Move force_uniceode to utils package

ebook_converter/ebooks/oeb/base.py

@@ -17,12 +17,12 @@ from lxml import etree
 from lxml import html
 
 from ebook_converter import constants as const
-from ebook_converter import force_unicode
 from ebook_converter.constants_old import filesystem_encoding, __version__
 from ebook_converter.ebooks.chardet import xml_to_unicode
 from ebook_converter.ebooks.conversion.preprocess import CSSPreProcessor
 from ebook_converter.ebooks.oeb import parse_utils
 from ebook_converter.utils.cleantext import clean_xml_chars
+from ebook_converter.utils import encoding as uenc
 from ebook_converter.utils.short_uuid import uuid4
 
 
@@ -1074,7 +1074,7 @@ class Manifest(object):
         def sort_key(self):
             href = self.href
             if isinstance(href, bytes):
-                href = force_unicode(href)
+                href = uenc.force_unicode(href)
 
             if isinstance(self.spine_position, numbers.Number):
                 sp = self.spine_position

ebook_converter/ebooks/oeb/parse_utils.py

@@ -5,9 +5,10 @@ from lxml import etree
 from lxml import html
 
 from ebook_converter import constants as const
-from ebook_converter import xml_replace_entities, force_unicode
+from ebook_converter import xml_replace_entities
 from ebook_converter.constants_old import filesystem_encoding
 from ebook_converter.ebooks.chardet import xml_to_unicode, strip_encoding_declarations
+from ebook_converter.utils import encoding as uenc
 
 
 RECOVER_PARSER = etree.XMLParser(recover=True, no_network=True,
@@ -159,7 +160,7 @@ def parse_html(data, log=None, decoder=None, preprocessor=None,
     if log is None:
         log = LOG
 
-    filename = force_unicode(filename, enc=filesystem_encoding)
+    filename = uenc.force_unicode(filename, enc=filesystem_encoding)
 
     if not isinstance(data, str):
         if decoder is not None:

ebook_converter/ebooks/oeb/polish/css.py

@@ -3,12 +3,12 @@ import functools
 
 from css_parser.css import CSSRule, CSSStyleDeclaration
 
-from ebook_converter import force_unicode
 from ebook_converter.css_selectors import parse, SelectorSyntaxError
 from ebook_converter.ebooks.oeb import base
 from ebook_converter.ebooks.oeb.polish import pretty
 from ebook_converter.utils.icu import numeric_sort_key
 from ebook_converter.css_selectors import Select, SelectorError
+from ebook_converter.utils import encoding as uenc
 
 
 def filter_used_rules(rules, log, select):
@@ -137,7 +137,7 @@ def remove_unused_css(container, report=None, remove_unused_classes=False,
                 if unused_rules:
                     num_of_removed_rules += len(unused_rules)
                     [sheet.cssRules.remove(r) for r in unused_rules]
-                    style.text = force_unicode(sheet.cssText, 'utf-8')
+                    style.text = uenc.force_unicode(sheet.cssText, 'utf-8')
                     pretty.pretty_script_or_style(container, style)
                     container.dirty(name)
 
@@ -241,7 +241,7 @@ def transform_inline_styles(container, name, transform_sheet, transform_style):
             sheet = container.parse_css(style.text)
             if transform_sheet(sheet):
                 changed = True
-                style.text = force_unicode(sheet.cssText, 'utf-8')
+                style.text = uenc.force_unicode(sheet.cssText, 'utf-8')
                 pretty.pretty_script_or_style(container, style)
     for elem in root.xpath('//*[@style]'):
         text = elem.get('style', None)
@@ -253,8 +253,9 @@ def transform_inline_styles(container, name, transform_sheet, transform_style):
                     del elem.attrib['style']
                 else:
                     elem.set('style',
-                             force_unicode(style.getCssText(separator=' '),
-                                           'utf-8'))
+                             uenc.force_unicode(style
+                                                .getCssText(separator=' '),
+                                                'utf-8'))
     return changed
 
 

ebook_converter/ebooks/oeb/polish/pretty.py

@@ -1,10 +1,10 @@
 import textwrap
 
 from ebook_converter import constants as const
-from ebook_converter import force_unicode
 from ebook_converter.ebooks.oeb import parse_utils
 from ebook_converter.ebooks.oeb import base
 from ebook_converter.ebooks.oeb.polish.utils import guess_type
+from ebook_converter.utils import encoding as uenc
 from ebook_converter.utils.icu import sort_key
 
 
@@ -157,8 +157,9 @@ def pretty_script_or_style(container, child):
     if child.text:
         indent = indent_for_tag(child)
         if child.tag.endswith('style'):
-            child.text = force_unicode(pretty_css(container, '', child.text),
-                                       'utf-8')
+            child.text = uenc.force_unicode(pretty_css(container, '',
+                                                       child.text),
+                                            'utf-8')
         child.text = textwrap.dedent(child.text)
         child.text = '\n' + '\n'.join([(indent + x) if x else ''
                                        for x in child.text.splitlines()])

ebook_converter/ebooks/oeb/stylizer.py

@@ -12,12 +12,12 @@ from css_parser import (profile as cssprofiles, parseString, parseStyle, log as
         css_parser_log, CSSParser, profiles, replaceUrls)
 
 from ebook_converter import constants as const
-from ebook_converter import force_unicode
 from ebook_converter.ebooks import unit_convert
 from ebook_converter.ebooks.oeb import base
 from ebook_converter.ebooks.oeb.normalize_css import DEFAULTS, normalizers
 from ebook_converter.css_selectors import Select, SelectorError, INAPPROPRIATE_PSEUDO_CLASSES
 from ebook_converter.tinycss.media3 import CSSMedia3Parser
+from ebook_converter.utils import encoding as uenc
 
 
 css_parser_log.setLevel(logging.WARN)
@@ -223,10 +223,10 @@ class Stylizer(object):
                 for x in elem:
                     t = getattr(x, 'text', None)
                     if t:
-                        text += '\n\n' + force_unicode(t, 'utf-8')
+                        text += '\n\n' + uenc.force_unicode(t, 'utf-8')
                     t = getattr(x, 'tail', None)
                     if t:
-                        text += '\n\n' + force_unicode(t, 'utf-8')
+                        text += '\n\n' + uenc.force_unicode(t, 'utf-8')
                 if text:
                     text = oeb.css_preprocessor(text)
                     # We handle @import rules separately

ebook_converter/ebooks/oeb/transforms/split.py

@@ -14,12 +14,12 @@ from lxml.etree import XPath as _XPath
 from lxml import etree
 
 from ebook_converter import constants as const
-from ebook_converter import force_unicode
 from ebook_converter.ebooks.epub import rules
 from ebook_converter.ebooks.oeb import base
 from ebook_converter.ebooks.oeb.polish.split import do_split
 from ebook_converter.polyglot.urllib import unquote
 from ebook_converter.css_selectors import Select, SelectorError
+from ebook_converter.utils import encoding as uenc
 
 
 __license__ = 'GPL v3'
@@ -86,10 +86,12 @@ class Split(object):
             stylesheets = [x.data for x in self.oeb.manifest if x.media_type in
                            base.OEB_STYLES]
             for rule in rules(stylesheets):
-                before = force_unicode(getattr(rule.style.getPropertyCSSValue(
-                    'page-break-before'), 'cssText', '').strip().lower())
-                after = force_unicode(getattr(rule.style.getPropertyCSSValue(
-                    'page-break-after'), 'cssText', '').strip().lower())
+                before = uenc.force_unicode(
+                    getattr(rule.style.getPropertyCSSValue(
+                        'page-break-before'), 'cssText', '').strip().lower())
+                after = uenc.force_unicode(
+                    getattr(rule.style.getPropertyCSSValue(
+                        'page-break-after'), 'cssText', '').strip().lower())
                 try:
                     if before and before not in {'avoid', 'auto', 'inherit'}:
                         self.page_break_selectors.add((rule.selectorText,