jobextra/itstool/0001-Fix-the-crash-from-912...

91 lines
3.9 KiB
Diff

From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Tanguy Ortolo <tanguy+debian@ortolo.eu>
Date: Fri, 7 Dec 2018 00:00:00 +0000
Subject: [PATCH] Fix the crash from #912099
ITS Tool 2.0.4 crashes when building some documentation, as reported in #912099.
This comes from translations with invalid XML markup, which ITS Tool fails to
merge (which is not abnormal), and to report these issues, needlessly encodes
the original msgstr from unicode to bytes, causing it to be recoded using the
default ascii codec, which fails when the msgstr contains anything out of ascii.
This patch removes the useless decoding, avoiding the failing subsequent
recoding. It also explicitly encodes the output strings to be able to print them
in all cases, even when the output encoding cannot be detected.
Bug: https://github.com/itstool/itstool/issues/25
Bug-Debian: https://bugs.debian.org/912099
Forwarded: https://github.com/itstool/itstool/issues/25
---
itstool.in | 21 +++++++++++++++++----
1 file changed, 17 insertions(+), 4 deletions(-)
diff --git a/itstool.in b/itstool.in
index c21ad4bfeb86..f34673581c88 100755
--- a/itstool.in
+++ b/itstool.in
@@ -44,9 +44,22 @@ if PY3:
else:
return str(s)
ustr_type = str
+ def pr_str(s):
+ """Return a string that can be safely print()ed"""
+ # Since print works on both bytes and unicode, just return the argument
+ return s
else:
string_types = basestring,
ustr = ustr_type = unicode
+ def pr_str(s):
+ """Return a string that can be safely print()ed"""
+ if isinstance(s, str):
+ # Since print works on str, just return the argument
+ return s
+ else:
+ # print may not work on unicode if the output encoding cannot be
+ # detected, so just encode with UTF-8
+ return unicode.encode(s, 'utf-8')
NS_ITS = 'http://www.w3.org/2005/11/its'
NS_ITST = 'http://itstool.org/extensions/'
@@ -1077,36 +1090,36 @@ class Document (object):
if strict:
raise
else:
- sys.stderr.write('Warning: Could not merge %stranslation for msgid:\n%s\n' % (
+ sys.stderr.write(pr_str('Warning: Could not merge %stranslation for msgid:\n%s\n' % (
(lang + ' ') if lang is not None else '',
- msgstr.encode('utf-8')))
+ msgstr)))
self._xml_err = ''
return node
def scan_node(node):
children = [child for child in xml_child_iter(node)]
for child in children:
if child.type != 'element':
continue
if child.ns() is not None and child.ns().content == NS_BLANK:
ph_node = msg.get_placeholder(child.name).node
if self.has_child_elements(ph_node):
self.merge_translations(translations, None, ph_node, strict=strict)
newnode = ph_node.copyNode(1)
newnode.setTreeDoc(self._doc)
child.replaceNode(newnode)
else:
repl = self.get_translated(ph_node, translations, strict=strict, lang=lang)
child.replaceNode(repl)
scan_node(child)
try:
scan_node(trnode)
except:
if strict:
raise
else:
- sys.stderr.write('Warning: Could not merge %stranslation for msgid:\n%s\n' % (
+ sys.stderr.write(pr_str('Warning: Could not merge %stranslation for msgid:\n%s\n' % (
(lang + ' ') if lang is not None else '',
- msgstr.encode('utf-8')))
+ msgstr)))
self._xml_err = ''
ctxt.doc().freeDoc()
return node