Review Request 108685: Make it possible to overwrite content-lenght using metaData

[Dawit Alemayehu]

> On Feb. 10, 2013, 8:43 p.m., Dawit Alemayehu wrote:
> > Why not simply use the KIO::http_post function that accepts a pointer to QIODevice ? All you have to do is set the pointer to NULL, pass the total total size to the ioslave and it should work for the scenario you outlined.
> > 
> > KIO::TransferJob* job = KIO::http_post(url, 0, size, flags);
> > job->setAsyncDataEnabled(true);
> > ...
> > ...
> >
> 
> Àlex Fiestas wrote:
>     Right I coul do that, nevertheless don't you think that being able to overwrite the content-length as you do with everything else could be useful and handy?

You can override a lot of things, but not everything. Anyhow, other than adding yet another HTTP meta-data which is really not that necessary at the moment, I have no other objection to adding this particular meta-data. I simply dislike the notion of adding a meta-data just for convenience.


- Dawit


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
http://git.reviewboard.kde.org/r/108685/#review27142
-----------------------------------------------------------


On Jan. 31, 2013, 4:15 p.m., Àlex Fiestas wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> http://git.reviewboard.kde.org/r/108685/
> -----------------------------------------------------------
> 
> (Updated Jan. 31, 2013, 4:15 p.m.)
> 
> 
> Review request for kdelibs, Dawit Alemayehu and Andreas Hartmetz.
> 
> 
> Description
> -------
> 
> When submitting a Content-Type: multipart/related using AsyncDataEnabled you need to specify the Content-Length of the whole thing you are trying to send, not just of the first chunk of the data. http::SendBody will append the Content-Length using the length of the first chunk of data automatically.
> 
> This patch adds the possibility of indicating the Content-Lenght using metadata "content-length".
> 
> Another option will be using the data set in "setTotalSize" in case it makes all the trip down to the kioslave.
> 
> Also, is there any case where we want AsyncDataEnabled but we don't need to specify the content-length? 
> 
> 
> Diffs
> -----
> 
>   kioslave/http/http.cpp 2d139a9 
> 
> Diff: http://git.reviewboard.kde.org/r/108685/diff/
> 
> 
> Testing
> -------
> 
> 
> Thanks,
> 
> Àlex Fiestas
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kde-core-devel/attachments/20130310/64b7f910/attachment.htm>