def convert_markdown_with_metadata(txt, title='', extensions=DEFAULT_MD_EXTENSIONS):
from calibre.ebooks.metadata.book.base import Metadata
from calibre.utils.date import parse_only_date
from calibre.db.write import get_series_values
if 'meta' not in extensions:
extensions.append('meta')
md = create_markdown_object(extensions)
html = md.convert(txt)
mi = Metadata(title or _('Unknown'))
m = md.Meta
for k, v in iteritems({'date':'pubdate', 'summary':'comments'}):
if v not in m and k in m:
m[v] = m.pop(k)
for k in 'title authors series tags pubdate comments publisher rating'.split():
val = m.get(k)
if val:
mf = mi.metadata_for_field(k)
if not mf.get('is_multiple'):
val = val[0]
if k == 'series':
val, si = get_series_values(val)
mi.series_index = 1 if si is None else si
if k == 'rating':
try:
val = max(0, min(int(float(val)), 10))
except Exception:
continue
if mf.get('datatype') == 'datetime':
try:
val = parse_only_date(val, assume_utc=False)
except Exception:
continue
setattr(mi, k, val)
return mi, HTML_TEMPLATE % (mi.title, html)
def metadata(self):
mi = Metadata(_('Unknown'))
dp_name, ap_name = self.get_document_properties_names()
if dp_name:
try:
raw = self.read(dp_name)
except KeyError:
pass
else:
read_doc_props(raw, mi, self.namespace.XPath)
if mi.is_null('language'):
try:
raw = self.read('word/styles.xml')
except KeyError:
pass
else:
read_default_style_language(raw, mi, self.namespace.XPath)
ap_name = self.relationships.get(self.namespace.names['APPPROPS'], None)
if ap_name:
try:
raw = self.read(ap_name)
except KeyError:
pass
else:
read_app_props(raw, mi)
return mi
def __init__(self, storage_id, lpath, other=None):
Metadata.__init__(self, _('Unknown'), other=other)
self.storage_id, self.lpath = storage_id, lpath
self.lpath = self.path = self.lpath.replace(os.sep, '/')
self.mtp_relpath = tuple([icu_lower(x) for x in self.lpath.split('/')])
self.datetime = utcnow().timetuple()
self.thumbail = None
def opdsToMetadata(self, opdsBookStructure):
authors = opdsBookStructure.author.replace(u'& ', u'&')
metadata = Metadata(opdsBookStructure.title, authors.split(u'&'))
metadata.uuid = opdsBookStructure.id.replace('urn:uuid:', '', 1)
rawTimestamp = opdsBookStructure.updated
parsableTimestamp = re.sub('((\.[0-9]+)?\+00:00|Z)$', '', rawTimestamp)
metadata.timestamp = datetime.datetime.strptime(parsableTimestamp, '%Y-%m-%dT%H:%M:%S')
tags = []
summary = opdsBookStructure.get(u'summary', u'')
summarylines = summary.splitlines()
for summaryline in summarylines:
if summaryline.startswith(u'TAGS: '):
tagsline = summaryline.replace(u'TAGS: ', u'')
tagsline = tagsline.replace(u'<br />',u'')
tagsline = tagsline.replace(u', ', u',')
tags = tagsline.split(u',')
metadata.tags = tags
bookDownloadUrls = []
links = opdsBookStructure.get('links', [])
for link in links:
url = link.get('href', '')
bookType = link.get('type', '')
# Skip covers and thumbnails
if not bookType.startswith('image/'):
if bookType == 'application/epub+zip':
# EPUB books are preferred and always put at the head of the list if found
bookDownloadUrls.insert(0, url)
else:
# Formats other than EPUB (eg. AZW), are appended as they are found
bookDownloadUrls.append(url)
metadata.links = bookDownloadUrls
return metadata
def test_legacy_adding_books(self): # {{{
'Test various adding books methods'
from calibre.ebooks.metadata.book.base import Metadata
legacy, old = self.init_legacy(self.cloned_library), self.init_old(self.cloned_library)
mi = Metadata('Added Book0', authors=('Added Author',))
with NamedTemporaryFile(suffix='.aff') as f:
f.write(b'xxx')
f.flush()
T = partial(ET, 'add_books', ([f.name], ['AFF'], [mi]), old=old, legacy=legacy)
T()(self)
book_id = T(kwargs={'return_ids':True})(self)[1][0]
self.assertEqual(legacy.new_api.formats(book_id), ('AFF',))
T(kwargs={'add_duplicates':False})(self)
mi.title = 'Added Book1'
mi.uuid = 'uuu'
T = partial(ET, 'import_book', (mi,[f.name]), old=old, legacy=legacy)
book_id = T()(self)
self.assertNotEqual(legacy.uuid(book_id, index_is_id=True), old.uuid(book_id, index_is_id=True))
book_id = T(kwargs={'preserve_uuid':True})(self)
self.assertEqual(legacy.uuid(book_id, index_is_id=True), old.uuid(book_id, index_is_id=True))
self.assertEqual(legacy.new_api.formats(book_id), ('AFF',))
with NamedTemporaryFile(suffix='.opf') as f:
f.write(b'zzzz')
f.flush()
T = partial(ET, 'import_book', (mi,[f.name]), old=old, legacy=legacy)
book_id = T()(self)
self.assertFalse(legacy.new_api.formats(book_id))
mi.title = 'Added Book2'
T = partial(ET, 'create_book_entry', (mi,), old=old, legacy=legacy)
T()
T({'add_duplicates':False})
T({'force_id':1000})
def to_metadata_for_single_entry(self, log, ozon_id, title, authors): # {{{
# parsing javascript data from the redirect page
mi = Metadata(title, authors)
mi.identifiers = {'ozon': ozon_id}
return mi
def add_catalog(cache, path, title, dbapi=None):
from calibre.ebooks.metadata.book.base import Metadata
from calibre.ebooks.metadata.meta import get_metadata
from calibre.utils.date import utcnow
fmt = os.path.splitext(path)[1][1:].lower()
new_book_added = False
with lopen(path, 'rb') as stream:
with cache.write_lock:
matches = cache._search('title:="%s" and tags:="%s"' % (title.replace('"', '\\"'), _('Catalog')), None)
db_id = None
if matches:
db_id = list(matches)[0]
try:
mi = get_metadata(stream, fmt)
mi.authors = ['calibre']
except:
mi = Metadata(title, ['calibre'])
mi.title, mi.authors = title, ['calibre']
mi.author_sort = 'calibre' # The MOBI/AZW3 format sets author sort to date
mi.tags = [_('Catalog')]
mi.pubdate = mi.timestamp = utcnow()
if fmt == 'mobi':
mi.cover, mi.cover_data = None, (None, None)
if db_id is None:
db_id = cache._create_book_entry(mi, apply_import_tags=False)
new_book_added = True
else:
cache._set_metadata(db_id, mi)
cache.add_format(db_id, fmt, stream, dbapi=dbapi) # Cant keep write lock since post-import hooks might run
return db_id, new_book_added
def get_metadata_from_detail(self, log, entry, title, authors, identifiers): # {{{
title = unicode(entry.xpath(u'normalize-space(.//h1[@itemprop="name"][1]/text())'))
# log.debug(u'Tile (from_detail): -----> %s' % title)
author = unicode(entry.xpath(u'normalize-space(.//a[contains(@href, "person")][1]/text())'))
# log.debug(u'Author (from_detail): -----> %s' % author)
norm_authors = map(_normalizeAuthorNameWithInitials, map(unicode.strip, unicode(author).split(u',')))
mi = Metadata(title, norm_authors)
ozon_id = entry.xpath(u'substring-before(substring-after(normalize-space(.//a[starts-with(@href, "/context/detail/id/")][1]/@href), "id/"), "/")')
if ozon_id:
# log.debug(u'ozon_id (from_detail): -----> %s' % ozon_id)
mi.identifiers = {'ozon':ozon_id}
mi.ozon_cover_url = None
cover = entry.xpath(u'normalize-space(.//img[1]/@src)')
if cover:
mi.ozon_cover_url = _translateToBigCoverUrl(cover)
# log.debug(u'mi.ozon_cover_url (from_detail): -----> %s' % mi.ozon_cover_url)
mi.rating = self.get_rating(entry)
# log.debug(u'mi.rating (from_detail): -----> %s' % mi.rating)
if not mi.rating:
log.debug('No rating (from_detail) found. ozon_id:%s'%ozon_id)
return mi
def test(scale=0.5):
from PyQt5.Qt import QLabel, QApplication, QPixmap, QMainWindow, QWidget, QScrollArea, QGridLayout
app = QApplication([])
mi = Metadata('xxx', ['Kovid Goyal', 'John Q. Doe', 'Author'])
mi.series = 'A series of styles'
m = QMainWindow()
sa = QScrollArea(m)
w = QWidget(m)
sa.setWidget(w)
l = QGridLayout(w)
w.setLayout(l), l.setSpacing(30)
labels = []
for r, color in enumerate(sorted(default_color_themes)):
for c, style in enumerate(sorted(all_styles())):
mi.series_index = c + 1
mi.title = 'An algorithmic cover [%s]' % color
prefs = override_prefs(cprefs, override_color_theme=color, override_style=style)
for x in ('cover_width', 'cover_height', 'title_font_size', 'subtitle_font_size', 'footer_font_size'):
prefs[x] = int(scale * prefs[x])
img = generate_cover(mi, prefs=prefs, as_qimage=True)
la = QLabel()
la.setPixmap(QPixmap.fromImage(img))
l.addWidget(la, r, c)
labels.append(la)
m.setCentralWidget(sa)
w.resize(w.sizeHint())
m.show()
app.exec_()
def test(scale=0.25):
from PyQt5.Qt import QLabel, QPixmap, QMainWindow, QWidget, QScrollArea, QGridLayout
from calibre.gui2 import Application
app = Application([])
mi = Metadata('Unknown', ['Kovid Goyal', 'John & Doe', 'Author'])
mi.series = 'A series & styles'
m = QMainWindow()
sa = QScrollArea(m)
w = QWidget(m)
sa.setWidget(w)
l = QGridLayout(w)
w.setLayout(l), l.setSpacing(30)
scale *= w.devicePixelRatioF()
labels = []
for r, color in enumerate(sorted(default_color_themes)):
for c, style in enumerate(sorted(all_styles())):
mi.series_index = c + 1
mi.title = 'An algorithmic cover [%s]' % color
prefs = override_prefs(cprefs, override_color_theme=color, override_style=style)
scale_cover(prefs, scale)
img = generate_cover(mi, prefs=prefs, as_qimage=True)
img.setDevicePixelRatio(w.devicePixelRatioF())
la = QLabel()
la.setPixmap(QPixmap.fromImage(img))
l.addWidget(la, r, c)
labels.append(la)
m.setCentralWidget(sa)
w.resize(w.sizeHint())
m.show()
app.exec_()
def add_catalog(cache, path, title):
from calibre.ebooks.metadata.book.base import Metadata
from calibre.ebooks.metadata.meta import get_metadata
from calibre.utils.date import utcnow
fmt = os.path.splitext(path)[1][1:].lower()
with lopen(path, 'rb') as stream, cache.write_lock:
matches = cache._search('title:="%s" and tags:="%s"' % (title.replace('"', '\\"'), _('Catalog')), None)
db_id = None
if matches:
db_id = list(matches)[0]
try:
mi = get_metadata(stream, fmt)
mi.authors = ['calibre']
except:
mi = Metadata(title, ['calibre'])
mi.title, mi.authors = title, ['calibre']
mi.tags = [_('Catalog')]
mi.pubdate = mi.timestamp = utcnow()
if fmt == 'mobi':
mi.cover, mi.cover_data = None, (None, None)
if db_id is None:
db_id = cache._create_book_entry(mi, apply_import_tags=False)
else:
cache._set_metadata(db_id, mi)
cache._add_format(db_id, fmt, stream)
return db_id
def get_series(title, authors, timeout=60):
mi = Metadata(title, authors)
if title and title[0] in _ignore_starts:
title = title[1:]
title = re.sub(r'^(A|The|An)\s+', '', title).strip()
if not title:
return mi
if isinstance(title, unicode):
title = title.encode('utf-8')
title = urllib.quote_plus(title)
author = authors[0].strip()
if not author:
return mi
if ',' in author:
author = author.split(',')[0]
else:
author = author.split()[-1]
url = URL.format(author, title)
br = browser()
try:
raw = br.open_novisit(url, timeout=timeout).read()
except URLError as e:
if isinstance(e.reason, socket.timeout):
raise Exception('KDL Server busy, try again later')
raise
if 'see the full results' not in raw:
return mi
raw = xml_to_unicode(raw)[0]
soup = BeautifulSoup(raw)
searcharea = soup.find('div', attrs={'class':'searcharea'})
if searcharea is None:
return mi
ss = searcharea.find('div', attrs={'class':'seriessearch'})
if ss is None:
return mi
a = ss.find('a', href=True)
if a is None:
return mi
href = a['href'].partition('?')[-1]
data = urlparse.parse_qs(href)
series = data.get('SeriesName', [])
if not series:
return mi
series = series[0]
series = re.sub(r' series$', '', series).strip()
if series:
mi.series = series
ns = ss.nextSibling
if ns.contents:
raw = unicode(ns.contents[0])
raw = raw.partition('.')[0].strip()
try:
mi.series_index = int(raw)
except:
pass
return mi
def test_legacy_adding_books(self): # {{{
'Test various adding books methods'
from calibre.ebooks.metadata.book.base import Metadata
legacy, old = self.init_legacy(self.cloned_library), self.init_old(self.cloned_library)
mi = Metadata('Added Book0', authors=('Added Author',))
with NamedTemporaryFile(suffix='.aff') as f:
f.write(b'xxx')
f.flush()
T = partial(ET, 'add_books', ([f.name], ['AFF'], [mi]), old=old, legacy=legacy)
T()(self)
book_id = T(kwargs={'return_ids':True})(self)[1][0]
self.assertEqual(legacy.new_api.formats(book_id), ('AFF',))
T(kwargs={'add_duplicates':False})(self)
mi.title = 'Added Book1'
mi.uuid = 'uuu'
T = partial(ET, 'import_book', (mi,[f.name]), old=old, legacy=legacy)
book_id = T()(self)
self.assertNotEqual(legacy.uuid(book_id, index_is_id=True), old.uuid(book_id, index_is_id=True))
book_id = T(kwargs={'preserve_uuid':True})(self)
self.assertEqual(legacy.uuid(book_id, index_is_id=True), old.uuid(book_id, index_is_id=True))
self.assertEqual(legacy.new_api.formats(book_id), ('AFF',))
T = partial(ET, 'add_format', old=old, legacy=legacy)
T((0, 'AFF', BytesIO(b'fffff')))(self)
T((0, 'AFF', BytesIO(b'fffff')))(self)
T((0, 'AFF', BytesIO(b'fffff')), {'replace':True})(self)
with NamedTemporaryFile(suffix='.opf') as f:
f.write(b'zzzz')
f.flush()
T = partial(ET, 'import_book', (mi,[f.name]), old=old, legacy=legacy)
book_id = T()(self)
self.assertFalse(legacy.new_api.formats(book_id))
mi.title = 'Added Book2'
T = partial(ET, 'create_book_entry', (mi,), old=old, legacy=legacy)
T()
T({'add_duplicates':False})
T({'force_id':1000})
with NamedTemporaryFile(suffix='.txt') as f:
f.write(b'tttttt')
f.seek(0)
bid = legacy.add_catalog(f.name, 'My Catalog')
self.assertEqual(old.add_catalog(f.name, 'My Catalog'), bid)
cache = legacy.new_api
self.assertEqual(cache.formats(bid), ('TXT',))
self.assertEqual(cache.field_for('title', bid), 'My Catalog')
self.assertEqual(cache.field_for('authors', bid), ('calibre',))
self.assertEqual(cache.field_for('tags', bid), (_('Catalog'),))
self.assertTrue(bid < legacy.add_catalog(f.name, 'Something else'))
self.assertEqual(legacy.add_catalog(f.name, 'My Catalog'), bid)
self.assertEqual(old.add_catalog(f.name, 'My Catalog'), bid)
bid = legacy.add_news(f.name, {'title':'Events', 'add_title_tag':True, 'custom_tags':('one', 'two')})
self.assertEqual(cache.formats(bid), ('TXT',))
self.assertEqual(cache.field_for('authors', bid), ('calibre',))
self.assertEqual(cache.field_for('tags', bid), (_('News'), 'Events', 'one', 'two'))
old.close()
def metadata_from_dict(src):
ans = Metadata('Unknown')
for key, value in src.iteritems():
if key == 'user_metadata':
ans.set_all_user_metadata(value)
else:
setattr(ans, key, value)
return ans
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 to_book_metadata(self):
mi = Metadata(_('Unknown'))
if self.db is None:
return mi
mi.set_all_user_metadata(self.db.field_metadata.custom_field_metadata())
for widget in self.basic_metadata_widgets:
widget.apply_to_metadata(mi)
for widget in getattr(self, 'custom_metadata_widgets', []):
widget.apply_to_metadata(mi)
return mi
def report_metadata_failure(self, group_id, details):
a = self.report.append
paths = self.file_groups[group_id]
a(""), a("-" * 70)
a(_("Failed to read metadata from the file(s):"))
[a("\t" + f) for f in paths]
a(_("With error:")), a(details)
mi = Metadata(_("Unknown"))
mi.read_metadata_failed = False
return mi
def report_metadata_failure(self, group_id, details):
a = self.report.append
paths = self.file_groups[group_id]
a(''), a('-' * 70)
a(_('Failed to read metadata from the file(s):'))
[a('\t' + f) for f in paths]
a(_('With error:')), a(details)
mi = Metadata(_('Unknown'))
mi.read_metadata_failed = False
return mi
def create_cover(title, authors, series=None, series_index=1, prefs=None, as_qimage=False):
' Create a cover from the specified title, author and series. Any user set'
' templates are ignored, to ensure that the specified metadata is used. '
mi = Metadata(title, authors)
if series:
mi.series, mi.series_index = series, series_index
d = cprefs.defaults
prefs = override_prefs(
prefs or cprefs, title_template=d['title_template'], subtitle_template=d['subtitle_template'], footer_template=d['footer_template'])
return generate_cover(mi, prefs=prefs, as_qimage=as_qimage)
def get_baike_metadata(self, title):
from baidubaike import Page
try: baike = Page(title)
except: return None
info = baike.get_info()
print "\n".join( "%s:\t%s" % v for v in info.items())
mi = Metadata(info['title'])
plat = info.get(u'首发网站', None)
if not plat:
plat = info.get(u'首发状态', "网络小说平台")
plat = plat.replace(u'首发', '')
mi.publisher = info.get(u'连载平台', plat)
mi.authors = [info[u'作者']]
mi.isbn = '0000000000001'
mi.tags = baike.get_tags()
mi.pubdate = datetime.datetime.now()
mi.timestamp = datetime.datetime.now()
mi.comments = baike.get_summary()
if u'完结' in info.get(u'连载状态', ""):
day = re.findall('\d*-\d*-\d*', info[u'连载状态'])
try: mi.pubdate = datetime.datetime.strptime(day[0], '%Y-%m-%d')
except: pass
return mi
def build_meta(log, issue_id):
"""Build metadata record based on comicvine issue_id"""
issue = pycomicvine.Issue(
issue_id,
field_list=[
"id",
"name",
"volume",
"issue_number",
"person_credits",
"description",
"store_date",
"cover_date",
],
)
if not issue or not issue.volume:
log.warn("Unable to load Issue(%d)" % issue_id)
return None
title = "%s #%s" % (issue.volume.name, issue.issue_number)
if issue.name:
title = title + ": %s" % (issue.name)
authors = [p.name for p in issue.person_credits]
meta = Metadata(title, authors)
meta.series = issue.volume.name
meta.series_index = str(issue.issue_number)
meta.set_identifier("comicvine", str(issue.id))
meta.comments = issue.description
meta.has_cover = False
if issue.volume.publisher:
meta.publisher = issue.volume.publisher.name
meta.pubdate = issue.store_date or issue.cover_date
return meta
def __init__(self):
Metadata.__init__(self, '')
self._new_book = False
self.size = 0
self.datetime = time.gmtime()
self.path = '' #the quitthyme book storage_id
self.thumbnail = None
self.quietthyme_id = -1
def test_rtf_metadata(self):
stream = BytesIO(br'{\rtf1\ansi\ansicpg1252}')
m = Metadata('Test ø̄title', ['Author One', 'Author БTwo'])
m.tags = 'tag1 見tag2'.split()
m.comments = '<p>some ⊹comments</p>'
m.publisher = 'publiSher'
set_metadata(stream, m)
stream.seek(0)
o = get_metadata(stream)
for attr in 'title authors publisher comments tags'.split():
self.assertEqual(getattr(m, attr), getattr(o, attr))
def identify(self,
log, result_queue, abort,
title=None, authors=None,
identifiers={}, timeout=30):
dlog("IDENTIFIERS")
dlog(identifiers)
dc = {}
for Getter in CGetter.getter_list:
dlog("USING GETTER: %s" % Getter)
dlog("my prefs:", prefs)
isvalid = True
for pref_key, is_valid_check in Getter.ls_required_pref_key:
dlog("validating: %s --> = %s" % (pref_key, prefs[pref_key]))
if not is_valid_check(prefs[pref_key]):
dlog("INVALID: %s" % prefs[pref_key])
isvalid = False
break
entry_id = identifiers.get(Getter.entry_identifier, None)
dlog("---------------", entry_id)
if entry_id and Getter.does_understand(entry_id):
dc = Getter.resolve(entry_id)
dlog("RESULT")
dlog(dc)
if dc['authors']: break
dlog(dc)
if not dc:
dlog("not dc: returning")
return None
dlog("OK")
mi = Metadata(dc["title"], dc["authors"])
for attr in ("pubdate",
"publisher",
"issue",
"abstract",
"keywords",
"volume",
"pages",
):
if hasattr(mi, attr) and attr in dc:
setattr(mi, attr, dc[attr])
mi.print_all_attributes()
self.clean_downloaded_metadata(mi)
result_queue.put(mi)
dlog("=" * 20)
dlog(dir(self))
dlog("=" * 20)
return None
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 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)
return ans
def generate_test_db(
library_path, # {{{
num_of_records=20000,
num_of_authors=6000,
num_of_tags=10000,
tag_length=7,
author_length=7,
title_length=10,
max_authors=10,
max_tags=10,
):
import random, string, os, sys, time
from calibre.constants import preferred_encoding
if not os.path.exists(library_path):
os.makedirs(library_path)
letters = string.letters.decode(preferred_encoding)
def randstr(length):
return "".join(random.choice(letters) for i in xrange(length))
all_tags = [randstr(tag_length) for j in xrange(num_of_tags)]
print "Generated", num_of_tags, "tags"
all_authors = [randstr(author_length) for j in xrange(num_of_authors)]
print "Generated", num_of_authors, "authors"
all_titles = [randstr(title_length) for j in xrange(num_of_records)]
print "Generated", num_of_records, "titles"
testdb = db(library_path)
print "Creating", num_of_records, "records..."
start = time.time()
for i, title in enumerate(all_titles):
print i + 1,
sys.stdout.flush()
authors = random.randint(1, max_authors)
authors = [random.choice(all_authors) for i in xrange(authors)]
tags = random.randint(0, max_tags)
tags = [random.choice(all_tags) for i in xrange(tags)]
from calibre.ebooks.metadata.book.base import Metadata
mi = Metadata(title, authors)
mi.tags = tags
testdb.import_book(mi, [])
t = time.time() - start
print "\nGenerated", num_of_records, "records in:", t, "seconds"
print "Time per record:", t / float(num_of_records)
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 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 to_metadata(browser, log, entry_, timeout): # {{{
from lxml import etree
XPath = partial(etree.XPath, namespaces=NAMESPACES)
# total_results = XPath('//openSearch:totalResults')
# start_index = XPath('//openSearch:startIndex')
# items_per_page = XPath('//openSearch:itemsPerPage')
entry = XPath('//atom:entry')
entry_id = XPath('descendant::atom:id')
creator = XPath('descendant::dc:creator')
identifier = XPath('descendant::dc:identifier')
title = XPath('descendant::dc:title')
date = XPath('descendant::dc:date')
publisher = XPath('descendant::dc:publisher')
subject = XPath('descendant::dc:subject')
description = XPath('descendant::dc:description')
language = XPath('descendant::dc:language')
rating = XPath('descendant::gd:rating[@average]')
def get_text(extra, x):
try:
ans = x(extra)
if ans:
ans = ans[0].text
if ans and ans.strip():
return ans.strip()
except:
log.exception('Programming error:')
return None
id_url = entry_id(entry_)[0].text
google_id = id_url.split('/')[-1]
title_ = ': '.join([x.text for x in title(entry_)]).strip()
authors = [x.text.strip() for x in creator(entry_) if x.text]
if not authors:
authors = [_('Unknown')]
if not id_url or not title:
# Silently discard this entry
return None
mi = Metadata(title_, authors)
mi.identifiers = {'google': google_id}
try:
raw = get_details(browser, id_url, timeout)
feed = etree.fromstring(
xml_to_unicode(clean_ascii_chars(raw),
strip_encoding_pats=True)[0])
extra = entry(feed)[0]
except:
log.exception('Failed to get additional details for', mi.title)
return mi
mi.comments = get_text(extra, description)
lang = canonicalize_lang(get_text(extra, language))
if lang:
mi.language = lang
mi.publisher = get_text(extra, publisher)
# ISBN
isbns = []
for x in identifier(extra):
t = str(x.text).strip()
if t[:5].upper() in ('ISBN:', 'LCCN:', 'OCLC:'):
if t[:5].upper() == 'ISBN:':
t = check_isbn(t[5:])
if t:
isbns.append(t)
if isbns:
mi.isbn = sorted(isbns, key=len)[-1]
mi.all_isbns = isbns
# Tags
try:
btags = [x.text for x in subject(extra) if x.text]
tags = []
for t in btags:
atags = [y.strip() for y in t.split('/')]
for tag in atags:
if tag not in tags:
tags.append(tag)
except:
log.exception('Failed to parse tags:')
tags = []
if tags:
mi.tags = [x.replace(',', ';') for x in tags]
# pubdate
pubdate = get_text(extra, date)
if pubdate:
from calibre.utils.date import parse_date, utcnow
try:
default = utcnow().replace(day=15)
mi.pubdate = parse_date(pubdate, assume_utc=True, default=default)
except:
log.error('Failed to parse pubdate %r' % pubdate)
# Ratings
for x in rating(extra):
try:
mi.rating = float(x.get('average'))
if mi.rating > 5:
mi.rating /= 2
except:
log.exception('Failed to parse rating')
# Cover
mi.has_google_cover = None
for x in extra.xpath(
'//*[@href and @rel="http://schemas.google.com/books/2008/thumbnail"]'
):
mi.has_google_cover = x.get('href')
break
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(f'//{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
class Book(Book_):
def __init__(self, prefix, lpath, title=None, authors=None, mime=None, date=None, ContentType=None,
thumbnail_name=None, size=None, other=None):
from calibre.utils.date import parse_date
# debug_print('Book::__init__ - title=', title)
show_debug = title is not None and title.lower().find("xxxxx") >= 0
if other is not None:
other.title = title
other.published_date = date
if show_debug:
debug_print("Book::__init__ - title=", title, 'authors=', authors)
debug_print("Book::__init__ - other=", other)
super(Book, self).__init__(prefix, lpath, size, other)
if title is not None and len(title) > 0:
self.title = title
if authors is not None and len(authors) > 0:
self.authors_from_string(authors)
if self.author_sort is None or self.author_sort == "Unknown":
self.author_sort = author_to_author_sort(authors)
self.mime = mime
self.size = size # will be set later if None
if ContentType == '6' and date is not None:
try:
self.datetime = time.strptime(date, "%Y-%m-%dT%H:%M:%S.%f")
except:
try:
self.datetime = time.strptime(date.split('+')[0], "%Y-%m-%dT%H:%M:%S")
except:
try:
self.datetime = time.strptime(date.split('+')[0], "%Y-%m-%d")
except:
try:
self.datetime = parse_date(date,
assume_utc=True).timetuple()
except:
try:
self.datetime = time.gmtime(os.path.getctime(self.path))
except:
self.datetime = time.gmtime()
self.kobo_metadata = Metadata(title, self.authors)
self.contentID = None
self.current_shelves = []
self.kobo_collections = []
self.can_put_on_shelves = True
self.kobo_series = None
self.kobo_series_number = None # Kobo stores the series number as string. And it can have a leading "#".
self.kobo_subtitle = None
if thumbnail_name is not None:
self.thumbnail = ImageWrapper(thumbnail_name)
if show_debug:
debug_print("Book::__init__ end - self=", self)
debug_print("Book::__init__ end - title=", title, 'authors=', authors)
@property
def is_sideloaded(self):
# If we don't have a content Id, we don't know what type it is.
return self.contentID and self.contentID.startswith("file")
@property
def is_purchased_kepub(self):
return self.contentID and not self.contentID.startswith("file")
def __unicode__(self):
'''
A string representation of this object, suitable for printing to
console
'''
ans = [u"Kobo metadata:"]
def fmt(x, y):
ans.append(u'%-20s: %s'%(unicode_type(x), unicode_type(y)))
if self.contentID:
fmt('Content ID', self.contentID)
if self.kobo_series:
fmt('Kobo Series', self.kobo_series + ' #%s'%self.kobo_series_number)
if self.kobo_subtitle:
fmt('Subtitle', self.kobo_subtitle)
if self.mime:
fmt('MimeType', self.mime)
ans = u'\n'.join(ans) + u"\n" + self.kobo_metadata.__unicode__()
return super(Book,self).__unicode__() + u"\n" + ans
def parse_details(self, raw, root):
asin = parse_asin(root, self.log, self.url)
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)
try:
title = self.parse_title(root)
except:
self.log.exception('Error parsing title for url: %r'%self.url)
title = None
try:
authors = self.parse_authors(root)
except:
self.log.exception('Error parsing authors for url: %r'%self.url)
authors = []
if not title or not authors or not 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
mi = Metadata(title, authors)
idtype = 'amazon' if self.domain == 'com' else 'amazon_'+self.domain
mi.set_identifier(idtype, asin)
self.amazon_id = asin
try:
mi.rating = self.parse_rating(root)
except:
self.log.exception('Error parsing ratings for url: %r'%self.url)
try:
mi.comments = self.parse_comments(root)
except:
self.log.exception('Error parsing comments for url: %r'%self.url)
try:
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)
try:
mi.tags = self.parse_tags(root)
except:
self.log.exception('Error parsing tags for url: %r'%self.url)
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)
non_hero = tuple(self.selector('div#bookDetails_container_div div#nonHeroSection'))
if non_hero:
# New style markup
try:
self.parse_new_details(root, mi, non_hero[0])
except:
self.log.exception('Failed to parse new-style book details section')
else:
pd = root.xpath(self.pd_xpath)
if pd:
pd = pd[0]
try:
isbn = self.parse_isbn(pd)
if isbn:
self.isbn = mi.isbn = isbn
except:
self.log.exception('Error parsing ISBN for url: %r'%self.url)
try:
mi.publisher = self.parse_publisher(pd)
except:
self.log.exception('Error parsing publisher for url: %r'%self.url)
try:
mi.pubdate = self.parse_pubdate(pd)
except:
self.log.exception('Error parsing publish date for url: %r'%self.url)
try:
lang = self.parse_language(pd)
if lang:
mi.language = lang
except:
self.log.exception('Error parsing language for url: %r'%self.url)
else:
self.log.warning('Failed to find product description for url: %r'%self.url)
mi.source_relevance = self.relevance
if self.amazon_id:
if self.isbn:
self.plugin.cache_isbn_to_identifier(self.isbn, self.amazon_id)
if self.cover_url:
self.plugin.cache_identifier_to_cover_url(self.amazon_id,
self.cover_url)
self.plugin.clean_downloaded_metadata(mi)
self.result_queue.put(mi)
def identify(
log,
abort, # {{{
title=None,
authors=None,
identifiers={},
timeout=30):
if title == _('Unknown'):
title = None
if authors == [_('Unknown')]:
authors = None
start_time = time.time()
plugins = [p for p in metadata_plugins(['identify']) if p.is_configured()]
kwargs = {
'title': title,
'authors': authors,
'identifiers': identifiers,
'timeout': timeout,
}
log('Running identify query with parameters:')
log(kwargs)
log('Using plugins:', ', '.join([p.name for p in plugins]))
log('The log from individual plugins is below')
workers = [Worker(p, kwargs, abort) for p in plugins]
for w in workers:
w.start()
first_result_at = None
results = {}
for p in plugins:
results[p] = []
logs = dict([(w.plugin, w.buf) for w in workers])
def get_results():
found = False
for w in workers:
try:
result = w.rq.get_nowait()
except Empty:
pass
else:
results[w.plugin].append(result)
found = True
return found
wait_time = msprefs['wait_after_first_identify_result']
while True:
time.sleep(0.2)
if get_results() and first_result_at is None:
first_result_at = time.time()
if not is_worker_alive(workers):
break
if (first_result_at is not None
and time.time() - first_result_at > wait_time):
log.warn('Not waiting any longer for more results. Still running'
' sources:')
for worker in workers:
if worker.is_alive():
log.debug('\t' + worker.name)
abort.set()
break
while not abort.is_set() and get_results():
pass
sort_kwargs = dict(kwargs)
for k in list(sort_kwargs.iterkeys()):
if k not in ('title', 'authors', 'identifiers'):
sort_kwargs.pop(k)
longest, lp = -1, ''
for plugin, presults in results.iteritems():
presults.sort(key=plugin.identify_results_keygen(**sort_kwargs))
# Throw away lower priority results from the same source that have exactly the same
# title and authors as a higher priority result
filter_results = set()
filtered_results = []
for r in presults:
key = (r.title, tuple(r.authors))
if key not in filter_results:
filtered_results.append(r)
filter_results.add(key)
results[plugin] = presults = filtered_results
plog = logs[plugin].getvalue().strip()
log('\n' + '*' * 30, plugin.name, '*' * 30)
log('Request extra headers:', plugin.browser.addheaders)
log('Found %d results' % len(presults))
time_spent = getattr(plugin, 'dl_time_spent', None)
if time_spent is None:
log('Downloading was aborted')
longest, lp = -1, plugin.name
else:
log('Downloading from', plugin.name, 'took', time_spent)
if time_spent > longest:
longest, lp = time_spent, plugin.name
for r in presults:
log('\n\n---')
log(unicode(r))
if plog:
log(plog)
log('\n' + '*' * 80)
dummy = Metadata(_('Unknown'))
for i, result in enumerate(presults):
for f in plugin.prefs['ignore_fields']:
if ':' not in f:
setattr(result, f, getattr(dummy, f))
if f == 'series':
result.series_index = dummy.series_index
result.relevance_in_source = i
result.has_cached_cover_url = (plugin.cached_cover_url_is_reliable
and plugin.get_cached_cover_url(
result.identifiers) is not None)
result.identify_plugin = plugin
if msprefs['txt_comments']:
if plugin.has_html_comments and result.comments:
result.comments = html2text(result.comments)
log('The identify phase took %.2f seconds' % (time.time() - start_time))
log('The longest time (%f) was taken by:' % longest, lp)
log('Merging results from different sources and finding earliest ',
'publication dates from the worldcat.org service')
start_time = time.time()
results = merge_identify_results(results, log)
log('We have %d merged results, merging took: %.2f seconds' %
(len(results), time.time() - start_time))
max_tags = msprefs['max_tags']
for r in results:
r.tags = r.tags[:max_tags]
if getattr(r.pubdate, 'year', 2000) <= UNDEFINED_DATE.year:
r.pubdate = None
if msprefs['swap_author_names']:
for r in results:
def swap_to_ln_fn(a):
if ',' in a:
return a
parts = a.split(None)
if len(parts) <= 1:
return a
surname = parts[-1]
return '%s, %s' % (surname, ' '.join(parts[:-1]))
r.authors = [swap_to_ln_fn(a) for a in r.authors]
return results
def _GoodreadsBook_to_Metadata(self, book):
# type: (_GoodreadsBook) -> Metadata
"""
:param book: _GoodreadsBook: book
:return: Metadata: Metadata
"""
mi = Metadata(book.title, book.authors)
mi.source_relevance = 0
mi.set_identifier('goodreads', book.id)
if self.prefs['NEVER_REPLACE_ISBN'] and mi.get_identifiers().get(
'isbn'):
mi.set_identifier('isbn', '')
if book.asin and not self.prefs['NEVER_REPLACE_AMAZONID']:
mi.set_identifier('amazon', book.asin)
if book.isbn and not self.prefs['NEVER_REPLACE_ISBN']:
try:
if len(book.isbn) == 10:
mi.isbn = check_isbn13(_ISBNConvert.convert(book.isbn))
else:
mi.isbn = check_isbn13(book.isbn)
except:
self.log.error("ISBN CONVERSION ERROR:", book.isbn)
self.log.exception()
if book.image_url:
self.log.info('cache_identifier_to_cover_url:', book.asin, ':',
book.image_url)
self.cache_identifier_to_cover_url(book.id, book.image_url)
if book.publisher:
self.log.info('book.publisher is:', book.publisher)
mi.publisher = book.publisher
if book.pubdate:
self.log.info('book.pubdate is:',
book.pubdate.strftime('%Y-%m-%d'))
mi.pubdate = book.pubdate
if book.comments:
self.log.info('book.editorial_review is:', book.comments)
mi.comments = book.comments
tags = self.prefs['ADD_THESE_TAGS'].split(',')
tags.extend(book.tags)
# tag_mappings = JSONConfig('plugins/GenreMappings')['genreMappings']
# mi.tags = list(set(sorted(filter(lambda x: tag_mappings.get(x, x), tags))))
if book.series:
mi.series = book.series
self.log.info(u'series:', book.series)
if book.series_index:
mi.series_index = book.series_index
self.log.info(u'series_index:',
"{0:.2f}".format(book.series_index))
else:
mi.series_index = 0
if book.average_rating:
mi.rating = book.average_rating
self.clean_downloaded_metadata(mi)
return mi
def validate(self, x):
from calibre.ebooks.metadata.book.base import Metadata
self.book = Metadata('')
return self.vformat(x, [], {})
def test_legacy_direct(self): # {{{
'Test read-only methods that are directly equivalent in the old and new interface'
from calibre.ebooks.metadata.book.base import Metadata
from datetime import timedelta
ndb = self.init_legacy(self.cloned_library)
db = self.init_old()
newstag = ndb.new_api.get_item_id('tags', 'news')
self.assertEqual(dict(db.prefs), dict(ndb.prefs))
for meth, args in {
'find_identical_books': [(Metadata('title one',
['author one']), ),
(Metadata('unknown'), ),
(Metadata('xxxx'), )],
'get_books_for_category': [('tags', newstag),
('#formats', 'FMT1')],
'get_next_series_num_for': [('A Series One', )],
'get_id_from_uuid': [('ddddd', ), (db.uuid(1, True), )],
'cover': [(0, ), (1, ), (2, )],
'get_author_id': [('author one', ), ('unknown', ),
('xxxxx', )],
'series_id': [(0, ), (1, ), (2, )],
'publisher_id': [(0, ), (1, ), (2, )],
'@tags_older_than': [
('News', None),
('Tag One', None),
('xxxx', None),
('Tag One', None, 'News'),
('News', None, 'xxxx'),
('News', None, None, ['xxxxxxx']),
('News', None, 'Tag One', ['Author Two', 'Author One']),
('News', timedelta(0), None, None),
('News', timedelta(100000)),
],
'format': [(1, 'FMT1', True), (2, 'FMT1', True),
(0, 'xxxxxx')],
'has_format': [(1, 'FMT1', True), (2, 'FMT1', True),
(0, 'xxxxxx')],
'sizeof_format': [(1, 'FMT1', True), (2, 'FMT1', True),
(0, 'xxxxxx')],
'@format_files': [(0, ), (1, ), (2, )],
'formats': [(0, ), (1, ), (2, )],
'max_size': [(0, ), (1, ), (2, )],
'format_hash': [(1, 'FMT1'), (1, 'FMT2'), (2, 'FMT1')],
'author_sort_from_authors':
[(['Author One', 'Author Two', 'Unknown'], )],
'has_book': [(Metadata('title one'), ),
(Metadata('xxxx1111'), )],
'has_id': [(1, ), (2, ), (3, ), (9999, )],
'id': [
(1, ),
(2, ),
(0, ),
],
'index': [
(1, ),
(2, ),
(3, ),
],
'row': [
(1, ),
(2, ),
(3, ),
],
'is_empty': [()],
'count': [()],
'all_author_names': [()],
'all_tag_names': [()],
'all_series_names': [()],
'all_publisher_names': [()],
'!all_authors': [()],
'!all_tags2': [()],
'@all_tags': [()],
'@get_all_identifier_types': [()],
'!all_publishers': [()],
'!all_titles': [()],
'!all_series': [()],
'standard_field_keys': [()],
'all_field_keys': [()],
'searchable_fields': [()],
'search_term_to_field_key': [('author', ), ('tag', )],
'metadata_for_field': [('title', ), ('tags', )],
'sortable_field_keys': [()],
'custom_field_keys': [(True, ), (False, )],
'!get_usage_count_by_id': [('authors', ), ('tags', ),
('series', ), ('publisher', ),
('#tags', ), ('languages', )],
'get_field': [(1, 'title'), (2, 'tags'), (0, 'rating'),
(1, 'authors'), (2, 'series'), (1, '#tags')],
'all_formats': [()],
'get_authors_with_ids': [()],
'!get_tags_with_ids': [()],
'!get_series_with_ids': [()],
'!get_publishers_with_ids': [()],
'!get_ratings_with_ids': [()],
'!get_languages_with_ids': [()],
'tag_name': [(3, )],
'author_name': [(3, )],
'series_name': [(3, )],
'authors_sort_strings': [(0, ), (1, ), (2, )],
'author_sort_from_book': [(0, ), (1, ), (2, )],
'authors_with_sort_strings': [(0, ), (1, ), (2, )],
'book_on_device_string': [(1, ), (2, ), (3, )],
'books_in_series_of': [(0, ), (1, ), (2, )],
'books_with_same_title': [(Metadata(db.title(0)), ),
(Metadata(db.title(1)), ),
(Metadata('1234'), )],
}.iteritems():
fmt = lambda x: x
if meth[0] in {'!', '@'}:
fmt = {'!': dict, '@': frozenset}[meth[0]]
meth = meth[1:]
elif meth == 'get_authors_with_ids':
fmt = lambda val: {x[0]: tuple(x[1:]) for x in val}
for a in args:
self.assertEqual(
fmt(getattr(db, meth)(*a)), fmt(getattr(ndb, meth)(*a)),
'The method: %s() returned different results for argument %s'
% (meth, a))
def f(
x, y
): # get_top_level_move_items is broken in the old db on case-insensitive file systems
x.discard('metadata_db_prefs_backup.json')
return x, y
self.assertEqual(f(*db.get_top_level_move_items()),
f(*ndb.get_top_level_move_items()))
d1, d2 = BytesIO(), BytesIO()
db.copy_cover_to(1, d1, True)
ndb.copy_cover_to(1, d2, True)
self.assertTrue(d1.getvalue() == d2.getvalue())
d1, d2 = BytesIO(), BytesIO()
db.copy_format_to(1, 'FMT1', d1, True)
ndb.copy_format_to(1, 'FMT1', d2, True)
self.assertTrue(d1.getvalue() == d2.getvalue())
old = db.get_data_as_dict(prefix='test-prefix')
new = ndb.get_data_as_dict(prefix='test-prefix')
for o, n in zip(old, new):
o = {
type('')(k) if isinstance(k, bytes) else k:
set(v) if isinstance(v, list) else v
for k, v in o.iteritems()
}
n = {
k: set(v) if isinstance(v, list) else v
for k, v in n.iteritems()
}
self.assertEqual(o, n)
ndb.search('title:Unknown')
db.search('title:Unknown')
self.assertEqual(db.row(3), ndb.row(3))
self.assertRaises(ValueError, ndb.row, 2)
self.assertRaises(ValueError, db.row, 2)
db.close()
def reset_info(self):
self.show_data(Metadata(_('Unknown')))
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 = from_base64_bytes(m[COVER_KEY])
fmt, w, h = identify(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 mi(self):
mi = Metadata(unicode(self.title.text()).strip() or _('Unknown'))
mi.authors = string_to_authors(unicode(
self.authors.text()).strip()) or [_('Unknown')]
mi.languages = self.languages.lang_codes or [get_lang()]
return mi
def to_metadata(log,gmetadata,ExHentai_Status): # {{{
title = gmetadata['title']
title_jpn = gmetadata['title_jpn']
tags = gmetadata['tags']
rating = gmetadata['rating']
category = gmetadata['category']
gid = gmetadata['gid']
token = gmetadata['token']
thumb = gmetadata['thumb']
# title
if title_jpn:
raw_title = title_jpn
else:
raw_title = title
pat1 = re.compile(r'(?P<comments>.*?\[(?P<author>(?:(?!汉化|漢化)[^\[\]])*)\](?:\s*(?:\[[^\(\)]+\]|\([^\[\]\(\)]+\))\s*)*(?P<title>[^\[\]\(\)]+).*)')
if re.findall(pat1,raw_title):
m = re.search(pat1, raw_title)
title_ = m.group('title').strip()
author = m.group('author').strip()
else:
title_ = raw_title.strip()
author = 'Unknown'
log.exception('Title match failed. Title is %s' % raw_title)
authors = [(author)]
mi = Metadata(title_, authors)
mi.identifiers = {'ehentai':'%s_%s_%d' % (str(gid),str(token),int(ExHentai_Status))}
# publisher
pat2 = re.compile(r'^\(([^\[\]\(\)]*)\)')
if re.findall(pat2, raw_title):
publisher = re.search(pat2, raw_title).group(1).strip()
mi.publisher = publisher
else:
mi.publisher = 'Unknown'
log.exception('Not Found publisher.')
# Tags
tags_ = []
for tag in tags:
if re.match('language',tag):
tag_ = re.sub('language:','',tag)
if tag_ != 'translated':
mi.language = tag_
else:
tags_.append(tag_)
# elif re.match('parody|group|character|artist', tag):
# log('drop tag %s' % tag)
# continue
elif not ':' in tag:
log('drop tag %s' % tag)
continue
else:
tags_.append(tag)
tags_.append(category)
mi.tags = tags_
# rating
mi.rating = float(rating)
# cover
mi.has_ehentai_cover = None
if thumb:
mi.has_ehentai_cover = thumb
return mi
def to_metadata(browser, log, entry_, timeout): # {{{
from lxml import etree
from calibre.ebooks.chardet import xml_to_unicode
from calibre.utils.date import parse_date, utcnow
from calibre.utils.cleantext import clean_ascii_chars
XPath = partial(etree.XPath, namespaces=NAMESPACES)
entry = XPath('//atom:entry')
entry_id = XPath('descendant::atom:id')
title = XPath('descendant::atom:title')
description = XPath('descendant::atom:summary')
publisher = XPath("descendant::db:attribute[@name='publisher']")
isbn = XPath("descendant::db:attribute[@name='isbn13']")
date = XPath("descendant::db:attribute[@name='pubdate']")
creator = XPath("descendant::db:attribute[@name='author']")
booktag = XPath("descendant::db:tag/attribute::name")
rating = XPath("descendant::gd:rating/attribute::average")
cover_url = XPath("descendant::atom:link[@rel='image']/attribute::href")
def get_text(extra, x):
try:
ans = x(extra)
if ans:
ans = ans[0].text
if ans and ans.strip():
return ans.strip()
except:
log.exception('Programming error:')
return None
id_url = entry_id(entry_)[0].text
douban_id = id_url.split('/')[-1]
title_ = ': '.join([x.text for x in title(entry_)]).strip()
authors = [x.text.strip() for x in creator(entry_) if x.text]
if not authors:
authors = [_('Unknown')]
if not id_url or not title:
# Silently discard this entry
return None
mi = Metadata(title_, authors)
mi.identifiers = {'douban': douban_id}
try:
raw = get_details(browser, id_url, timeout)
feed = etree.fromstring(
xml_to_unicode(clean_ascii_chars(raw),
strip_encoding_pats=True)[0])
extra = entry(feed)[0]
except:
log.exception('Failed to get additional details for', mi.title)
return mi
mi.comments = get_text(extra, description)
mi.publisher = get_text(extra, publisher)
# ISBN
isbns = []
for x in [t.text for t in isbn(extra)]:
if check_isbn(x):
isbns.append(x)
if isbns:
mi.isbn = sorted(isbns, key=len)[-1]
mi.all_isbns = isbns
# Tags
try:
btags = [x for x in booktag(extra) if x]
tags = []
for t in btags:
atags = [y.strip() for y in t.split('/')]
for tag in atags:
if tag not in tags:
tags.append(tag)
except:
log.exception('Failed to parse tags:')
tags = []
if tags:
mi.tags = [x.replace(',', ';') for x in tags]
# pubdate
pubdate = get_text(extra, date)
if pubdate:
try:
default = utcnow().replace(day=15)
mi.pubdate = parse_date(pubdate, assume_utc=True, default=default)
except:
log.error('Failed to parse pubdate %r' % pubdate)
# Ratings
if rating(extra):
try:
mi.rating = float(rating(extra)[0]) / 2.0
except:
log.exception('Failed to parse rating')
mi.rating = 0
# Cover
mi.has_douban_cover = None
u = cover_url(extra)
if u:
u = u[0].replace('/spic/', '/lpic/')
# If URL contains "book-default", the book doesn't have a cover
if u.find('book-default') == -1:
mi.has_douban_cover = u
return mi
def _metadata(self, baike):
from calibre.ebooks.metadata.book.base import Metadata
from cStringIO import StringIO
info = baike.get_info()
print "\n".join("%s:\t%s" % v for v in info.items())
mi = Metadata(info['title'])
plat = "網絡小說平台"
plat = info.get(u'首發狀態', plat)
plat = info.get(u'首發網站', plat)
plat = plat.replace(u'首發', '')
mi.publisher = info.get(u'連載平台', plat)
mi.authors = [info.get(u'作者', u'佚名')]
mi.author_sort = mi.authors[0]
mi.isbn = BAIKE_ISBN
mi.tags = baike.get_tags()
mi.pubdate = datetime.datetime.now()
mi.timestamp = datetime.datetime.now()
mi.cover_url = baike.get_image()
mi.comments = re.sub(r'\[\d+\]$', "", baike.get_summary())
mi.website = baike.http.url
mi.source = u'百度百科'
if self.copy_image:
img = StringIO(urlopen(mi.cover_url).read())
img_fmt = mi.cover_url.split(".")[-1]
mi.cover_data = (img_fmt, img)
if u'完結' in info.get(u'連載狀態', ""):
day = re.findall('\d*-\d*-\d*', info[u'連載狀態'])
try:
mi.pubdate = datetime.datetime.strptime(day[0], '%Y-%m-%d')
except:
pass
return mi
if fmt == 'azw3':
with TemporaryDirectory('create-azw3') as tdir, CurrentDir(tdir):
for name, data in ((opf_name, opf), (html_name, HTML), (toc_name,
ncx)):
with open(name, 'wb') as f:
f.write(data)
c = Container(os.path.dirname(os.path.abspath(opf_name)), opf_name,
DevNull())
opf_to_azw3(opf_name, path, c)
else:
with ZipFile(path, 'w', compression=ZIP_STORED) as zf:
zf.writestr('mimetype',
b'application/epub+zip',
compression=ZIP_STORED)
zf.writestr('META-INF/', b'', 0o755)
zf.writestr('META-INF/container.xml', CONTAINER)
zf.writestr(opf_name, opf)
zf.writestr(html_name, HTML)
zf.writestr(toc_name, ncx)
if __name__ == '__main__':
from calibre.ebooks.metadata.book.base import Metadata
mi = Metadata('Test book', authors=('Kovid Goyal', ))
path = sys.argv[-1]
ext = path.rpartition('.')[-1].lower()
if ext not in valid_empty_formats:
print(('Unsupported format:', ext))
raise SystemExit(1)
create_book(mi, path, fmt=ext)
def build_meta(log, issue_id):
'''Build metadata record based on comicvine issue_id'''
issue = pycomicvine.Issue(issue_id, field_list=[
'id', 'name', 'volume', 'issue_number', 'person_credits', 'description',
'store_date', 'cover_date'])
if not issue or not issue.volume:
log.warn('Unable to load Issue(%d)' % issue_id)
return None
title = '%s #%s' % (issue.volume.name, issue.issue_number)
if issue.name:
title = title + ': %s' % (issue.name)
authors = [p.name for p in issue.person_credits]
meta = Metadata(title, authors)
meta.series = issue.volume.name
meta.series_index = str(issue.issue_number)
meta.set_identifier('comicvine', str(issue.id))
meta.set_identifier('comicvine-volume', str(issue.volume.id))
meta.comments = issue.description
meta.has_cover = False
if issue.volume.publisher:
meta.publisher = issue.volume.publisher.name
meta.pubdate = issue.store_date or issue.cover_date
return meta
def parse(self, raw):
from calibre.ebooks.metadata.book.base import Metadata
from calibre.utils.date import parse_only_date, UNDEFINED_DATE
from css_selectors import Select
root = parse_html(raw)
selector = Select(root)
sku = next(selector('div.sku.attGroup'))
info = sku.getparent()
top = info.getparent().getparent()
banner = top.find('div')
spans = banner.findall('span')
title = ''
for i, span in enumerate(spans):
if i == 0 or '12pt' in span.get('style', ''):
title += astext(span)
else:
break
authors = [re.sub(r'\(.*\)', '', x).strip() for x in astext(spans[-1]).split(',')]
mi = Metadata(title.strip(), authors)
# Identifiers
isbns = [check_isbn(x.strip()) for x in astext(sku).split(',')]
for isbn in isbns:
if isbn:
self.plugin.cache_isbn_to_identifier(isbn, self.sku)
isbns = sorted(isbns, key=lambda x:len(x) if x else 0, reverse=True)
if isbns and isbns[0]:
mi.isbn = isbns[0]
mi.set_identifier('edelweiss', self.sku)
# Tags
bisac = tuple(selector('div.bisac.attGroup'))
if bisac:
bisac = astext(bisac[0])
mi.tags = [x.strip() for x in bisac.split(',')]
mi.tags = [t[1:].strip() if t.startswith('&') else t for t in mi.tags]
# Publisher
pub = tuple(selector('div.supplier.attGroup'))
if pub:
pub = astext(pub[0])
mi.publisher = pub
# Pubdate
pub = tuple(selector('div.shipDate.attGroupItem'))
if pub:
pub = astext(pub[0])
parts = pub.partition(':')[0::2]
pub = parts[1] or parts[0]
try:
if ', Ship Date:' in pub:
pub = pub.partition(', Ship Date:')[0]
q = parse_only_date(pub, assume_utc=True)
if q.year != UNDEFINED_DATE:
mi.pubdate = q
except:
self.log.exception('Error parsing published date: %r'%pub)
# Comments
comm = ''
general = tuple(selector('div#pd-general-overview-content'))
if general:
q = self.render_comments(general[0])
if q != '<p>No title summary available. </p>':
comm += q
general = tuple(selector('div#pd-general-contributor-content'))
if general:
comm += self.render_comments(general[0])
general = tuple(selector('div#pd-general-quotes-content'))
if general:
comm += self.render_comments(general[0])
if comm:
mi.comments = comm
# Cover
img = tuple(selector('img.title-image[src]'))
if img:
href = img[0].get('src').replace('jacket_covers/medium/',
'jacket_covers/flyout/')
self.plugin.cache_identifier_to_cover_url(self.sku, href)
mi.has_cover = self.plugin.cached_identifier_to_cover_url(self.sku) is not None
return mi
def get_metadata_(src, encoding=None):
# Meta data definitions as in
# http://www.mobileread.com/forums/showpost.php?p=712544&postcount=9
if isbytestring(src):
if not encoding:
src = xml_to_unicode(src)[0]
else:
src = src.decode(encoding, 'replace')
src = src[:150000] # Searching shouldn't take too long
comment_tags = parse_comment_tags(src)
meta_tags = parse_meta_tags(src)
def get(field):
ans = comment_tags.get(field, meta_tags.get(field, None))
if ans:
ans = ans.strip()
if not ans:
ans = None
return ans
# Title
title = get('title')
if not title:
pat = re.compile('<title>([^<>]+?)</title>', re.IGNORECASE)
match = pat.search(src)
if match:
title = replace_entities(match.group(1))
# Author
authors = get('authors') or _('Unknown')
# Create MetaInformation with Title and Author
mi = Metadata(title or _('Unknown'), string_to_authors(authors))
for field in ('publisher', 'isbn', 'language', 'comments'):
val = get(field)
if val:
setattr(mi, field, val)
for field in ('pubdate', 'timestamp'):
try:
val = parse_date(get(field))
except:
pass
else:
if not is_date_undefined(val):
setattr(mi, field, val)
# SERIES
series = get('series')
if series:
pat = re.compile(r'\[([.0-9]+)\]$')
match = pat.search(series)
series_index = None
if match is not None:
try:
series_index = float(match.group(1))
except:
pass
series = series.replace(match.group(), '').strip()
mi.series = series
if series_index is None:
series_index = get('series_index')
try:
series_index = float(series_index)
except:
pass
if series_index is not None:
mi.series_index = series_index
# RATING
rating = get('rating')
if rating:
try:
mi.rating = float(rating)
if mi.rating < 0:
mi.rating = 0
if mi.rating > 5:
mi.rating /= 2.
if mi.rating > 5:
mi.rating = 0
except:
pass
# TAGS
tags = get('tags')
if tags:
tags = [x.strip() for x in tags.split(',') if x.strip()]
if tags:
mi.tags = tags
return mi
def 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 merge(self, results, min_year, do_asr=True):
ans = Metadata(_('Unknown'))
# We assume the shortest title has the least cruft in it
ans.title = self.length_merge('title', results, null_value=ans.title)
# No harm in having extra authors, maybe something useful like an
# editor or translator
ans.authors = self.length_merge('authors',
results,
null_value=ans.authors,
shortest=False)
# We assume the shortest publisher has the least cruft in it
ans.publisher = self.length_merge('publisher',
results,
null_value=ans.publisher)
# We assume the smallest set of tags has the least cruft in it
ans.tags = self.length_merge('tags',
results,
null_value=ans.tags,
shortest=msprefs['fewer_tags'])
# We assume the longest series has the most info in it
ans.series = self.length_merge('series',
results,
null_value=ans.series,
shortest=False)
for r in results:
if r.series and r.series == ans.series:
ans.series_index = r.series_index
break
# Average the rating over all sources
ratings = []
for r in results:
rating = r.rating
if rating and rating > 0 and rating <= 5:
ratings.append(rating)
if ratings:
ans.rating = sum(ratings) / len(ratings)
# Smallest language is likely to be valid
ans.language = self.length_merge('language',
results,
null_value=ans.language)
# Choose longest comments
ans.comments = self.length_merge('comments',
results,
null_value=ans.comments,
shortest=False)
# Published date
if min_year:
for r in results:
year = getattr(r.pubdate, 'year', None)
if year == min_year:
ans.pubdate = r.pubdate
break
if getattr(ans.pubdate, 'year', None) == min_year:
min_date = datetime(min_year,
ans.pubdate.month,
ans.pubdate.day,
tzinfo=utc_tz)
else:
min_date = datetime(min_year, 1, 2, tzinfo=utc_tz)
ans.pubdate = min_date
else:
min_date = datetime(3001, 1, 1, tzinfo=utc_tz)
for r in results:
if r.pubdate is not None:
candidate = as_utc(r.pubdate)
if candidate < min_date:
min_date = candidate
if min_date.year < 3000:
ans.pubdate = min_date
# Identifiers
for r in results:
ans.identifiers.update(r.identifiers)
# Cover URL
ans.has_cached_cover_url = bool(
[r for r in results if getattr(r, 'has_cached_cover_url', False)])
# Merge any other fields with no special handling (random merge)
touched_fields = set()
for r in results:
if hasattr(r, 'identify_plugin'):
touched_fields |= r.identify_plugin.touched_fields
for f in touched_fields:
if f.startswith('identifier:') or not ans.is_null(f):
continue
setattr(ans, f,
self.random_merge(f, results, null_value=getattr(ans, f)))
if do_asr:
avg = [x.relevance_in_source for x in results]
avg = sum(avg) / len(avg)
ans.average_source_relevance = avg
return ans
def __init__(self,
parent,
text,
mi=None,
fm=None,
color_field=None,
icon_field_key=None,
icon_rule_kind=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
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)
self.colored_field.setCurrentIndex(
self.colored_field.findData(color_field))
elif self.iconing:
self.icon_layout.setVisible(True)
for n1, k1 in cols:
self.icon_field.addItem(n1, 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()
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=QVariant(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 mi:
self.mi = mi
else:
self.mi = Metadata(_('Title'), [_('Author')])
self.mi.author_sort = _('Author Sort')
self.mi.series = _('Series')
self.mi.series_index = 3
self.mi.rating = 4.0
self.mi.tags = [_('Tag 1'), _('Tag 2')]
self.mi.languages = ['eng']
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
self.mi.set_all_user_metadata(get_gui(
).current_db.new_api.field_metadata.custom_field_metadata())
# Remove help icon on title bar
icon = self.windowIcon()
self.setWindowFlags(self.windowFlags()
& (~Qt.WindowContextHelpButtonHint))
self.setWindowIcon(icon)
self.last_text = ''
self.highlighter = TemplateHighlighter(self.textbox.document())
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:
self.textbox.setPlainText(text)
self.buttonBox.button(QDialogButtonBox.Ok).setText(_('&OK'))
self.buttonBox.button(QDialogButtonBox.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 = {}
self.funcs = formatter_functions().get_functions()
self.builtins = formatter_functions().get_builtins()
func_names = sorted(self.funcs)
self.function.clear()
self.function.addItem('')
self.function.addItems(func_names)
self.function.setCurrentIndex(0)
self.function.currentIndexChanged[str].connect(self.function_changed)
self.textbox_changed()
self.rule = (None, '')
tt = _('Template language tutorial')
self.template_tutorial.setText(
'<a href="http://manual.calibre-ebook.com/template_lang.html">'
'%s</a>' % tt)
tt = _('Template function reference')
self.template_func_reference.setText(
'<a href="http://manual.calibre-ebook.com/template_ref.html">'
'%s</a>' % tt)
self.font_size_box.setValue(gprefs['gpm_template_editor_font_size'])
self.font_size_box.valueChanged.connect(self.font_size_changed)
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 __init__(self,
prefix,
lpath,
title=None,
authors=None,
mime=None,
date=None,
ContentType=None,
thumbnail_name=None,
size=None,
other=None):
from calibre.utils.date import parse_date
# debug_print('Book::__init__ - title=', title)
show_debug = title is not None and title.lower().find("xxxxx") >= 0
if other is not None:
other.title = title
other.published_date = date
if show_debug:
debug_print("Book::__init__ - title=", title, 'authors=', authors)
debug_print("Book::__init__ - other=", other)
super(Book, self).__init__(prefix, lpath, size, other)
if title is not None and len(title) > 0:
self.title = title
if authors is not None and len(authors) > 0:
self.authors_from_string(authors)
if self.author_sort is None or self.author_sort == "Unknown":
self.author_sort = author_to_author_sort(authors)
self.mime = mime
self.size = size # will be set later if None
if ContentType == '6' and date is not None:
try:
self.datetime = time.strptime(date, "%Y-%m-%dT%H:%M:%S.%f")
except:
try:
self.datetime = time.strptime(
date.split('+')[0], "%Y-%m-%dT%H:%M:%S")
except:
try:
self.datetime = time.strptime(
date.split('+')[0], "%Y-%m-%d")
except:
try:
self.datetime = parse_date(
date, assume_utc=True).timetuple()
except:
try:
self.datetime = time.gmtime(
os.path.getctime(self.path))
except:
self.datetime = time.gmtime()
self.kobo_metadata = Metadata(title, self.authors)
self.contentID = None
self.current_shelves = []
self.kobo_collections = []
self.can_put_on_shelves = True
self.kobo_series = None
self.kobo_series_number = None # Kobo stores the series number as string. And it can have a leading "#".
self.kobo_subtitle = None
if thumbnail_name is not None:
self.thumbnail = ImageWrapper(thumbnail_name)
if show_debug:
debug_print("Book::__init__ end - self=", self)
debug_print("Book::__init__ end - title=", title, 'authors=',
authors)
def sample_results(self):
m1 = Metadata('The Great Gatsby', ['Francis Scott Fitzgerald'])
m2 = Metadata(
'The Great Gatsby - An extra long title to test resizing',
['F. Scott Fitzgerald'])
m1.has_cached_cover_url = True
m2.has_cached_cover_url = False
m1.comments = 'Some comments ' * 10
m1.tags = ['tag%d' % i for i in range(20)]
m1.rating = 4.4
m1.language = 'en'
m2.language = 'fr'
m1.pubdate = utcnow()
m2.pubdate = fromordinal(1000000)
m1.publisher = 'Publisher 1'
m2.publisher = 'Publisher 2'
return [m1, m2]
def test_legacy_adding_books(self): # {{{
'Test various adding/deleting books methods'
from calibre.ebooks.metadata.book.base import Metadata
from calibre.ptempfile import TemporaryFile
legacy, old = self.init_legacy(self.cloned_library), self.init_old(
self.cloned_library)
mi = Metadata('Added Book0', authors=('Added Author', ))
with TemporaryFile(suffix='.aff') as name:
with open(name, 'wb') as f:
f.write(b'xxx')
T = partial(ET,
'add_books', ([name], ['AFF'], [mi]),
old=old,
legacy=legacy)
T()(self)
book_id = T(kwargs={'return_ids': True})(self)[1][0]
self.assertEqual(legacy.new_api.formats(book_id), ('AFF', ))
T(kwargs={'add_duplicates': False})(self)
mi.title = 'Added Book1'
mi.uuid = 'uuu'
T = partial(ET,
'import_book', (mi, [name]),
old=old,
legacy=legacy)
book_id = T()(self)
self.assertNotEqual(legacy.uuid(book_id, index_is_id=True),
old.uuid(book_id, index_is_id=True))
book_id = T(kwargs={'preserve_uuid': True})(self)
self.assertEqual(legacy.uuid(book_id, index_is_id=True),
old.uuid(book_id, index_is_id=True))
self.assertEqual(legacy.new_api.formats(book_id), ('AFF', ))
T = partial(ET, 'add_format', old=old, legacy=legacy)
T((0, 'AFF', BytesIO(b'fffff')))(self)
T((0, 'AFF', BytesIO(b'fffff')))(self)
T((0, 'AFF', BytesIO(b'fffff')), {'replace': True})(self)
with TemporaryFile(suffix='.opf') as name:
with open(name, 'wb') as f:
f.write(b'zzzz')
T = partial(ET,
'import_book', (mi, [name]),
old=old,
legacy=legacy)
book_id = T()(self)
self.assertFalse(legacy.new_api.formats(book_id))
mi.title = 'Added Book2'
T = partial(ET, 'create_book_entry', (mi, ), old=old, legacy=legacy)
T()
T({'add_duplicates': False})
T({'force_id': 1000})
with TemporaryFile(suffix='.txt') as name:
with open(name, 'wb') as f:
f.write(b'tttttt')
bid = legacy.add_catalog(name, 'My Catalog')
self.assertEqual(old.add_catalog(name, 'My Catalog'), bid)
cache = legacy.new_api
self.assertEqual(cache.formats(bid), ('TXT', ))
self.assertEqual(cache.field_for('title', bid), 'My Catalog')
self.assertEqual(cache.field_for('authors', bid), ('calibre', ))
self.assertEqual(cache.field_for('tags', bid), (_('Catalog'), ))
self.assertTrue(bid < legacy.add_catalog(name, 'Something else'))
self.assertEqual(legacy.add_catalog(name, 'My Catalog'), bid)
self.assertEqual(old.add_catalog(name, 'My Catalog'), bid)
bid = legacy.add_news(
name, {
'title': 'Events',
'add_title_tag': True,
'custom_tags': ('one', 'two')
})
self.assertEqual(cache.formats(bid), ('TXT', ))
self.assertEqual(cache.field_for('authors', bid), ('calibre', ))
self.assertEqual(cache.field_for('tags', bid),
(_('News'), 'Events', 'one', 'two'))
self.assertTrue(legacy.cover(1, index_is_id=True))
origcov = legacy.cover(1, index_is_id=True)
self.assertTrue(legacy.has_cover(1))
legacy.remove_cover(1)
self.assertFalse(legacy.has_cover(1))
self.assertFalse(legacy.cover(1, index_is_id=True))
legacy.set_cover(3, origcov)
self.assertEqual(legacy.cover(3, index_is_id=True), origcov)
self.assertTrue(legacy.has_cover(3))
self.assertTrue(legacy.format(1, 'FMT1', index_is_id=True))
legacy.remove_format(1, 'FMT1', index_is_id=True)
self.assertIsNone(legacy.format(1, 'FMT1', index_is_id=True))
legacy.delete_book(1)
old.delete_book(1)
self.assertNotIn(1, legacy.all_ids())
legacy.dump_metadata((2, 3))
old.close()
class TemplateDialog(QDialog, Ui_TemplateDialog):
def __init__(self,
parent,
text,
mi=None,
fm=None,
color_field=None,
icon_field_key=None,
icon_rule_kind=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
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)
self.colored_field.setCurrentIndex(
self.colored_field.findData(color_field))
elif self.iconing:
self.icon_layout.setVisible(True)
for n1, k1 in cols:
self.icon_field.addItem(n1, 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()
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=QVariant(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 mi:
self.mi = mi
else:
self.mi = Metadata(_('Title'), [_('Author')])
self.mi.author_sort = _('Author Sort')
self.mi.series = _('Series')
self.mi.series_index = 3
self.mi.rating = 4.0
self.mi.tags = [_('Tag 1'), _('Tag 2')]
self.mi.languages = ['eng']
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
self.mi.set_all_user_metadata(get_gui(
).current_db.new_api.field_metadata.custom_field_metadata())
# Remove help icon on title bar
icon = self.windowIcon()
self.setWindowFlags(self.windowFlags()
& (~Qt.WindowContextHelpButtonHint))
self.setWindowIcon(icon)
self.last_text = ''
self.highlighter = TemplateHighlighter(self.textbox.document())
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:
self.textbox.setPlainText(text)
self.buttonBox.button(QDialogButtonBox.Ok).setText(_('&OK'))
self.buttonBox.button(QDialogButtonBox.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 = {}
self.funcs = formatter_functions().get_functions()
self.builtins = formatter_functions().get_builtins()
func_names = sorted(self.funcs)
self.function.clear()
self.function.addItem('')
self.function.addItems(func_names)
self.function.setCurrentIndex(0)
self.function.currentIndexChanged[str].connect(self.function_changed)
self.textbox_changed()
self.rule = (None, '')
tt = _('Template language tutorial')
self.template_tutorial.setText(
'<a href="http://manual.calibre-ebook.com/template_lang.html">'
'%s</a>' % tt)
tt = _('Template function reference')
self.template_func_reference.setText(
'<a href="http://manual.calibre-ebook.com/template_ref.html">'
'%s</a>' % tt)
self.font_size_box.setValue(gprefs['gpm_template_editor_font_size'])
self.font_size_box.valueChanged.connect(self.font_size_changed)
def font_size_changed(self, toWhat):
gprefs['gpm_template_editor_font_size'] = toWhat
self.highlighter.initializeFormats()
self.highlighter.rehighlight()
def filename_button_clicked(self):
try:
path = choose_files(self,
'choose_category_icon',
_('Select Icon'),
filters=[('Images',
['png', 'gif', 'jpg', 'jpeg'])],
all_files=False,
select_only_single_file=True)
if path:
icon_path = path[0]
icon_name = sanitize_file_name_unicode(
os.path.splitext(os.path.basename(icon_path))[0] + '.png')
if icon_name not in self.icon_file_names:
self.icon_file_names.append(icon_name)
self.update_filename_box()
try:
p = QIcon(icon_path).pixmap(QSize(128, 128))
d = os.path.join(config_dir, 'cc_icons')
if not os.path.exists(os.path.join(d, icon_name)):
if not os.path.exists(d):
os.makedirs(d)
with open(os.path.join(d, icon_name), 'wb') as f:
f.write(pixmap_to_data(p, format='PNG'))
except:
traceback.print_exc()
self.icon_files.setCurrentIndex(
self.icon_files.findText(icon_name))
self.icon_files.adjustSize()
except:
traceback.print_exc()
return
def update_filename_box(self):
self.icon_files.clear()
self.icon_file_names.sort(key=sort_key)
self.icon_files.addItem('')
self.icon_files.addItems(self.icon_file_names)
for i, filename in enumerate(self.icon_file_names):
icon = QIcon(os.path.join(config_dir, 'cc_icons', filename))
self.icon_files.setItemIcon(i + 1, icon)
def color_to_clipboard(self):
app = QApplication.instance()
c = app.clipboard()
c.setText(unicode(self.color_name.color))
def icon_to_clipboard(self):
app = QApplication.instance()
c = app.clipboard()
c.setText(unicode(self.icon_files.currentText()))
def textbox_changed(self):
cur_text = unicode(self.textbox.toPlainText())
if self.last_text != cur_text:
self.last_text = cur_text
self.highlighter.regenerate_paren_positions()
self.text_cursor_changed()
self.template_value.setText(SafeFormat().safe_format(
cur_text, self.mi, _('EXCEPTION: '), self.mi))
def text_cursor_changed(self):
cursor = self.textbox.textCursor()
position = cursor.position()
t = unicode(self.textbox.toPlainText())
if position > 0 and position <= len(t):
block_number = cursor.blockNumber()
pos_in_block = cursor.positionInBlock() - 1
self.highlighter.check_cursor_pos(t[position - 1], block_number,
pos_in_block)
def function_changed(self, toWhat):
name = unicode(toWhat)
self.source_code.clear()
self.documentation.clear()
if name in self.funcs:
self.documentation.setPlainText(self.funcs[name].doc)
if name in self.builtins and name in self.builtin_source_dict:
self.source_code.setPlainText(self.builtin_source_dict[name])
else:
self.source_code.setPlainText(self.funcs[name].program_text)
def accept(self):
txt = unicode(self.textbox.toPlainText()).rstrip()
if self.coloring:
if self.colored_field.currentIndex() == -1:
error_dialog(self,
_('No column chosen'),
_('You must specify a column to be colored'),
show=True)
return
if not txt:
error_dialog(self,
_('No template provided'),
_('The template box cannot be empty'),
show=True)
return
self.rule = (unicode(
self.colored_field.itemData(
self.colored_field.currentIndex()).toString()), txt)
elif self.iconing:
rt = unicode(
self.icon_kind.itemData(
self.icon_kind.currentIndex()).toString())
self.rule = (rt,
unicode(
self.icon_field.itemData(
self.icon_field.currentIndex()).toString()),
txt)
else:
self.rule = ('', txt)
QDialog.accept(self)
def test_legacy_setters(self): # {{{
'Test methods that are directly equivalent in the old and new interface'
from calibre.ebooks.metadata.book.base import Metadata
from calibre.utils.date import now
n = now()
ndb = self.init_legacy(self.cloned_library)
amap = ndb.new_api.get_id_map('authors')
sorts = [(aid, 's%d' % aid) for aid in amap]
db = self.init_old(self.cloned_library)
run_funcs(self, db, ndb, (
('+format_metadata', 1, 'FMT1', itemgetter('size')),
('+format_metadata', 1, 'FMT2', itemgetter('size')),
('+format_metadata', 2, 'FMT1', itemgetter('size')),
('get_tags', 0),
('get_tags', 1),
('get_tags', 2),
('is_tag_used', 'News'),
('is_tag_used', 'xchkjgfh'),
('bulk_modify_tags', (1, ), ['t1'], ['News']),
('bulk_modify_tags', (2, ), ['t1'], ['Tag One', 'Tag Two']),
('bulk_modify_tags', (3, ), ['t1', 't2', 't3']),
(db.clean, ),
('@all_tags', ),
('@tags', 0),
('@tags', 1),
('@tags', 2),
('unapply_tags', 1, ['t1']),
('unapply_tags', 2, ['xxxx']),
('unapply_tags', 3, ['t2', 't3']),
(db.clean, ),
('@all_tags', ),
('@tags', 0),
('@tags', 1),
('@tags', 2),
('update_last_modified', (1, ), True, n),
('update_last_modified', (3, ), True, n),
('metadata_last_modified', 1, True),
('metadata_last_modified', 3, True),
('set_sort_field_for_author', sorts[0][0], sorts[0][1]),
('set_sort_field_for_author', sorts[1][0], sorts[1][1]),
('set_sort_field_for_author', sorts[2][0], sorts[2][1]),
('set_link_field_for_author', sorts[0][0], sorts[0][1]),
('set_link_field_for_author', sorts[1][0], sorts[1][1]),
('set_link_field_for_author', sorts[2][0], sorts[2][1]),
(db.refresh, ),
('author_sort', 0),
('author_sort', 1),
('author_sort', 2),
))
omi = [db.get_metadata(x) for x in (0, 1, 2)]
nmi = [ndb.get_metadata(x) for x in (0, 1, 2)]
self.assertEqual([x.author_sort_map for x in omi],
[x.author_sort_map for x in nmi])
self.assertEqual([x.author_link_map for x in omi],
[x.author_link_map for x in nmi])
db.close()
ndb = self.init_legacy(self.cloned_library)
db = self.init_old(self.cloned_library)
run_funcs(self, db, ndb, (
(
'set_authors',
1,
('author one', ),
),
('set_authors', 2, ('author two', ), True, True, True),
('set_author_sort', 3, 'new_aus'),
('set_comment', 1, ''),
('set_comment', 2, None),
('set_comment', 3, '<p>a comment</p>'),
('set_has_cover', 1, True),
('set_has_cover', 2, True),
('set_has_cover', 3, 1),
('set_identifiers', 2, {
'test': '',
'a': 'b'
}),
('set_identifiers', 3, {
'id': '1',
'isbn': '9783161484100'
}),
('set_identifiers', 1, {}),
('set_languages', 1, ('en', )),
('set_languages', 2, ()),
('set_languages', 3, ('deu', 'spa', 'fra')),
('set_pubdate', 1, None),
('set_pubdate', 2, '2011-1-7'),
('set_series', 1, 'a series one'),
('set_series', 2, 'another series [7]'),
('set_series', 3, 'a third series'),
('set_publisher', 1, 'publisher two'),
('set_publisher', 2, None),
('set_publisher', 3, 'a third puB'),
('set_rating', 1, 2.3),
('set_rating', 2, 0),
('set_rating', 3, 8),
('set_timestamp', 1, None),
('set_timestamp', 2, '2011-1-7'),
('set_uuid', 1, None),
('set_uuid', 2, 'a test uuid'),
('set_title', 1, 'title two'),
('set_title', 2, None),
('set_title', 3, 'The Test Title'),
('set_tags', 1, ['a1', 'a2'], True),
('set_tags', 2, ['b1', 'tag one'], False, False, False, True),
('set_tags', 3, ['A1']),
(db.refresh, ),
('title', 0),
('title', 1),
('title', 2),
('title_sort', 0),
('title_sort', 1),
('title_sort', 2),
('authors', 0),
('authors', 1),
('authors', 2),
('author_sort', 0),
('author_sort', 1),
('author_sort', 2),
('has_cover', 3),
('has_cover', 1),
('has_cover', 2),
('get_identifiers', 0),
('get_identifiers', 1),
('get_identifiers', 2),
('pubdate', 0),
('pubdate', 1),
('pubdate', 2),
('timestamp', 0),
('timestamp', 1),
('timestamp', 2),
('publisher', 0),
('publisher', 1),
('publisher', 2),
('rating', 0),
('+rating', 1, lambda x: x or 0),
('rating', 2),
('series', 0),
('series', 1),
('series', 2),
('series_index', 0),
('series_index', 1),
('series_index', 2),
('uuid', 0),
('uuid', 1),
('uuid', 2),
('isbn', 0),
('isbn', 1),
('isbn', 2),
('@tags', 0),
('@tags', 1),
('@tags', 2),
('@all_tags', ),
('@get_all_identifier_types', ),
('set_title_sort', 1, 'Title Two'),
('set_title_sort', 2, None),
('set_title_sort', 3, 'The Test Title_sort'),
('set_series_index', 1, 2.3),
('set_series_index', 2, 0),
('set_series_index', 3, 8),
('set_identifier', 1, 'moose', 'val'),
('set_identifier', 2, 'test', ''),
('set_identifier', 3, '', ''),
(db.refresh, ),
('series_index', 0),
('series_index', 1),
('series_index', 2),
('title_sort', 0),
('title_sort', 1),
('title_sort', 2),
('get_identifiers', 0),
('get_identifiers', 1),
('get_identifiers', 2),
('@get_all_identifier_types', ),
('set_metadata', 1, Metadata(
'title', ('a1', )), False, False, False, True, True),
('set_metadata', 3, Metadata('title', ('a1', ))),
(db.refresh, ),
('title', 0),
('title', 1),
('title', 2),
('title_sort', 0),
('title_sort', 1),
('title_sort', 2),
('authors', 0),
('authors', 1),
('authors', 2),
('author_sort', 0),
('author_sort', 1),
('author_sort', 2),
('@tags', 0),
('@tags', 1),
('@tags', 2),
('@all_tags', ),
('@get_all_identifier_types', ),
))
db.close()
ndb = self.init_legacy(self.cloned_library)
db = self.init_old(self.cloned_library)
run_funcs(self, db, ndb, (
('set', 0, 'title', 'newtitle'),
('set', 0, 'tags', 't1,t2,tag one', True),
('set', 0, 'authors', 'author one & Author Two', True),
('set', 0, 'rating', 3.2),
('set', 0, 'publisher', 'publisher one', False),
(db.refresh, ),
('title', 0),
('rating', 0),
('#tags', 0),
('#tags', 1),
('#tags', 2),
('authors', 0),
('authors', 1),
('authors', 2),
('publisher', 0),
('publisher', 1),
('publisher', 2),
('delete_tag', 'T1'),
('delete_tag', 'T2'),
('delete_tag', 'Tag one'),
('delete_tag', 'News'),
(db.clean, ),
(db.refresh, ),
('@all_tags', ),
('#tags', 0),
('#tags', 1),
('#tags', 2),
))
db.close()
ndb = self.init_legacy(self.cloned_library)
db = self.init_old(self.cloned_library)
run_funcs(self, db, ndb, (
('remove_all_tags', (1, 2, 3)),
(db.clean, ),
('@all_tags', ),
('@tags', 0),
('@tags', 1),
('@tags', 2),
))
db.close()
ndb = self.init_legacy(self.cloned_library)
db = self.init_old(self.cloned_library)
a = {v: k
for k, v in ndb.new_api.get_id_map('authors').iteritems()
}['Author One']
t = {v: k
for k, v in ndb.new_api.get_id_map('tags').iteritems()}['Tag One']
s = {v: k
for k, v in ndb.new_api.get_id_map('series').iteritems()
}['A Series One']
p = {v: k
for k, v in ndb.new_api.get_id_map('publisher').iteritems()
}['Publisher One']
run_funcs(self, db, ndb, (
('rename_author', a, 'Author Two'),
('rename_tag', t, 'News'),
('rename_series', s, 'ss'),
('rename_publisher', p, 'publisher one'),
(db.clean, ),
(db.refresh, ),
('@all_tags', ),
('tags', 0),
('tags', 1),
('tags', 2),
('series', 0),
('series', 1),
('series', 2),
('publisher', 0),
('publisher', 1),
('publisher', 2),
('series_index', 0),
('series_index', 1),
('series_index', 2),
('authors', 0),
('authors', 1),
('authors', 2),
('author_sort', 0),
('author_sort', 1),
('author_sort', 2),
))
db.close()
def parse_details(self, root):
try:
kyobobook_id = self.parse_kyobobook_id(self.url)
except:
self.log.exception('Error parsing Kyobobook id for url: %r' %
self.url)
kyobobook_id = None
try:
(title, series, series_index) = self.parse_title_series(root)
except:
self.log.exception('Error parsing title and series for url: %r' %
self.url)
title = series = series_index = None
try:
authors = self.parse_authors(root)
except:
self.log.exception('Error parsing authors for url: %r' % self.url)
authors = []
if not title or not authors or not kyobobook_id:
self.log.error('Could not find title/authors/kyobobook id for %r' %
self.url)
self.log.error('Kyobobook: %r Title: %r Authors: %r' %
(kyobobook_id, title, authors))
return
mi = Metadata(title, authors)
if series:
mi.series = series
mi.series_index = series_index
mi.set_identifier('kyobobook', kyobobook_id)
self.kyobobook_id = kyobobook_id
try:
isbn = self.parse_isbn(root)
if isbn:
self.isbn = mi.isbn = isbn
except:
self.log.exception('Error parsing ISBN for url: %r' % self.url)
try:
mi.rating = self.parse_rating(root)
except:
self.log.exception('Error parsing ratings for url: %r' % self.url)
try:
mi.comments = self.parse_comments(root)
except:
self.log.exception('Error parsing comments for url: %r' % self.url)
try:
self.cover_url = self.parse_cover(root)
except:
self.log.exception('Error parsing cover for url: %r' % self.url)
mi.has_cover = bool(self.cover_url)
try:
tags = self.parse_tags(root)
if tags:
mi.tags = tags
except:
self.log.exception('Error parsing tags for url: %r' % self.url)
try:
mi.publisher, mi.pubdate = self.parse_publisher_and_date(root)
except:
self.log.exception('Error parsing publisher and date for url: %r' %
self.url)
try:
lang = self._parse_language(root)
if lang:
mi.language = lang
except:
self.log.exception('Error parsing language for url: %r' % self.url)
mi.source_relevance = self.relevance
if self.kyobobook_id:
if self.isbn:
self.plugin.cache_isbn_to_identifier(self.isbn,
self.kyobobook_id)
if self.cover_url:
self.plugin.cache_identifier_to_cover_url(
self.kyobobook_id, self.cover_url)
self.plugin.clean_downloaded_metadata(mi)
self.result_queue.put(mi)
def parse_feed(self, feed, seen, orig_title, orig_authors, identifiers):
from lxml import etree
def tostring(x):
if x is None:
return ''
return etree.tostring(x, method='text', encoding=unicode).strip()
orig_isbn = identifiers.get('isbn', None)
title_tokens = list(self.get_title_tokens(orig_title))
author_tokens = list(self.get_author_tokens(orig_authors))
results = []
def ismatch(title, authors):
authors = lower(' '.join(authors))
title = lower(title)
match = not title_tokens
for t in title_tokens:
if lower(t) in title:
match = True
break
amatch = not author_tokens
for a in author_tokens:
if lower(a) in authors:
amatch = True
break
if not author_tokens: amatch = True
return match and amatch
bl = feed.find('BookList')
if bl is None:
err = tostring(feed.find('errormessage'))
raise ValueError('ISBNDb query failed:' + err)
total_results = int(bl.get('total_results'))
shown_results = int(bl.get('shown_results'))
for bd in bl.xpath('.//BookData'):
isbn = check_isbn(bd.get('isbn', None))
isbn13 = check_isbn(bd.get('isbn13', None))
if not isbn and not isbn13:
continue
if orig_isbn and orig_isbn not in {isbn, isbn13}:
continue
title = tostring(bd.find('Title'))
if not title:
continue
authors = []
for au in bd.xpath('.//Authors/Person'):
au = tostring(au)
if au:
if ',' in au:
ln, _, fn = au.partition(',')
au = fn.strip() + ' ' + ln.strip()
authors.append(au)
if not authors:
continue
comments = tostring(bd.find('Summary'))
id_ = (title, tuple(authors))
if id_ in seen:
continue
seen.add(id_)
if not ismatch(title, authors):
continue
publisher = tostring(bd.find('PublisherText'))
if not publisher: publisher = None
if publisher and 'audio' in publisher.lower():
continue
mi = Metadata(title, authors)
mi.isbn = isbn
mi.publisher = publisher
mi.comments = comments
results.append(mi)
return total_results, shown_results, results
def start(self, title, authors, identifiers):
book = Metadata(title, authors)
book.identifiers = identifiers
self.covers_widget.start(book, self.current_cover, title, authors, {})
return self.exec_()