def apply_diff(self, identical_msg, cache, syntax_map, changed_names, renamed_names, removed_names, added_names):
self.view.clear()
self.apply_diff_calls = calls = []
def add(args, kwargs):
self.view.add_diff(*args, **kwargs)
calls.append((args, kwargs))
if len(changed_names) + len(renamed_names) + len(removed_names) + len(added_names) < 1:
info_dialog(self, _('No changes found'), identical_msg, show=True)
return True
kwargs = lambda name: {'context':self.context, 'beautify':self.beautify, 'syntax':syntax_map.get(name, None)}
if isinstance(changed_names, dict):
for name, other_name in sorted(changed_names.iteritems(), key=lambda x:numeric_sort_key(x[0])):
args = (name, other_name, cache.left(name), cache.right(other_name))
add(args, kwargs(name))
else:
for name in sorted(changed_names, key=numeric_sort_key):
args = (name, name, cache.left(name), cache.right(name))
add(args, kwargs(name))
for name in sorted(added_names, key=numeric_sort_key):
args = (_('[%s was added]') % name, name, None, cache.right(name))
add(args, kwargs(name))
for name in sorted(removed_names, key=numeric_sort_key):
args = (name, _('[%s was removed]') % name, cache.left(name), None)
add(args, kwargs(name))
for name, new_name in sorted(renamed_names.iteritems(), key=lambda x:numeric_sort_key(x[0])):
args = (name, new_name, None, None)
add(args, kwargs(name))
def apply_diff(self, identical_msg, cache, syntax_map, changed_names, renamed_names, removed_names, added_names):
self.view.clear()
self.apply_diff_calls = calls = []
def add(args, kwargs):
self.view.add_diff(*args, **kwargs)
calls.append((args, kwargs))
if len(changed_names) + len(renamed_names) + len(removed_names) + len(added_names) < 1:
self.busy.setVisible(False)
info_dialog(self, _('No changes found'), identical_msg, show=True)
self.busy.setVisible(True)
return True
kwargs = lambda name: {'context':self.context, 'beautify':self.beautify, 'syntax':syntax_map.get(name, None)}
if isinstance(changed_names, dict):
for name, other_name in sorted(changed_names.iteritems(), key=lambda x:numeric_sort_key(x[0])):
args = (name, other_name, cache.left(name), cache.right(other_name))
add(args, kwargs(name))
else:
for name in sorted(changed_names, key=numeric_sort_key):
args = (name, name, cache.left(name), cache.right(name))
add(args, kwargs(name))
for name in sorted(added_names, key=numeric_sort_key):
args = (_('[%s was added]') % name, name, None, cache.right(name))
add(args, kwargs(name))
for name in sorted(removed_names, key=numeric_sort_key):
args = (name, _('[%s was removed]') % name, cache.left(name), None)
add(args, kwargs(name))
for name, new_name in sorted(renamed_names.iteritems(), key=lambda x:numeric_sort_key(x[0])):
args = (name, new_name, None, None)
add(args, kwargs(name))
def serialize(self, styles):
for style in sorted(
self.block_styles,
key=lambda s:
(s is not self.normal_block_style, numeric_sort_key(s.id))):
style.serialize(styles, self.normal_block_style)
for style in sorted(
self.text_styles,
key=lambda s:
(s is not self.normal_text_style, numeric_sort_key(s.id))):
style.serialize(styles, self.normal_text_style)
def test_sorting(self):
" Test the various sorting APIs "
german = """Sonntag Montag Dienstag Januar Februar März Fuße Fluße Flusse flusse fluße flüße flüsse""".split()
german_good = (
"""Dienstag Februar flusse Flusse fluße Fluße flüsse flüße Fuße Januar März Montag Sonntag""".split()
)
french = """dimanche lundi mardi janvier février mars déjà Meme deja même dejà bpef bœg Boef Mémé bœf boef bnef pêche pèché pêché pêche pêché""".split()
french_good = """bnef boef Boef bœf bœg bpef deja dejà déjà dimanche février janvier lundi mardi mars Meme Mémé même pèché pêche pêche pêché pêché""".split() # noqa
# Test corner cases
sort_key = icu.sort_key
s = "\U0001f431"
self.ae(
sort_key(s),
sort_key(s.encode(sys.getdefaultencoding())),
"UTF-8 encoded object not correctly decoded to generate sort key",
)
self.ae(s.encode("utf-16"), s.encode("utf-16"), "Undecodable bytestring not returned as itself")
self.ae(b"", sort_key(None))
self.ae(0, icu.strcmp(None, b""))
self.ae(0, icu.strcmp(s, s.encode(sys.getdefaultencoding())))
# Test locales
with make_collation_func("dsk", "de", func="sort_key") as dsk:
self.ae(german_good, sorted(german, key=dsk))
with make_collation_func("dcmp", "de", template="_strcmp_template") as dcmp:
for x in german:
for y in german:
self.ae(cmp(dsk(x), dsk(y)), dcmp(x, y))
with make_collation_func("fsk", "fr", func="sort_key") as fsk:
self.ae(french_good, sorted(french, key=fsk))
with make_collation_func("fcmp", "fr", template="_strcmp_template") as fcmp:
for x in french:
for y in french:
self.ae(cmp(fsk(x), fsk(y)), fcmp(x, y))
with make_collation_func("ssk", "es", func="sort_key") as ssk:
self.assertNotEqual(ssk("peña"), ssk("pena"))
with make_collation_func("scmp", "es", template="_strcmp_template") as scmp:
self.assertNotEqual(0, scmp("pena", "peña"))
for k, v in {"pèché": "peche", "flüße": "Flusse", "Štepánek": "ŠtepaneK"}.iteritems():
self.ae(0, icu.primary_strcmp(k, v))
# Test different types of collation
self.ae(icu.primary_sort_key("Aä"), icu.primary_sort_key("aa"))
self.assertLess(icu.numeric_sort_key("something 2"), icu.numeric_sort_key("something 11"))
self.assertLess(icu.case_sensitive_sort_key("A"), icu.case_sensitive_sort_key("a"))
self.ae(0, icu.strcmp("a", "A"))
self.ae(cmp("a", "A"), icu.case_sensitive_strcmp("a", "A"))
self.ae(0, icu.primary_strcmp("ä", "A"))
def setup_export_panel(self):
self.export_panel = w = QWidget(self)
self.stack.addWidget(w)
w.l = l = QVBoxLayout(w)
w.la = la = QLabel(
_('Select which libraries you want to export below'))
la.setWordWrap(True), l.addWidget(la)
self.lib_list = ll = QListWidget(self)
l.addWidget(ll)
ll.setSelectionMode(QAbstractItemView.SelectionMode.ExtendedSelection)
ll.setStyleSheet('QListView::item { padding: 5px }')
ll.setAlternatingRowColors(True)
lpaths = all_known_libraries()
for lpath in sorted(
lpaths, key=lambda x: numeric_sort_key(os.path.basename(x))):
i = QListWidgetItem(self.export_lib_text(lpath), ll)
i.setData(Qt.ItemDataRole.UserRole, lpath)
i.setData(Qt.ItemDataRole.UserRole + 1, lpaths[lpath])
i.setIcon(QIcon(I('lt.png')))
i.setSelected(True)
self.update_disk_usage.connect(
(lambda i, sz: self.lib_list.item(i).setText(
self.export_lib_text(
self.lib_list.item(i).data(Qt.ItemDataRole.UserRole), sz))
),
type=Qt.ConnectionType.QueuedConnection)
def flatten_spine(self):
names = defaultdict(int)
styles, pseudo_styles = {}, defaultdict(dict)
for item in self.items:
html = item.data
stylizer = self.stylizers[item]
if self.specializer is not None:
self.specializer(item, stylizer)
body = html.find(XHTML('body'))
fsize = self.context.dest.fbase
self.flatten_node(body, stylizer, names, styles, pseudo_styles,
fsize, item.id)
items = sorted(((key, val) for (val, key) in iteritems(styles)),
key=lambda x: numeric_sort_key(x[0]))
# :hover must come after link and :active must come after :hover
psels = sorted(pseudo_styles,
key=lambda x: {
'hover': 1,
'active': 2
}.get(x, 0))
for psel in psels:
styles = pseudo_styles[psel]
if not styles:
continue
x = sorted(((k + ':' + psel, v) for v, k in iteritems(styles)))
items.extend(x)
css = ''.join(".%s {\n%s;\n}\n\n" % (key, val) for key, val in items)
href = self.replace_css(css)
global_css = self.collect_global_css()
for item in self.items:
stylizer = self.stylizers[item]
self.flatten_head(item, href, global_css[item])
def find_pages(dir, sort_on_mtime=False, verbose=False):
'''
Find valid comic pages in a previously un-archived comic.
:param dir: Directory in which extracted comic lives
:param sort_on_mtime: If True sort pages based on their last modified time.
Otherwise, sort alphabetically.
'''
extensions = {'jpeg', 'jpg', 'gif', 'png', 'webp'}
pages = []
for datum in os.walk(dir):
for name in datum[-1]:
path = os.path.abspath(os.path.join(datum[0], name))
if '__MACOSX' in path:
continue
for ext in extensions:
if path.lower().endswith('.'+ext):
pages.append(path)
break
sep_counts = {x.replace(os.sep, '/').count('/') for x in pages}
# Use the full path to sort unless the files are in folders of different
# levels, in which case simply use the filenames.
basename = os.path.basename if len(sep_counts) > 1 else lambda x: x
if sort_on_mtime:
key = lambda x:os.stat(x).st_mtime
else:
key = lambda x:numeric_sort_key(basename(x))
pages.sort(key=key)
if verbose:
prints('Found comic pages...')
prints('\t'+'\n\t'.join([os.path.relpath(p, dir) for p in pages]))
return pages
def find_pages(dir, sort_on_mtime=False, verbose=False):
'''
Find valid comic pages in a previously un-archived comic.
:param dir: Directory in which extracted comic lives
:param sort_on_mtime: If True sort pages based on their last modified time.
Otherwise, sort alphabetically.
'''
extensions = {'jpeg', 'jpg', 'gif', 'png', 'webp'}
pages = []
for datum in os.walk(dir):
for name in datum[-1]:
path = os.path.join(datum[0], name)
if '__MACOSX' in path:
continue
for ext in extensions:
if path.lower().endswith('.'+ext):
pages.append(path)
break
if sort_on_mtime:
key = lambda x:os.stat(x).st_mtime
else:
key = lambda x:numeric_sort_key(os.path.basename(x))
pages.sort(key=key)
if verbose:
prints('Found comic pages...')
prints('\t'+'\n\t'.join([os.path.basename(p) for p in pages]))
return pages
def find_pages(dir, sort_on_mtime=False, verbose=False):
'''
Find valid comic pages in a previously un-archived comic.
:param dir: Directory in which extracted comic lives
:param sort_on_mtime: If True sort pages based on their last modified time.
Otherwise, sort alphabetically.
'''
extensions = {'jpeg', 'jpg', 'gif', 'png', 'webp'}
pages = []
for datum in os.walk(dir):
for name in datum[-1]:
path = os.path.abspath(os.path.join(datum[0], name))
if '__MACOSX' in path:
continue
for ext in extensions:
if path.lower().endswith('.' + ext):
pages.append(path)
break
sep_counts = {x.replace(os.sep, '/').count('/') for x in pages}
# Use the full path to sort unless the files are in folders of different
# levels, in which case simply use the filenames.
basename = os.path.basename if len(sep_counts) > 1 else lambda x: x
if sort_on_mtime:
key = lambda x: os.stat(x).st_mtime
else:
key = lambda x: numeric_sort_key(basename(x))
pages.sort(key=key)
if verbose:
prints('Found comic pages...')
prints('\t' + '\n\t'.join([os.path.relpath(p, dir) for p in pages]))
return pages
def test_sorting(self):
' Test the various sorting APIs '
german = '''Sonntag Montag Dienstag Januar Februar März Fuße Fluße Flusse flusse fluße flüße flüsse'''.split()
german_good = '''Dienstag Februar flusse Flusse fluße Fluße flüsse flüße Fuße Januar März Montag Sonntag'''.split()
french = '''dimanche lundi mardi janvier février mars déjà Meme deja même dejà bpef bœg Boef Mémé bœf boef bnef pêche pèché pêché pêche pêché'''.split()
french_good = '''bnef boef Boef bœf bœg bpef deja dejà déjà dimanche février janvier lundi mardi mars Meme Mémé même pèché pêche pêche pêché pêché'''.split() # noqa
# Test corner cases
sort_key = icu.sort_key
s = '\U0001f431'
self.ae(sort_key(s), sort_key(s.encode(sys.getdefaultencoding())), 'UTF-8 encoded object not correctly decoded to generate sort key')
self.ae(s.encode('utf-16'), s.encode('utf-16'), 'Undecodable bytestring not returned as itself')
self.ae(b'', sort_key(None))
self.ae(0, icu.strcmp(None, b''))
self.ae(0, icu.strcmp(s, s.encode(sys.getdefaultencoding())))
# Test locales
with make_collation_func('dsk', 'de', func='sort_key') as dsk:
self.ae(german_good, sorted(german, key=dsk))
with make_collation_func('dcmp', 'de', template='_strcmp_template') as dcmp:
for x in german:
for y in german:
self.ae(cmp(dsk(x), dsk(y)), dcmp(x, y))
with make_collation_func('fsk', 'fr', func='sort_key') as fsk:
self.ae(french_good, sorted(french, key=fsk))
with make_collation_func('fcmp', 'fr', template='_strcmp_template') as fcmp:
for x in french:
for y in french:
self.ae(cmp(fsk(x), fsk(y)), fcmp(x, y))
with make_collation_func('ssk', 'es', func='sort_key') as ssk:
self.assertNotEqual(ssk('peña'), ssk('pena'))
with make_collation_func('scmp', 'es', template='_strcmp_template') as scmp:
self.assertNotEqual(0, scmp('pena', 'peña'))
for k, v in {u'pèché': u'peche', u'flüße':u'Flusse', u'Štepánek':u'ŠtepaneK'}.iteritems():
self.ae(0, icu.primary_strcmp(k, v))
# Test different types of collation
self.ae(icu.primary_sort_key('Aä'), icu.primary_sort_key('aa'))
self.assertLess(icu.numeric_sort_key('something 2'), icu.numeric_sort_key('something 11'))
self.assertLess(icu.case_sensitive_sort_key('A'), icu.case_sensitive_sort_key('a'))
self.ae(0, icu.strcmp('a', 'A'))
self.ae(cmp('a', 'A'), icu.case_sensitive_strcmp('a', 'A'))
self.ae(0, icu.primary_strcmp('ä', 'A'))
def test_sorting(self):
' Test the various sorting APIs '
german = '''Sonntag Montag Dienstag Januar Februar März Fuße Fluße Flusse flusse fluße flüße flüsse'''.split()
german_good = '''Dienstag Februar flusse Flusse fluße Fluße flüsse flüße Fuße Januar März Montag Sonntag'''.split()
french = '''dimanche lundi mardi janvier février mars déjà Meme deja même dejà bpef bœg Boef Mémé bœf boef bnef pêche pèché pêché pêche pêché'''.split()
french_good = '''bnef boef Boef bœf bœg bpef deja dejà déjà dimanche février janvier lundi mardi mars Meme Mémé même pèché pêche pêche pêché pêché'''.split() # noqa
# Test corner cases
sort_key = icu.sort_key
s = '\U0001f431'
self.ae(sort_key(s), sort_key(s.encode(sys.getdefaultencoding())), 'UTF-8 encoded object not correctly decoded to generate sort key')
self.ae(s.encode('utf-16'), s.encode('utf-16'), 'Undecodable bytestring not returned as itself')
self.ae(b'', sort_key(None))
self.ae(0, icu.strcmp(None, b''))
self.ae(0, icu.strcmp(s, s.encode(sys.getdefaultencoding())))
# Test locales
with make_collation_func('dsk', 'de', func='sort_key') as dsk:
self.ae(german_good, sorted(german, key=dsk))
with make_collation_func('dcmp', 'de', template='_strcmp_template') as dcmp:
for x in german:
for y in german:
self.ae(cmp(dsk(x), dsk(y)), dcmp(x, y))
with make_collation_func('fsk', 'fr', func='sort_key') as fsk:
self.ae(french_good, sorted(french, key=fsk))
with make_collation_func('fcmp', 'fr', template='_strcmp_template') as fcmp:
for x in french:
for y in french:
self.ae(cmp(fsk(x), fsk(y)), fcmp(x, y))
with make_collation_func('ssk', 'es', func='sort_key') as ssk:
self.assertNotEqual(ssk('peña'), ssk('pena'))
with make_collation_func('scmp', 'es', template='_strcmp_template') as scmp:
self.assertNotEqual(0, scmp('pena', 'peña'))
for k, v in iteritems({u'pèché': u'peche', u'flüße':u'Flusse', u'Štepánek':u'ŠtepaneK'}):
self.ae(0, icu.primary_strcmp(k, v))
# Test different types of collation
self.ae(icu.primary_sort_key('Aä'), icu.primary_sort_key('aa'))
self.assertLess(icu.numeric_sort_key('something 2'), icu.numeric_sort_key('something 11'))
self.assertLess(icu.case_sensitive_sort_key('A'), icu.case_sensitive_sort_key('a'))
self.ae(0, icu.strcmp('a', 'A'))
self.ae(cmp('a', 'A'), icu.case_sensitive_strcmp('a', 'A'))
self.ae(0, icu.primary_strcmp('ä', 'A'))
def resort(self):
if self.sort_alphabetically.isChecked():
sorted_locations = sorted(self.locations, key=lambda (name, loc): numeric_sort_key(name))
else:
sorted_locations = self.locations
self.items.clear()
for name, loc in sorted_locations:
i = QListWidgetItem(name, self.items)
i.setData(Qt.UserRole, loc)
self.items.setCurrentRow(0)
def resort(self):
if self.sort_alphabetically.isChecked():
sorted_locations = sorted(self.locations, key=lambda name_loc: numeric_sort_key(name_loc[0]))
else:
sorted_locations = self.locations
self.items.clear()
for name, loc in sorted_locations:
i = QListWidgetItem(name, self.items)
i.setData(Qt.UserRole, loc)
self.items.setCurrentRow(0)
def setup_select_libraries_panel(self):
self.imported_lib_widgets = []
self.frames = []
l = self.slp.layout()
for lpath in sorted(self.importer.metadata['libraries'], key=lambda x:numeric_sort_key(os.path.basename(x))):
f = QFrame(self)
self.frames.append(f)
l.addWidget(f)
f.setFrameShape(f.HLine)
w = ImportLocation(lpath, self.slp)
l.addWidget(w)
self.imported_lib_widgets.append(w)
l.addStretch()
def populate_anchors(self, name):
if name not in self.anchor_cache:
from calibre.ebooks.oeb.base import XHTML_NS
root = self.container.parsed(name)
ac = self.anchor_cache[name] = []
for item in set(root.xpath('//*[@id]')) | set(root.xpath('//h:a[@name]', namespaces={'h':XHTML_NS})):
frag = item.get('id', None) or item.get('name')
if not frag:
continue
text = lead_text(item, num_words=4)
ac.append((text, frag))
ac.sort(key=lambda text_frag: numeric_sort_key(text_frag[0]))
self.anchor_names.model().set_names(self.anchor_cache[name])
self.update_target()
def populate_anchors(self, name):
if name not in self.anchor_cache:
from calibre.ebooks.oeb.base import XHTML_NS
root = self.container.parsed(name)
ac = self.anchor_cache[name] = []
for item in set(root.xpath('//*[@id]')) | set(root.xpath('//h:a[@name]', namespaces={'h':XHTML_NS})):
frag = item.get('id', None) or item.get('name')
if not frag:
continue
text = lead_text(item, num_words=4)
ac.append((text, frag))
ac.sort(key=lambda text_frag: numeric_sort_key(text_frag[0]))
self.anchor_names.model().set_names(self.anchor_cache[name])
self.update_target()
def set_display_name(name, item):
if tprefs['file_list_shows_full_pathname']:
text = name
else:
if name in processed:
# We have an exact duplicate (can happen if there are
# duplicates in the spine)
item.setText(0, processed[name].text(0))
item.setText(1, processed[name].text(1))
return
parts = name.split('/')
text = parts.pop()
while text in seen and parts:
text = parts.pop() + '/' + text
seen[text] = item
item.setText(0, text)
item.setText(1, as_hex_unicode(numeric_sort_key(text)))
def set_display_name(name, item):
if tprefs['file_list_shows_full_pathname']:
text = name
else:
if name in processed:
# We have an exact duplicate (can happen if there are
# duplicates in the spine)
item.setText(0, processed[name].text(0))
item.setText(1, processed[name].text(1))
return
parts = name.split('/')
text = parts.pop()
while text in seen and parts:
text = parts.pop() + '/' + text
seen[text] = item
item.setText(0, text)
item.setText(1, as_hex_unicode(numeric_sort_key(text)))
def setup_export_panel(self):
self.export_panel = w = QWidget(self)
self.stack.addWidget(w)
w.l = l = QVBoxLayout(w)
w.la = la = QLabel(_('Select which libraries you want to export below'))
la.setWordWrap(True), l.addWidget(la)
self.lib_list = ll = QListWidget(self)
l.addWidget(ll)
ll.setSelectionMode(ll.ExtendedSelection)
ll.setStyleSheet('QListView::item { padding: 5px }')
ll.setAlternatingRowColors(True)
lpaths = all_known_libraries()
for lpath in sorted(lpaths, key=lambda x:numeric_sort_key(os.path.basename(x))):
i = QListWidgetItem(self.export_lib_text(lpath), ll)
i.setData(Qt.UserRole, lpath)
i.setData(Qt.UserRole+1, lpaths[lpath])
i.setIcon(QIcon(I('lt.png')))
i.setSelected(True)
self.update_disk_usage.connect((
lambda i, sz: self.lib_list.item(i).setText(self.export_lib_text(self.lib_list.item(i).data(Qt.UserRole), sz))), type=Qt.QueuedConnection)
def css_data(container, book_locale, result_data, *args):
import tinycss
from tinycss.css21 import RuleSet, ImportRule
def css_rules(file_name, rules, sourceline=0):
ans = []
for rule in rules:
if isinstance(rule, RuleSet):
selector = rule.selector.as_css()
ans.append(CSSRule(selector, RuleLocation(file_name, sourceline + rule.line, rule.column)))
elif isinstance(rule, ImportRule):
import_name = safe_href_to_name(container, rule.uri, file_name)
if import_name and container.exists(import_name):
ans.append(import_name)
elif getattr(rule, 'rules', False):
ans.extend(css_rules(file_name, rule.rules, sourceline))
return ans
parser = tinycss.make_full_parser()
importable_sheets = {}
html_sheets = {}
spine_names = {name for name, is_linear in container.spine_names}
style_path, link_path = XPath('//h:style'), XPath('//h:link/@href')
for name, mt in container.mime_map.iteritems():
if mt in OEB_STYLES:
importable_sheets[name] = css_rules(name, parser.parse_stylesheet(container.raw_data(name)).rules)
elif mt in OEB_DOCS and name in spine_names:
html_sheets[name] = []
for style in style_path(container.parsed(name)):
if style.get('type', 'text/css') == 'text/css' and style.text:
html_sheets[name].append(
css_rules(name, parser.parse_stylesheet(force_unicode(style.text, 'utf-8')).rules, style.sourceline - 1))
rule_map = defaultdict(lambda : defaultdict(list))
def rules_in_sheet(sheet):
for rule in sheet:
if isinstance(rule, CSSRule):
yield rule
else: # @import rule
isheet = importable_sheets.get(rule)
if isheet is not None:
for irule in rules_in_sheet(isheet):
yield irule
def sheets_for_html(name, root):
for href in link_path(root):
tname = safe_href_to_name(container, href, name)
sheet = importable_sheets.get(tname)
if sheet is not None:
yield sheet
tt_cache = {}
def tag_text(elem):
ans = tt_cache.get(elem)
if ans is None:
tag = elem.tag.rpartition('}')[-1]
if elem.attrib:
attribs = ' '.join('%s="%s"' % (k, prepare_string_for_xml(elem.get(k, ''), True)) for k in elem.keys())
return '<%s %s>' % (tag, attribs)
ans = tt_cache[elem] = '<%s>' % tag
def matches_for_selector(selector, select, class_map, rule):
lsel = selector.lower()
try:
matches = tuple(select(selector))
except SelectorError:
return ()
for elem in matches:
for cls in elem.get('class', '').split():
if '.' + cls.lower() in lsel:
class_map[cls][elem].append(rule)
return (MatchLocation(tag_text(elem), elem.sourceline) for elem in matches)
class_map = defaultdict(lambda : defaultdict(list))
for name, inline_sheets in html_sheets.iteritems():
root = container.parsed(name)
cmap = defaultdict(lambda : defaultdict(list))
for elem in root.xpath('//*[@class]'):
for cls in elem.get('class', '').split():
cmap[cls][elem] = []
select = Select(root, ignore_inappropriate_pseudo_classes=True)
for sheet in chain(sheets_for_html(name, root), inline_sheets):
for rule in rules_in_sheet(sheet):
rule_map[rule][name].extend(matches_for_selector(rule.selector, select, cmap, rule))
for cls, elem_map in cmap.iteritems():
class_elements = class_map[cls][name]
for elem, usage in elem_map.iteritems():
class_elements.append(
ClassElement(name, elem.sourceline, elem.get('class'), tag_text(elem), tuple(usage)))
result_data['classes'] = ans = []
for cls, name_map in class_map.iteritems():
la = tuple(ClassFileMatch(name, tuple(class_elements), numeric_sort_key(name)) for name, class_elements in name_map.iteritems() if class_elements)
num_of_matches = sum(sum(len(ce.matched_rules) for ce in cfm.class_elements) for cfm in la)
ans.append(ClassEntry(cls, num_of_matches, la, numeric_sort_key(cls)))
ans = []
for rule, loc_map in rule_map.iteritems():
la = tuple(CSSFileMatch(name, tuple(locations), numeric_sort_key(name)) for name, locations in loc_map.iteritems() if locations)
count = sum(len(fm.locations) for fm in la)
ans.append(CSSEntry(rule, count, la, numeric_sort_key(rule.selector)))
return ans
def sort_key(name):
return nmap.get(name, len(nmap)), numeric_sort_key(name)
def text_sort_key(x):
return numeric_sort_key(str(x or ''))
def key(account_key):
return numeric_sort_key(account_key)
def key(account_key):
return numeric_sort_key(
as_unicode(self.accounts[account_key][0]) or '')
def key(account_key):
return numeric_sort_key(self.aliases.get(account_key) or '')
def css_data(container, book_locale, result_data, *args):
import tinycss
from tinycss.css21 import RuleSet, ImportRule
def css_rules(file_name, rules, sourceline=0):
ans = []
for rule in rules:
if isinstance(rule, RuleSet):
selector = rule.selector.as_css()
ans.append(
CSSRule(
selector,
RuleLocation(file_name, sourceline + rule.line,
rule.column)))
elif isinstance(rule, ImportRule):
import_name = safe_href_to_name(container, rule.uri, file_name)
if import_name and container.exists(import_name):
ans.append(import_name)
elif getattr(rule, 'rules', False):
ans.extend(css_rules(file_name, rule.rules, sourceline))
return ans
parser = tinycss.make_full_parser()
importable_sheets = {}
html_sheets = {}
spine_names = {name for name, is_linear in container.spine_names}
style_path, link_path = XPath('//h:style'), XPath('//h:link/@href')
for name, mt in iteritems(container.mime_map):
if mt in OEB_STYLES:
importable_sheets[name] = css_rules(
name,
parser.parse_stylesheet(container.raw_data(name)).rules)
elif mt in OEB_DOCS and name in spine_names:
html_sheets[name] = []
for style in style_path(container.parsed(name)):
if style.get('type', 'text/css') == 'text/css' and style.text:
html_sheets[name].append(
css_rules(
name,
parser.parse_stylesheet(
force_unicode(style.text, 'utf-8')).rules,
style.sourceline - 1))
rule_map = defaultdict(lambda: defaultdict(list))
def rules_in_sheet(sheet):
for rule in sheet:
if isinstance(rule, CSSRule):
yield rule
else: # @import rule
isheet = importable_sheets.get(rule)
if isheet is not None:
for irule in rules_in_sheet(isheet):
yield irule
def sheets_for_html(name, root):
for href in link_path(root):
tname = safe_href_to_name(container, href, name)
sheet = importable_sheets.get(tname)
if sheet is not None:
yield sheet
tt_cache = {}
def tag_text(elem):
ans = tt_cache.get(elem)
if ans is None:
tag = elem.tag.rpartition('}')[-1]
if elem.attrib:
attribs = ' '.join(
'%s="%s"' %
(k, prepare_string_for_xml(elem.get(k, ''), True))
for k in elem.keys())
return '<%s %s>' % (tag, attribs)
ans = tt_cache[elem] = '<%s>' % tag
def matches_for_selector(selector, select, class_map, rule):
lsel = selector.lower()
try:
matches = tuple(select(selector))
except SelectorError:
return ()
for elem in matches:
for cls in elem.get('class', '').split():
if '.' + cls.lower() in lsel:
class_map[cls][elem].append(rule)
return (MatchLocation(tag_text(elem), elem.sourceline)
for elem in matches)
class_map = defaultdict(lambda: defaultdict(list))
for name, inline_sheets in iteritems(html_sheets):
root = container.parsed(name)
cmap = defaultdict(lambda: defaultdict(list))
for elem in root.xpath('//*[@class]'):
for cls in elem.get('class', '').split():
cmap[cls][elem] = []
select = Select(root, ignore_inappropriate_pseudo_classes=True)
for sheet in chain(sheets_for_html(name, root), inline_sheets):
for rule in rules_in_sheet(sheet):
rule_map[rule][name].extend(
matches_for_selector(rule.selector, select, cmap, rule))
for cls, elem_map in iteritems(cmap):
class_elements = class_map[cls][name]
for elem, usage in iteritems(elem_map):
class_elements.append(
ClassElement(name, elem.sourceline, elem.get('class'),
tag_text(elem), tuple(usage)))
result_data['classes'] = ans = []
for cls, name_map in iteritems(class_map):
la = tuple(
ClassFileMatch(name, tuple(class_elements), numeric_sort_key(name))
for name, class_elements in iteritems(name_map) if class_elements)
num_of_matches = sum(
sum(len(ce.matched_rules) for ce in cfm.class_elements)
for cfm in la)
ans.append(ClassEntry(cls, num_of_matches, la, numeric_sort_key(cls)))
ans = []
for rule, loc_map in iteritems(rule_map):
la = tuple(
CSSFileMatch(name, tuple(locations), numeric_sort_key(name))
for name, locations in iteritems(loc_map) if locations)
count = sum(len(fm.locations) for fm in la)
ans.append(CSSEntry(rule, count, la, numeric_sort_key(rule.selector)))
return ans
def key(account_key):
return numeric_sort_key(self.tags.get(account_key) or '')
def serialize(self, styles):
for style in sorted(self.block_styles, key=lambda s:(s is not self.normal_block_style, numeric_sort_key(s.id))):
style.serialize(styles, self.normal_block_style)
for style in sorted(self.text_styles, key=lambda s:(s is not self.normal_text_style, numeric_sort_key(s.id))):
style.serialize(styles, self.normal_text_style)
def key(account_key):
return numeric_sort_key(self.aliases.get(account_key) or '')
def key(account_key):
return numeric_sort_key(self.subjects.get(account_key) or '')
def sort_key(filename):
bn, ext = filename.rpartition('.')[::2]
if not bn and ext:
bn, ext = ext, bn
return (numeric_sort_key(bn), numeric_sort_key(ext))
def sort_key(l):
return (nmap.get(l.name, len(nmap)), numeric_sort_key(l.name), l.line_number)
def key(account_key):
return numeric_sort_key(type(u'')(self.accounts[account_key][0]) or '')
def sort_key(name):
return nmap.get(name, len(nmap)), numeric_sort_key(name)
def key(account_key):
return numeric_sort_key(self.subjects.get(account_key) or '')
def sort_key(l):
return (nmap.get(l.name,
len(nmap)), numeric_sort_key(l.name), l.line_number)
def sort_key(filename):
bn, ext = filename.rpartition('.')[::2]
if not bn and ext:
bn, ext = ext, bn
return (numeric_sort_key(bn), numeric_sort_key(ext))
def key(account_key):
return numeric_sort_key(account_key)
def text_sort_key(x):
return numeric_sort_key(unicode_type(x or ''))
def flatten_spine(self):
names = defaultdict(int)
styles, pseudo_styles = {}, defaultdict(dict)
for item in self.items:
html = item.data
stylizer = self.stylizers[item]
if self.specializer is not None:
self.specializer(item, stylizer)
body = html.find(XHTML('body'))
fsize = self.context.dest.fbase
self.flatten_node(body, stylizer, names, styles, pseudo_styles, fsize, item.id)
items = sorted(((key, val) for (val, key) in iteritems(styles)), key=lambda x:numeric_sort_key(x[0]))
# :hover must come after link and :active must come after :hover
psels = sorted(pseudo_styles, key=lambda x :
{'hover':1, 'active':2}.get(x, 0))
for psel in psels:
styles = pseudo_styles[psel]
if not styles:
continue
x = sorted(((k+':'+psel, v) for v, k in iteritems(styles)))
items.extend(x)
css = ''.join(".%s {\n%s;\n}\n\n" % (key, val) for key, val in items)
href = self.replace_css(css)
global_css = self.collect_global_css()
for item in self.items:
stylizer = self.stylizers[item]
self.flatten_head(item, href, global_css[item])
def key(account_key):
return numeric_sort_key(self.accounts[account_key][0] or '')
def key(account_key):
return numeric_sort_key(self.accounts[account_key][0] or '')
def text_sort_key(x):
return numeric_sort_key(unicode(x or ''))
def key(account_key):
return numeric_sort_key(
type(u'')(self.accounts[account_key][0]) or '')
def css_data(container, book_locale):
import tinycss
from tinycss.css21 import RuleSet, ImportRule
def css_rules(file_name, rules, sourceline=0):
ans = []
for rule in rules:
if isinstance(rule, RuleSet):
selector = rule.selector.as_css()
ans.append(CSSRule(selector, RuleLocation(file_name, sourceline + rule.line, rule.column)))
elif isinstance(rule, ImportRule):
import_name = safe_href_to_name(container, rule.uri, file_name)
if import_name and container.exists(import_name):
ans.append(import_name)
elif getattr(rule, 'rules', False):
ans.extend(css_rules(file_name, rule.rules, sourceline))
return ans
parser = tinycss.make_full_parser()
importable_sheets = {}
html_sheets = {}
spine_names = {name for name, is_linear in container.spine_names}
style_path, link_path = XPath('//h:style'), XPath('//h:link/@href')
for name, mt in container.mime_map.iteritems():
if mt in OEB_STYLES:
importable_sheets[name] = css_rules(name, parser.parse_stylesheet(container.raw_data(name)).rules)
elif mt in OEB_DOCS and name in spine_names:
html_sheets[name] = []
for style in style_path(container.parsed(name)):
if style.get('type', 'text/css') == 'text/css' and style.text:
html_sheets[name].append(
css_rules(name, parser.parse_stylesheet(force_unicode(style.text, 'utf-8')).rules, style.sourceline - 1))
rule_map = defaultdict(lambda : defaultdict(list))
pseudo_pat = re.compile(PSEUDO_PAT, re.I)
cache = {}
def rules_in_sheet(sheet):
for rule in sheet:
if isinstance(rule, CSSRule):
yield rule
else: # @import rule
isheet = importable_sheets.get(rule)
if isheet is not None:
for irule in rules_in_sheet(isheet):
yield irule
def sheets_for_html(name, root):
for href in link_path(root):
tname = safe_href_to_name(container, href, name)
sheet = importable_sheets.get(tname)
if sheet is not None:
yield sheet
def tag_text(elem):
tag = elem.tag.rpartition('}')[-1]
if elem.attrib:
attribs = ' '.join('%s="%s"' % (k, prepare_string_for_xml(elem.get(k, ''), True)) for k in elem.keys())
return '<%s %s>' % (tag, attribs)
return '<%s>' % tag
def matches_for_selector(selector, root):
selector = pseudo_pat.sub('', selector)
selector = MIN_SPACE_RE.sub(r'\1', selector)
try:
xp = cache[(True, selector)]
except KeyError:
xp = cache[(True, selector)] = build_selector(selector)
try:
matches = xp(root)
except Exception:
return ()
if not matches:
try:
xp = cache[(False, selector)]
except KeyError:
xp = cache[(False, selector)] = build_selector(selector, case_sensitive=False)
try:
matches = xp(root)
except Exception:
return ()
return (MatchLocation(tag_text(elem), elem.sourceline) for elem in matches)
for name, inline_sheets in html_sheets.iteritems():
root = container.parsed(name)
for sheet in chain(sheets_for_html(name, root), inline_sheets):
for rule in rules_in_sheet(sheet):
rule_map[rule][name].extend(matches_for_selector(rule.selector, root))
ans = []
for rule, loc_map in rule_map.iteritems():
la = tuple(CSSFileMatch(name, tuple(locations), numeric_sort_key(name)) for name, locations in loc_map.iteritems() if locations)
count = sum(len(fm.locations) for fm in la)
ans.append(CSSEntry(rule, count, la, numeric_sort_key(rule.selector)))
return ans
def key(account_key):
return numeric_sort_key(self.tags.get(account_key) or '')
def key(account_key):
return numeric_sort_key(as_unicode(self.accounts[account_key][0]) or '')
