[jira] [Commented] (INFRA-7153) mirrors.cgi doesn't ignore all HTML comments

["Konstantin Kolinko (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/INFRA-7153?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14281288#comment-14281288 ] 

Konstantin Kolinko commented on INFRA-7153:
-------------------------------------------

According to documentation for ezt tool that is used by mirrors.cgi, the '[' symbol can be escaped as '[[]'
https://code.google.com/p/ezt/wiki/Syntax

Thus I think that <!--[[]if lt IE 9]> should work for you.

I think it is not really a job for a simple template engine to bother with HTML / XML syntax. Processing the text of comments is useful, e.g. to allow variable substitutions in them.

> mirrors.cgi doesn't ignore all HTML comments
> --------------------------------------------
>
>                 Key: INFRA-7153
>                 URL: https://issues.apache.org/jira/browse/INFRA-7153
>             Project: Infrastructure
>          Issue Type: Bug
>          Components: Mirrors, Other/Misc
>            Reporter: Kanak Biscuitwala
>             Fix For: Aug 2014
>
>
> We use the following setup for the Apache Helix website:
> 1. Source files are mirrored to http://helix.incubator.apache.org from https://svn.apache.org/repos/asf/incubator/helix/site-content
> 2. The source HTML files may include comments, specifically our download page https://svn.apache.org/repos/asf/incubator/helix/site-content/download.html contains the following due to the tool we use to generate the pages:
> <!-- Le HTML5 shim, for IE6-8 support of HTML5 elements -->
> <!--[if lt IE 9]>
>     <script src="http://html5shim.googlecode.com/svn/trunk/html5.js"></script>
> <![endif]-->
> 3. Our .htaccess contains the following line:
> Redirect /download.html /download.cgi
> 4. Our download.cgi contains the following line:
> exec /www/www.apache.org/dyn/mirrors/mirrors.cgi $*
> 5. When visiting download.html (http://helix.incubator.apache.org/download.html), we get a Python traceback:
> Traceback (most recent call last):
>   File "/www/www.apache.org/dyn/mirrors/mirrors.cgi", line 143, in mirrorwrap
>     return mirrorsapp(environ, start_response)
>   File "/www/www.apache.org/dyn/mirrors/mirrors.cgi", line 251, in mirrorsapp
>     template.generate(output, data)
>   File "/www/www.apache.org/dyn/mirrors/ezt.py", line 141, in generate
>     self._execute(self.program, fp, ctx)
>   File "/www/www.apache.org/dyn/mirrors/ezt.py", line 301, in _execute
>     method(method_args, fp, ctx, filename, line_number)
>   File "/www/www.apache.org/dyn/mirrors/ezt.py", line 321, in _cmd_subst
>     fmt = _get_value(valref, ctx, filename, line_number)
>   File "/www/www.apache.org/dyn/mirrors/ezt.py", line 486, in _get_value
>     raise UnknownReference(refname, filename, line_number)
> UnknownReference: if in file /x1/www/helix.apache.org/download.html at line 30
> 6. Manually removing the commented lines I noted above fixes the issue.
> The mirror script should ignore all commented HTML when it's trying to generate the webpage, right?



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)