")
self.log.info(" self: ", self)
self.log.info(" self.url: ", self.url)
try:
raw = self.browser.open_novisit(
self.url, timeout=self.timeout).read().strip()
self.log.info(raw)
except Exception as e:
if callable(getattr(e, 'getcode', None)) and e.getcode() == 404:
self.log.error('URL malformed: %r' % self.url)
return
attr = getattr(e, 'args', [None])
attr = attr if attr else [None]
if isinstance(attr[0], socket.timeout):
msg = 'Bookmeta for biblionet timed out. Try again later.'
self.log.error(msg)
else:
msg = 'Failed to make details query: %r' % self.url
self.log.exception(msg)
return
if '<title>404 - ' in raw:
self.log.error('URL malformed: %r' % self.url)
return
try:
# root = fromstring(clean_ascii_chars(raw))
root = json.loads(raw)
self.log.info(root)
except:
msg = 'Failed to parse book detail page: %r' % self.url
self.log.exception(msg)
return
try:
self.biblionetid = root['biblionetid']
except:
self.log.exception('Error parsing book id for url: %r' % self.url)
self.biblionetid = None
try:
self.title = root['title'].strip()
except:
self.log.exception('Error parsing title for url: %r' % self.url)
self.title = None
self.series_index = None
try:
self.authors = [root['authors'].strip()]
self.log.info(self.authors)
except:
self.log.exception('Error parsing authors for url: %r' % self.url)
self.authors = None
try:
self.cover_url = root['cover_url']
self.log.info('Parsed URL for cover:%r' % self.cover_url)
self.plugin.cache_identifier_to_cover_url(self.biblionetid,
self.cover_url)
except:
self.log.exception('Error parsing cover for url: %r' % self.url)
self.has_cover = bool(self.cover_url)
try:
self.publisher = root['publisher']
self.log.info('Parsed publisher:%s' % self.publisher)
except:
self.log.exception('Error parsing publisher for url: %r' %
self.url)
try:
self.tags = root['categories'].replace('DDC: ', 'DDC:').replace(
'-', '').split()[:-1]
self.log.info('Parsed tags:%s' % self.tags)
except:
self.log.exception('Error parsing tags for url: %r' % self.url)
try:
self.pubdate = root['yr_published']
self.log.info('Parsed publication date:%s' % self.pubdate)
except:
self.log.exception('Error parsing published date for url: %r' %
self.url)
mi = Metadata(self.title, self.authors)
mi.set_identifier('biblionet', self.biblionetid)
if self.series_index:
try:
mi.series_index = float(self.series_index)
except:
self.log.exception('Error loading series')
if self.relevance:
try:
mi.source_relevance = self.relevance
except:
self.log.exception('Error loading relevance')
if self.cover_url:
try:
mi.cover_url = self.cover_url
except:
self.log.exception('Error loading cover_url')
if self.publisher:
try:
mi.publisher = self.publisher
except:
self.log.exception('Error loading publisher')
if self.tags:
try:
mi.tags = self.tags
except:
self.log.exception('Error loading tags')
if self.pubdate:
try:
if self.pubdate not in (self.yr_msg1, self.yr_msg2):
d = datetime.date(int(self.pubdate), 1, 1)
mi.pubdate = d
except:
self.log.exception('Error loading pubdate')
self.plugin.clean_downloaded_metadata(mi)
self.result_queue.put(mi)
def _get_metadata(self, book_id, get_user_categories=True): # {{{
mi = Metadata(None, template_cache=self.formatter_template_cache)
author_ids = self._field_ids_for('authors', book_id)
aut_list = [self._author_data(i) for i in author_ids]
aum = []
aus = {}
aul = {}
for rec in aut_list:
aut = rec['name']
aum.append(aut)
aus[aut] = rec['sort']
aul[aut] = rec['link']
mi.title = self._field_for('title',
book_id,
default_value=_('Unknown'))
mi.authors = aum
mi.author_sort = self._field_for('author_sort',
book_id,
default_value=_('Unknown'))
mi.author_sort_map = aus
mi.author_link_map = aul
mi.comments = self._field_for('comments', book_id)
mi.publisher = self._field_for('publisher', book_id)
n = nowf()
mi.timestamp = self._field_for('timestamp', book_id, default_value=n)
mi.pubdate = self._field_for('pubdate', book_id, default_value=n)
mi.uuid = self._field_for('uuid', book_id, default_value='dummy')
mi.title_sort = self._field_for('sort',
book_id,
default_value=_('Unknown'))
mi.book_size = self._field_for('size', book_id, default_value=0)
mi.ondevice_col = self._field_for('ondevice',
book_id,
default_value='')
mi.last_modified = self._field_for('last_modified',
book_id,
default_value=n)
formats = self._field_for('formats', book_id)
mi.format_metadata = {}
mi.languages = list(self._field_for('languages', book_id))
if not formats:
good_formats = None
else:
mi.format_metadata = FormatMetadata(self, book_id, formats)
good_formats = FormatsList(formats, mi.format_metadata)
mi.formats = good_formats
mi.has_cover = _('Yes') if self._field_for(
'cover', book_id, default_value=False) else ''
mi.tags = list(self._field_for('tags', book_id, default_value=()))
mi.series = self._field_for('series', book_id)
if mi.series:
mi.series_index = self._field_for('series_index',
book_id,
default_value=1.0)
mi.rating = self._field_for('rating', book_id)
mi.set_identifiers(
self._field_for('identifiers', book_id, default_value={}))
mi.application_id = book_id
mi.id = book_id
composites = []
for key, meta in self.field_metadata.custom_iteritems():
mi.set_user_metadata(key, meta)
if meta['datatype'] == 'composite':
composites.append(key)
else:
val = self._field_for(key, book_id)
if isinstance(val, tuple):
val = list(val)
extra = self._field_for(key + '_index', book_id)
mi.set(key, val=val, extra=extra)
for key in composites:
mi.set(key, val=self._composite_for(key, book_id, mi))
user_cat_vals = {}
if get_user_categories:
user_cats = self.backend.prefs['user_categories']
for ucat in user_cats:
res = []
for name, cat, ign in user_cats[ucat]:
v = mi.get(cat, None)
if isinstance(v, list):
if name in v:
res.append([name, cat])
elif name == v:
res.append([name, cat])
user_cat_vals[ucat] = res
mi.user_categories = user_cat_vals
return mi
def get_metadata_(src, encoding=None):
# Meta data definitions as in
# http://www.mobileread.com/forums/showpost.php?p=712544&postcount=9
if isbytestring(src):
if not encoding:
src = xml_to_unicode(src)[0]
else:
src = src.decode(encoding, 'replace')
src = src[:150000] # Searching shouldn't take too long
comment_tags = parse_comment_tags(src)
meta_tags = parse_meta_tags(src)
def get(field):
ans = comment_tags.get(field, meta_tags.get(field, None))
if ans:
ans = ans.strip()
if not ans:
ans = None
return ans
# Title
title = get('title')
if not title:
pat = re.compile('<title>([^<>]+?)</title>', re.IGNORECASE)
match = pat.search(src)
if match:
title = replace_entities(match.group(1))
# Author
authors = get('authors') or _('Unknown')
# Create MetaInformation with Title and Author
mi = Metadata(title or _('Unknown'), string_to_authors(authors))
for field in ('publisher', 'isbn', 'language', 'comments'):
val = get(field)
if val:
setattr(mi, field, val)
for field in ('pubdate', 'timestamp'):
try:
val = parse_date(get(field))
except:
pass
else:
if not is_date_undefined(val):
setattr(mi, field, val)
# SERIES
series = get('series')
if series:
pat = re.compile(r'\[([.0-9]+)\]$')
match = pat.search(series)
series_index = None
if match is not None:
try:
series_index = float(match.group(1))
except:
pass
series = series.replace(match.group(), '').strip()
mi.series = series
if series_index is None:
series_index = get('series_index')
try:
series_index = float(series_index)
except:
pass
if series_index is not None:
mi.series_index = series_index
# RATING
rating = get('rating')
if rating:
try:
mi.rating = float(rating)
if mi.rating < 0:
mi.rating = 0
if mi.rating > 5:
mi.rating /= 2.
if mi.rating > 5:
mi.rating = 0
except:
pass
# TAGS
tags = get('tags')
if tags:
tags = [x.strip() for x in tags.split(',') if x.strip()]
if tags:
mi.tags = tags
return mi
def parse_details(self, root):
try:
goodreads_id = self.parse_goodreads_id(self.url)
except:
self.log.exception("Error parsing goodreads id for url: %r" % self.url)
goodreads_id = None
try:
(title, series, series_index) = self.parse_title_series(root)
except:
self.log.exception("Error parsing title and series for url: %r" % self.url)
title = series = series_index = None
try:
authors = self.parse_authors(root)
except:
self.log.exception("Error parsing authors for url: %r" % self.url)
authors = []
if not title or not authors or not goodreads_id:
self.log.error("Could not find title/authors/goodreads id for %r" % self.url)
self.log.error("Goodreads: %r Title: %r Authors: %r" % (goodreads_id, title, authors))
return
mi = Metadata(title, authors)
if series:
mi.series = series
mi.series_index = series_index
mi.set_identifier("goodreads", goodreads_id)
self.goodreads_id = goodreads_id
try:
isbn = self.parse_isbn(root)
if isbn:
self.isbn = mi.isbn = isbn
except:
self.log.exception("Error parsing ISBN for url: %r" % self.url)
try:
mi.rating = self.parse_rating(root)
except:
self.log.exception("Error parsing ratings for url: %r" % self.url)
try:
mi.comments = self.parse_comments(root)
except:
self.log.exception("Error parsing comments for url: %r" % self.url)
try:
self.cover_url = self.parse_cover(root)
except:
self.log.exception("Error parsing cover for url: %r" % self.url)
mi.has_cover = bool(self.cover_url)
try:
tags = self.parse_tags(root)
if tags:
mi.tags = tags
except:
self.log.exception("Error parsing tags for url: %r" % self.url)
try:
mi.publisher, mi.pubdate = self.parse_publisher_and_date(root)
except:
self.log.exception("Error parsing publisher and date for url: %r" % self.url)
mi.source_relevance = self.relevance
if self.goodreads_id:
if self.isbn:
self.plugin.cache_isbn_to_identifier(self.isbn, self.goodreads_id)
if self.cover_url:
self.plugin.cache_identifier_to_cover_url(self.goodreads_id, self.cover_url)
self.plugin.clean_downloaded_metadata(mi)
self.result_queue.put(mi)
#!/usr/bin/env python
def get_metadata_(src, encoding=None):
# Meta data definitions as in
# https://www.mobileread.com/forums/showpost.php?p=712544&postcount=9
if isbytestring(src):
if not encoding:
src = xml_to_unicode(src)[0]
else:
src = src.decode(encoding, 'replace')
src = src[:150000] # Searching shouldn't take too long
comment_tags, meta_tags, meta_tag_ids, title_tag = parse_metadata(src)
def get_all(field):
ans = comment_tags.get(field, meta_tags.get(field, None))
if ans:
ans = [x.strip() for x in ans if x.strip()]
if not ans:
ans = None
return ans
def get(field):
ans = get_all(field)
if ans:
ans = ans[0]
return ans
# Title
title = get('title') or title_tag.strip() or _('Unknown')
# Author
authors = authors_to_string(get_all('authors')) or _('Unknown')
# Create MetaInformation with Title and Author
mi = Metadata(title, string_to_authors(authors))
# Single-value text fields
for field in ('publisher', 'isbn'):
val = get(field)
if val:
setattr(mi, field, val)
# Multi-value text fields
for field in ('languages', ):
val = get_all(field)
if val:
setattr(mi, field, val)
# HTML fields
for field in ('comments', ):
val = get(field)
if val:
setattr(
mi, field,
val.replace('&', '&').replace('<', '<').replace(
'>', '>').replace('"', '"').replace("'", '''))
# Date fields
for field in ('pubdate', 'timestamp'):
try:
val = parse_date(get(field))
except:
pass
else:
if not is_date_undefined(val):
setattr(mi, field, val)
# SERIES
series = get('series')
if series:
pat = re.compile(r'\[([.0-9]+)\]$')
match = pat.search(series)
series_index = None
if match is not None:
try:
series_index = float(match.group(1))
except:
pass
series = series.replace(match.group(), '').strip()
mi.series = series
if series_index is None:
series_index = get('series_index')
try:
series_index = float(series_index)
except:
pass
if series_index is not None:
mi.series_index = series_index
# RATING
rating = get('rating')
if rating:
try:
mi.rating = float(rating)
if mi.rating < 0:
mi.rating = 0
if mi.rating > 10:
mi.rating = 0
except:
pass
# TAGS
tags = get_all('tags')
if tags:
tags = [x.strip() for s in tags for x in s.split(',') if x.strip()]
if tags:
mi.tags = tags
# IDENTIFIERS
for (k, v) in iteritems(meta_tag_ids):
v = [x.strip() for x in v if x.strip()]
if v:
mi.set_identifier(k, v[0])
return mi
def _get_metadata(self, book_id, get_user_categories=True): # {{{
mi = Metadata(None, template_cache=self.formatter_template_cache)
author_ids = self._field_ids_for('authors', book_id)
aut_list = [self._author_data(i) for i in author_ids]
aum = []
aus = {}
aul = {}
for rec in aut_list:
aut = rec['name']
aum.append(aut)
aus[aut] = rec['sort']
aul[aut] = rec['link']
mi.title = self._field_for('title', book_id,
default_value=_('Unknown'))
mi.authors = aum
mi.author_sort = self._field_for('author_sort', book_id,
default_value=_('Unknown'))
mi.author_sort_map = aus
mi.author_link_map = aul
mi.comments = self._field_for('comments', book_id)
mi.publisher = self._field_for('publisher', book_id)
n = nowf()
mi.timestamp = self._field_for('timestamp', book_id, default_value=n)
mi.pubdate = self._field_for('pubdate', book_id, default_value=n)
mi.uuid = self._field_for('uuid', book_id,
default_value='dummy')
mi.title_sort = self._field_for('sort', book_id,
default_value=_('Unknown'))
mi.book_size = self._field_for('size', book_id, default_value=0)
mi.ondevice_col = self._field_for('ondevice', book_id, default_value='')
mi.last_modified = self._field_for('last_modified', book_id,
default_value=n)
formats = self._field_for('formats', book_id)
mi.format_metadata = {}
mi.languages = list(self._field_for('languages', book_id))
if not formats:
good_formats = None
else:
mi.format_metadata = FormatMetadata(self, book_id, formats)
good_formats = FormatsList(formats, mi.format_metadata)
mi.formats = good_formats
mi.has_cover = _('Yes') if self._field_for('cover', book_id,
default_value=False) else ''
mi.tags = list(self._field_for('tags', book_id, default_value=()))
mi.series = self._field_for('series', book_id)
if mi.series:
mi.series_index = self._field_for('series_index', book_id,
default_value=1.0)
mi.rating = self._field_for('rating', book_id)
mi.set_identifiers(self._field_for('identifiers', book_id,
default_value={}))
mi.application_id = book_id
mi.id = book_id
composites = []
for key, meta in self.field_metadata.custom_iteritems():
mi.set_user_metadata(key, meta)
if meta['datatype'] == 'composite':
composites.append(key)
else:
val = self._field_for(key, book_id)
if isinstance(val, tuple):
val = list(val)
extra = self._field_for(key+'_index', book_id)
mi.set(key, val=val, extra=extra)
for key in composites:
mi.set(key, val=self._composite_for(key, book_id, mi))
user_cat_vals = {}
if get_user_categories:
user_cats = self.backend.prefs['user_categories']
for ucat in user_cats:
res = []
for name,cat,ign in user_cats[ucat]:
v = mi.get(cat, None)
if isinstance(v, list):
if name in v:
res.append([name,cat])
elif name == v:
res.append([name,cat])
user_cat_vals[ucat] = res
mi.user_categories = user_cat_vals
return mi
def parse_response(cls, response, log):
metadata_items = []
tags = []
series = u''
title = u''
translators = []
series_index = 0
authors = []
resp = urllib2.urlopen(response)
page = html.parse(resp)
e = page.getroot().find_class('_ga1_on_').pop()
e.find("noindex").drop_tree()
for i in e.xpath('//ol/li/a/text()'):
tags.append(unicode(i))
for i in e.xpath(u"//div[@class='_ga1_on_']/br[position()=1]/preceding-sibling::a[contains(@href,'/a/')]/text()"):
authors.append(unicode(i))
for i in e.xpath(u"//div[@class='_ga1_on_']/br[position()=1]/preceding-sibling::a[preceding::text()[contains(.,'перевод:')]]/text()"):
translators.append(unicode(i))
for i in e.xpath("(//div[@class='_ga1_on_']/div[@id='z0']/following-sibling::text())[1]"):
title += i
for i in e.xpath("//div[@class='_ga1_on_']/h8"):
series = i.text_content()
series_index = re.findall('\d+', i.tail)[0]
for i in e.xpath("./a[contains(@href,'/s/')]/text()"):
tags.append(unicode(i))
for i in e.xpath("//div[@class='genre']/a/@href"):
tags.append(unicode(i.split('/')[-1]))
log.info(u'Found %s/%s: %s' % (series, series_index, title))
if tags and series in tags:
tags.remove(series)
if translators:
for t in translators:
if t in authors:
authors.remove(t)
metadata_item = Metadata(title, authors)
if tags:
metadata_item.tags = tags
if series != '':
metadata_item.series = series
if series_index is not 0:
metadata_item.series_index = series_index
metadata_items.append(metadata_item)
log.info(series, metadata_item.series)
if u'Игрушечный дом' == series:
log.info('1')
if u'Игрушечный дом' == metadata_item.series:
log.info('2')
if series == metadata_item.series:
log.info('3')
return metadata_items
def parse_details(self, root):
try:
antik_id = self.parse_antik_id(root)
self.log.info('Parsed Antikvarium identifier: %s' % antik_id)
except:
self.log.exception('Error parsing Antikvarium id for url: %r' %
self.url)
antik_id = None
try:
title = self.parse_title(root)
self.log.info('Parsed title: %s' % title)
except:
self.log.exception('Error parsing title for url: %r' % self.url)
title = None
try:
authors = self.parse_authors(root)
self.log.info('Parsed authors: %s' % authors)
except:
self.log.exception('Error parsing authors for url: %r' % self.url)
authors = []
if not title or not authors or not antik_id:
self.log.error(
'Could not find title/authors/Antikvarium.hu id for %r' %
self.url)
self.log.error('Antikvarium.hu id: %r Title: %r Authors: %r' %
(antik_id, title, authors))
return
mi = Metadata(title, authors)
mi.set_identifier('antik_hu', antik_id)
self.antik_id = antik_id
try:
isbn = self.parse_isbn(root)
self.log.info('Parsed ISBN: %s' % isbn)
if isbn:
self.isbn = mi.isbn = isbn
except:
self.log.exception('Error parsing ISBN for url: %r' % self.url)
try:
series = self.parse_series(root)
self.log.info('Parsed series: %s' % series)
except:
self.log.exception('Error parsing series for url: %r' % self.url)
series = None
try:
mi.series_index = self.parse_series_index(root)
self.log.info('Parsed series index: %s' % mi.series_index)
except:
self.log.exception('Error parsing series for url: %r' % self.url)
mi.series_index = None
try:
mi.comments = self.parse_comments(root)
self.log.info('Parsed comments: %s' % mi.comments)
except:
self.log.exception('Error parsing comments for url: %r' % self.url)
try:
self.cover_url = self.parse_cover(root)
self.log.info('Parsed URL for cover: %r' % self.cover_url)
self.plugin.cache_identifier_to_cover_url(self.antik_id,
self.cover_url)
mi.has_cover = bool(self.cover_url)
except:
self.log.exception('Error parsing cover for url: %r' % self.url)
try:
mi.publisher = self.parse_publisher(root)
self.log.info('Parsed publisher: %s' % mi.publisher)
except:
self.log.exception('Error parsing publisher for url: %r' %
self.url)
try:
mi.tags = self.parse_tags(root)
self.log.info('Parsed tags: %s' % mi.tags)
except:
self.log.exception('Error parsing tags for url: %r' % self.url)
try:
mi.pubdate = self.parse_published_date(root)
self.log.info('Parsed publication date: %s' % mi.pubdate)
except:
self.log.exception('Error parsing published date for url: %r' %
self.url)
try:
mi.languages = self.parse_languages(root)
self.log.info('Parsed languages: %r' % mi.languages)
except:
self.log.exception('Error parsing languages for url: %r' %
self.url)
mi.source_relevance = self.relevance
if series:
mi.series = series
if self.antik_id and self.isbn:
self.plugin.cache_isbn_to_identifier(self.isbn, self.antik_id)
self.plugin.clean_downloaded_metadata(mi)
self.result_queue.put(mi)
def parse_details(self, root):
try:
legie_id = self.parse_legie_id(self.url)
except:
self.log.exception('Error parsing Legie id for url: %r' % self.url)
legie_id = None
try:
title = self.parse_title(root)
except:
self.log.exception('Error parsing title for url: %r' % self.url)
title = None
try:
authors = self.parse_authors(root)
except:
self.log.exception('Error parsing authors for url: %r' % self.url)
authors = []
if not title or not authors or not legie_id:
self.log.error('Could not find title/authors/Legie id for %r' %
self.url)
self.log.error('Legie: %r Title: %r Authors: %r' %
(legie_id, title, authors))
return
self.legie_id = legie_id
rating = comments = series = series_index = None
try:
rating = self.parse_rating(root)
except:
self.log.exception('Error parsing ratings for url: %r' % self.url)
try:
comments = self.parse_comments(root)
except:
self.log.exception('Error parsing comments for url: %r' % self.url)
try:
(series, series_index) = self.parse_series(root)
except:
self.log.info('Series not found.')
try:
tags = self.parse_tags(root)
except:
self.log.exception('Error parsing tags for url: %r' % self.url)
tags = None
if legie_id:
editions = self.get_editions()
if editions:
num_editions = len(editions)
self.log.info('Nalezeno %d vydani' % num_editions)
for edition in editions:
(year, cover_url, publisher, isbn) = edition
mi = Metadata(title, authors)
self.legie_id = "%s#%s" % (legie_id, year)
mi.set_identifier('legie', self.legie_id)
mi.source_relevance = self.relevance
mi.rating = rating
mi.comments = comments
mi.series = series
mi.series_index = series_index
if cover_url:
mi.cover_url = self.cover_url = cover_url
self.plugin.cache_identifier_to_cover_url(
self.legie_id, self.cover_url)
if tags:
mi.tags = tags
mi.has_cover = bool(self.cover_url)
mi.publisher = publisher
mi.isbn = isbn
mi.pubdate = self.prepare_date(int(year))
mi.language = "ces"
self.result_queue.put(mi)
else:
mi = Metadata(title, authors)
mi.set_identifier('legie', self.legie_id)
mi.source_relevance = self.relevance
mi.rating = rating
mi.comments = comments
mi.series = series
mi.series_index = series_index
try:
self.cover_url = self.parse_cover(root)
except:
self.log.exception('Error parsing cover for url: %r' %
self.url)
if tags:
mi.tags = tags
mi.has_cover = bool(self.cover_url)
mi.publisher = publisher
mi.isbn = isbn
mi.pubdate = self.prepare_date(int(year))
mi.language = "ces"
self.result_queue.put(mi)
if self.legie_id:
if self.cover_url:
self.plugin.cache_identifier_to_cover_url(
self.legie_id, self.cover_url)
def parse_details(self, root):
try:
yes24_id = self.parse_yes24_id(self.url)
except:
self.log.exception('Error parsing YES24 id for url: %r'%self.url)
yes24_id = None
try:
(title, series, series_index) = self.parse_title_series(root)
except:
self.log.exception('Error parsing title and series for url: %r'%self.url)
title = series = series_index = None
try:
authors = self.parse_authors(root)
except:
self.log.exception('Error parsing authors for url: %r'%self.url)
authors = []
if not title or not authors or not yes24_id:
self.log.error('Could not find title/authors/YES24 id for %r'%self.url)
self.log.error('YES24: %r Title: %r Authors: %r'%(yes24_id, title,
authors))
return
mi = Metadata(title, authors)
if series:
mi.series = series
mi.series_index = series_index
mi.set_identifier('yes24', yes24_id)
self.yes24_id = yes24_id
try:
isbn = self.parse_isbn(root)
if isbn:
self.isbn = mi.isbn = isbn
except:
self.log.exception('Error parsing ISBN for url: %r'%self.url)
try:
mi.comments = self.parse_comments(root)
except:
self.log.exception('Error parsing comments for url: %r'%self.url)
try:
self.cover_url = self.parse_cover(root)
except:
self.log.exception('Error parsing cover for url: %r'%self.url)
mi.has_cover = bool(self.cover_url)
mi.cover_url = self.cover_url # This is purely so we can run a test for it!!!
try:
mi.publisher = self.parse_publisher(root)
except:
self.log.exception('Error parsing publisher for url: %r'%self.url)
try:
mi.pubdate = self.parse_published_date(root)
except:
self.log.exception('Error parsing published date for url: %r'%self.url)
mi.language = 'ko'
mi.source_relevance = self.relevance
if self.yes24_id:
if self.isbn:
self.plugin.cache_isbn_to_identifier(self.isbn, self.yes24_id)
self.plugin.clean_downloaded_metadata(mi)
self.result_queue.put(mi)
def _GoodreadsBook_to_Metadata(self, book):
# type: (_GoodreadsBook) -> Metadata
"""
:param book: _GoodreadsBook: book
:return: Metadata: Metadata
"""
mi = Metadata(book.title, book.authors)
mi.source_relevance = 0
mi.set_identifier('goodreads', book.id)
if self.prefs['NEVER_REPLACE_ISBN'] and mi.get_identifiers().get(
'isbn'):
mi.set_identifier('isbn', '')
if book.asin and not self.prefs['NEVER_REPLACE_AMAZONID']:
mi.set_identifier('amazon', book.asin)
if book.isbn and not self.prefs['NEVER_REPLACE_ISBN']:
try:
if len(book.isbn) == 10:
mi.isbn = check_isbn13(_ISBNConvert.convert(book.isbn))
else:
mi.isbn = check_isbn13(book.isbn)
except:
self.log.error("ISBN CONVERSION ERROR:", book.isbn)
self.log.exception()
if book.image_url:
self.log.info('cache_identifier_to_cover_url:', book.asin, ':',
book.image_url)
self.cache_identifier_to_cover_url(book.id, book.image_url)
if book.publisher:
self.log.info('book.publisher is:', book.publisher)
mi.publisher = book.publisher
if book.pubdate:
self.log.info('book.pubdate is:',
book.pubdate.strftime('%Y-%m-%d'))
mi.pubdate = book.pubdate
if book.comments:
self.log.info('book.editorial_review is:', book.comments)
mi.comments = book.comments
tags = self.prefs['ADD_THESE_TAGS'].split(',')
tags.extend(book.tags)
# tag_mappings = JSONConfig('plugins/GenreMappings')['genreMappings']
# mi.tags = list(set(sorted(filter(lambda x: tag_mappings.get(x, x), tags))))
if book.series:
mi.series = book.series
self.log.info(u'series:', book.series)
if book.series_index:
mi.series_index = book.series_index
self.log.info(u'series_index:',
"{0:.2f}".format(book.series_index))
else:
mi.series_index = 0
if book.average_rating:
mi.rating = book.average_rating
self.clean_downloaded_metadata(mi)
return mi
def get_details(self):
'''
The get_details() function for stripping the website for all information
'''
self.log.info(" Worker.get_details:")
self.log.info(" self: ", self)
self.log.info(" self.url: ", self.url)
# Parse the html code from the website
try:
raw = self.browser.open_novisit(
self.url, timeout=self.timeout).read().strip()
# Do some error handling if it fails to read data
except Exception as e:
if callable(getattr(e, 'getcode', None)) and e.getcode() == 404:
self.log.error('URL malformed: %r' % self.url)
return
attr = getattr(e, 'args', [None])
attr = attr if attr else [None]
if isinstance(attr[0], socket.timeout):
msg = 'Bookmeta for saxo timed out. Try again later.'
self.log.error(msg)
else:
msg = 'Failed to make details query: %r' % self.url
self.log.exception(msg)
return
# Do some error handling if the html code returned 404
if "<title>404 - " == raw:
self.log.error('URL malformed: %r' % self.url)
return
# Clean the html data a little
try:
root = parse(raw)
except:
self.log.error("Error cleaning HTML")
return
# Get the title of the book
try:
title_node = root.xpath('//span[@itemprop="name"]')
self.title = title_node[0].text
except:
self.log.exception('Error parsing title for url: %r' % self.url)
# Get the author of the book
try:
author_node = root.xpath('//span[@class="expandAuthorName"]')
author_strings = author_node[0].text.split(",")
#print(author_strings)
for name in author_strings:
self.authors.append(name)
except:
self.log.exception('Error parsing authors for url: %r' % self.url)
self.authors = None
# Get the series of the book
try:
series_node = root.xpath('//b[contains(text(), "Serie")]/a')
if len(series_node) > 0:
self.series = series_node[0].text.split(": ")[0].strip()
self.series_index = series_node[0].text.split(": ")[-1].strip()
# print("'%s'" % self.series)
# print("'%s'" % self.series_index)
except:
self.log.exception('Error parsing series for url: %r' % self.url)
# Some books have ratings, let's use them.
try:
self.rating = 0.0
except:
self.log.exception('Error parsing rating for url: %r' % self.url)
self.rating = 0.0
# Get the ISBN number from the site
try:
isbn_node = root.xpath(
'//div[@class="eBookContainer"]/b/span[@itemprop="identifier"]'
)
if len(isbn_node) > 0:
self.isbn = isbn_node[0].text.replace("ISBN: ", "").strip()
except:
self.log.exception('Error parsing isbn for url: %r' % self.url)
self.isbn = None
# Get the comments/blurb for the book
try:
comment_node = root.xpath('//meta[@name="description"]/@content')
self.comments = comment_node[0]
except:
self.log.exception('Error parsing comments for url: %r' % self.url)
self.comments = None
# Parse the cover url for downloading the cover.
try:
cover_node = root.xpath(
'//div[@class="bookDetailCoverCover"]/img/@src')
self.cover_url = "https://mofibo.com" + cover_node[0]
self.log.info(' Parsed URL for cover: %r' % self.cover_url)
self.plugin.cache_identifier_to_cover_url(self.isbn,
self.cover_url)
except:
self.log.exception('Error parsing cover for url: %r' % self.url)
self.has_cover = bool(self.cover_url)
# Get the publisher name
try:
publisher_node = root.xpath(
'//div[@class="eBookContainer"]/b/span/a[@itemprop="brand"]')
if len(publisher_node) > 0:
self.publisher = publisher_node[0].text
except:
self.log.exception('Error parsing publisher for url: %r' %
self.url)
# Get the language of the book. Only english and danish are supported tho
try:
language_node = root.xpath('//b[@class="expanderLanguage"]')
language = language_node[0].text.strip().replace("Sprog:",
"").replace(
" ", "")
language = self.lang_map.get(language, None)
self.language = language
except:
self.log.exception('Error parsing language for url: %r' % self.url)
# Get the publisher date
try:
pubdate_node = root.xpath(
'//div[@class="eBookContainer"]/b[contains(text(),"Udgivet:")]'
)
if len(pubdate_node) > 0:
date_str = pubdate_node[0].text.replace("Udgivet:", "").strip()
format_str = '%Y-%m-%d' # The format
self.pubdate = datetime.datetime.strptime(date_str, format_str)
except:
self.log.exception('Error parsing published date for url: %r' %
self.url)
# Get the tags
try:
tags = []
tags_node = root.xpath('//span[@itemprop="category"]')
tags.append(tags_node[0].text.strip())
self.tags = tags
except:
self.log.exception('Error parsing tags for url: %r' % self.url)
# Setup the metadata
meta_data = Metadata(self.title, self.authors)
meta_data.set_identifier('isbn', self.isbn)
meta_data.set_identifier('mofibo', self.url)
# Set rating
if self.series:
try:
meta_data.series = self.series
meta_data.series_index = self.series_index
except:
self.log.exception('Error loading series')
# Set ISBN
if self.isbn:
try:
meta_data.isbn = self.isbn
except:
self.log.exception('Error loading ISBN')
# Set relevance
if self.relevance:
try:
meta_data.source_relevance = self.relevance
except:
self.log.exception('Error loading relevance')
# Set cover url
if self.cover_url:
try:
meta_data.cover_url = self.cover_url
except:
self.log.exception('Error loading cover_url')
# Set publisher
if self.publisher:
try:
meta_data.publisher = self.publisher
except:
self.log.exception('Error loading publisher')
# Set language
if self.language:
try:
meta_data.language = self.language
except:
self.log.exception('Error loading language')
# Set comments/blurb
if self.comments:
try:
meta_data.comments = self.comments
except:
self.log.exception("Error loading comments")
# Set publisher data
if self.pubdate:
try:
meta_data.pubdate = self.pubdate
except:
self.log.exception('Error loading pubdate')
# Set tags data
if self.tags:
try:
meta_data.tags = self.tags
except:
self.log.exception('Error loading tags')
# Put meta data
self.plugin.clean_downloaded_metadata(meta_data)
self.result_queue.put(meta_data)
def identify(self, log, result_queue, abort, title=None, authors=[], identifiers={}, timeout=30):
self.load_config()
if authors is None:
authors=[]
# get identifying tags from book
idn = identifiers.get('dnb-idn', None)
isbn = check_isbn(identifiers.get('isbn', None))
# ignore unknown authors
ignored_authors = [ "V. A.", "V.A.", "Unknown", "Unbekannt" ]
for i in ignored_authors:
authors = [ x for x in authors if x != i ]
if (isbn is None) and (idn is None) and (title is None) and (authors is None):
log.info("This plugin requires at least either ISBN, IDN, Title or Author(s).")
return None
queries=[]
# DNB does not do an exact search when searching for a idn or isbn, so we have to filter the results
exact_search = {}
if idn is not None:
exact_search['idn'] = idn
# in case look for a IDN only search for the IDN and skip all the other stuff
queries.append('num='+idn)
else:
authors_v = []
title_v = []
# create some variants of given authors
if authors != []:
authors_v.append(' '.join(self.get_author_tokens(authors,only_first_author=False))) # concat all author names ("Peter Meier Luise Stark")
authors_v.append(' '.join(self.get_author_tokens(authors,only_first_author=True))) # use only first author
for a in authors:
authors_v.append(a) # use all authors, one by one
# remove duplicates
unique_authors_v = []
for i in authors_v:
if i not in unique_authors_v:
unique_authors_v.append(i)
# create some variants of given title
if title is not None:
title_v.append(title) # simply use given title
title_v.append(' '.join(self.get_title_tokens(title,strip_joiners=False,strip_subtitle=False))) # remove some punctation characters
title_v.append(' '.join(self.get_title_tokens(title,strip_joiners=False,strip_subtitle=True))) # remove subtitle (everything after " : ")
title_v.append(' '.join(self.get_title_tokens(title,strip_joiners=True,strip_subtitle=False))) # remove some punctation characters and joiners ("and", "&", ...)
title_v.append(' '.join(self.get_title_tokens(title,strip_joiners=True,strip_subtitle=True))) # remove subtitle (everything after " : ") and joiners ("and", "&", ...)
# TODO: remove subtitle after " - "
# remove duplicates
unique_title_v = []
for i in title_v:
if i not in unique_title_v:
unique_title_v.append(i)
# title and author
if authors_v != [] and title_v != []:
for a in authors_v:
for t in title_v:
if isbn is not None:
queries.append('tit="' + t + '" AND per="' + a + '" AND num="' + isbn + '"')
else:
queries.append('tit="' + t + '" AND per="' + a + '"')
# try with first author as title and title (without subtitle) as author
if isbn is not None:
queries.append('per="' + ' '.join(self.get_title_tokens(title,strip_joiners=True,strip_subtitle=True)) + '" AND tit="' + ' '.join(self.get_author_tokens(authors,only_first_author=True)) + '" AND num="'+isbn+'"')
else:
queries.append('per="' + ' '.join(self.get_title_tokens(title,strip_joiners=True,strip_subtitle=True)) + '" AND tit="' + ' '.join(self.get_author_tokens(authors,only_first_author=True)) + '"')
# try with author and title (without subtitle) in any index
if isbn is not None:
queries.append('"' + ' '.join(self.get_title_tokens(title,strip_joiners=True,strip_subtitle=True)) + '" AND "' + ' '.join(self.get_author_tokens(authors,only_first_author=True)) + '" AND num="'+isbn+'"')
else:
queries.append('"' + ' '.join(self.get_title_tokens(title,strip_joiners=True,strip_subtitle=True)) + '" AND "' + ' '.join(self.get_author_tokens(authors,only_first_author=True)) + '"')
# author but no title
elif authors_v != [] and title_v == []:
for i in authors_v:
if isbn is not None:
queries.append('per="'+ i +'" AND num="' + isbn + '"')
else:
queries.append('per="'+ i +'"')
# try with author as title
if isbn is not None:
queries.append('tit="' + ' '.join(self.get_author_tokens(authors,only_first_author=True)) + '" AND num="' + isbn + '"')
else:
queries.append('tit="' + ' '.join(self.get_author_tokens(authors,only_first_author=True)) + '"')
# title but no author
elif authors_v == [] and title_v != []:
for i in title_v:
if isbn is not None:
queries.append('tit="' + i + '" AND num="' + isbn + '"')
else:
queries.append('tit="' + i + '"')
# try with title as author
if isbn is not None:
queries.append('per="' + ' '.join(self.get_title_tokens(title,strip_joiners=True,strip_subtitle=True)) + '" AND num="' + isbn + '"')
else:
queries.append('per="' + ' '.join(self.get_title_tokens(title,strip_joiners=True,strip_subtitle=True)) + '"')
# as last resort only use isbn
if isbn is not None:
queries.append('num=' + isbn)
# remove duplicate queries
uniqueQueries=[]
for i in queries:
if i not in uniqueQueries:
uniqueQueries.append(i)
# Process queries
results = None
for query in uniqueQueries:
# SRU does not work with "+" or "?" characters in query, so we simply remove them
query = re.sub('[\+\?]','',query)
query = query + ' NOT (mat=film OR mat=music OR mat=microfiches OR cod=tt)'
log.info(query)
if self.cfg_dnb_token is None:
results = self.getSearchResultsByScraping(log, query, timeout)
else:
results = self.getSearchResults(log, query, timeout)
if results is None:
continue
log.info("Parsing records")
ns = { 'marc21' : 'http://www.loc.gov/MARC21/slim' }
for record in results:
series = None
series_index = None
publisher = None
pubdate = None
languages = []
title = None
title_sort = None
authors = []
author_sort = None
edition = None
comments = None
idn = None
urn = None
isbn = None
ddc = []
subjects_gnd = []
subjects_non_gnd = []
publisher_name = None
publisher_location = None
##### Field 264 #####
# Publisher Name and Location
fields = record.xpath(".//marc21:datafield[@tag='264']/marc21:subfield[@code='b' and string-length(text())>0]/../marc21:subfield[@code='a' and string-length(text())>0]/..",namespaces=ns)
if len(fields)>0:
publisher_name = fields[0].xpath(".//marc21:subfield[@code='b' and string-length(text())>0]",namespaces=ns)[0].text.strip();
publisher_location = fields[0].xpath(".//marc21:subfield[@code='a' and string-length(text())>0]",namespaces=ns)[0].text.strip();
else:
fields = record.xpath(".//marc21:datafield[@tag='264']/marc21:subfield[@code='b' and string-length(text())>0]/../..",namespaces=ns)
if len(fields)>0:
publisher_name = fields[0].xpath(".//marc21:subfield[@code='b' and string-length(text())>0]",namespaces=ns)[0].text.strip();
else:
fields = record.xpath(".//marc21:datafield[@tag='264']/marc21:subfield[@code='a' and string-length(text())>0]/../..",namespaces=ns)
if len(fields)>0:
publisher_location = fields[0].xpath(".//marc21:subfield[@code='a' and string-length(text())>0]",namespaces=ns)[0].text.strip();
# Publishing Date
for i in record.xpath(".//marc21:datafield[@tag='264']/marc21:subfield[@code='c' and string-length(text())>=4]",namespaces=ns):
match = re.search("(\d{4})", i.text.strip())
if match is not None:
year = match.group(1)
pubdate = datetime.datetime(int(year), 1, 1, 12 , 30, 0)
break
# Log
if publisher_name is not None:
log.info("Extracted Publisher: %s" % publisher_name)
if publisher_location is not None:
log.info("Extracted Publisher Location: %s" % publisher_location)
if pubdate is not None:
log.info("Extracted Publication Year: %s" % pubdate)
##### Field 245 ####
# Title/Series/Series_Index
title_parts = []
for i in record.xpath(".//marc21:datafield[@tag='245']/marc21:subfield[@code='a' and string-length(text())>0]/..",namespaces=ns):
# if a,n,p,n,p,n,p exist: series = a + n0 + " - " + p0 + n1 + " - " + p1, series_index = n2, title = p2
# if a,n,p,n,p exist: series = a + n0 + " - " + p0, series_index = n1, title = p1 (Example: dnb-id 1008774839)
# if a,n,p exist: series = a, series_index = n, title = p
# if a exist: title = a
# TODO: a,n,p,n (i.e. 956375146)
code_p = []
code_n = []
code_a = []
for j in i.xpath(".//marc21:subfield[@code='p']",namespaces=ns):
code_p.append(j.text.strip())
for j in i.xpath(".//marc21:subfield[@code='n']",namespaces=ns):
match = re.search("(\d+[,\.\d+]?)", j.text.strip())
if match:
code_n.append(match.group(1))
else:
code_n.append("0") # looks like sometimes DNB does not know the series index and uses something like "[...]"
for j in i.xpath(".//marc21:subfield[@code='a']",namespaces=ns):
code_a.append(j.text.strip())
if len(code_p) == 0:
title_parts = title_parts + code_a
elif len(code_p)>0 and len(code_p) == len(code_n):
series = " : ".join(code_a) # I've never seen more than one code_a, but who knows...
for i in range (0,len(code_p)-1):
series = series + " " + code_n[i] + " " + code_p[i]
series_index = code_n[-1]
title_parts.append(code_p[-1])
# subtitle 1: Field 245
for i in record.xpath(".//marc21:datafield[@tag='245']/marc21:subfield[@code='b' and string-length(text())>0]",namespaces=ns):
title_parts.append(i.text.strip())
break
# subtitle 2
#for i in record.xpath(".//marc21:datafield[@tag='245']/marc21:subfield[@code='c' and string-length(text())>0]",namespaces=ns):
# title = title + " / " + i.text.strip()
# break
title = " : ".join(title_parts)
# Log
if series_index is not None:
log.info("Extracted Series_Index from Field 245: %s" % series_index)
if series is not None:
log.info("Extracted Series from Field 245: %s" % series)
series = self.cleanUpSeries(log, series, publisher_name)
if title is not None:
log.info("Extracted Title: %s" % title)
title = self.cleanUpTitle(log, title)
# Title_Sort
if len(title_parts)>0:
title_sort_parts = list(title_parts)
title_sort_regex = re.match('^(.*?)('+chr(152)+'.*'+chr(156)+')?(.*?)$',title_parts[0])
sortword = title_sort_regex.group(2)
if sortword:
title_sort_parts[0] = ''.join(filter(None,[title_sort_regex.group(1).strip(),title_sort_regex.group(3).strip(),", "+sortword]))
title_sort = " : ".join(title_sort_parts)
# Log
if title_sort is not None:
log.info("Extracted Title_Sort: %s" % title_sort)
##### Field 100 and Field 700 #####
# Authors
for i in record.xpath(".//marc21:datafield[@tag='100']/marc21:subfield[@code='4' and text()='aut']/../marc21:subfield[@code='a' and string-length(text())>0]",namespaces=ns): # primary authors
name = re.sub(" \[.*\]$","",i.text.strip());
authors.append(name)
for i in record.xpath(".//marc21:datafield[@tag='700']/marc21:subfield[@code='4' and text()='aut']/../marc21:subfield[@code='a' and string-length(text())>0]",namespaces=ns): # secondary authors
name = re.sub(" \[.*\]$","",i.text.strip());
authors.append(name)
if len(authors)==0: # if no "real" autor was found take all persons involved
for i in record.xpath(".//marc21:datafield[@tag='700']/marc21:subfield[@code='a' and string-length(text())>0]",namespaces=ns): # secondary authors
name = re.sub(" \[.*\]$","",i.text.strip());
authors.append(name)
if len(authors)>0:
author_sort = authors[0]
# Log
if len(authors)>0:
log.info("Extracted Authors: %s" % " & ".join(authors))
if author_sort is not None:
log.info("Extracted Author_Sort: %s" % " & ".join(authors))
##### Field 856 #####
# Comments
for i in record.xpath(".//marc21:datafield[@tag='856']/marc21:subfield[@code='u' and string-length(text())>0]",namespaces=ns):
if i.text.startswith("http://deposit.dnb.de/"):
br = self.browser
log.info('Downloading Comments from: %s' % i.text)
try:
comments = br.open_novisit(i.text, timeout=30).read()
comments = re.sub('(\s|<br>|<p>|\n)*Angaben aus der Verlagsmeldung(\s|<br>|<p>|\n)*(<h3>.*?</h3>)?(\s|<br>|<p>|\n)*','',comments,flags=re.IGNORECASE)
comments = sanitize_comments_html(comments)
break
except:
log.info("Could not download Comments from %s" % i)
# Log
if comments is not None:
log.info('Comments: %s' % comments)
# If no comments are found for this edition, look at other editions of this book (Fields 776)
# TODO: Make this configurable (default: yes)
if comments is None:
# get all other issues
for i in record.xpath(".//marc21:datafield[@tag='776']/marc21:subfield[@code='w' and string-length(text())>0]",namespaces=ns):
other_idn = re.sub("^\(.*\)","",i.text.strip());
subquery = 'num='+other_idn+' NOT (mat=film OR mat=music OR mat=microfiches OR cod=tt)'
log.info(subquery)
if self.cfg_dnb_token is None:
subresults = self.getSearchResultsByScraping(log, subquery, timeout)
else:
subresults = self.getSearchResults(log, subquery, timeout)
if subresults is None:
continue
for subrecord in subresults:
for i in subrecord.xpath(".//marc21:datafield[@tag='856']/marc21:subfield[@code='u' and string-length(text())>0]",namespaces=ns):
if i.text.startswith("http://deposit.dnb.de/"):
br = self.browser
log.info('Downloading Comments from: %s' % i.text)
try:
comments = br.open_novisit(i.text, timeout=30).read()
comments = re.sub('(\s|<br>|<p>|\n)*Angaben aus der Verlagsmeldung(\s|<br>|<p>|\n)*(<h3>.*?</h3>)?(\s|<br>|<p>|\n)*','',comments,flags=re.IGNORECASE)
comments = sanitize_comments_html(comments)
break
except:
log.info("Could not download Comments from %s" % i)
if comments is not None:
log.info('Comments from other issue: %s' % comments)
break
##### Field 16 #####
# ID: IDN
for i in record.xpath(".//marc21:datafield[@tag='016']/marc21:subfield[@code='a' and string-length(text())>0]",namespaces=ns):
idn = i.text.strip()
break
# Log
if idn is not None:
log.info("Extracted ID IDN: %s" % idn)
##### Field 24 #####
# ID: URN
for i in record.xpath(".//marc21:datafield[@tag='024']/marc21:subfield[@code='2' and text()='urn']/../marc21:subfield[@code='a' and string-length(text())>0]",namespaces=ns):
urn = i.text.strip()
break
# Log
if urn is not None:
log.info("Extracted ID URN: %s" % urn)
##### Field 20 #####
# ID: ISBN
for i in record.xpath(".//marc21:datafield[@tag='020']/marc21:subfield[@code='a' and string-length(text())>0]",namespaces=ns):
isbn_regex = "(?:ISBN(?:-1[03])?:? )?(?=[-0-9 ]{17}|[-0-9X ]{13}|[0-9X]{10})(?:97[89][- ]?)?[0-9]{1,5}[- ]?(?:[0-9]+[- ]?){2}[0-9X]"
match = re.search(isbn_regex, i.text.strip())
isbn = match.group()
isbn = isbn.replace('-','')
break
# Log
if isbn is not None:
log.info("Extracted ID ISBN: %s" % isbn)
# When doing an exact search for a given ISBN skip books with wrong ISBNs
if isbn is not None and "isbn" in exact_search:
if isbn != exact_search["isbn"]:
log.info("Extracted ISBN does not match book's ISBN, skipping record")
continue
##### Field 82 #####
# ID: Sachgruppe (DDC)
for i in record.xpath(".//marc21:datafield[@tag='082']/marc21:subfield[@code='a' and string-length(text())>0]",namespaces=ns):
ddc.append(i.text.strip())
# Log
if len(ddc)>0:
log.info("Extracted ID DDC: %s" % ",".join(ddc))
##### Field 490 #####
# In theory this field is not used for "real" book series, use field 830 instead. But it is used.
# Series and Series_Index
if series is None or (series is not None and series_index == "0"):
for i in record.xpath(".//marc21:datafield[@tag='490']/marc21:subfield[@code='v' and string-length(text())>0]/../marc21:subfield[@code='a' and string-length(text())>0]/..",namespaces=ns):
# "v" either "Nr. 220" or "This great Seriestitle : Nr. 220" - if available use this instead of attribute a
attr_v = i.xpath(".//marc21:subfield[@code='v']",namespaces=ns)[0].text.strip()
parts = re.split(" : ",attr_v)
if len(parts)==2:
if bool(re.search("\d",parts[0])) != bool(re.search("\d",parts[1])):
# figure out which part contains the index
if bool(re.search("\d",parts[0])):
indexpart = parts[0]
textpart = parts[1]
else:
indexpart = parts[1]
textpart = parts[0]
match = re.search("(\d+[,\.\d+]?)", indexpart)
if match is not None:
series_index = match.group(1)
series = textpart.strip()
else:
match = re.search("(\d+[,\.\d+]?)", attr_v)
if match is not None:
series_index = match.group(1)
else:
series_index = "0"
series_index = series_index.replace(',','.')
# Use Series Name from attribute "a" if not already found in attribute "v"
if series is None:
series = i.xpath(".//marc21:subfield[@code='a']",namespaces=ns)[0].text.strip()
# Log
if series_index is not None:
log.info("Extracted Series Index from Field 490: %s" % series_index)
if series is not None:
log.info("Extracted Series from Field 490: %s" % series)
series = self.cleanUpSeries(log, series, publisher_name)
if series is not None:
break
##### Field 246 #####
# Series and Series_Index
if series is None or (series is not None and series_index == "0"):
for i in record.xpath(".//marc21:datafield[@tag='246']/marc21:subfield[@code='a' and string-length(text())>0]",namespaces=ns):
match = re.search("^(.+?) ; (\d+[,\.\d+]?)$", i.text.strip())
if match is not None:
series = match.group(1)
series_index = match.group(2)
# Log
if series_index is not None:
log.info("Extracted Series Index from Field 246: %s" % series_index)
if series is not None:
log.info("Extracted Series from Field 246: %s" % series)
series = self.cleanUpSeries(log, series, publisher_name)
if series is not None:
break
##### Field 800 #####
# Series and Series_Index
if series is None or (series is not None and series_index == "0"):
for i in record.xpath(".//marc21:datafield[@tag='800']/marc21:subfield[@code='v' and string-length(text())>0]/../marc21:subfield[@code='t' and string-length(text())>0]/..",namespaces=ns):
# Series Index
series_index = i.xpath(".//marc21:subfield[@code='v']",namespaces=ns)[0].text.strip()
match = re.search("(\d+[,\.\d+]?)", series_index)
if match is not None:
series_index = match.group(1)
else:
series_index = "0"
series_index = series_index.replace(',','.')
# Series
series = i.xpath(".//marc21:subfield[@code='t']",namespaces=ns)[0].text.strip()
# Log
if series_index is not None:
log.info("Extracted Series Index from Field 800: %s" % series_index)
if series is not None:
log.info("Extracted Series from Field 800: %s" % series)
series = self.cleanUpSeries(log, series, publisher_name)
if series is not None:
break
##### Field 830 #####
# Series and Series_Index
if series is None or (series is not None and series_index == "0"):
for i in record.xpath(".//marc21:datafield[@tag='830']/marc21:subfield[@code='v' and string-length(text())>0]/../marc21:subfield[@code='a' and string-length(text())>0]/..",namespaces=ns):
# Series Index
series_index = i.xpath(".//marc21:subfield[@code='v']",namespaces=ns)[0].text.strip()
match = re.search("(\d+[,\.\d+]?)", series_index)
if match is not None:
series_index = match.group(1)
else:
series_index = "0"
series_index = series_index.replace(',','.')
# Series
series = i.xpath(".//marc21:subfield[@code='a']",namespaces=ns)[0].text.strip()
# Log
if series_index is not None:
log.info("Extracted Series Index from Field 830: %s" % series_index)
if series is not None:
log.info("Extracted Series from Field 830: %s" % series)
series = self.cleanUpSeries(log, series, publisher_name)
if series is not None:
break
##### Field 689 #####
# GND Subjects
for i in record.xpath(".//marc21:datafield[@tag='689']/marc21:subfield[@code='a' and string-length(text())>0]",namespaces=ns):
subjects_gnd.append(i.text.strip())
for f in range(600,656):
for i in record.xpath(".//marc21:datafield[@tag='"+str(f)+"']/marc21:subfield[@code='2' and text()='gnd']/../marc21:subfield[@code='a' and string-length(text())>0]",namespaces=ns):
if i.text.startswith("("):
continue
subjects_gnd.append(i.text)
# Log
if len(subjects_gnd)>0:
log.info("Extracted GND Subjects: %s" % " ".join(subjects_gnd))
##### Fields 600-655 #####
# Non-GND subjects
for f in range(600,656):
for i in record.xpath(".//marc21:datafield[@tag='"+str(f)+"']/marc21:subfield[@code='a' and string-length(text())>0]",namespaces=ns):
# ignore entries starting with "(":
if i.text.startswith("("):
continue
subjects_non_gnd.extend(re.split(',|;',i.text))
# remove one-character subjects:
for i in subjects_non_gnd:
if len(i)<2:
subjects_non_gnd.remove(i)
# Log
if len(subjects_non_gnd)>0:
log.info("Extracted non-GND Subjects: %s" % " ".join(subjects_non_gnd))
##### Field 250 #####
# Edition
for i in record.xpath(".//marc21:datafield[@tag='250']/marc21:subfield[@code='a' and string-length(text())>0]",namespaces=ns):
edition = i.text.strip()
break
# Log
if edition is not None:
log.info("Extracted Edition: %s" % edition)
##### Field 41 #####
# Languages
for i in record.xpath(".//marc21:datafield[@tag='041']/marc21:subfield[@code='a' and string-length(text())>0]",namespaces=ns):
languages.append(i.text.strip())
# Log
if languages is not None:
log.info("Extracted Languages: %s" % ",".join(languages))
##### If configured: Try to separate Series, Series Index and Title from the fetched title #####
#if self.cfg_guess_series is True:
if (series is None or (series is not None and series_index == "0")) and self.cfg_guess_series is True:
guessed_series = None
guessed_series_index = None
guessed_title = None
log.info("Starting Series Guesser")
parts = re.split("[:]",self.removeSortingCharacters(title))
if len(parts)==2:
log.info("Title has two parts")
# make sure only one part of the two parts contains digits
if bool(re.search("\d",parts[0])) != bool(re.search("\d",parts[1])):
log.info("only one title part contains digits")
# figure out which part contains the index
if bool(re.search("\d",parts[0])):
indexpart = parts[0]
textpart = parts[1]
else:
indexpart = parts[1]
textpart = parts[0]
# Look at the part without digits:
match = re.match("^[\s\-–:]*(.+?)[\s\-–:]*$",textpart) # remove odd characters from start and end of the text part
if match:
textpart = match.group(1)
# Look at the part with digits:
# for Titleparts like: "Name of the series - Episode 2"
match = re.match("^\s*(\S\D*?[a-zA-Z]\D*?)\W[\(\/\.,\s\-–:]*(?:#|Nr\.|Episode|Bd\.|Sammelband|[B|b]and|Part|Teil|Folge)[,\-–:\s#\(]*(\d+\.?\d*)[\)\s\-–:]*$",indexpart)
if match:
guessed_series_index = match.group(2)
guessed_series = match.group(1)
if guessed_series is None:
guessed_series = textpart
guessed_title = textpart + " : Band " + guessed_series_index
else:
guessed_title = textpart
#log.info("ALGO1: guessed_title: " + guessed_title)
#log.info("ALGO1: guessed_series: " + guessed_series)
#log.info("ALGO1: guessed_series_index: " + guessed_series_index)
else:
# for Titleparts like: "Episode 2 Name of the series"
match = re.match("^\s*(?:#|Nr\.|Episode|Bd\.|Sammelband|[B|b]and|Part|Teil|Folge)[,\-–:\s#\(]*(\d+\.?\d*)[\)\s\-–:]*(\S\D*?[a-zA-Z]\D*?)[\/\.,\-–\s]*$",indexpart)
if match:
guessed_series_index = match.group(1)
guessed_series = match.group(2)
if guessed_series is None:
# sometimes books with multiple volumes are detected as series without name -> Add the volume to the title
guessed_series = textpart
guessed_title = textpart + " : Band " + guessed_series_index
else:
guessed_title = textpart
#log.info("ALGO2: guessed_title: " + guessed_title)
#log.info("ALGO2: guessed_series: " + guessed_series)
#log.info("ALGO2: guessed_series_index: " + guessed_series_index)
else:
# for titleparts like: "Band 2"
match = re.match("^[\s\(]*(?:#|Nr\.|Episode|Bd\.|Sammelband|[B|b]and|Part|Teil|Folge)[,\-–:\s#\(]*(\d+\.?\d*)[\)\s\-–:]*[\/\.,\-–\s]*$",indexpart)
if match:
guessed_series_index = match.group(1)
# ...with textpart like NAME OF SERIES\s[\-\.;:]\sNAME OF TITLE
# some false positives
match = re.match("^\s*(\w+.+?)\s?[\.;\-–:]+\s(\w+.+)\s*$",textpart)
if match:
guessed_series = match.group(1)
guessed_title = match.group(2)
log.info("ALGO3: guessed_title: " + guessed_title)
log.info("ALGO3: guessed_series: " + guessed_series)
log.info("ALGO3: guessed_series_index: " + guessed_series_index)
elif len(parts)==1:
log.info("Title has one part")
# for Titles like: "Name of the series - Title (Episode 2)"
match = re.match("^\s*(\S.+?) \- (\S.+?) [\(\/\.,\s\-–:](?:#|Nr\.|Episode|Bd\.|Sammelband|[B|b]and|Part|Teil|Folge)[,\-–:\s#\(]*(\d+\.?\d*)[\)\s\-–:]*$",parts[0])
if match:
guessed_series_index = match.group(3)
guessed_series = match.group(1)
guessed_title = match.group(2)
#log.info("ALGO4: guessed_title: " + guessed_title)
#log.info("ALGO4: guessed_series: " + guessed_series)
#log.info("ALGO4: guessed_series_index: " + guessed_series_index)
else:
# for Titles like: "Name of the series - Episode 2"
match = re.match("^\s*(\S.+?)[\(\/\.,\s\-–:]*(?:#|Nr\.|Episode|Bd\.|Sammelband|[B|b]and|Part|Teil|Folge)[,\-–:\s#\(]*(\d+\.?\d*)[\)\s\-–:]*$",parts[0])
if match:
guessed_series_index = match.group(2)
guessed_series = match.group(1)
guessed_title = guessed_series + " : Band " + guessed_series_index
#log.info("ALGO5: guessed_title: " + guessed_title)
#log.info("ALGO5: guessed_series: " + guessed_series)
#log.info("ALGO5: guessed_series_index: " + guessed_series_index)
# Log
if guessed_series is not None:
log.info("Guessed Series: %s" % guessed_series)
#guessed_series = self.cleanUpSeries(log, guessed_series, publisher_name)
if guessed_series_index is not None:
log.info("Guessed Series Index: %s" % guessed_series_index)
if guessed_title is not None:
log.info("Guessed Title: %s" % guessed_title)
guessed_title = self.cleanUpTitle(log, guessed_title)
if guessed_series is not None and guessed_series_index is not None and guessed_title is not None:
title = guessed_title
series = guessed_series
series_index = guessed_series_index
##### Filter exact searches #####
# When doing an exact search for a given IDN skip books with wrong IDNs
# TODO: Currently exact_search for ISBN is not implemented. Would require ISBN-10 and ISBN-13 conversions
if idn is not None and "idn" in exact_search:
if idn != exact_search["idn"]:
log.info("Extracted IDN does not match book's IDN, skipping record")
continue
##### Put it all together #####
if self.cfg_append_edition_to_title == True and edition is not None:
title = title + " : " + edition
mi = Metadata(self.removeSortingCharacters(title), map(lambda i: self.removeSortingCharacters(i), authors))
mi.title_sort = self.removeSortingCharacters(title_sort)
mi.author_sort = self.removeSortingCharacters(author_sort)
mi.languages = languages
mi.pubdate = pubdate
mi.publisher = " ; ".join(filter(None,[publisher_location, self.removeSortingCharacters(publisher_name)]))
mi.series = self.removeSortingCharacters(series)
mi.series_index = series_index
mi.comments = comments
mi.isbn = isbn # also required for cover download
mi.set_identifier('urn',urn)
mi.set_identifier('dnb-idn',idn)
mi.set_identifier('ddc', ",".join(ddc))
# cfg_subjects:
# 0: use only subjects_gnd
if self.cfg_fetch_subjects == 0:
mi.tags = self.uniq(subjects_gnd)
# 1: use only subjects_gnd if found, else subjects_non_gnd
elif self.cfg_fetch_subjects == 1:
if len(subjects_gnd)>0:
mi.tags = self.uniq(subjects_gnd)
else:
mi.tags = self.uniq(subjects_non_gnd)
# 2: subjects_gnd and subjects_non_gnd
elif self.cfg_fetch_subjects == 2:
mi.tags = self.uniq(subjects_gnd + subjects_non_gnd)
# 3: use only subjects_non_gnd if found, else subjects_gnd
elif self.cfg_fetch_subjects == 3:
if len(subjects_non_gnd)>0:
mi.tags = self.uniq(subjects_non_gnd)
else:
mi.tags = self.uniq(subjects_gnd)
# 4: use only subjects_non_gnd
elif self.cfg_fetch_subjects == 4:
mi.tags = self.uniq(subjects_non_gnd)
# 5: use no subjects at all
elif self.cfg_fetch_subjects == 5:
mi.tags = []
# put current result's metdata into result queue
log.info("Final formatted result: \n%s" % mi)
result_queue.put(mi)
def parse_details(self, root):
try:
moly_id = self.parse_moly_id(self.url)
self.log.info('Parsed moly.hu identifier: %s' % moly_id)
except:
self.log.exception(
'Error parsing moly.hu id for url: %r' % self.url)
moly_id = None
try:
title = self.parse_title(root)
self.log.info('Parsed title: %s' % title)
except:
self.log.exception('Error parsing title for url: %r' % self.url)
title = None
try:
authors = self.parse_authors(root)
self.log.info('Parsed authors: %s' % authors)
except:
self.log.exception('Error parsing authors for url: %r' % self.url)
authors = []
if not title or not authors or not moly_id:
self.log.error(
'Could not find title/authors/moly.hu id for %r' % self.url)
self.log.error('Moly.hu id: %r Title: %r Authors: %r' %
(moly_id, title, authors))
return
mi = Metadata(title, authors)
mi.set_identifier('moly_hu', moly_id)
self.moly_id = moly_id
try:
isbn = self.parse_isbn(root)
self.log.info('Parsed ISBN: %s' % isbn)
if isbn:
self.isbn = mi.isbn = isbn
except:
self.log.exception('Error parsing ISBN for url: %r' % self.url)
try:
series_info = self.parse_series(root)
if series_info is not None:
mi.series = series_info[0]
mi.series_index = int(series_info[1])
self.log.info('Parsed series: %s, series index: %f' %
(mi.series, mi.series_index))
except:
self.log.exception('Error parsing series for url: %r' % self.url)
try:
mi.comments = self.parse_comments(root)
self.log.info('Parsed comments: %s' % mi.comments)
except:
self.log.exception('Error parsing comments for url: %r' % self.url)
try:
self.cover_url = self.parse_covers(root)
self.log.info('Parsed URL for cover: %r' % self.cover_url)
self.plugin.cache_identifier_to_cover_url(
self.moly_id, self.cover_url)
mi.has_cover = bool(self.cover_url)
except:
self.log.exception('Error parsing cover for url: %r' % self.url)
try:
mi.tags = self.parse_tags(root)
self.log.info('Parsed tags: %s' % mi.tags)
except:
self.log.exception('Error parsing tags for url: %r' % self.url)
try:
mi.languages = self.parse_languages(mi.tags)
self.log.info('Parsed languages: %r' % mi.languages)
except:
self.log.exception('Error parsing language for url: %r' % self.url)
try:
mi.publisher = self.parse_publisher(root)
self.log.info('Parsed publisher: %s' % mi.publisher)
except:
self.log.exception(
'Error parsing publisher for url: %r' % self.url)
try:
mi.pubdate = self.parse_published_date(root)
self.log.info('Parsed publication date: %s' % mi.pubdate)
except:
self.log.exception(
'Error parsing published date for url: %r' % self.url)
try:
mi.rating = self.parse_rating(root)
self.log.info('Parsed rating: %s\n\n' % mi.rating)
except:
self.log.exception('Error parsing tags for url: %r\n\n' % self.url)
mi.source_relevance = self.relevance
if self.moly_id and self.isbn:
self.plugin.cache_isbn_to_identifier(self.isbn, self.moly_id)
self.plugin.clean_downloaded_metadata(mi)
self.result_queue.put(mi)
def parse_details(self, root):
try:
isbn = self.extract_isbn(self.url)
except:
self.log.exception('No ISBN in URL: %r'%self.url)
isbn = None
try:
(title, series, series_index) = self.parse_title_series(root)
except:
self.log.exception('Error parsing title and series for url: %r'%self.url)
title = series = series_index = None
try:
authors = self.parse_authors(root)
except:
self.log.exception('Error parsing authors for url: %r'%self.url)
authors = []
if not title or not authors or not isbn:
self.log.error('Could not find title/authors/Aladin id for %r'%self.url)
self.log.error('Aladin: %r Title: %r Authors: %r'%(isbn, title, authors))
return
mi = Metadata(title, authors)
if series:
mi.series = series
mi.series_index = series_index
#mi.set_identifier('isbn', isbn)
mi.isbn = isbn
self.isbn = isbn
# ISBN-13
try:
isbn = self.parse_isbn(root)
if isbn:
self.isbn = mi.isbn = isbn
except:
self.log.exception('Error parsing ISBN for url: %r'%self.url)
try:
mi.comments = self.parse_comments(root)
except:
self.log.exception('Error parsing comments for url: %r'%self.url)
try:
self.cover_url = self.parse_cover(root)
except:
self.log.exception('Error parsing cover for url: %r'%self.url)
mi.has_cover = bool(self.cover_url)
mi.cover_url = self.cover_url # This is purely so we can run a test for it!!!
if mi.has_cover:
self.log.info('Cover URL: '+mi.cover_url)
try:
mi.publisher = self.parse_publisher(root)
except:
self.log.exception('Error parsing publisher for url: %r'%self.url)
try:
mi.pubdate = self.parse_published_date(root)
except:
self.log.exception('Error parsing published date for url: %r'%self.url)
mi.language = 'ko'
mi.source_relevance = self.relevance
self.plugin.clean_downloaded_metadata(mi)
self.result_queue.put(mi)
def parse_details(self, root):
try:
kyobobook_id = self.parse_kyobobook_id(self.url)
except:
self.log.exception('Error parsing Kyobobook id for url: %r'%self.url)
kyobobook_id = None
try:
(title, series, series_index) = self.parse_title_series(root)
except:
self.log.exception('Error parsing title and series for url: %r'%self.url)
title = series = series_index = None
try:
authors = self.parse_authors(root)
except:
self.log.exception('Error parsing authors for url: %r'%self.url)
authors = []
if not title or not authors or not kyobobook_id:
self.log.error('Could not find title/authors/kyobobook id for %r'%self.url)
self.log.error('Kyobobook: %r Title: %r Authors: %r'%(kyobobook_id, title,
authors))
return
mi = Metadata(title, authors)
if series:
mi.series = series
mi.series_index = series_index
mi.set_identifier('kyobobook', kyobobook_id)
self.kyobobook_id = kyobobook_id
try:
isbn = self.parse_isbn(root)
if isbn:
self.isbn = mi.isbn = isbn
except:
self.log.exception('Error parsing ISBN for url: %r'%self.url)
try:
mi.rating = self.parse_rating(root)
except:
self.log.exception('Error parsing ratings for url: %r'%self.url)
try:
mi.comments = self.parse_comments(root)
except:
self.log.exception('Error parsing comments for url: %r'%self.url)
try:
self.cover_url = self.parse_cover(root)
except:
self.log.exception('Error parsing cover for url: %r'%self.url)
mi.has_cover = bool(self.cover_url)
try:
tags = self.parse_tags(root)
if tags:
mi.tags = tags
except:
self.log.exception('Error parsing tags for url: %r'%self.url)
try:
mi.publisher, mi.pubdate = self.parse_publisher_and_date(root)
except:
self.log.exception('Error parsing publisher and date for url: %r'%self.url)
try:
lang = self._parse_language(root)
if lang:
mi.language = lang
except:
self.log.exception('Error parsing language for url: %r'%self.url)
mi.source_relevance = self.relevance
if self.kyobobook_id:
if self.isbn:
self.plugin.cache_isbn_to_identifier(self.isbn, self.kyobobook_id)
if self.cover_url:
self.plugin.cache_identifier_to_cover_url(self.kyobobook_id,
self.cover_url)
self.plugin.clean_downloaded_metadata(mi)
self.result_queue.put(mi)
def get_details(self):
self.log.info(" Worker.get_details:")
self.log.info(" self: ", self)
self.log.info(" self.url: ", self.url)
# We should not even be here if we are not processing an ebook hit
if self.url.find("/ebook/") == -1:
return
try:
raw = self.browser.open_novisit(self.url, timeout=self.timeout).read().strip()
except Exception as e:
if callable(getattr(e, 'getcode', None)) and e.getcode() == 404:
self.log.error('URL malformed: %r' % self.url)
return
attr = getattr(e, 'args', [None])
attr = attr if attr else [None]
if isinstance(attr[0], socket.timeout):
msg = 'Beam Ebooks timed out. Try again later.'
self.log.error(msg)
else:
msg = 'Failed to make details query: %r' % self.url
self.log.exception(msg)
return
# raw = raw.decode('utf-8', errors='replace')
raw = raw.decode('iso-8859-1', errors='replace')
# open('D:\\work\\calibre-dump-book-details.html', 'wb').write(raw)
if '<title>404 - ' in raw:
self.log.error('URL malformed: %r' % self.url)
return
try:
# root = fromstring(clean_ascii_chars(raw))
root = fromstring(raw)
except:
msg = 'Failed to parse beam ebooks details page: %r' % self.url
self.log.exception(msg)
return
try:
self.beam_ebooks_id = self.parse_beam_ebooks_id(self.url)
except:
self.log.exception('Error parsing beam ebooks id for url: %r' % self.url)
self.beam_ebooks_id = None
try:
(self.title, self.series_index) = self.parse_title(root)
except:
self.log.exception('Error parsing title for url: %r' % self.url)
self.title = None
self.series_index = None
try:
self.authors = self.parse_authors(root)
except:
self.log.exception('Error parsing authors for url: %r' % self.url)
self.authors = None
mi = Metadata(self.title, self.authors)
mi.set_identifier('beam-ebooks', self.beam_ebooks_id)
if self.series_index:
mi.series_index = float(self.series_index)
self._determine_perry_rhodan_cycle_name(mi)
mi.source_relevance = self.relevance
self.plugin.clean_downloaded_metadata(mi)
print(mi)
self.result_queue.put(mi)
def load_details(self, url, timeout):
def _format_item(str):
return re.sub('^"(.*)"$', '\\1', unescape(str))
def _format_list(str):
return [_.strip() for _ in _format_item(str).split(',')]
def _find_meta(node, property):
return [_.get('content') for _ in node if _.get('property') == property][0]
def _format_date(date_text):
year = int(date_text[0:4])
month = int(date_text[4:6])
day = int(date_text[6:])
return datetime.datetime(year, month, day, tzinfo=utc_tz)
try:
response = self.browser.open(url, timeout=timeout)
root = lxml.html.fromstring(response.read())
# <meta> tag에서 불러오는 항목
# 책ID, 제목, ISBN, 이미지URL, 평점
meta = root.xpath('//meta[starts-with(@property, "og") or starts-with(@property, "books")]')
# schema.org JSON에서 불러오는 항목
# 제목, 저자, 책소개, 출판사
ld_json = root.xpath('//script[@type="application/ld+json"]/text()')
ld = [json.loads(_) for _ in ld_json]
book_info = [_ for _ in ld if _['@type'] == 'Book'][0]
except Exception as e:
self.log.exception(e)
ridibooks_id = re.search('id=([0-9]+)', url).group(1)
isbn = _find_meta(meta, 'books:isbn')
cover_url = _find_meta(meta, 'og:image')
title = _find_meta(meta, 'og:title')
authors = _format_list(book_info['author']['name'])
if book_info.has_key('translator'):
authors.extend([_ + u'(역자)' for _ in _format_list(book_info['translator']['name'])])
mi = Metadata(title, authors)
mi.set_identifier('ridibooks', ridibooks_id)
mi.cover_url = cover_url
mi.has_cover = bool(cover_url)
mi.publisher = _format_item(book_info['publisher']['name'])
mi.pubdate = _format_date(book_info['datePublished'])
mi.comments = _format_item(book_info['description'])
mi.rating = float(_find_meta(meta, 'books:rating:normalized_value'))
series = re.search(u'(.*)\s*(\d+)권', title)
if series:
mi.series = series.group(1)
mi.series_index = float(series.group(2))
mi.language = 'Korean'
mi.source_relevance = self.relevance
if ridibooks_id:
if isbn:
self.plugin.cache_isbn_to_identifier(isbn, ridibooks_id)
if cover_url:
self.plugin.cache_identifier_to_cover_url(ridibooks_id, cover_url)
self.plugin.clean_downloaded_metadata(mi)
self.result_queue.put(mi)
def load_details(self, url, timeout):
def _format_item(str):
return re.sub('^"(.*)"$', '\\1', unescape(str))
def _format_list(str):
return [_.strip() for _ in _format_item(str).split(',')]
def _find_meta(node, property):
return [
_.get('content') for _ in node if _.get('property') == property
][0]
def _format_date(date_text):
year = int(date_text[0:4])
month = int(date_text[4:6])
day = int(date_text[6:])
return datetime.datetime(year, month, day, tzinfo=utc_tz)
try:
response = self.browser.open(url, timeout=timeout)
root = lxml.html.fromstring(response.read())
# <meta> tag에서 불러오는 항목
# 책ID, 제목, ISBN, 이미지URL, 평점
meta = root.xpath(
'//meta[starts-with(@property, "og") or starts-with(@property, "books")]'
)
# schema.org JSON에서 불러오는 항목
# 제목, 저자, 책소개, 출판사
ld_json = root.xpath(
'//script[@type="application/ld+json"]/text()')
ld = [json.loads(_) for _ in ld_json]
book_info = [_ for _ in ld if _['@type'] == 'Book'][0]
except Exception as e:
self.log.exception(e)
ridibooks_id = re.search('id=([0-9]+)', url).group(1)
isbn = _find_meta(meta, 'books:isbn')
cover_url = _find_meta(meta, 'og:image')
title = _find_meta(meta, 'og:title')
authors = _format_list(book_info['author']['name'])
if book_info.has_key('translator'):
authors.extend([
_ + u'(역자)'
for _ in _format_list(book_info['translator']['name'])
])
mi = Metadata(title, authors)
mi.set_identifier('ridibooks', ridibooks_id)
mi.cover_url = cover_url
mi.has_cover = bool(cover_url)
mi.publisher = _format_item(book_info['publisher']['name'])
mi.pubdate = _format_date(book_info['datePublished'])
mi.comments = _format_item(book_info['description'])
mi.rating = float(_find_meta(meta, 'books:rating:normalized_value'))
series = re.search(u'(.*)\s*(\d+)권', title)
if series:
mi.series = series.group(1)
mi.series_index = float(series.group(2))
mi.language = 'Korean'
mi.source_relevance = self.relevance
if ridibooks_id:
if isbn:
self.plugin.cache_isbn_to_identifier(isbn, ridibooks_id)
if cover_url:
self.plugin.cache_identifier_to_cover_url(
ridibooks_id, cover_url)
self.plugin.clean_downloaded_metadata(mi)
self.result_queue.put(mi)
def merge(self, results, min_year, do_asr=True):
ans = Metadata(_('Unknown'))
# We assume the shortest title has the least cruft in it
ans.title = self.length_merge('title', results, null_value=ans.title)
# No harm in having extra authors, maybe something useful like an
# editor or translator
ans.authors = self.length_merge('authors', results,
null_value=ans.authors, shortest=False)
# We assume the shortest publisher has the least cruft in it
ans.publisher = self.length_merge('publisher', results,
null_value=ans.publisher)
# We assume the smallest set of tags has the least cruft in it
ans.tags = self.length_merge('tags', results,
null_value=ans.tags, shortest=msprefs['fewer_tags'])
# We assume the longest series has the most info in it
ans.series = self.length_merge('series', results,
null_value=ans.series, shortest=False)
for r in results:
if r.series and r.series == ans.series:
ans.series_index = r.series_index
break
# Average the rating over all sources
ratings = []
for r in results:
rating = r.rating
if rating and rating > 0 and rating <= 5:
ratings.append(rating)
if ratings:
ans.rating = int(round(sum(ratings)/len(ratings)))
# Smallest language is likely to be valid
ans.language = self.length_merge('language', results,
null_value=ans.language)
# Choose longest comments
ans.comments = self.length_merge('comments', results,
null_value=ans.comments, shortest=False)
# Published date
if min_year:
for r in results:
year = getattr(r.pubdate, 'year', None)
if year == min_year:
ans.pubdate = r.pubdate
break
if getattr(ans.pubdate, 'year', None) == min_year:
min_date = datetime(min_year, ans.pubdate.month, ans.pubdate.day,
tzinfo=utc_tz)
else:
min_date = datetime(min_year, 1, 2, tzinfo=utc_tz)
ans.pubdate = min_date
else:
min_date = datetime(3001, 1, 1, tzinfo=utc_tz)
for r in results:
if r.pubdate is not None:
candidate = as_utc(r.pubdate)
if candidate < min_date:
min_date = candidate
if min_date.year < 3000:
ans.pubdate = min_date
# Identifiers
for r in results:
ans.identifiers.update(r.identifiers)
# Cover URL
ans.has_cached_cover_url = bool([r for r in results if
getattr(r, 'has_cached_cover_url', False)])
# Merge any other fields with no special handling (random merge)
touched_fields = set()
for r in results:
if hasattr(r, 'identify_plugin'):
touched_fields |= r.identify_plugin.touched_fields
for f in touched_fields:
if f.startswith('identifier:') or not ans.is_null(f):
continue
setattr(ans, f, self.random_merge(f, results,
null_value=getattr(ans, f)))
if do_asr:
avg = [x.relevance_in_source for x in results]
avg = sum(avg)/len(avg)
ans.average_source_relevance = avg
return ans
def set_mi(self, mi, fm):
'''
This sets the metadata for the test result books table. It doesn't reset
the contents of the field selectors for editing rules.
'''
self.fm = fm
if mi:
if not isinstance(mi, list):
mi = (mi, )
else:
mi = Metadata(_('Title'), [_('Author')])
mi.author_sort = _('Author Sort')
mi.series = ngettext('Series', 'Series', 1)
mi.series_index = 3
mi.rating = 4.0
mi.tags = [_('Tag 1'), _('Tag 2')]
mi.languages = ['eng']
mi.id = 1
if self.fm is not None:
mi.set_all_user_metadata(self.fm.custom_field_metadata())
else:
# No field metadata. Grab a copy from the current library so
# that we can validate any custom column names. The values for
# the columns will all be empty, which in some very unusual
# cases might cause formatter errors. We can live with that.
from calibre.gui2.ui import get_gui
fm = get_gui().current_db.new_api.field_metadata
mi.set_all_user_metadata(fm.custom_field_metadata())
for col in mi.get_all_user_metadata(False):
if fm[col]['datatype'] == 'datetime':
mi.set(col, DEFAULT_DATE)
elif fm[col]['datatype'] in ('int', 'float', 'rating'):
mi.set(col, 2)
elif fm[col]['datatype'] == 'bool':
mi.set(col, False)
elif fm[col]['is_multiple']:
mi.set(col, (col, ))
else:
mi.set(col, col, 1)
mi = (mi, )
self.mi = mi
tv = self.template_value
tv.setColumnCount(2)
tv.setHorizontalHeaderLabels((_('Book title'), _('Template value')))
tv.horizontalHeader().setStretchLastSection(True)
tv.horizontalHeader().sectionResized.connect(self.table_column_resized)
tv.setRowCount(len(mi))
# Set the height of the table
h = tv.rowHeight(0) * min(len(mi), 5)
h += 2 * tv.frameWidth() + tv.horizontalHeader().height()
tv.setMinimumHeight(h)
tv.setMaximumHeight(h)
# Set the size of the title column
if self.table_column_widths:
tv.setColumnWidth(0, self.table_column_widths[0])
else:
tv.setColumnWidth(0, tv.fontMetrics().averageCharWidth() * 10)
tv.setSelectionBehavior(QAbstractItemView.SelectionBehavior.SelectRows)
tv.setRowCount(len(mi))
# Use our own widget to get rid of elision. setTextElideMode() doesn't work
for r in range(0, len(mi)):
w = QLineEdit(tv)
w.setReadOnly(True)
tv.setCellWidget(r, 0, w)
w = QLineEdit(tv)
w.setReadOnly(True)
tv.setCellWidget(r, 1, w)
self.display_values('')
def parse_book_page(self, url):
# TODO: Support for login-based rating fetching
# TODO: Move all parsing logic to methods in order to avoid dangling variables
# TODO: Saving metadata in custom columns
# TODO: Configurable embedding metadata in comment
# TODO: missing items
# original language, first polish publish date, publisher serie, form
self.log.info('INFO: Downloading book page: {}'.format(url))
root_tag = self.get_lxml_root(url)
if not root_tag:
return None
book_tag = self.get_book_tag(root_tag)
if self.prefs['title']:
book_title = self.parse_title(root_tag, book_tag, url)
else:
book_title = self.title
if self.prefs['authors']:
book_authors = self.parse_authors(root_tag, book_tag, url)
else:
book_authors = self.authors
mi = Metadata(book_title, book_authors)
additional_meta = {}
if self.enabled('languages'):
languages = self.parse_languages(root_tag, book_tag, url)
if languages:
mi.languages = languages
if self.enabled('rating'):
rating = self.parse_rating(root_tag, book_tag, url)
if rating != None:
mi.rating = rating
if self.enabled('tags'):
tags = self.parse_tags(root_tag, book_tag, url)
if tags:
mi.tags = tags
if self.enabled('identifier'):
identifier = self.parse_identifier(root_tag, book_tag, url)
if identifier:
mi.set_identifier(IDENTIFIER, identifier)
if self.enabled('pubdate'):
pubdate = self.parse_pubdate(root_tag, book_tag, url)
if pubdate:
mi.pubdate = pubdate
if self.enabled('covers'):
covers = self.parse_covers(root_tag, book_tag, url)
if covers:
mi.has_cover = True
self.plugin.cached_identifier_to_cover_url('urls').extend(
covers)
else:
self.plugin.cache_identifier_to_cover_url('nocover', True)
# TODO: is this necessary?
if self.enabled('series'):
series = self.parse_series(root_tag, book_tag, url)
if series:
additional_meta['series'] = [
self.get_series_string(name, index)
for name, index in series
]
name, index = series[0]
mi.series = name
if index is not None:
mi.series_index = index
if self.enabled('translators'):
translators = self.parse_translators(root_tag, book_tag, url)
if translators:
additional_meta['translators'] = translators
if self.enabled('original_title'):
original_title = self.parse_original_title(root_tag, book_tag, url)
if original_title:
additional_meta['original_title'] = original_title
if self.enabled('categories'):
categories = self.parse_categories(root_tag, book_tag, url)
if categories:
additional_meta['categories'] = categories
if self.enabled('genres'):
genres = self.parse_genres(root_tag, book_tag, url)
if genres:
additional_meta['genres'] = genres
if self.enabled('comments'):
comments = self.parse_comments(root_tag, book_tag, url) or ''
additional_comments = self.format_additional_comment(
additional_meta)
if comments or additional_comments:
mi.comments = comments + additional_comments
self.log.info('INFO: Parsing book page completed')
return mi
def merge(self, results, min_year, do_asr=True):
ans = Metadata(_('Unknown'))
# We assume the shortest title has the least cruft in it
ans.title = self.length_merge('title', results, null_value=ans.title)
# No harm in having extra authors, maybe something useful like an
# editor or translator
ans.authors = self.length_merge('authors',
results,
null_value=ans.authors,
shortest=False)
# We assume the shortest publisher has the least cruft in it
ans.publisher = self.length_merge('publisher',
results,
null_value=ans.publisher)
# We assume the smallest set of tags has the least cruft in it
ans.tags = self.length_merge('tags',
results,
null_value=ans.tags,
shortest=msprefs['fewer_tags'])
# We assume the longest series has the most info in it
ans.series = self.length_merge('series',
results,
null_value=ans.series,
shortest=False)
for r in results:
if r.series and r.series == ans.series:
ans.series_index = r.series_index
break
# Average the rating over all sources
ratings = []
for r in results:
rating = r.rating
if rating and rating > 0 and rating <= 5:
ratings.append(rating)
if ratings:
ans.rating = int(round(sum(ratings) / len(ratings)))
# Smallest language is likely to be valid
ans.language = self.length_merge('language',
results,
null_value=ans.language)
# Choose longest comments
ans.comments = self.length_merge('comments',
results,
null_value=ans.comments,
shortest=False)
# Published date
if min_year:
for r in results:
year = getattr(r.pubdate, 'year', None)
if year == min_year:
ans.pubdate = r.pubdate
break
if getattr(ans.pubdate, 'year', None) == min_year:
min_date = datetime(min_year,
ans.pubdate.month,
ans.pubdate.day,
tzinfo=utc_tz)
else:
min_date = datetime(min_year, 1, 2, tzinfo=utc_tz)
ans.pubdate = min_date
else:
min_date = datetime(3001, 1, 1, tzinfo=utc_tz)
for r in results:
if r.pubdate is not None:
candidate = as_utc(r.pubdate)
if candidate < min_date:
min_date = candidate
if min_date.year < 3000:
ans.pubdate = min_date
# Identifiers
for r in results:
ans.identifiers.update(r.identifiers)
# Cover URL
ans.has_cached_cover_url = bool(
[r for r in results if getattr(r, 'has_cached_cover_url', False)])
# Merge any other fields with no special handling (random merge)
touched_fields = set()
for r in results:
if hasattr(r, 'identify_plugin'):
touched_fields |= r.identify_plugin.touched_fields
for f in touched_fields:
if f.startswith('identifier:') or not ans.is_null(f):
continue
setattr(ans, f,
self.random_merge(f, results, null_value=getattr(ans, f)))
if do_asr:
avg = [x.relevance_in_source for x in results]
avg = sum(avg) / len(avg)
ans.average_source_relevance = avg
return ans
def parse_details(self, root):
try:
kyobobook_id = self.parse_kyobobook_id(self.url)
except:
self.log.exception('Error parsing Kyobobook id for url: %r' %
self.url)
kyobobook_id = None
try:
(title, series, series_index) = self.parse_title_series(root)
except:
self.log.exception('Error parsing title and series for url: %r' %
self.url)
title = series = series_index = None
try:
authors = self.parse_authors(root)
except:
self.log.exception('Error parsing authors for url: %r' % self.url)
authors = []
if not title or not authors or not kyobobook_id:
self.log.error('Could not find title/authors/kyobobook id for %r' %
self.url)
self.log.error('Kyobobook: %r Title: %r Authors: %r' %
(kyobobook_id, title, authors))
return
mi = Metadata(title, authors)
if series:
mi.series = series
mi.series_index = series_index
mi.set_identifier('kyobobook', kyobobook_id)
self.kyobobook_id = kyobobook_id
try:
isbn = self.parse_isbn(root)
if isbn:
self.isbn = mi.isbn = isbn
except:
self.log.exception('Error parsing ISBN for url: %r' % self.url)
try:
mi.rating = self.parse_rating(root)
except:
self.log.exception('Error parsing ratings for url: %r' % self.url)
try:
mi.comments = self.parse_comments(root)
except:
self.log.exception('Error parsing comments for url: %r' % self.url)
try:
self.cover_url = self.parse_cover(root)
except:
self.log.exception('Error parsing cover for url: %r' % self.url)
mi.has_cover = bool(self.cover_url)
try:
tags = self.parse_tags(root)
if tags:
mi.tags = tags
except:
self.log.exception('Error parsing tags for url: %r' % self.url)
try:
mi.publisher, mi.pubdate = self.parse_publisher_and_date(root)
except:
self.log.exception('Error parsing publisher and date for url: %r' %
self.url)
try:
lang = self._parse_language(root)
if lang:
mi.language = lang
except:
self.log.exception('Error parsing language for url: %r' % self.url)
mi.source_relevance = self.relevance
if self.kyobobook_id:
if self.isbn:
self.plugin.cache_isbn_to_identifier(self.isbn,
self.kyobobook_id)
if self.cover_url:
self.plugin.cache_identifier_to_cover_url(
self.kyobobook_id, self.cover_url)
self.plugin.clean_downloaded_metadata(mi)
self.result_queue.put(mi)
def __init__(self,
parent,
text,
mi=None,
fm=None,
color_field=None,
icon_field_key=None,
icon_rule_kind=None,
doing_emblem=False,
text_is_placeholder=False,
dialog_is_st_editor=False,
global_vars=None,
all_functions=None,
builtin_functions=None):
QDialog.__init__(self, parent)
Ui_TemplateDialog.__init__(self)
self.setupUi(self)
self.coloring = color_field is not None
self.iconing = icon_field_key is not None
self.embleming = doing_emblem
self.dialog_is_st_editor = dialog_is_st_editor
if global_vars is None:
self.global_vars = {}
else:
self.global_vars = global_vars
cols = []
if fm is not None:
for key in sorted(
displayable_columns(fm),
key=lambda k: sort_key(fm[k]['name']
if k != color_row_key else 0)):
if key == color_row_key and not self.coloring:
continue
from calibre.gui2.preferences.coloring import all_columns_string
name = all_columns_string if key == color_row_key else fm[key][
'name']
if name:
cols.append((name, key))
self.color_layout.setVisible(False)
self.icon_layout.setVisible(False)
if self.coloring:
self.color_layout.setVisible(True)
for n1, k1 in cols:
self.colored_field.addItem(
n1 + (' (' + k1 + ')' if k1 != color_row_key else ''), k1)
self.colored_field.setCurrentIndex(
self.colored_field.findData(color_field))
elif self.iconing or self.embleming:
self.icon_layout.setVisible(True)
if self.embleming:
self.icon_kind_label.setVisible(False)
self.icon_kind.setVisible(False)
self.icon_chooser_label.setVisible(False)
self.icon_field.setVisible(False)
for n1, k1 in cols:
self.icon_field.addItem('{} ({})'.format(n1, k1), k1)
self.icon_file_names = []
d = os.path.join(config_dir, 'cc_icons')
if os.path.exists(d):
for icon_file in os.listdir(d):
icon_file = icu_lower(icon_file)
if os.path.exists(os.path.join(d, icon_file)):
if icon_file.endswith('.png'):
self.icon_file_names.append(icon_file)
self.icon_file_names.sort(key=sort_key)
self.update_filename_box()
if self.iconing:
dex = 0
from calibre.gui2.preferences.coloring import icon_rule_kinds
for i, tup in enumerate(icon_rule_kinds):
txt, val = tup
self.icon_kind.addItem(txt, userData=(val))
if val == icon_rule_kind:
dex = i
self.icon_kind.setCurrentIndex(dex)
self.icon_field.setCurrentIndex(
self.icon_field.findData(icon_field_key))
if dialog_is_st_editor:
self.buttonBox.setVisible(False)
else:
self.new_doc_label.setVisible(False)
self.new_doc.setVisible(False)
self.template_name_label.setVisible(False)
self.template_name.setVisible(False)
if mi:
if not isinstance(mi, list):
mi = (mi, )
else:
mi = Metadata(_('Title'), [_('Author')])
mi.author_sort = _('Author Sort')
mi.series = ngettext('Series', 'Series', 1)
mi.series_index = 3
mi.rating = 4.0
mi.tags = [_('Tag 1'), _('Tag 2')]
mi.languages = ['eng']
mi.id = 1
if fm is not None:
mi.set_all_user_metadata(fm.custom_field_metadata())
else:
# No field metadata. Grab a copy from the current library so
# that we can validate any custom column names. The values for
# the columns will all be empty, which in some very unusual
# cases might cause formatter errors. We can live with that.
from calibre.gui2.ui import get_gui
mi.set_all_user_metadata(get_gui(
).current_db.new_api.field_metadata.custom_field_metadata())
for col in mi.get_all_user_metadata(False):
mi.set(col, (col, ), 0)
mi = (mi, )
self.mi = mi
# Set up the display table
self.table_column_widths = None
try:
self.table_column_widths = \
gprefs.get('template_editor_table_widths', None)
except:
pass
tv = self.template_value
tv.setRowCount(len(mi))
tv.setColumnCount(2)
tv.setHorizontalHeaderLabels((_('Book title'), _('Template value')))
tv.horizontalHeader().setStretchLastSection(True)
tv.horizontalHeader().sectionResized.connect(self.table_column_resized)
# Set the height of the table
h = tv.rowHeight(0) * min(len(mi), 5)
h += 2 * tv.frameWidth() + tv.horizontalHeader().height()
tv.setMinimumHeight(h)
tv.setMaximumHeight(h)
# Set the size of the title column
if self.table_column_widths:
tv.setColumnWidth(0, self.table_column_widths[0])
else:
tv.setColumnWidth(0, tv.fontMetrics().averageCharWidth() * 10)
# Use our own widget to get rid of elision. setTextElideMode() doesn't work
for r in range(0, len(mi)):
w = QLineEdit(tv)
w.setReadOnly(True)
tv.setCellWidget(r, 0, w)
w = QLineEdit(tv)
w.setReadOnly(True)
tv.setCellWidget(r, 1, w)
tv.setSelectionBehavior(QAbstractItemView.SelectionBehavior.SelectRows)
# Remove help icon on title bar
icon = self.windowIcon()
self.setWindowFlags(self.windowFlags()
& (~Qt.WindowType.WindowContextHelpButtonHint))
self.setWindowIcon(icon)
self.all_functions = all_functions if all_functions else formatter_functions(
).get_functions()
self.builtins = (builtin_functions if builtin_functions else
formatter_functions().get_builtins_and_aliases())
self.last_text = ''
self.highlighter = TemplateHighlighter(self.textbox.document(),
builtin_functions=self.builtins)
self.textbox.cursorPositionChanged.connect(self.text_cursor_changed)
self.textbox.textChanged.connect(self.textbox_changed)
self.textbox.setFont(self.get_current_font())
self.textbox.setTabStopWidth(10)
self.source_code.setTabStopWidth(10)
self.documentation.setReadOnly(True)
self.source_code.setReadOnly(True)
if text is not None:
if text_is_placeholder:
self.textbox.setPlaceholderText(text)
self.textbox.clear()
text = ''
else:
self.textbox.setPlainText(text)
else:
text = ''
self.buttonBox.button(QDialogButtonBox.StandardButton.Ok).setText(
_('&OK'))
self.buttonBox.button(QDialogButtonBox.StandardButton.Cancel).setText(
_('&Cancel'))
self.color_copy_button.clicked.connect(self.color_to_clipboard)
self.filename_button.clicked.connect(self.filename_button_clicked)
self.icon_copy_button.clicked.connect(self.icon_to_clipboard)
try:
with open(P('template-functions.json'), 'rb') as f:
self.builtin_source_dict = json.load(f, encoding='utf-8')
except:
self.builtin_source_dict = {}
func_names = sorted(self.all_functions)
self.function.clear()
self.function.addItem('')
for f in func_names:
self.function.addItem(
'{} -- {}'.format(
f, self.function_type_string(f, longform=False)), f)
self.function.setCurrentIndex(0)
self.function.currentIndexChanged.connect(self.function_changed)
self.display_values(text)
self.rule = (None, '')
tt = _('Template language tutorial')
self.template_tutorial.setText(
'<a href="%s">%s</a>' % (localize_user_manual_link(
'https://manual.calibre-ebook.com/template_lang.html'), tt))
tt = _('Template function reference')
self.template_func_reference.setText(
'<a href="%s">%s</a>' % (localize_user_manual_link(
'https://manual.calibre-ebook.com/generated/en/template_ref.html'
), tt))
s = gprefs.get('template_editor_break_on_print', False)
self.go_button.setEnabled(s)
self.remove_all_button.setEnabled(s)
self.set_all_button.setEnabled(s)
self.toggle_button.setEnabled(s)
self.breakpoint_line_box.setEnabled(s)
self.breakpoint_line_box_label.setEnabled(s)
self.break_box.setChecked(s)
self.break_box.stateChanged.connect(self.break_box_changed)
self.go_button.clicked.connect(self.go_button_pressed)
self.textbox.setFocus()
self.set_up_font_boxes()
self.toggle_button.clicked.connect(self.toggle_button_pressed)
self.remove_all_button.clicked.connect(self.remove_all_button_pressed)
self.set_all_button.clicked.connect(self.set_all_button_pressed)
self.load_button.clicked.connect(self.load_template)
self.save_button.clicked.connect(self.save_template)
# Now geometry
try:
geom = gprefs.get('template_editor_dialog_geometry', None)
if geom is not None:
QApplication.instance().safe_restore_geometry(
self, QByteArray(geom))
except Exception:
pass
def get_metadata_(src, encoding=None):
# Meta data definitions as in
# https://www.mobileread.com/forums/showpost.php?p=712544&postcount=9
if isbytestring(src):
if not encoding:
src = xml_to_unicode(src)[0]
else:
src = src.decode(encoding, 'replace')
src = src[:150000] # Searching shouldn't take too long
comment_tags = parse_comment_tags(src)
meta_tags = parse_meta_tags(src)
def get(field):
ans = comment_tags.get(field, meta_tags.get(field, None))
if ans:
ans = ans.strip()
if not ans:
ans = None
return ans
# Title
title = get('title')
if not title:
pat = re.compile('<title>([^<>]+?)</title>', re.IGNORECASE)
match = pat.search(src)
if match:
title = replace_entities(match.group(1))
# Author
authors = get('authors') or _('Unknown')
# Create MetaInformation with Title and Author
mi = Metadata(title or _('Unknown'), string_to_authors(authors))
for field in ('publisher', 'isbn', 'language', 'comments'):
val = get(field)
if val:
setattr(mi, field, val)
for field in ('pubdate', 'timestamp'):
try:
val = parse_date(get(field))
except:
pass
else:
if not is_date_undefined(val):
setattr(mi, field, val)
# SERIES
series = get('series')
if series:
pat = re.compile(r'\[([.0-9]+)\]$')
match = pat.search(series)
series_index = None
if match is not None:
try:
series_index = float(match.group(1))
except:
pass
series = series.replace(match.group(), '').strip()
mi.series = series
if series_index is None:
series_index = get('series_index')
try:
series_index = float(series_index)
except:
pass
if series_index is not None:
mi.series_index = series_index
# RATING
rating = get('rating')
if rating:
try:
mi.rating = float(rating)
if mi.rating < 0:
mi.rating = 0
if mi.rating > 5:
mi.rating /= 2.
if mi.rating > 5:
mi.rating = 0
except:
pass
# TAGS
tags = get('tags')
if tags:
tags = [x.strip() for x in tags.split(',') if x.strip()]
if tags:
mi.tags = tags
return mi
def extract_vol_details(self, vol_url):
# Here we extract and format the information from the choosen volume.
# - The first name and last name to populate author and author sort : vol_auteur_prenom and vol_auteur_nom
# - The title of the volume : vol_title
# - The serie name the volume is part of : vol_serie
# - The sequence number in the serie : vol_serie_seq # missing
# - The editor of this volume : vol_editor
# - The editor's collection of this volume : vol_coll
# - The collection serial code of this volume : vol_coll_srl
# - The "dépot légal" date (the publication date is vastly unknown) : vol_dp_lgl # date format to be computed
# - The ISBN number assoi-ciated with the volume : vol_isbn
# - The volume tags : vol_genre
# - The url pointer to the volume cover image : vol_cover_index
# - The comments includes various info about the book : vol_comment_soup
# . reference, an url pointer to noosfere
# . couverture, an url pointer to noosfere, cover may be real smal, but is accurate to the volume
# . first edition information
# . serie (cycle) name and number
# . this volume editor info
# . Resume (quatrième de couverture)
# . Critiques
# . Sommaire detailing what novels are in the volume when it is an anthology
# . Critiques about the serie and/or about another volume of the book
#
debug = self.dbg_lvl & 2
self.log.info(self.who, "\nIn extract_vol_details(soup)")
if debug:
self.log.info(self.who, "vol_url : ", vol_url)
if debug:
self.log.info(
self.who,
"calling ret_soup(log, dbg_lvl, br, url, rkt=None, who='[__init__]')"
)
self.log.info(self.who, "vol_url : ", vol_url, "who : ", self.who)
rsp = ret_soup(self.log, self.dbg_lvl, self.br, vol_url, who=self.who)
soup = rsp[0]
url_vrai = rsp[1].replace("&Tri=3", "")
# if debug: self.log.info(self.who,soup.prettify()) # useful but too big...
self.nsfr_id = self.nsfr_id + "$vl$" + url_vrai.replace(
'?', '&').replace('=', '&').split('&')[2]
# self.nsfr_id = (self.nfsr_id).strip("$") # If I use this form, it gives this error: 'Worker' object has no attribute 'nfsr_id' ???
tmp = self.nsfr_id
self.nsfr_id = tmp.strip('$')
if debug:
self.log.info(self.who, "self.nsfr_id, type() : ", self.nsfr_id,
type(self.nsfr_id))
tmp_lst = []
vol_info = {}
vol_title = ""
vol_auteur = ""
vol_auteur_prenom = ""
vol_auteur_nom = ""
vol_serie = ""
vol_serie_seq = ""
vol_editor = ""
vol_coll = ""
vol_coll_srl = ""
vol_dp_lgl = ""
vol_isbn = ""
vol_genre = ""
vol_cover_index = ""
comment_generic = None
comment_resume = None
comment_Critiques = None
comment_Sommaire = None
comment_AutresCritique = None
comment_cover = None
comment_decoupage_annexe = None
# add volume address as a reference in the comment
vol_comment_soup = BS(
'<div><p>Référence: <a href="' + url_vrai + '">' + url_vrai +
'</a></p></div>', "lxml")
if debug: self.log.info(self.who, "vol reference processed")
if soup.select("span[class='TitreNiourf']"):
vol_title = soup.select(
"span[class='TitreNiourf']")[0].text.strip()
if debug: self.log.info(self.who, "vol_title processed : ", vol_title)
if soup.select("span[class='AuteurNiourf']"):
vol_auteur = soup.select(
"span[class='AuteurNiourf']")[0].text.replace("\n",
"").strip()
if debug:
self.log.info(self.who, "vol_auteur processed : ", vol_auteur)
for i in range(len(vol_auteur.split())):
if not vol_auteur.split()[i].isupper():
vol_auteur_prenom += " " + vol_auteur.split()[i]
else:
vol_auteur_nom += " " + vol_auteur.split()[i].title()
vol_auteur = vol_auteur.title()
vol_auteur_prenom = vol_auteur_prenom.strip()
if debug:
self.log.info(self.who, "vol_auteur_prenom processed : ",
vol_auteur_prenom)
vol_auteur_nom = vol_auteur_nom.strip()
if debug:
self.log.info(self.who, "vol_auteur_nom processed : ",
vol_auteur_nom)
if soup.select("a[href*='serie.asp']"):
if soup.select("a[href*='serie.asp']")[0].find_parent(
"span", {"class": "ficheNiourf"}):
vol_serie = soup.select("a[href*='serie.asp']")[0].text
tmp_vss = [
x for x in soup.select("a[href*='serie.asp']")
[0].parent.stripped_strings
]
for i in range(len(tmp_vss)):
if "vol." in tmp_vss[i]:
if not vol_serie_seq:
vol_serie_seq = tmp_vss[i].replace("vol.",
"").strip()
if "découpage" in tmp_vss[i]:
dec_anx_url = "https://www.noosfere.org/livres/" + soup.select(
"a[href*='serie.asp']")[0]['href']
comment_pre_decoupage_annexe = BS(
'<div><p> </p><p style="font-weight: 600; font-size: 18px"> Découpage annexe</p><hr style="color:CCC;"/></div>',
"lxml")
comment_decoupage_annexe = self.get_decoupage_annexe(
dec_anx_url)
if debug:
self.log.info(self.who,
"vol_serie, vol_serie_seq processed : ",
vol_serie, ",", vol_serie_seq)
comment_generic = soup.select("span[class='ficheNiourf']")[0]
new_div = soup.new_tag('div')
comment_generic = comment_generic.wrap(new_div)
if debug: self.log.info(self.who, "comment_generic processed")
if soup.select("a[href*='editeur.asp']"):
vol_editor = soup.select("a[href*='editeur.asp']")[0].text
if debug:
self.log.info(self.who, "vol_editor processed : ", vol_editor)
if soup.select("a[href*='collection.asp']"):
vol_coll = soup.select("a[href*='collection.asp']")[0].text
if debug: self.log.info(self.who, "vol_coll : ", vol_coll)
for i in comment_generic.stripped_strings:
tmp_lst.append(str(i))
vol_coll_srl = tmp_lst[len(tmp_lst) - 1]
if "n°" in vol_coll_srl:
for k in ["n°", "(", ")"]:
if k in vol_coll_srl:
vol_coll_srl = vol_coll_srl.replace(k, "")
vol_coll_srl = vol_coll_srl.strip()
vol_coll_srl = vol_coll_srl.split("/")[0]
if vol_coll_srl[0].isnumeric():
vol_coll_srl = ("0" * 5 + vol_coll_srl)[-6:]
else:
vol_coll_srl = ""
if debug:
self.log.info(self.who, "vol_coll_srl processed : ", vol_coll_srl)
# publication date is largely ignored in noosfere, but we have the "dépot legal" date and I use it instead
# note that I 'calculate' the missing day of the month and even sometimes the missing month
ms = ("janvier", "février", "mars", "avril", "mai", "juin", "juillet",
"août", "septembre", "octobre", "novembre", "décembre")
for elemnt in soup.select_one(
"span[class='sousFicheNiourf']").stripped_strings:
if debug: self.log.info(self.who, "elemnt : ", elemnt)
if not vol_dp_lgl:
elemn = (elemnt.replace("Dépôt légal :",
"").split(','))[0].strip()
if elemn:
if elemn.isnumeric() and len(elemn) == 4:
vol_dp_lgl = datetime.datetime.strptime(
"175 " + elemn, "%j %Y")
elif "semestre" in elemn:
ele = elemn.split()
vol_dp_lgl = datetime.datetime.strptime(
("000" + str((int(ele[0][0]) - 1) * 175 + 97))[-3:]
+ " " + ele[2], "%j %Y")
elif "trimestre" in elemn:
ele = elemn.split()
vol_dp_lgl = datetime.datetime.strptime(
("000" + str((int(ele[0][0]) - 1) * 91 + 47))[-3:]
+ " " + ele[2], "%j %Y")
else:
for i in range(len(ms)):
if ms[i] in elemn:
ele = elemn.split()
vol_dp_lgl = datetime.datetime.strptime(
("000" + str(10 + 31 * i))[-3:] + " " +
ele[1], "%j %Y")
break
if debug:
self.log.info(self.who, "vol_dp_lgl : ", vol_dp_lgl)
if "ISBN" in elemnt:
vol_isbn = elemnt.lower().replace(" ", "").replace('isbn:', '')
if "néant" in vol_isbn: vol_isbn = ""
if debug:
self.log.info(self.who, "vol_isbn processed : ", vol_isbn)
if "Genre" in elemnt:
vol_genre = elemnt.lstrip("Genre : ")
if debug:
self.log.info(self.who, "vol_genre processed : ",
vol_genre)
if soup.select("img[name='couverture']"):
for elemnt in repr(
soup.select("img[name='couverture']")[0]).split('"'):
if "http" in elemnt:
if not vol_cover_index:
vol_cover_index = elemnt
if debug:
self.log.info(self.who,
"vol_cover_index processed : ",
vol_cover_index)
# add cover image address as a reference in the comment
if vol_cover_index:
comment_cover = BS(
'<div><p>Couverture: <a href="' + vol_cover_index + '">' +
vol_cover_index + '</a></p></div>', "lxml")
# select the fields I want... More exist such as film adaptations or references to advises to read
# but that is not quite consistant around all the books (noosfere is a common database from many people)
# and beside I have enough info like that AND I do NOT want to take out the noosfere's business
tmp_comm_lst = soup.select("span[class='AuteurNiourf']")
if debug: self.log.info(self.who, tmp_comm_lst) #usefull but too long
for i in range(len(tmp_comm_lst)):
if "Quatrième de couverture" in str(tmp_comm_lst[i]):
comment_resume = tmp_comm_lst[i].find_parents(
"div", {'class': 'sousbloc'})[0]
if debug: self.log.info(self.who, "comment_resume processed")
if "Critiques" in str(tmp_comm_lst[i]):
if not "autres" in str(tmp_comm_lst[i]):
comment_Critiques = tmp_comm_lst[i].find_parents(
"div", {'class': 'sousbloc'})[0]
if debug:
self.log.info(self.who, "comment_Critiques processed")
if "Sommaire" in str(tmp_comm_lst[i]):
comment_Sommaire = tmp_comm_lst[i].find_parents(
"div", {'class': 'sousbloc'})[0]
if debug: self.log.info(self.who, "comment_Sommaire processed")
if "Critiques des autres" in str(tmp_comm_lst[i]):
comment_AutresCritique = tmp_comm_lst[i].find_parents(
"div", {'class': 'sousbloc'})[0]
if comment_AutresCritique.select('a[href*="serie.asp"]') and (
"Critique de la série" in comment_AutresCritique.
select('a[href*="serie.asp"]')[0].text):
critic_url = "https://www.noosfere.org/livres/" + comment_AutresCritique.select(
'a[href*="serie.asp"]')[0]['href']
try:
more_comment_AutresCritique = self.get_Critique_de_la_serie(
critic_url)
comment_AutresCritique.append(
more_comment_AutresCritique)
except:
self.log.exception(
"get_Critique_de_la_serie failed for url: ",
critic_url)
if debug:
self.log.info(self.who, "comment_AutresCritique processed")
# group in a big bundle all the fields I think I want... (It is difficult not to include more... :-))
if comment_cover:
vol_comment_soup.append(comment_cover)
if comment_generic:
vol_comment_soup.append(comment_generic)
if comment_resume:
vol_comment_soup.append(comment_resume)
if comment_Critiques:
vol_comment_soup.append(comment_Critiques)
if comment_Sommaire:
vol_comment_soup.append(comment_Sommaire)
if comment_AutresCritique:
vol_comment_soup.append(comment_AutresCritique)
if comment_decoupage_annexe:
vol_comment_soup.append(
comment_pre_decoupage_annexe) # this is the title
vol_comment_soup.append(comment_decoupage_annexe)
#
# Make a minimum of "repair" over vol_comment_soup so that it displays correctly (how I like it) in the comments and in my catalogs
# - I hate justify when it makes margin "float" around the correct position (in fact when space are used instead of absolute positioning)
# - I like to have functional url when they exist
# - I like to find out the next and/or previous books in a serie (simulated arrows are link :-) )
for elemnt in vol_comment_soup.select('[align="justify"]'):
del elemnt['align']
# remove all double or triple 'br' to improve presentation.
# Note: tmp1 and tmp2 must contain a different value from any possible first elemnt. (yes, I am lrp and I am unique :-) )
#
# ouais, et alors, si je modifie comment_generic APRES l'avoir integré à vol_comment_soup, il n'y a qu'une seule version en mémoire...
# donc vol_comment_soup est modifié...
#
tmp1 = tmp2 = "lrp_the_unique"
for elemnt in vol_comment_soup.findAll():
tmp1, tmp2 = tmp2, elemnt
if tmp1 == tmp2:
elemnt.extract()
br = soup.new_tag('br')
for elemnt in vol_comment_soup.select('.AuteurNiourf'):
elemnt.insert(0, br)
elemnt["style"] = "font-weight: 600; font-size: 18px"
if debug:
for elemnt in vol_comment_soup.select("a[href*='.asp']"):
if 'http' not in elemnt.get('href'):
self.log.info(self.who, "url incomplet avant correction: ",
elemnt)
for elemnt in vol_comment_soup.select("a[href*='/livres/auteur.asp']"):
if 'http' not in elemnt.get('href'):
elemnt["href"] = elemnt["href"].replace(
"/livres/auteur.asp",
"https://www.noosfere.org/livres/auteur.asp")
for elemnt in vol_comment_soup.select("a[href*='/livres/niourf.asp']"):
if 'http' not in elemnt.get('href'):
elemnt["href"] = elemnt["href"].replace(
"/livres/niourf.asp",
"https://www.noosfere.org/livres/niourf.asp")
for elemnt in vol_comment_soup.select("a[href*='/heberg/']"):
if 'http' not in elemnt.get('href'):
elemnt["href"] = elemnt["href"].replace(
"/heberg/", "https://www.noosfere.org/heberg/")
for elemnt in vol_comment_soup.select(
"a[href*='./EditionsLivre.asp']"):
if 'http' not in elemnt.get('href'):
elemnt["href"] = elemnt["href"].replace(
"./EditionsLivre.asp",
"https://www.noosfere.org/livres/EditionsLivre.asp")
for elemnt in vol_comment_soup.select("a[href*='./niourf.asp']"):
if 'http' not in elemnt.get('href'):
elemnt["href"] = elemnt["href"].replace(
"./niourf.asp",
"https://www.noosfere.org/livres/niourf.asp")
for elemnt in vol_comment_soup.select("a[href*='heberg']"):
if 'http' not in elemnt.get('href'):
elemnt["href"] = elemnt["href"].replace(
"../../heberg", "https://www.noosfere.org/heberg")
for elemnt in vol_comment_soup.select("a[href*='../bd']"):
if 'http' not in elemnt.get('href'):
elemnt["href"] = elemnt["href"].replace(
"../bd", "https://www.noosfere.org/bd")
for elemnt in vol_comment_soup.select("a[href*='auteur.asp']"):
if 'http' not in elemnt.get('href'):
elemnt["href"] = elemnt["href"].replace(
"auteur.asp", "https://www.noosfere.org/livres/auteur.asp")
for elemnt in vol_comment_soup.select("a[href*='collection.asp']"):
if 'http' not in elemnt.get('href'):
elemnt["href"] = elemnt["href"].replace(
"collection.asp",
"https://www.noosfere.org/livres/collection.asp")
for elemnt in vol_comment_soup.select("a[href*='critsign.asp']"):
if 'http' not in elemnt.get('href'):
elemnt["href"] = elemnt["href"].replace(
"critsign.asp",
"https://www.noosfere.org/livres/critsign.asp")
for elemnt in vol_comment_soup.select("a[href*='EditionsLivre.asp']"):
if 'http' not in elemnt.get('href'):
elemnt["href"] = elemnt["href"].replace(
"EditionsLivre.asp",
"https://www.noosfere.org/livres/EditionsLivre.asp")
for elemnt in vol_comment_soup.select("a[href*='editeur.asp']"):
if 'http' not in elemnt.get('href'):
elemnt["href"] = elemnt["href"].replace(
"editeur.asp",
"https://www.noosfere.org/livres/editeur.asp")
for elemnt in vol_comment_soup.select("a[href*='editionslivre.asp']"):
if 'http' not in elemnt.get('href'):
elemnt["href"] = elemnt["href"].replace(
"editionslivre.asp",
"https://www.noosfere.org/livres/editionslivre.asp")
for elemnt in vol_comment_soup.select("a[href*='niourf.asp']"):
if 'http' not in elemnt.get('href'):
elemnt["href"] = elemnt["href"].replace(
"niourf.asp", "https://www.noosfere.org/livres/niourf.asp")
for elemnt in vol_comment_soup.select("a[href*='serie.asp']"):
if 'http' not in elemnt.get('href'):
elemnt["href"] = elemnt["href"].replace(
"serie.asp", "https://www.noosfere.org/livres/serie.asp")
if debug:
for elemnt in vol_comment_soup.select("a[href*='.asp']"):
if 'http' not in elemnt.get('href'):
self.log.info(self.who, "url incomplet apres correction: ",
elemnt)
fg, fd = "<<==", "==>>" #chr(0x21D0),chr(0x21D2) #chr(0x27f8),chr(0x27f9)
for elemnt in vol_comment_soup.select("img[src*='arrow_left']"):
elemnt.replace_with(fg)
for elemnt in vol_comment_soup.select("img[src*='arrow_right']"):
elemnt.replace_with(fd)
# depending on the tick box, make a fat publisher using seperators that have a very low probability to pop up (§ and €)
# only set vol_coll_srl if vol_coll exists
# the idea is to use search and replace in the edit Metadata in bulk window.
if self.extended_publisher:
if debug:
self.log.info(
self.who,
"""flag : "Ajoute collection et son numéro d'ordre au champ èditeur" set"""
)
if vol_coll:
if debug: self.log.info(self.who, 'add collection')
vol_editor = vol_editor + ('§') + vol_coll
if vol_coll_srl:
if debug: self.log.info(self.who, 'add collection number')
vol_editor = vol_editor + ('€') + vol_coll_srl
if vol_serie:
if vol_serie_seq.isnumeric(): vol_serie_seq = float(vol_serie_seq)
else: vol_serie_seq = 1.0
# UTF-8 characters may be serialized different ways, only xmlcharrefreplace produces xml compatible strings
# any other non ascii character with another utf-8 byte representation will make calibre behave with the messsage:
# ValueError: All strings must be XML compatible: Unicode or ASCII, no NULL bytes or control characters
# Side note:
# I have no real good url structure(i once got html 3 times, div a sibling of html...), but calibre does not seems to care (nice :-) )
#
# Ca m'a pris un temps fou pour trouver, par hazard, que encode('ascii','xmlcharrefreplace') aidait bien...
# (enfin, quasi par hazard, j' ai essayé tout ce qui pouvait ameliorer la compatibilité avec xml... mais je
# lisais mal et je pensais à une incompatibilité avec la structure xml),
#
vol_comment_soup = vol_comment_soup.encode('ascii',
'xmlcharrefreplace')
self.log.info(self.who, "+++" * 25)
self.log.info(self.who,
"nsfr_id, type() : ", self.nsfr_id,
type(self.nsfr_id)) # must be <class 'str'>
self.log.info(self.who,
"relevance, type() : ", self.relevance,
type(self.relevance)) # must be <class 'float'>
self.log.info(self.who, "vol_title, type() : ", vol_title,
type(vol_title)) # must be <class 'str'>
self.log.info(
self.who, "vol_auteur, type() : ", vol_auteur,
type(vol_auteur)) # must be <class 'list'> of <class 'str'>
self.log.info(self.who,
"vol_auteur_prenom, type() : ", vol_auteur_prenom,
type(vol_auteur_prenom)) # must be <class 'str'>
self.log.info(self.who,
"vol_auteur_nom, type() : ", vol_auteur_nom,
type(vol_auteur_nom)) # must be <class 'str'>
if vol_serie:
self.log.info(self.who, "vol_serie, type() : ",
vol_serie, type(vol_serie)) # must be <class 'str'>
self.log.info(self.who,
"vol_serie_seq, type() : ", vol_serie_seq,
type(vol_serie_seq)) # must be <class 'float'>
self.log.info(self.who, "vol_editor, type() : ",
vol_editor, type(vol_editor)) # must be <class 'str'>
self.log.info(self.who, "vol_coll, type() : ", vol_coll,
type(vol_coll)) # must be <class 'str'>
self.log.info(self.who,
"vol_coll_srl, type() : ", vol_coll_srl,
type(vol_coll_srl)) # must be <class 'str'>
self.log.info(
self.who, "vol_dp_lgl, type() : ", vol_dp_lgl,
type(vol_dp_lgl)
) # must be <class 'datetime.datetime'> ('renderer=isoformat')
self.log.info(self.who, "vol_isbn, type() : ", vol_isbn,
type(vol_isbn)) # must be <class 'str'>
self.log.info(
self.who, "vol_genre, type() : ", vol_genre,
type(vol_genre)) # must be <class 'list'> of <class 'str'>
self.log.info(self.who, "vol_cover_index, type() : ",
vol_cover_index, type(vol_cover_index)) # must be
self.log.info(self.who, "type(vol_comment_soup) : ",
type(vol_comment_soup)
) # must be byte encoded (start with b'blablabla...
# self.log.info(self.who,"vol_comment_soup :\n",vol_comment_soup) # Maybe a bit long sometimes
# language must be <class 'str'>
if vol_cover_index:
self.plugin.cache_identifier_to_cover_url(self.nsfr_id,
vol_cover_index)
if vol_isbn:
self.plugin.cache_isbn_to_identifier(vol_isbn, self.nsfr_id)
mi = Metadata(vol_title, [vol_auteur])
mi.set_identifier('nsfr_id', self.nsfr_id)
mi.publisher = vol_editor
mi.isbn = vol_isbn
mi.tags = [vol_genre]
mi.source_relevance = self.relevance
mi.has_cover = bool(vol_cover_index)
if vol_dp_lgl:
mi.pubdate = vol_dp_lgl
if vol_serie:
mi.series = vol_serie
mi.series_index = vol_serie_seq
mi.language = "fra"
mi.comments = vol_comment_soup
if debug: self.log.info(self.who, "mi\n", mi, "\n")
self.plugin.clean_downloaded_metadata(mi)
self.result_queue.put(mi)