[Akonadi] [Bug 456923] Outgoing e-mail from GMail account to smtp.gmail.com sometimes hangs

2022-11-20 Thread malcolm macleod
https://bugs.kde.org/show_bug.cgi?id=456923

malcolm macleod  changed:

   What|Removed |Added

 CC||mmacl...@webmail.co.za

--- Comment #3 from malcolm macleod  ---
I have to restart akonadictl every time I want to send an email out of kmail
from one of my gmail accounts. This is a usability nightmare.
Is there any info that can be given to get this resolved?

-- 
You are receiving this mail because:
You are watching all bug changes.

[kdevelop] [Bug 393808] New: Breakpoints are sometimes set using a path that confuses GDB (and potentially other tools?)

2018-05-03 Thread malcolm macleod
https://bugs.kde.org/show_bug.cgi?id=393808

Bug ID: 393808
   Summary: Breakpoints are sometimes set using a path that
confuses GDB (and potentially other tools?)
   Product: kdevelop
   Version: 5.2.1
  Platform: Other
OS: Linux
Status: UNCONFIRMED
  Severity: normal
  Priority: NOR
 Component: general
  Assignee: kdevelop-bugs-n...@kde.org
  Reporter: mmacl...@webmail.co.za
  Target Milestone: ---

Today I ran into a situation where no breakpoints were working inside kdevelop,
breakpoints all just said "pending" even though gdb was otherwise functioning.

After some bother I managed to track this down to my use of a linux "bind
mount"
1) I have a folder "/home/foobar/blah/Development"
2) I have a secondary mount of this folder "/Development"
3) Kdevelop project and all tools etc. use "/Development" everywhere
4) However I accidentally opened my kdevelop project through
"/home/foobar/blah/Development" instead of "/Development"
5) When setting breakpoints kdevelop is passing the path from which the project
was opened to gdb "/home/foobar/blah/Development/src/foo.cpp" instead of
"/Development/src/foo.cpp"
6) This confusingly leaves gdb unable to find the file, even though both paths
are valid (presumably because the paths are hard compiled into the debug
information or similar)

While this is how it presented in this specific case, I do wonder if other
tools may have similar issues - so perhaps (ideally) kdevelop should warn if
the paths inside the project file and the path from which the project are
opened differ from one another.

-- 
You are receiving this mail because:
You are watching all bug changes.

[kdevelop] [Bug 386103] Frame stack depth hard coded to 20

2017-10-23 Thread malcolm macleod
https://bugs.kde.org/show_bug.cgi?id=386103

--- Comment #1 from malcolm macleod  ---
On closer inspection I see I slightly misdiagnosed the issue.

If I change kdevelop into a small windowed mode where "20 stack items" is
taller than the window, then I get a scrollbar and the additional stack items
appear upon scrolling.

If I am running fullscreen then 20 stack items fit comfortably without a
scrollbar showing, and at this point there is no way to get more stack items
(except to shrink the window).

So I guess the real issue then is that this should be handled differently for
larger window sizes - though I still think a way to just increase the default
might also be good.

-- 
You are receiving this mail because:
You are watching all bug changes.

[kdevelop] [Bug 386103] New: Frame stack depth hard coded to 20

2017-10-23 Thread malcolm macleod
https://bugs.kde.org/show_bug.cgi?id=386103

Bug ID: 386103
   Summary: Frame stack depth hard coded to 20
   Product: kdevelop
   Version: git master
  Platform: Other
OS: Linux
Status: UNCONFIRMED
  Severity: minor
  Priority: NOR
 Component: UI: general
  Assignee: kdevelop-bugs-n...@kde.org
  Reporter: mmacl...@webmail.co.za
  Target Milestone: ---

It seems the frame stack depth to display is hardcoded here:
https://github.com/KDE/kdevelop/blob/f5b25083620dc518f5fa5ed3e3e237390614d7db/kdevplatform/debugger/framestack/framestackmodel.cpp#L295

In template heavy code 20 is often not enough to see what is going on properly,
especially when e.g. breaking a multithreaded program mid execution and then
wanting to step through all of the threads to view what is going on at the
time.

It would be good as a simple solution if at a minimum this "upper limit" value
could be increased via the settings/preferences, while 20 is sufficient for
many people I'd prefer to set it to 40.

A more complex solution could perhaps both have a configurable "upper limit"
value as well as a "..." button at the bottom of the stack where the user can
click to fetch more items.

-- 
You are receiving this mail because:
You are watching all bug changes.