A ton of changes by Dale. Lots of new features. See VersionHistory for version 7.3.0 #184
Conversation
…a couple of dialogs
…rking better and added knife plot. May not be completely accurate yet. More testing in progress. but this save current state.
… try changing display to Q3D bar graph.
… when input wave length is 550. Added hot keys I and ctrl-o. Limited profile y offset to just one wave from when multiple are selected. Change percent control to a button.
…y help. Added auto collimation function to ronchi and foucault page.
…ectories. Added user settable prefixes to wave front movie generation.
… for menus. Updated some help and tool tips. Added L hot key to open igram directory.
…g back from auto collimation in Foucault view.
…rredtly and astig and wave front movies can both be made.
…age 3 of the report. Added log display to the process wizard window and various QApplication::processevents so the screen updates the displays during the process for better user feedback.
…ile can be moved up or down in y value on profile plot. Letting any profile be selected to be moved.
…ot headings to be sort enabled. Disabled ability to edit polar plot table entries.
|
That's a lot of changes. Kindly let me fix the CI scripts and do a quick code review before merging and releasing (release cannot work without script anyway). Might take until this week end as I'm quite busy right now. In the meantime, you can update the releaseNotes file |
|
If you have some time, can you look at the newly introduced build warnings ? : https://github.com/githubdoe/DFTFringe/actions/runs/15024125124 They are not particularly bad, but they may mask really bad warnings so they need to be gone. edit: maybe not newly introduced |
I looked at them all. Well what I mean is I compiled it myself in the qtcreator ide and got 29 warnings. I looked at all 29. There may be one that you linked to that isn't showing up with compiler settings I'm using. Anyway nothing serious. I could fix most of them quite quickly I admit. I don't think I want to at this point. When I'm building something and I see warnings, I fix them but I only notice them if they are happening on the files I'm editing (because only those ones get built). And Dale seems to be ignoring them anyway. So I don't plan to fix them at this time. |
atsju
left a comment
atsju
left a comment
There was a problem hiding this comment.
the PR is too long to be review-able.
I will to look at the 3 comments I mentioned. Please delay the release just a little bit.
|
|
||
| // Set the logging format | ||
| spdlog::get("logger")->set_pattern("[%Y-%m-%d %H:%M:%S.%e] [%^%l%$] %v"); | ||
| //spdlog::get("logger")->set_pattern("[%Y-%m-%d %H:%M:%S.%e] [%^%l%$] %v"); |
There was a problem hiding this comment.
this must be uncommented in release
There was a problem hiding this comment.
Sorry it was part of my attempt to disable that for my debugging. I don't remember but maybe the ifndef technique I used for part of that did not work there for reasons I did not understand nor could get to work. IIRC
There was a problem hiding this comment.
I had noticed this but when I looked at the log output there was date and time included. Maybe I should have looked in the log file? I didn't think to.
| } | ||
| catch(std::exception& e) { | ||
|
|
||
| qDebug() << "Exception writing video " << e.what(); |
There was a problem hiding this comment.
should probably be qWarning (better for logging purpose)
| //For some reason the original starter code of makeChart was in this function but it caused a crash. | ||
| // I could never figure out why because that code was taken from a working Qt example. When | ||
| // moved to inside another function it worked. So there you go. Exact same code with no changes worked there but not here. | ||
| makeChart(); |
There was a problem hiding this comment.
this makes me super uncomfortable. Might just break later on if we don't understand it.
There was a problem hiding this comment.
Me too but I did a lot of debugging that I did not list in the comment. I narrowed the issue down to the line where it attaches the radial axis to the plot. At that point there is no debug code to show what happens inside that call. Like I said that code works everywhere else except in that routine. At that point I had to punt. Others are welcome to dig deeper.
There was a problem hiding this comment.
I tested on linux and got no issue. I will try the linux build also. So far I did NOT reproduce your issue.
|
I spent an hour testing. No problems so far. Want to test a few more things. |
|
I don't see any log file at all 😢 |
No description provided.