"%3f" shouldn't be unescaped in HTTP URLs

"%3f" shouldn't be unescaped in HTTP URLs

[Sebastien Delafond]

Quoting from Debian bug #573186[0]:

  Some HTTP URLs have literal '?' in them.  Since '?' also separates the
  path from query arguments, it needs to be escaped in the first case.
  These are examples of the two cases:

  - http://en.wikipedia.org/wiki/Who%27s_Afraid_of_Virginia_Woolf%3F
    (literal '?')
  - http://www.google.com/search?q=org+mode (query separator)

  These are the URLs generated by Org mode when exporting them to HTML:

  - http://en.wikipedia.org/wiki/Who%27s_Afraid_of_Virginia_Woolf?
  - http://www.google.com/search?q=org+mode

  I.e. '?' is left as is, but '%3f' is unescaped, producing the wrong
  URL (don't be fooled by the fact that Wikipedia actually accepts it ;)
  ).  This makes impossible to correctly export an HTTP URL with a
  literal question mark in it.

  The solution would be to leave '%3f' as is, too, in the same way as
  '%27' has been left as is in the previous example.

What do you guys think ?

Cheers,

--Seb

[0] http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=573186

Re: "%3f" shouldn't be unescaped in HTTP URLs

[Ivan Vilata i Balaguer]

[-- Attachment #1.1: Type: text/plain, Size: 1354 bytes --]

Sebastien Delafond (el 2010-03-09 a les 17:36:16 +0000) va dir::

> Quoting from Debian bug #573186[0]:
> 
>   Some HTTP URLs have literal '?' in them.  Since '?' also separates the
>   path from query arguments, it needs to be escaped in the first case.
>   These are examples of the two cases:
> 
>   - http://en.wikipedia.org/wiki/Who%27s_Afraid_of_Virginia_Woolf%3F
>     (literal '?')
>   - http://www.google.com/search?q=org+mode (query separator)
> 
>   These are the URLs generated by Org mode when exporting them to HTML:
> 
>   - http://en.wikipedia.org/wiki/Who%27s_Afraid_of_Virginia_Woolf?
>   - http://www.google.com/search?q=org+mode
> 
>   I.e. '?' is left as is, but '%3f' is unescaped, producing the wrong
>   URL (don't be fooled by the fact that Wikipedia actually accepts it ;)
>   ).  This makes impossible to correctly export an HTTP URL with a
>   literal question mark in it.
> 
>   The solution would be to leave '%3f' as is, too, in the same way as
>   '%27' has been left as is in the previous example.

Hasn't anyone come across this behaviour?  I think it's a bug which should be
fixed since it makes some URLs impossible to write.  And according to that last
sentence, the solution doesn't seem too hard.

Thanks and cheers,

::

  Ivan Vilata i Balaguer -- http://ivan.lovesgazpacho.net/

[-- Attachment #1.2: Digital signature --]
[-- Type: application/pgp-signature, Size: 316 bytes --]

[-- Attachment #2: Type: text/plain, Size: 201 bytes --]

_______________________________________________
Emacs-orgmode mailing list
Please use `Reply All' to send replies to the list.
Emacs-orgmode@gnu.org
http://lists.gnu.org/mailman/listinfo/emacs-orgmode

Re: "%3f" shouldn't be unescaped in HTTP URLs

[Carsten Dominik]

On Mar 15, 2010, at 8:17 PM, Ivan Vilata i Balaguer wrote:

> Sebastien Delafond (el 2010-03-09 a les 17:36:16 +0000) va dir::
>
>> Quoting from Debian bug #573186[0]:
>>
>>  Some HTTP URLs have literal '?' in them.  Since '?' also separates  
>> the
>>  path from query arguments, it needs to be escaped in the first case.
>>  These are examples of the two cases:
>>
>>  - http://en.wikipedia.org/wiki/Who%27s_Afraid_of_Virginia_Woolf%3F
>>    (literal '?')
>>  - http://www.google.com/search?q=org+mode (query separator)
>>
>>  These are the URLs generated by Org mode when exporting them to  
>> HTML:
>>
>>  - http://en.wikipedia.org/wiki/Who%27s_Afraid_of_Virginia_Woolf?
>>  - http://www.google.com/search?q=org+mode
>>
>>  I.e. '?' is left as is, but '%3f' is unescaped, producing the wrong
>>  URL (don't be fooled by the fact that Wikipedia actually accepts  
>> it ;)
>>  ).  This makes impossible to correctly export an HTTP URL with a
>>  literal question mark in it.
>>
>>  The solution would be to leave '%3f' as is, too, in the same way as
>>  '%27' has been left as is in the previous example.
>
> Hasn't anyone come across this behaviour?  I think it's a bug which  
> should be
> fixed since it makes some URLs impossible to write.  And according  
> to that last
> sentence, the solution doesn't seem too hard.

Hi Ivan,

the bug has not yet been fixed because my time is limited.  Please  
consider submitting a patch.

- Carsten