Re: Doc string for svn_wc_status misleading?

[Julian Foad <ju...@btopenworld.com>]

Arild Fines wrote:
> cmpilato@localhost.localdomain wrote:
> 
>>"Arild Fines" <ar...@broadpark.no> writes:
>>
>>>Looking at the code, there does not seem to be any way for
>>>text_status to be set to svn_wc_status_none for an ignored
>>>file. This means it is impossible to detect an ignored file using
>>>svn_wc_status alone.
>>
>>The docstring is just out of date, I think.  Here's a code comment
>>that seems to explain the state of the real world:
>>
>>      /* If this path has no entry, but IS present on disk, it's
>>         unversioned.  If this file is being explicitly ignored (due
>>         to matching an ignore-pattern), the text_status is set to
>>         svn_wc_status_ignored.  Otherwise the text_status is set to
>>         svn_wc_status_unversioned. */
> 
> Yes, I saw that commment in assemble_status. However, the following block
> performs a check for the is_ignored parameter, which is always passed from
> svn_wc_status as FALSE:

So, is the doc string or that code comment (or both) actually wrong?  If so, we should first fix the comment (perhaps with a note that the behaviour is not what we would really like).  Then discuss fixing the API.

- Julian


>   SVN_ERR (assemble_status (status, path, adm_access, entry, parent_entry,
>                             svn_node_unknown, TRUE, FALSE, pool));
> 
> The last FALSE is is_ignored. That brings us back to my original point: It
> is impossible to get information about ignore status out of svn_wc_status.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org