Petr Pisar
Petr Pisar
Actually I'm able to reproduce it on ARMv6 where I have interactive access. It raises SIGILL there. I'm not sure if this is the same issue: ``` Program received signal...
I see it's the bundled ZPAQ. Unfortunately the bundled 5.0 version if very old. Upstream has 7.05. It would be great if were able to rebase the bundle.
The JIT is supported on x86 and x86_64 only. One needs to pass CPPFLAGS=-DNOJIT to configure on other platforms.
On Tue, Nov 10, 2015 at 10:27:43AM -0800, Naveed Massjouni wrote: > I wanted to make those deps "optional". Maybe using "eval 'require XML::Hash::LX'" to add or remove the subs...
Following patch adjusts the test input as I'm not sure whether the canonization was from point of view of mojomojo intentional or not. ``` From 8cab2f071958c9570b22e074251a6dd41999e205 Mon Sep 17 00:00:00...
Actually one need to adjust the test too: ``` From 3cb8ddd23808ce1797738e14c877e5409b1af764 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Petr=20P=C3=ADsa=C5=99?= Date: Thu, 24 Jul 2014 11:27:22 +0200 Subject: [PATCH] Provide canonical URL...
This can be the same bug as in issue #63 or #118.
MojoMojo::Schema::ResultSet::Page::normalize_name() does: ``` return ( Encode::decode_utf8(URI::Escape::uri_unescape($name_orig)), Encode::decode_utf8(URI::Escape::uri_unescape($name)), ); ``` while input variable is an Unicode string. uri_unescape() unfortunately preserves the UTF-8 flag despite the output of URI-unescaping is a byte-string...
I can reproduce it with perl 5.18.2 and current Text-Xslate's master HEAD. This happens if the host is heavily loaded and it's a race in the test. If you comment...