X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/295a1350b069ee750d2f7430e779e5ac6f23a673..3a0c83394d5783715fab2be29fa1a9cfc3574e28:/src/librarian/elements/base.py diff --git a/src/librarian/elements/base.py b/src/librarian/elements/base.py index 646067e..863436c 100644 --- a/src/librarian/elements/base.py +++ b/src/librarian/elements/base.py @@ -1,3 +1,6 @@ +# This file is part of Librarian, licensed under GNU Affero GPLv3 or later. +# Copyright © Fundacja Wolne Lektury. See NOTICE for more information. +# import copy import re from lxml import etree @@ -39,11 +42,11 @@ class WLElement(etree.ElementBase): STRIP = False text_substitutions = [ - (u'---', u'—'), - (u'--', u'–'), - #(u'...', u'…'), # Temporary turnoff for epub - (u',,', u'„'), - (u'"', u'”'), + ('---', '—'), + ('--', '–'), + #('...', '…'), # Temporary turnoff for epub + (',,', '„'), + ('"', '”'), ('\ufeff', ''), ("'", "\u2019"), # This was enabled for epub. @@ -113,11 +116,11 @@ class WLElement(etree.ElementBase): newt = '' wlist = re.compile(r'\w+|[^\w]', re.UNICODE).findall(text) for w in wlist: - newt += builder.hyphenator.inserted(w, u'\u00AD') + newt += builder.hyphenator.inserted(w, '\u00AD') text = newt if builder.orphans: - text = re.sub(r'(?<=\s\w)\s+', u'\u00A0', text) + text = re.sub(r'(?<=\s\w)\s+', '\u00A0', text) return text @@ -278,8 +281,6 @@ class WLElement(etree.ElementBase): # do we dare go up? parent = self.getparent() if parent is not None and parent.CAN_HAVE_TEXT: - print(etree.tostring(self, encoding='unicode')) - assert False words, parsnip = parent.snip(words, before=self) return words, parsnip[:-1] + snippet + parsnip[-1:]