gryf/ebook-converter
1
0
Fork 0
mirror of https://github.com/gryf/ebook-converter.git synced 2026-04-28 09:34:05 +02:00
Code Issues Projects Releases Wiki Activity

Get rid of icu string functions in favor of native ones.

Browse Source
This commit is contained in:
gryf
2020-05-03 20:19:11 +02:00
parent 212cb56d42
commit da010d7841
10 changed files with 138 additions and 185 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ebook_converter/ebooks/oeb/base.py
+2 -1
View File
@@ -6,6 +6,7 @@ from collections import defaultdict
from itertools import count
from operator import attrgetter
import urllib.parse
import string
from lxml import etree, html
from ebook_converter import force_unicode
@@ -763,7 +764,7 @@ class Metadata(object):
key = barename(key)
attrib[key] = prefixname(value, nsrmap)
if namespace(self.term) == DC11_NS:
name = DC(icu_title(barename(self.term)))
name = DC(string.capwords(barename(self.term)))
elem = element(dcmeta, name, attrib=attrib)
elem.text = self.value
else:
ebook_converter/ebooks/oeb/polish/css.py
+9 -5
View File
@@ -93,10 +93,14 @@ def remove_unused_css(container, report=None, remove_unused_classes=False, merge
if num:
container.dirty(name)
num_merged += num
import_map = {name:get_imported_sheets(name, container, sheets) for name in sheets}
import_map = {name: get_imported_sheets(name, container, sheets)
for name in sheets}
if remove_unused_classes:
class_map = {name:{icu_lower(x) for x in classes_in_rule_list(sheet.cssRules)} for name, sheet in sheets.items()}
style_rules = {name:tuple(sheet.cssRules.rulesOfType(CSSRule.STYLE_RULE)) for name, sheet in sheets.items()}
class_map = {name: {x.lower() for x in
classes_in_rule_list(sheet.cssRules)}
for name, sheet in sheets.items()}
style_rules = {name: tuple(sheet.cssRules.rulesOfType(CSSRule.STYLE_RULE))
for name, sheet in sheets.items()}
num_of_removed_rules = num_of_removed_classes = 0
@@ -115,7 +119,7 @@ def remove_unused_css(container, report=None, remove_unused_classes=False, merge
num_merged += num
container.dirty(name)
if remove_unused_classes:
used_classes |= {icu_lower(x) for x in classes_in_rule_list(sheet.cssRules)}
used_classes |= {x.lower() for x in classes_in_rule_list(sheet.cssRules)}
imports = get_imported_sheets(name, container, sheets, sheet=sheet)
for imported_sheet in imports:
style_rules[imported_sheet] = tuple(filter_used_rules(style_rules[imported_sheet], container.log, select))
@@ -147,7 +151,7 @@ def remove_unused_css(container, report=None, remove_unused_classes=False, merge
for elem in root.xpath('//*[@class]'):
original_classes, classes = elem.get('class', '').split(), []
for x in original_classes:
if icu_lower(x) in used_classes:
if x.lower() in used_classes:
classes.append(x)
if len(classes) != len(original_classes):
if classes:
ebook_converter/ebooks/oeb/polish/utils.py
+32 -17
View File
@@ -1,5 +1,6 @@
import re, os
from bisect import bisect
import bisect
import os
import re
from ebook_converter import guess_type as _guess_type, replace_entities
@@ -8,6 +9,10 @@ __license__ = 'GPL v3'
__copyright__ = '2013, Kovid Goyal <kovid at kovidgoyal.net>'
def _upper(string):
return string.upper()
def guess_type(x):
return _guess_type(x)[0] or 'application/octet-stream'
@@ -23,7 +28,8 @@ def setup_css_parser_serialization(tab_width=2):
def actual_case_for_name(container, name):
from ebook_converter.utils.filenames import samefile
if not container.exists(name):
raise ValueError('Cannot get actual case for %s as it does not exist' % name)
raise ValueError('Cannot get actual case for %s as it does not '
'exist' % name)
parts = name.split('/')
base = ''
ans = []
@@ -55,9 +61,12 @@ def corrected_case_for_name(container, name):
correctx = x
else:
try:
candidates = {q for q in os.listdir(os.path.dirname(container.name_to_abspath(base)))}
dirname = os.path.dirname(container.name_to_abspath(base))
candidates = {q for q in os.listdir(dirname)}
except EnvironmentError:
return None # one of the non-terminal components of name is a file instead of a directory
# one of the non-terminal components of name is a file instead
# of a directory
return None
for q in candidates:
if q.lower() == x.lower():
correctx = q
@@ -75,7 +84,7 @@ class PositionFinder(object):
self.new_lines = tuple(m.start() + 1 for m in re.finditer(pat, raw))
def __call__(self, pos):
lnum = bisect(self.new_lines, pos)
lnum = bisect.bisect(self.new_lines, pos)
try:
offset = abs(pos - self.new_lines[lnum - 1])
except IndexError:
@@ -94,7 +103,7 @@ class CommentFinder(object):
def __call__(self, offset):
if not self.starts:
return False
q = bisect(self.starts, offset) - 1
q = bisect.bisect(self.starts, offset) - 1
return q >= 0 and self.starts[q] <= offset <= self.ends[q]
@@ -182,13 +191,16 @@ def handle_entities(text, func):
return func(replace_entities(text))
def apply_func_to_match_groups(match, func=icu_upper, handle_entities=handle_entities):
'''Apply the specified function to individual groups in the match object (the result of re.search() or
the whole match if no groups were defined. Returns the replaced string.'''
def apply_func_to_match_groups(match, func=_upper,
handle_entities=handle_entities):
"""
Apply the specified function to individual groups in the match object (the
result of re.search() or
the whole match if no groups were defined. Returns the replaced string.
"""
found_groups = False
i = 0
parts, pos = [], match.start()
f = lambda text:handle_entities(text, func)
while True:
i += 1
try:
@@ -198,19 +210,22 @@ def apply_func_to_match_groups(match, func=icu_upper, handle_entities=handle_ent
found_groups = True
if start > -1:
parts.append(match.string[pos:start])
parts.append(f(match.string[start:end]))
parts.append(handle_entities(match.string[start:end], func))
pos = end
if not found_groups:
return f(match.group())
return handle_entities(match.group(), func)
parts.append(match.string[pos:match.end()])
return ''.join(parts)
def apply_func_to_html_text(match, func=icu_upper, handle_entities=handle_entities):
''' Apply the specified function only to text between HTML tag definitions. '''
f = lambda text:handle_entities(text, func)
def apply_func_to_html_text(match, func=_upper,
handle_entities=handle_entities):
"""
Apply the specified function only to text between HTML tag definitions.
"""
parts = re.split(r'(<[^>]+>)', match.group())
parts = (x if x.startswith('<') else f(x) for x in parts)
parts = (x if x.startswith('<') else handle_entities(x, func)
for x in parts)
return ''.join(parts)
ebook_converter/ebooks/oeb/transforms/manglecase.py
+8 -4
View File
@@ -1,7 +1,10 @@
"""
CSS case-mangling transform.
"""
import string
from lxml import etree
from ebook_converter.ebooks.oeb.base import XHTML, XHTML_NS
from ebook_converter.ebooks.oeb.base import CSS_MIME
from ebook_converter.ebooks.oeb.base import namespace
@@ -46,16 +49,17 @@ class CaseMangler(object):
relhref = item.relhref(href)
etree.SubElement(html.find(XHTML('head')), XHTML('link'),
rel='stylesheet', href=relhref, type=CSS_MIME)
stylizer = Stylizer(html, item.href, self.oeb, self.opts, self.profile)
stylizer = Stylizer(html, item.href, self.oeb, self.opts,
self.profile)
self.mangle_elem(html.find(XHTML('body')), stylizer)
def text_transform(self, transform, text):
if transform == 'capitalize':
return icu_title(text)
return string.capwords(text)
elif transform == 'uppercase':
return icu_upper(text)
return text.upper()
elif transform == 'lowercase':
return icu_lower(text)
return text.lower()
return text
def split_text(self, text):