Review Request 130117: Fix detection of EOF

[Martin Tobias Holmedahl Sandsmark]

> On May 8, 2017, 7:40 a.m., Oswald Buddenhagen wrote:
> > this looks familiar ... https://phabricator.kde.org/D4975
> > 
> > so assuming the condition is valid, this requires an _extensive_ comment.
> > also, shouldn't this just be unified with the solaris code path?

No, Solaris has a very different behavior that is described by the already existing comment (needing to "drain" the empty bytes with read(..., 0); etc.). It won't work on Linux.

What stuff do you want in the comment? I described how detecting an EOF is supposed to be done (by actually trying to read).


- Martin Tobias Holmedahl


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/130117/#review103203
-----------------------------------------------------------


On May 6, 2017, 6:09 p.m., Martin Tobias Holmedahl Sandsmark wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://git.reviewboard.kde.org/r/130117/
> -----------------------------------------------------------
> 
> (Updated May 6, 2017, 6:09 p.m.)
> 
> 
> Review request for Konsole, Kurt Hindenburg, Oswald Buddenhagen, and Peter Wu.
> 
> 
> Bugs: 372991
>     https://bugs.kde.org/show_bug.cgi?id=372991
> 
> 
> Repository: kpty
> 
> 
> Description
> -------
> 
> Don't just assume that 0 bytes read means EOF.
> 
> 
> Diffs
> -----
> 
>   src/kptydevice.cpp 22233a5 
> 
> Diff: https://git.reviewboard.kde.org/r/130117/diff/
> 
> 
> Testing
> -------
> 
> Fixes the konsole issue from https://phabricator.kde.org/D4975 and https://bugs.kde.org/show_bug.cgi?id=372991 and the autotest works.
> 
> 
> Thanks,
> 
> Martin Tobias Holmedahl Sandsmark
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/konsole-devel/attachments/20170508/43b9d72b/attachment.html>