From e014f4f9442f9a5f57ef3af8038f69b6608a482a Mon Sep 17 00:00:00 2001
From: Todd Zullinger <tmz@pobox.com>
Date: Sat, 16 Jun 2018 00:52:17 -0400
Subject: [PATCH] Add some a2x decoding fixes from upstream PR#5

PR: https://github.com/asciidoc/asciidoc-py3/pull/5

Commits:
6469317 Remove unnecessary decode in a2x (Matthew Peveler)
684913e Fix decoding of file that specifies encoding in header tag in a2x (Matthew Peveler)
8369a97 re-add --nonet option (Matthew Peveler)

These changes fix failures in building cgit's documentation (and
hopefully other similar decoding failures).
---
 asciidoc-python3-a2x-decode-fix.patch | 40 +++++++++++++++++++++++++++
 asciidoc.spec                         | 14 +++++++++-
 2 files changed, 53 insertions(+), 1 deletion(-)
 create mode 100644 asciidoc-python3-a2x-decode-fix.patch

diff --git a/asciidoc-python3-a2x-decode-fix.patch b/asciidoc-python3-a2x-decode-fix.patch
new file mode 100644
index 0000000..507fa87
--- /dev/null
+++ b/asciidoc-python3-a2x-decode-fix.patch
@@ -0,0 +1,40 @@
+Taken from upstream PR#5 (https://github.com/asciidoc/asciidoc-py3/pull/5)
+
+6469317 Remove unnecessary decode in a2x (Matthew Peveler)
+684913e Fix decoding of file that specifies encoding in header tag in a2x (Matthew Peveler)
+8369a97 re-add --nonet option (Matthew Peveler)
+
+diff --git c/a2x.py w/a2x.py
+index 55eb57e..c015079 100755
+--- c/a2x.py
++++ w/a2x.py
+@@ -254,15 +254,11 @@ def find_resources(files, tagname, attrname, filter=None):
+         if OPTIONS.dry_run:
+             continue
+         parser = FindResources()
+-        # HTMLParser has problems with non-ASCII strings.
+-        # See http://bugs.python.org/issue3932
+-        contents = read_file(filename)
+-        mo = re.search(r'\A<\?xml.* encoding="(.*?)"', contents)
+-        if mo:
+-            encoding = mo.group(1)
+-            parser.feed(contents.decode(encoding))
+-        else:
+-            parser.feed(contents)
++        with open(filename, 'rb') as open_file:
++            contents = open_file.read()
++        mo = re.search(b'\A<\?xml.* encoding="(.*?)"', contents)
++        contents = contents.decode(mo.group(1).decode('utf-8') if mo else 'utf-8')
++        parser.feed(contents)
+         parser.close()
+     result = list(set(result))   # Drop duplicate values.
+     result.sort()
+@@ -337,7 +333,7 @@ def get_source_options(asciidoc_file):
+     result = []
+     if os.path.isfile(asciidoc_file):
+         options = ''
+-        with open(asciidoc_file) as f:
++        with open(asciidoc_file, encoding='utf-8') as f:
+             for line in f:
+                 mo = re.search(r'^//\s*a2x:', line)
+                 if mo:
diff --git a/asciidoc.spec b/asciidoc.spec
index 2b7d3db..f56cf67 100644
--- a/asciidoc.spec
+++ b/asciidoc.spec
@@ -5,7 +5,7 @@
 Summary: Text based document generation
 Name: asciidoc
 Version: 8.6.10
-Release: 0.4.20180605git%{shortcommit}%{?dist}
+Release: 0.5.20180605git%{shortcommit}%{?dist}
 # The python code does not specify a version.
 # The javascript example code is GPLv2+.
 License: GPL+ and GPLv2+
@@ -15,6 +15,14 @@ Source0: https://github.com/%{name}/%{name}-py3/archive/%{commit}/%{name}-py3-%{
 
 Patch1: asciidoc-python3.patch
 
+# https://github.com/asciidoc/asciidoc-py3/pull/5
+# https://github.com/asciidoc/asciidoc-py3/issues/16
+# Commits:
+# 6469317 Remove unnecessary decode in a2x
+# 684913e Fix decoding of file that specifies encoding in header tag in a2x
+# 8369a97 re-add --nonet option
+Patch2: asciidoc-python3-a2x-decode-fix.patch
+
 BuildRequires: python3-devel
 BuildRequires: dblatex
 BuildRequires: docbook-style-xsl
@@ -174,6 +182,10 @@ cd tests
 %{_sysconfdir}/asciidoc/filters/music/*.py
 
 %changelog
+* Sat Jun 16 2018 Todd Zullinger <tmz@pobox.com> - 8.6.10-0.5.20180605git986f99d
+- Add some a2x decoding fixes from upstream PR#5
+  (https://github.com/asciidoc/asciidoc-py3/pull/5)
+
 * Fri Jun 15 2018 Todd Zullinger <tmz@pobox.com> - 8.6.10-0.4.20180605git986f99d
 - Restore BUGS.txt and CHANGELOG.txt doc files