def convert(self, oeb_book, output, input_plugin, opts, log):
"""Convert from calibre's internal format to KePub."""
self.epub_output_plugin.convert(oeb_book, output, input_plugin, opts, log)
container = KEPubContainer(output, default_log)
if container.is_drm_encumbered:
return
# Write the details file
o = {
"kepub_output_version": ".".join([str(n) for n in self.version]),
"kepub_output_currenttime": datetime.utcnow().ctime(),
}
kte_data_file = self.temporary_file("_KePubOutputPluginInfo")
kte_data_file.write(json.dumps(o))
kte_data_file.close()
container.copy_file_to_container(
kte_data_file.name, name="plugininfo.kte", mt="application/json"
)
title = container.opf_xpath("./opf:metadata/dc:title/text()")
if len(title) > 0:
title = title[0]
else:
title = NULL_VALUES["title"]
authors = container.opf_xpath(
'./opf:metadata/dc:creator[@opf:role="aut"]/text()'
)
if len(authors) < 1:
authors = NULL_VALUES["authors"]
mi = Metadata(title, authors)
language = container.opf_xpath("./opf:metadata/dc:language/text()")
if len(language) > 0:
mi.languages = language
language = language[0]
else:
mi.languages = NULL_VALUES["languages"]
language = NULL_VALUES["language"]
mi.language
modify_epub(
container,
output,
metadata=mi,
opts={
"clean_markup": opts.kepub_clean_markup,
"hyphenate": opts.kepub_hyphenate,
"no-hyphens": opts.kepub_disable_hyphenation,
"smarten_punctuation": False,
"extended_kepub_features": True,
},
)
def convert(self, oeb_book, output, input_plugin, opts, log):
self.epub_output_plugin.convert(oeb_book, output, input_plugin, opts,
log)
container = KEPubContainer(output, default_log)
if container.is_drm_encumbered:
return
# Write the details file
o = {
'kepub_output_version': ".".join([str(n) for n in self.version]),
'kepub_output_currenttime': datetime.utcnow().ctime()
}
kte_data_file = self.temporary_file('_KePubOutputPluginInfo')
kte_data_file.write(json.dumps(o))
kte_data_file.close()
container.copy_file_to_container(kte_data_file.name,
name='plugininfo.kte',
mt='application/json')
title = container.opf_xpath("./opf:metadata/dc:title/text()")
if len(title) > 0:
title = title[0]
else:
title = NULL_VALUES['title']
authors = container.opf_xpath(
'./opf:metadata/dc:creator[@opf:role="aut"]/text()')
if len(authors) < 1:
authors = NULL_VALUES['authors']
mi = Metadata(title, authors)
language = container.opf_xpath("./opf:metadata/dc:language/text()")
if len(language) > 0:
mi.languages = language
language = language[0]
else:
mi.languages = NULL_VALUES['languages']
language = NULL_VALUES['language']
mi.language
modify_epub(container,
output,
metadata=mi,
opts={
'clean_markup': opts.kepub_clean_markup,
'hyphenate': opts.kepub_hyphenate,
'no-hyphens': opts.kepub_disable_hyphenation,
'replace_lang': opts.kepub_replace_lang,
'smarten_punctuation': False,
'extended_kepub_features': True
})
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 parse(self, xml_detail):
data = xml_detail.split('\n')[1].split("|")
self.log(data)
title = data[1]
authors = [data[0]]
comments = data[13]
isbn = data[3]
publisher = data[6]
pub_date_tmp = data[34].split('-')
pub_date = datetime.datetime(int(pub_date_tmp[0]), int(pub_date_tmp[1]), int(pub_date_tmp[2]), tzinfo=utc_tz)
if isbn is not None:
isbn_tmp = re.sub("-", "", isbn)
cover = "%s/images/covers/%s.jpg"%(self.plugin.BASE_URL, isbn_tmp)
else:
cover = None
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.publisher = publisher
mi.pubdate = pub_date
mi.isbn = isbn
mi.cover_url = cover
if cover:
self.plugin.cache_identifier_to_cover_url(self.ident, cover)
return mi
else:
return None
def read_metadata(root):
ans = Metadata(_('Unknown'), [_('Unknown')])
prefixes, refines = read_prefixes(root), read_refines(root)
identifiers = read_identifiers(root, prefixes, refines)
ids = {}
for key, vals in identifiers.iteritems():
if key == 'calibre':
ans.application_id = vals[0]
elif key != 'uuid':
ids[key] = vals[0]
ans.set_identifiers(ids)
ans.title = read_title(root, prefixes, refines) or ans.title
ans.title_sort = read_title_sort(root, prefixes, refines) or ans.title_sort
ans.languages = read_languages(root, prefixes, refines) or ans.languages
auts, aus = [], []
for a in read_authors(root, prefixes, refines):
auts.append(a.name), aus.append(a.sort)
ans.authors = auts or ans.authors
ans.author_sort = authors_to_string(aus) or ans.author_sort
bkp = read_book_producers(root, prefixes, refines)
if bkp:
ans.book_producer = bkp[0]
pd = read_pubdate(root, prefixes, refines)
if not is_date_undefined(pd):
ans.pubdate = pd
ts = read_timestamp(root, prefixes, refines)
if not is_date_undefined(ts):
ans.timestamp = ts
lm = read_last_modified(root, prefixes, refines)
if not is_date_undefined(lm):
ans.last_modified = lm
return ans
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 read_metadata_kfx(stream, read_cover=True):
' Read the metadata.kfx file that is found in the sdr book folder for KFX files '
c = Container(stream.read())
m = extract_metadata(c.decode())
# dump_metadata(m)
def has(x):
return m[x] and m[x][0]
def get(x, single=True):
ans = m[x]
if single:
ans = clean_xml_chars(ans[0]) if ans else ''
else:
ans = [clean_xml_chars(y) for y in ans]
return ans
title = get('title') or _('Unknown')
authors = get('authors', False) or [_('Unknown')]
auth_pat = re.compile(r'([^,]+?)\s*,\s+([^,]+)$')
def fix_author(x):
if tweaks['author_sort_copy_method'] != 'copy':
m = auth_pat.match(x.strip())
if m is not None:
return m.group(2) + ' ' + m.group(1)
return x
mi = Metadata(title, [fix_author(x) for x in authors])
if has('author'):
mi.author_sort = get('author')
if has('ASIN'):
mi.set_identifier('mobi-asin', get('ASIN'))
elif has('content_id'):
mi.set_identifier('mobi-asin', get('content_id'))
if has('languages'):
langs = list(filter(None, (canonicalize_lang(x) for x in get('languages', False))))
if langs:
mi.languages = langs
if has('issue_date'):
try:
mi.pubdate = parse_only_date(get('issue_date'))
except Exception:
pass
if has('publisher') and get('publisher') != 'Unknown':
mi.publisher = get('publisher')
if read_cover and m[COVER_KEY]:
try:
data = base64.standard_b64decode(m[COVER_KEY])
w, h, fmt = identify_data(data)
except Exception:
w, h, fmt = 0, 0, None
if fmt and w and h:
mi.cover_data = (fmt, data)
return mi
def read_metadata_kfx(stream, read_cover=True):
" Read the metadata.kfx file that is found in the sdr book folder for KFX files "
c = Container(stream.read())
m = extract_metadata(c.decode())
# dump_metadata(m)
def has(x):
return m[x] and m[x][0]
def get(x, single=True):
ans = m[x]
if single:
ans = clean_xml_chars(ans[0]) if ans else ""
else:
ans = [clean_xml_chars(y) for y in ans]
return ans
title = get("title") or _("Unknown")
authors = get("authors", False) or [_("Unknown")]
auth_pat = re.compile(r"([^,]+?)\s*,\s+([^,]+)$")
def fix_author(x):
if tweaks["author_sort_copy_method"] != "copy":
m = auth_pat.match(x.strip())
if m is not None:
return m.group(2) + " " + m.group(1)
return x
mi = Metadata(title, [fix_author(x) for x in authors])
if has("author"):
mi.author_sort = get("author")
if has("ASIN"):
mi.set_identifier("mobi-asin", get("ASIN"))
elif has("content_id"):
mi.set_identifier("mobi-asin", get("content_id"))
if has("languages"):
langs = list(filter(None, (canonicalize_lang(x) for x in get("languages", False))))
if langs:
mi.languages = langs
if has("issue_date"):
try:
mi.pubdate = parse_only_date(get("issue_date"))
except Exception:
pass
if has("publisher") and get("publisher") != "Unknown":
mi.publisher = get("publisher")
if read_cover and m[COVER_KEY]:
try:
data = base64.standard_b64decode(m[COVER_KEY])
fmt, w, h = identify(bytes(data))
except Exception:
w, h, fmt = 0, 0, None
if fmt and w > -1 and h > -1:
mi.cover_data = (fmt, data)
return mi
def read_metadata(root, ver=None, return_extra_data=False):
ans = Metadata(_('Unknown'), [_('Unknown')])
prefixes, refines = read_prefixes(root), read_refines(root)
identifiers = read_identifiers(root, prefixes, refines)
ids = {}
for key, vals in iteritems(identifiers):
if key == 'calibre':
ans.application_id = vals[0]
elif key == 'uuid':
ans.uuid = vals[0]
else:
ids[key] = vals[0]
ans.set_identifiers(ids)
ans.title = read_title(root, prefixes, refines) or ans.title
ans.title_sort = read_title_sort(root, prefixes, refines) or ans.title_sort
ans.languages = read_languages(root, prefixes, refines) or ans.languages
auts, aus = [], []
for a in read_authors(root, prefixes, refines):
auts.append(a.name), aus.append(a.sort)
ans.authors = auts or ans.authors
ans.author_sort = authors_to_string(aus) or ans.author_sort
bkp = read_book_producers(root, prefixes, refines)
if bkp:
if bkp[0]:
ans.book_producer = bkp[0]
pd = read_pubdate(root, prefixes, refines)
if not is_date_undefined(pd):
ans.pubdate = pd
ts = read_timestamp(root, prefixes, refines)
if not is_date_undefined(ts):
ans.timestamp = ts
lm = read_last_modified(root, prefixes, refines)
if not is_date_undefined(lm):
ans.last_modified = lm
ans.comments = read_comments(root, prefixes, refines) or ans.comments
ans.publisher = read_publisher(root, prefixes, refines) or ans.publisher
ans.tags = read_tags(root, prefixes, refines) or ans.tags
ans.rating = read_rating(root, prefixes, refines) or ans.rating
s, si = read_series(root, prefixes, refines)
if s:
ans.series, ans.series_index = s, si
ans.author_link_map = read_author_link_map(root, prefixes,
refines) or ans.author_link_map
ans.user_categories = read_user_categories(root, prefixes,
refines) or ans.user_categories
for name, fm in iteritems(
read_user_metadata(root, prefixes, refines) or {}):
ans.set_user_metadata(name, fm)
if return_extra_data:
ans = ans, ver, read_raster_cover(root, prefixes,
refines), first_spine_item(
root, prefixes, refines)
return ans
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': 'https://calibre-ebook.com'}
mi.languages = ['eng', 'fra']
mi.pubdate = mi.timestamp = now()
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 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 test_input_meta_single(self):
stream_meta = get_metadata(self.get_stream('meta_single'))
canon_meta = Metadata('A Meta Tag & Title Ⓒ',
['George Washington'])
canon_meta.publisher = 'Publisher A'
canon_meta.languages = ['English']
canon_meta.pubdate = parse_date('2019-01-01')
canon_meta.timestamp = parse_date('2018-01-01')
canon_meta.series = 'Meta Series'
canon_meta.series_index = float(1)
# canon_meta.rating = float(0)
# canon_meta.comments = ''
canon_meta.tags = ['tag a', 'tag b']
canon_meta.set_identifiers({'isbn': '1234567890'})
self.compare_metadata(stream_meta, canon_meta)
def read_metadata(root, ver=None, return_extra_data=False):
ans = Metadata(_('Unknown'), [_('Unknown')])
prefixes, refines = read_prefixes(root), read_refines(root)
identifiers = read_identifiers(root, prefixes, refines)
ids = {}
for key, vals in identifiers.iteritems():
if key == 'calibre':
ans.application_id = vals[0]
elif key == 'uuid':
ans.uuid = vals[0]
else:
ids[key] = vals[0]
ans.set_identifiers(ids)
ans.title = read_title(root, prefixes, refines) or ans.title
ans.title_sort = read_title_sort(root, prefixes, refines) or ans.title_sort
ans.languages = read_languages(root, prefixes, refines) or ans.languages
auts, aus = [], []
for a in read_authors(root, prefixes, refines):
auts.append(a.name), aus.append(a.sort)
ans.authors = auts or ans.authors
ans.author_sort = authors_to_string(aus) or ans.author_sort
bkp = read_book_producers(root, prefixes, refines)
if bkp:
if bkp[0]:
ans.book_producer = bkp[0]
pd = read_pubdate(root, prefixes, refines)
if not is_date_undefined(pd):
ans.pubdate = pd
ts = read_timestamp(root, prefixes, refines)
if not is_date_undefined(ts):
ans.timestamp = ts
lm = read_last_modified(root, prefixes, refines)
if not is_date_undefined(lm):
ans.last_modified = lm
ans.comments = read_comments(root, prefixes, refines) or ans.comments
ans.publisher = read_publisher(root, prefixes, refines) or ans.publisher
ans.tags = read_tags(root, prefixes, refines) or ans.tags
ans.rating = read_rating(root, prefixes, refines) or ans.rating
s, si = read_series(root, prefixes, refines)
if s:
ans.series, ans.series_index = s, si
ans.author_link_map = read_author_link_map(root, prefixes, refines) or ans.author_link_map
ans.user_categories = read_user_categories(root, prefixes, refines) or ans.user_categories
for name, fm in (read_user_metadata(root, prefixes, refines) or {}).iteritems():
ans.set_user_metadata(name, fm)
if return_extra_data:
ans = ans, ver, read_raster_cover(root, prefixes, refines), first_spine_item(root, prefixes, refines)
return ans
def read_metadata(root):
ans = Metadata(_('Unknown'), [_('Unknown')])
prefixes, refines = read_prefixes(root), read_refines(root)
identifiers = read_identifiers(root, prefixes, refines)
ids = {}
for key, vals in identifiers.iteritems():
if key == 'calibre':
ans.application_id = vals[0]
elif key != 'uuid':
ids[key] = vals[0]
ans.set_identifiers(ids)
ans.title = read_title(root, prefixes, refines) or ans.title
ans.title_sort = read_title_sort(root, prefixes, refines) or ans.title_sort
ans.languages = read_languages(root, prefixes, refines) or ans.languages
return ans
def identify(self,
log,
result_queue,
abort,
title=None,
authors=None,
identifiers={},
timeout=30):
isbn = check_isbn(identifiers.get('isbn', None))
if isbn is None:
log.info("The plugin can only look up data by ISBN")
return None
url = self.QURL % isbn
log.info('Querying: %s' % url)
data = json.loads(
self.browser.open_novisit(url, timeout=timeout).read())
log.info('Got JSON data: %s' % data)
if data.get('stat', None) != 'ok':
return None
data = data.get('list', [])
if len(data) == 0:
return None
data = data[0]
log.info("Book JSON data: %s" % data)
title = data.get("title", None)
authors = data.get("author", None)
if title is None or authors is None:
return None
authors = authors.rstrip('.').replace(' and ',
',').replace(';', ',').split(',')
mi = Metadata(title, [a.strip() for a in authors])
mi.publisher = data.get("publisher", None)
mi.languages = data.get("lang", None).split(',')
year = data.get("year", None)
if year is not None:
mi.pubdate = datetime.datetime(int(year), 1, 1)
log.info("Final formatted result: %s" % mi)
result_queue.put(mi)
def test_input_comment_single(self):
stream_meta = get_metadata(self.get_stream('comment_single'))
canon_meta = Metadata('A Comment Tag & Title Ⓒ',
['James Madison', 'James Monroe'])
canon_meta.publisher = 'Publisher C'
canon_meta.languages = ['French']
canon_meta.pubdate = parse_date('2015-01-01')
canon_meta.timestamp = parse_date('2014-01-01')
canon_meta.series = 'Comment Series'
canon_meta.series_index = float(3)
canon_meta.rating = float(0)
canon_meta.comments = 'comment "comments" ♥ HTML -- too &amp;'
canon_meta.tags = ['tag d']
canon_meta.set_identifiers({
'isbn': '3456789012',
'url': 'http://google.com/search?q=calibre'
})
self.compare_metadata(stream_meta, canon_meta)
def test_input_meta_multi(self):
stream_meta = get_metadata(self.get_stream('meta_multi'))
canon_meta = Metadata(
'A Meta Tag & Title Ⓒ',
['George Washington', 'John Adams', 'Thomas Jefferson'])
canon_meta.publisher = 'Publisher A'
canon_meta.languages = ['English', 'Spanish']
canon_meta.pubdate = parse_date('2019-01-01')
canon_meta.timestamp = parse_date('2018-01-01')
canon_meta.series = 'Meta Series'
canon_meta.series_index = float(1)
canon_meta.rating = float(8)
canon_meta.comments = 'meta "comments" ♥ HTML &amp;'
canon_meta.tags = ['tag a', 'tag b', 'tag c']
canon_meta.set_identifiers({
'isbn': '1234567890',
'url': 'http://google.com/search?q=calibre'
})
self.compare_metadata(stream_meta, canon_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)
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 metadata_from_xmp_packet(raw_bytes):
root = parse_xmp_packet(raw_bytes)
mi = Metadata(_('Unknown'))
title = first_alt('//dc:title', root)
if title:
mi.title = title
authors = multiple_sequences('//dc:creator', root)
if authors:
mi.authors = authors
tags = multiple_sequences('//dc:subject', root) or multiple_sequences('//pdf:Keywords', root)
if tags:
mi.tags = tags
comments = first_alt('//dc:description', root)
if comments:
mi.comments = comments
publishers = multiple_sequences('//dc:publisher', root)
if publishers:
mi.publisher = publishers[0]
try:
pubdate = parse_date(first_sequence('//dc:date', root) or first_simple('//xmp:CreateDate', root), assume_utc=False)
except:
pass
else:
mi.pubdate = pubdate
bkp = first_simple('//xmp:CreatorTool', root)
if bkp:
mi.book_producer = bkp
rating = first_simple('//calibre:rating', root)
if rating is not None:
try:
rating = float(rating)
if 0 <= rating <= 10:
mi.rating = rating
except (ValueError, TypeError):
pass
series, series_index = read_series(root)
if series:
mi.series, mi.series_index = series, series_index
for x in ('title_sort', 'author_sort'):
for elem in XPath('//calibre:' + x)(root):
val = read_simple_property(elem)
if val:
setattr(mi, x, val)
break
languages = multiple_sequences('//dc:language', root)
if languages:
languages = filter(None, map(canonicalize_lang, languages))
if languages:
mi.languages = languages
identifiers = {}
for xmpid in XPath('//xmp:Identifier')(root):
for scheme, value in read_xmp_identifers(xmpid):
if scheme and value:
identifiers[scheme.lower()] = value
for namespace in ('prism', 'pdfx'):
for scheme in KNOWN_ID_SCHEMES:
if scheme not in identifiers:
val = first_simple('//%s:%s' % (namespace, scheme), root)
scheme = scheme.lower()
if scheme == 'isbn':
val = check_isbn(val)
elif scheme == 'doi':
val = check_doi(val)
if val:
identifiers[scheme] = val
# Check Dublin Core for recognizable identifier types
for scheme, check_func in {'doi':check_doi, 'isbn':check_isbn}.iteritems():
if scheme not in identifiers:
val = check_func(first_simple('//dc:identifier', root))
if val:
identifiers['doi'] = val
if identifiers:
mi.set_identifiers(identifiers)
return mi
def parse_details(self, root):
try:
title = self.parse_title(root)
except:
self.log.exception('Error parsing title for query: %r' %
self.query)
title = None
if not title:
self.log.error('Could not find title for %r' % self.query)
try:
authors = self.parse_authors(root)
except:
self.log.exception('Error parsing authors for query: %r' %
self.query)
authors = []
if not authors:
self.log.error('Could not find authors for %r' % self.query)
return
mi = Metadata(title, authors)
try:
isbn = self.parse_isbn(root)
if isbn:
# match 10 of 13 getallen aan het begin, gevolgd door een spatie of niets
p = re.compile('^([0-9]{13}|[0-9]{10})(?= |\Z)')
if isinstance(isbn, str):
m = p.match(isbn)
if m:
mi.isbn = m.group()
else:
m = p.match(isbn[0])
if m:
mi.isbn = m.group()
except:
self.log.exception('Error parsing ISBN for url: %r' % self.url)
try:
lang = self.parse_language(root)
if lang:
mi.languages = lang
except:
self.log.exception('Error parsing language for url: %r' % self.url)
try:
lccn = self.parse_lccn(root)
if lccn:
if isinstance(lccn, str):
mi.set_identifier('lccn', lccn)
else:
for identifier in lccn:
mi.set_identifier('lccn', identifier)
except:
self.log.exception('Error parsing LCCN for url: %r' % self.url)
try:
ddc = self.parse_ddc(root)
if ddc:
if isinstance(ddc, str):
mi.set_identifier('ddc', ddc)
else:
for identifier in ddc:
mi.set_identifier('ddc', identifier)
except:
self.log.exception('Error parsing DDC for url: %r' % self.url)
try:
lcc = self.parse_lcc(root)
if lcc:
if isinstance(lcc, str):
mi.set_identifier('lcc', lcc)
else:
for identifier in lcc:
mi.set_identifier('lcc', identifier)
except:
self.log.exception('Error parsing LCC for url: %r' % self.url)
mi.source_relevance = self.relevance
self.result_queue.put(mi)
def convert(self, oeb_book, output, input_plugin, opts, logger):
"""Convert from calibre's internal format to KePub."""
common.log.debug("Running ePub conversion")
self.epub_output_plugin.convert(
oeb_book, output, input_plugin, opts, common.log
)
common.log.debug("Done ePub conversion")
container = KEPubContainer(output, common.log, opts.kepub_clean_markup)
if container.is_drm_encumbered:
common.log.error("DRM-encumbered container, skipping conversion")
return
# Write the details file
o = {
"kepub_output_version": ".".join([str(n) for n in self.version]),
"kepub_output_currenttime": datetime.utcnow().ctime(),
}
kte_data_file = self.temporary_file("_KePubOutputPluginInfo")
kte_data_file.write(json.dumps(o).encode("UTF-8"))
kte_data_file.close()
container.copy_file_to_container(
kte_data_file.name, name="plugininfo.kte", mt="application/json"
)
title = container.opf_xpath("./opf:metadata/dc:title/text()")
if len(title) > 0:
title = title[0]
else:
title = NULL_VALUES["title"]
authors = container.opf_xpath(
'./opf:metadata/dc:creator[@opf:role="aut"]/text()'
)
if len(authors) < 1:
authors = NULL_VALUES["authors"]
mi = Metadata(title, authors)
language = container.opf_xpath("./opf:metadata/dc:language/text()")
if len(language) > 0:
mi.languages = language
language = language[0]
else:
mi.languages = NULL_VALUES["languages"]
language = NULL_VALUES["language"]
try:
common.modify_epub(
container,
output,
metadata=mi,
opts={
"hyphenate": opts.kepub_hyphenate,
"hyphen_min_chars": opts.kepub_hyphenate_chars,
"hyphen_min_chars_before": opts.kepub_hyphenate_chars_before,
"hyphen_min_chars_after": opts.kepub_hyphenate_chars_after,
"hyphen_limit_lines": opts.kepub_hyphenate_limit_lines,
"no-hyphens": opts.kepub_disable_hyphenation,
"smarten_punctuation": False,
"extended_kepub_features": True,
},
)
except Exception:
common.log.exception("Failed converting!")
raise
def metadata_from_xmp_packet(raw_bytes):
root = parse_xmp_packet(raw_bytes)
mi = Metadata(_('Unknown'))
title = first_alt('//dc:title', root)
if title:
mi.title = title
authors = multiple_sequences('//dc:creator', root)
if authors:
mi.authors = authors
tags = multiple_sequences('//dc:subject', root) or multiple_sequences(
'//pdf:Keywords', root)
if tags:
mi.tags = tags
comments = first_alt('//dc:description', root)
if comments:
mi.comments = comments
publishers = multiple_sequences('//dc:publisher', root)
if publishers:
mi.publisher = publishers[0]
try:
pubdate = parse_date(first_sequence('//dc:date', root)
or first_simple('//xmp:CreateDate', root),
assume_utc=False)
except:
pass
else:
mi.pubdate = pubdate
bkp = first_simple('//xmp:CreatorTool', root)
if bkp:
mi.book_producer = bkp
md = first_simple('//xmp:MetadataDate', root)
if md:
try:
mi.metadata_date = parse_date(md)
except:
pass
rating = first_simple('//calibre:rating', root)
if rating is not None:
try:
rating = float(rating)
if 0 <= rating <= 10:
mi.rating = rating
except (ValueError, TypeError):
pass
series, series_index = read_series(root)
if series:
mi.series, mi.series_index = series, series_index
for x in ('title_sort', 'author_sort'):
for elem in XPath('//calibre:' + x)(root):
val = read_simple_property(elem)
if val:
setattr(mi, x, val)
break
for x in ('author_link_map', 'user_categories'):
val = first_simple('//calibre:' + x, root)
if val:
try:
setattr(mi, x, json.loads(val))
except:
pass
languages = multiple_sequences('//dc:language', root)
if languages:
languages = filter(None, map(canonicalize_lang, languages))
if languages:
mi.languages = languages
identifiers = {}
for xmpid in XPath('//xmp:Identifier')(root):
for scheme, value in read_xmp_identifers(xmpid):
if scheme and value:
identifiers[scheme.lower()] = value
for namespace in ('prism', 'pdfx'):
for scheme in KNOWN_ID_SCHEMES:
if scheme not in identifiers:
val = first_simple('//%s:%s' % (namespace, scheme), root)
scheme = scheme.lower()
if scheme == 'isbn':
val = check_isbn(val)
elif scheme == 'doi':
val = check_doi(val)
if val:
identifiers[scheme] = val
# Check Dublin Core for recognizable identifier types
for scheme, check_func in {
'doi': check_doi,
'isbn': check_isbn
}.iteritems():
if scheme not in identifiers:
val = check_func(first_simple('//dc:identifier', root))
if val:
identifiers['doi'] = val
if identifiers:
mi.set_identifiers(identifiers)
read_user_metadata(mi, root)
return mi
def mi(self):
mi = Metadata(unicode_type(self.title.text()).strip() or _('Unknown'))
mi.authors = string_to_authors(unicode_type(self.authors.text()).strip()) or [_('Unknown')]
mi.languages = self.languages.lang_codes or [get_lang()]
return 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:
self.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)
# 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.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))
self.set_up_font_boxes()
self.textbox.setFocus()
# 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(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_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_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 identify(self,
log,
result_queue,
abort,
title=None,
authors=None,
identifiers={},
timeout=30):
self.load_config()
# get identifying tags from book
idn = identifiers.get('dnb-idn', None)
isbn = check_isbn(identifiers.get('isbn', None))
# ignore unknown authors
if authors is "V. A." or authors is "V.A." or authors is "Unknown" or authors is "Unbekannt":
authors = None
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:
queries.append('num=' + idn)
exact_search['idn'] = idn
else:
authors_v = []
title_v = []
if authors is not None:
authors_v.append(' '.join(authors))
authors_v.append(' '.join(
self.get_author_tokens(authors, only_first_author=False)))
authors_v.append(' '.join(
self.get_author_tokens(authors, only_first_author=True)))
if title is not None:
title_v.append(title)
title_v.append(' '.join(
self.get_title_tokens(title,
strip_joiners=False,
strip_subtitle=False)))
title_v.append(' '.join(
self.get_title_tokens(title,
strip_joiners=False,
strip_subtitle=True)))
if isbn is not None:
exact_search['isbn'] = isbn
# title and author
if authors is not None and title is not None:
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 author and title swapped
if isbn is not None:
queries.append('per="' + title + '" AND tit="' +
authors[0] + '" AND num="' + isbn + '"')
else:
queries.append('per="' + title + '" AND tit="' +
authors[0] + '"')
# title but no author
elif authors is not None and title is None:
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 and title swapped
if isbn is not None:
queries.append('tit="' + authors[0] + '" AND num="' +
isbn + '"')
else:
queries.append('tit="' + authors[0] + '"')
# author but no title
elif authors is None and title is not None:
for i in title_v:
if isbn is not None:
queries.append('tit="' + i + '" AND num="' + isbn +
'"')
else:
queries.append('tit="' + i + '"')
# try with author and title swapped
if isbn is not None:
queries.append('per="' + title + '" AND num="' + isbn +
'"')
else:
queries.append('per="' + title + '"')
# as last resort only use isbn
if isbn is not None:
queries.append('num=' + isbn)
# Sort queries descending by length (assumption: longer query -> less but better results)
#queries.sort(key=len)
#queries.reverse()
# remove duplicate queries
uniqueQueries = []
for i in queries:
if i not in uniqueQueries:
uniqueQueries.append(i)
# Process queries
results = None
for query in uniqueQueries:
query = query + ' NOT (mat=film OR mat=music OR mat=microfiches)'
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
edition = None
comments = None
idn = None
urn = None
isbn = None
ddc = []
subjects_gnd = []
subjects_non_gnd = []
# Title: Field 245
title_parts = []
# if a,n,p exist: series = a, series_index = n, title = p
for i in record.xpath(
".//marc21:datafield[@tag='245']/marc21:subfield[@code='a' and string-length(text())>0]/../marc21:subfield[@code='n' and string-length(text())>0]/../marc21:subfield[@code='p' and string-length(text())>0]/..",
namespaces=ns):
series_index = i.xpath(".//marc21:subfield[@code='n']",
namespaces=ns)[0].text.strip()
match = re.search("(\d+[,\.\d+]?)", series_index)
if match:
series_index = match.group(1)
else:
series_index = "0" # looks like sometimes DNB does not know the series index and uses something like "[...]"
series_index = series_index.replace(',', '.')
series = i.xpath(".//marc21:subfield[@code='a']",
namespaces=ns)[0].text.strip()
title_parts.append(
i.xpath(".//marc21:subfield[@code='p']",
namespaces=ns)[0].text.strip())
log.info("Extracted Series: %s" % series)
log.info("Extracted Series Index: %s" % series_index)
break
# otherwise: title = a
if len(title_parts) == 0:
for i in record.xpath(
".//marc21:datafield[@tag='245']/marc21:subfield[@code='a' and string-length(text())>0]",
namespaces=ns):
title_parts.append(i.text.strip())
break
# subtitle 1
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.info("Extracted Title: %s" % title)
# Title_Sort
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.info("Extracted Title_Sort: %s" % title_sort)
# Authors
authors = []
author_sort = None
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.info("Extracted Authors: %s" % " & ".join(authors))
# 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 = sanitize_comments_html(comments)
log.info('Comments: %s' % comments)
break
except:
log.info("Could not download Comments from %s" % i)
# Publisher Name and Location
publisher_name = None
publisher_location = None
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()
log.info("Extracted Publisher: %s" % publisher_name)
log.info("Extracted Publisher Location: %s" %
publisher_location)
# 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, 2)
break
log.info("Extracted Publication Year: %s" % pubdate)
# 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.info("Extracted ID IDN: %s" % idn)
if "idn" in exact_search:
if idn != exact_search["idn"]:
log.info(
"Extracted IDN does not match book's IDN, skipping record"
)
continue
# 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.info("Extracted ID URN: %s" % urn)
# 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.info("Extracted ID ISBN: %s" % isbn)
if "isbn" in exact_search:
if isbn != exact_search["isbn"]:
log.info(
"Extracted ISBN does not match book's ISBN, skipping record"
)
continue
# 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.info("Extracted ID DDC: %s" % ",".join(ddc))
# Series and Series_Index
if series is None and series_index is None:
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(',', '.')
log.info("Extracted Series Index: %s" % series_index)
# Series
series = i.xpath(".//marc21:subfield[@code='a']",
namespaces=ns)[0].text.strip()
log.info("Extracted Series: %s" % series)
break
# Try to extract Series, Series Index and Title from the fetched title.
# Caution: This overwrites DNB's series/series_index and modifies the title!
if self.cfg_guess_series is True:
guessed_series = None
guessed_series_index = None
parts = re.split("[:]",
self.removeSortingCharacters(title))
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.match(
"^[\s\-–:]*(.+?)[\s\-–:]*$", textpart
) # remove odd characters from start and end of the text part
if match:
textpart = match.group(1)
# from Titleparts like: "Name of the series - Episode 2" OK
match = re.match(
"^\s*(\S.*?)[\(\/\.,\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
title = textpart + " : Band " + guessed_series_index
else:
title = textpart
else:
# from 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.*?)[\/\.,\-–\s]*$",
indexpart)
if match:
guessed_series_index = match.group(1)
guessed_series = match.group(2)
if guessed_series is None:
guessed_series = textpart
title = textpart + " : Band " + guessed_series_index
else:
title = textpart
elif len(parts) == 1:
# from 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)
title = match.group(2)
else:
# from 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)
title = guessed_series + " : Band " + guessed_series_index
if guessed_series is not None and guessed_series_index is not None:
series = guessed_series
series_index = guessed_series_index
log.info("Guessed Series: %s" % series)
log.info("Guessed Series Index: %s" % series_index)
# GND Subjects from 689
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())
# GND Subjects from 600-655
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.info("Extracted GND Subjects: %s" % " ".join(subjects_gnd))
# Non-GND subjects from 600-655
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.info("Extracted non-GND Subjects: %s" %
" ".join(subjects_non_gnd))
# 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.info("Extracted Edition: %s" % edition)
# 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())
if languages is not None:
log.info("Extracted Languages: %s" % ",".join(languages))
# 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))
if self.cfg_fetch_subjects == 0:
mi.tags = self.uniq(subjects_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)
elif self.cfg_fetch_subjects == 2:
mi.tags = self.uniq(subjects_gnd + subjects_non_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)
elif self.cfg_fetch_subjects == 4:
mi.tags = self.uniq(subjects_non_gnd)
elif self.cfg_fetch_subjects == 5:
mi.tags = []
# put current result's metdata into result queue
log.info("Final formatted result: %s" % mi)
result_queue.put(mi)
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 _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 metadata_from_xmp_packet(raw_bytes):
root = parse_xmp_packet(raw_bytes)
mi = Metadata(_("Unknown"))
title = first_alt("//dc:title", root)
if title:
mi.title = title
authors = multiple_sequences("//dc:creator", root)
if authors:
mi.authors = authors
tags = multiple_sequences("//dc:subject", root) or multiple_sequences("//pdf:Keywords", root)
if tags:
mi.tags = tags
comments = first_alt("//dc:description", root)
if comments:
mi.comments = comments
publishers = multiple_sequences("//dc:publisher", root)
if publishers:
mi.publisher = publishers[0]
try:
pubdate = parse_date(
first_sequence("//dc:date", root) or first_simple("//xmp:CreateDate", root), assume_utc=False
)
except:
pass
else:
mi.pubdate = pubdate
bkp = first_simple("//xmp:CreatorTool", root)
if bkp:
mi.book_producer = bkp
md = first_simple("//xmp:MetadataDate", root)
if md:
try:
mi.metadata_date = parse_date(md)
except:
pass
rating = first_simple("//calibre:rating", root)
if rating is not None:
try:
rating = float(rating)
if 0 <= rating <= 10:
mi.rating = rating
except (ValueError, TypeError):
pass
series, series_index = read_series(root)
if series:
mi.series, mi.series_index = series, series_index
for x in ("title_sort", "author_sort"):
for elem in XPath("//calibre:" + x)(root):
val = read_simple_property(elem)
if val:
setattr(mi, x, val)
break
for x in ("author_link_map", "user_categories"):
val = first_simple("//calibre:" + x, root)
if val:
try:
setattr(mi, x, json.loads(val))
except:
pass
languages = multiple_sequences("//dc:language", root)
if languages:
languages = filter(None, map(canonicalize_lang, languages))
if languages:
mi.languages = languages
identifiers = {}
for xmpid in XPath("//xmp:Identifier")(root):
for scheme, value in read_xmp_identifers(xmpid):
if scheme and value:
identifiers[scheme.lower()] = value
for namespace in ("prism", "pdfx"):
for scheme in KNOWN_ID_SCHEMES:
if scheme not in identifiers:
val = first_simple("//%s:%s" % (namespace, scheme), root)
scheme = scheme.lower()
if scheme == "isbn":
val = check_isbn(val)
elif scheme == "doi":
val = check_doi(val)
if val:
identifiers[scheme] = val
# Check Dublin Core for recognizable identifier types
for scheme, check_func in {"doi": check_doi, "isbn": check_isbn}.iteritems():
if scheme not in identifiers:
val = check_func(first_simple("//dc:identifier", root))
if val:
identifiers["doi"] = val
if identifiers:
mi.set_identifiers(identifiers)
read_user_metadata(mi, root)
return mi
def metadata_from_xmp_packet(raw_bytes):
root = parse_xmp_packet(raw_bytes)
mi = Metadata(_('Unknown'))
title = first_alt('//dc:title', root)
if title:
if title.startswith(r'\376\377'):
# corrupted XMP packet generated by Nitro PDF. See
# https://bugs.launchpad.net/calibre/+bug/1541981
raise ValueError(
'Corrupted XMP metadata packet detected, probably generated by Nitro PDF'
)
mi.title = title
authors = multiple_sequences('//dc:creator', root)
if authors:
mi.authors = [au for aus in authors for au in string_to_authors(aus)]
tags = multiple_sequences('//dc:subject', root) or multiple_sequences(
'//pdf:Keywords', root)
if tags:
mi.tags = tags
comments = first_alt('//dc:description', root)
if comments:
mi.comments = comments
publishers = multiple_sequences('//dc:publisher', root)
if publishers:
mi.publisher = publishers[0]
try:
pubdate = parse_date(first_sequence('//dc:date', root)
or first_simple('//xmp:CreateDate', root),
assume_utc=False)
except:
pass
else:
mi.pubdate = pubdate
bkp = first_simple('//xmp:CreatorTool', root)
if bkp:
mi.book_producer = bkp
md = safe_parse_date(first_simple('//xmp:MetadataDate', root))
mod = safe_parse_date(first_simple('//xmp:ModifyDate', root))
fd = more_recent(md, mod)
if fd is not None:
mi.metadata_date = fd
rating = first_simple('//calibre:rating', root)
if rating is not None:
try:
rating = float(rating)
if 0 <= rating <= 10:
mi.rating = rating
except (ValueError, TypeError):
pass
series, series_index = read_series(root)
if series:
mi.series, mi.series_index = series, series_index
for x in ('title_sort', 'author_sort'):
for elem in XPath('//calibre:' + x)(root):
val = read_simple_property(elem)
if val:
setattr(mi, x, val)
break
for x in ('author_link_map', 'user_categories'):
val = first_simple('//calibre:' + x, root)
if val:
try:
setattr(mi, x, json.loads(val))
except:
pass
languages = multiple_sequences('//dc:language', root)
if languages:
languages = list(filter(None, map(canonicalize_lang, languages)))
if languages:
mi.languages = languages
identifiers = {}
for xmpid in XPath('//xmp:Identifier')(root):
for scheme, value in read_xmp_identifers(xmpid):
if scheme and value:
identifiers[scheme.lower()] = value
for namespace in ('prism', 'pdfx'):
for scheme in KNOWN_ID_SCHEMES:
if scheme not in identifiers:
val = first_simple('//%s:%s' % (namespace, scheme), root)
scheme = scheme.lower()
if scheme == 'isbn':
val = check_isbn(val)
elif scheme == 'doi':
val = check_doi(val)
if val:
identifiers[scheme] = val
# Check Dublin Core for recognizable identifier types
for scheme, check_func in iteritems({
'doi': check_doi,
'isbn': check_isbn
}):
if scheme not in identifiers:
val = check_func(first_simple('//dc:identifier', root))
if val:
identifiers['doi'] = val
if identifiers:
mi.set_identifiers(identifiers)
read_user_metadata(mi, root)
return mi
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, raw, root):
#解析元数据各字段数据
#self.log.info("=====")
try:
asin = self.parse_asin(root)
except:
self.log.exception('Error parsing asin for url: %r' % self.url)
asin = None
if self.testing:
import tempfile, uuid
with tempfile.NamedTemporaryFile(
prefix=(asin or str(uuid.uuid4())) + '_',
suffix='.html',
delete=False) as f:
f.write(raw)
print('Downloaded html for', asin, 'saved in', f.name)
# 分析取得书名
try:
title = self.parse_title(root)
except:
self.log.exception('Error parsing title for url: %r' % self.url)
title = None
#分析取得作者
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 asin:
self.log.error('Could not find title/authors/asin for %r' %
self.url)
self.log.error('ASIN: %r Title: %r Authors: %r' %
(asin, title, authors))
return
#以书名，作者为元数据对象mi，用于设置元数据
mi = Metadata(title, authors)
#设置Bookid
idtype = '17k'
mi.set_identifier(idtype, asin)
self.k17k_id = asin
#设备注释（简介）
try:
mi.comments = self.parse_comments(root)
except:
self.log.exception('Error parsing comments for url: %r' % self.url)
#设置丛书系列
try:
series, series_index = self.parse_series(root)
if series:
mi.series, mi.series_index = series, series_index
elif self.testing:
mi.series, mi.series_index = 'Dummy series for testing', 1
except:
self.log.exception('Error parsing series for url: %r' % self.url)
#设置标签
try:
mi.tags = self.parse_tags(root)
except:
self.log.exception('Error parsing tags for url: %r' % self.url)
#设置最后更新日期
# try:
# mi.last_modified = self.parse_last_modified(root)
# except:
# self.log.exception('Error parsing last_modified for url: %r'%self.url)
#设置封面
try:
self.cover_url = self.parse_cover(root, raw)
except:
self.log.exception('Error parsing cover for url: %r' % self.url)
mi.has_cover = bool(self.cover_url)
mi.source_relevance = self.relevance
mi.languages = [
u'中文',
]
if self.k17k_id:
if self.isbn:
self.plugin.cache_isbn_to_identifier(self.isbn, self.k17k_id)
if self.cover_url:
self.plugin.cache_identifier_to_cover_url(
self.k17k_id, self.cover_url)
self.plugin.clean_downloaded_metadata(mi)
self.result_queue.put(mi)
def metadata_from_xmp_packet(raw_bytes):
root = parse_xmp_packet(raw_bytes)
mi = Metadata(_('Unknown'))
title = first_alt('//dc:title', root)
if title.startswith(r'\376\377'):
# corrupted XMP packet generated by Nitro PDF. See
# https://bugs.launchpad.net/calibre/+bug/1541981
raise ValueError('Corrupted XMP metadata packet detected, probably generated by Nitro PDF')
if title:
mi.title = title
authors = multiple_sequences('//dc:creator', root)
if authors:
mi.authors = authors
tags = multiple_sequences('//dc:subject', root) or multiple_sequences('//pdf:Keywords', root)
if tags:
mi.tags = tags
comments = first_alt('//dc:description', root)
if comments:
mi.comments = comments
publishers = multiple_sequences('//dc:publisher', root)
if publishers:
mi.publisher = publishers[0]
try:
pubdate = parse_date(first_sequence('//dc:date', root) or first_simple('//xmp:CreateDate', root), assume_utc=False)
except:
pass
else:
mi.pubdate = pubdate
bkp = first_simple('//xmp:CreatorTool', root)
if bkp:
mi.book_producer = bkp
md = safe_parse_date(first_simple('//xmp:MetadataDate', root))
mod = safe_parse_date(first_simple('//xmp:ModifyDate', root))
fd = more_recent(md, mod)
if fd is not None:
mi.metadata_date = fd
rating = first_simple('//calibre:rating', root)
if rating is not None:
try:
rating = float(rating)
if 0 <= rating <= 10:
mi.rating = rating
except (ValueError, TypeError):
pass
series, series_index = read_series(root)
if series:
mi.series, mi.series_index = series, series_index
for x in ('title_sort', 'author_sort'):
for elem in XPath('//calibre:' + x)(root):
val = read_simple_property(elem)
if val:
setattr(mi, x, val)
break
for x in ('author_link_map', 'user_categories'):
val = first_simple('//calibre:'+x, root)
if val:
try:
setattr(mi, x, json.loads(val))
except:
pass
languages = multiple_sequences('//dc:language', root)
if languages:
languages = filter(None, map(canonicalize_lang, languages))
if languages:
mi.languages = languages
identifiers = {}
for xmpid in XPath('//xmp:Identifier')(root):
for scheme, value in read_xmp_identifers(xmpid):
if scheme and value:
identifiers[scheme.lower()] = value
for namespace in ('prism', 'pdfx'):
for scheme in KNOWN_ID_SCHEMES:
if scheme not in identifiers:
val = first_simple('//%s:%s' % (namespace, scheme), root)
scheme = scheme.lower()
if scheme == 'isbn':
val = check_isbn(val)
elif scheme == 'doi':
val = check_doi(val)
if val:
identifiers[scheme] = val
# Check Dublin Core for recognizable identifier types
for scheme, check_func in {'doi':check_doi, 'isbn':check_isbn}.iteritems():
if scheme not in identifiers:
val = check_func(first_simple('//dc:identifier', root))
if val:
identifiers['doi'] = val
if identifiers:
mi.set_identifiers(identifiers)
read_user_metadata(mi, root)
return mi
def read_metadata_kfx(stream, read_cover=True):
' Read the metadata.kfx file that is found in the sdr book folder for KFX files '
c = Container(stream.read())
m = extract_metadata(c.decode())
# dump_metadata(m)
def has(x):
return m[x] and m[x][0]
def get(x, single=True):
ans = m[x]
if single:
ans = clean_xml_chars(ans[0]) if ans else ''
else:
ans = [clean_xml_chars(y) for y in ans]
return ans
title = get('title') or _('Unknown')
authors = get('author', False) or [_('Unknown')]
auth_pat = re.compile(r'([^,]+?)\s*,\s+([^,]+)$')
def fix_author(x):
if tweaks['author_sort_copy_method'] != 'copy':
m = auth_pat.match(x.strip())
if m is not None:
return m.group(2) + ' ' + m.group(1)
return x
unique_authors = [] # remove duplicates while retaining order
for f in [fix_author(x) for x in authors]:
if f not in unique_authors:
unique_authors.append(f)
mi = Metadata(title, unique_authors)
if has('author'):
mi.author_sort = get('author')
if has('ASIN'):
mi.set_identifier('mobi-asin', get('ASIN'))
elif has('content_id'):
mi.set_identifier('mobi-asin', get('content_id'))
if has('languages'):
langs = list(
filter(None,
(canonicalize_lang(x) for x in get('languages', False))))
if langs:
mi.languages = langs
if has('issue_date'):
try:
mi.pubdate = parse_only_date(get('issue_date'))
except Exception:
pass
if has('publisher') and get('publisher') != 'Unknown':
mi.publisher = get('publisher')
if read_cover and m[COVER_KEY]:
try:
data = base64.standard_b64decode(m[COVER_KEY])
fmt, w, h = identify(bytes(data))
except Exception:
w, h, fmt = 0, 0, None
if fmt and w > -1 and h > -1:
mi.cover_data = (fmt, data)
return mi
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)
