On Sun, Mar 12, 2017 at 11:41 PM, Dirk Hohndel <[email protected]> wrote:
> Hi Tomaz, > > It's great to see that you are working on things! I had assumed that I'd > have more time this weekend to work on things, but it turned out I got > side tracked with all the pull requests and with fixing the unit system > change bug in the mobile apps... > > On Fri, Mar 10, 2017 at 11:53:16AM +0100, Tomaz Canabrava wrote: > > My time is crazy and I fear I'm taking too long to finish this because I > > just don't find the time to work, sorry. > > I know how that feels. > > > This is the current WIP, there's still a bit of things to be done, I > would > > appreciate some help finishing this. > > It would help if you told us what help you need. > > this is *NOT* a rework like the preferences that added a lot of code for > > little gain, this doesn't adds a lot of code (actually, almost nothing) > and > > the gain is big. > > And it would help to understand what that gain is. > The current code holds 6 tabs with all of the logic for saving / updating restoring a dive, a trip, an dive plan, copy/pasting, etc. all this in one file. This series tries to rework a few things: - Each tab has it's own header / source, so it's easier to make sure we are actually changing what we wanna change - Less interconectivity with the MainWindow and other widgets, trying to keep the state more sane. - Smaller ui files, because a single ui file is massive and sometimes we need to choose strange names on the widges to not have clashes. - a better, overall, code sanity. > > The series doesn't apply cleanly for me because my tooling tries to fix > some of the obvious white space garbage as it applies things, and then of > course patches that contain the white space garbage no longer apply. > > It would make my life so much easier if you just used Qt Creator with the > same settings as the rest of us, but I guess that ship has sailed and I > just should stop even asking for that. Which means it might be easier if > you just point me to the branch on github that I can pull from (because > then my tooling won't mess things up). > Sorry - I have a lot of different projects here (work, kde, subsurface), and all them uses a different code style. The better approach should be QtCreator / KDevelop, whatever, to have different styles per project, but currently it's a no-go. I'll squash them and use clang-format to have a sane code that you can apply. > > Anyway, after I applied the 42 patches things compile and I get a mostly > working UI - but again, I'm not sure what I should be looking at, what I > should be working on. > I'm not changing the code from master, mostly I'm trying to reason with the master code and split it into the different files that I created, so in theory things should still work. but there are tons of TODOs that I added to the code. > > Clearly the Equipment tab and Extra Info tab are non-functional. > Statistics and Photo tab seems unchanged (or identical). > There are small functional issues with the Notes tab (specifically, the > Depth and Duration fields that should only be shown if this is a manually > entered dive are always shown), but other than that this seems reasonably > complete. > > I can easily push this into a branch on github, but unless the rest of us > understand better > > - what are you doing? > Splitting maintab.cpp because it's a mess > - why are you doing it (i.e., what's the benefit)? > Smaller functions, better code quality, better to understand what's going on. > - where would it make sense for us to try to help (so that we aren't > changing code that you are working on right now)? > > I'm not quite sure what to do with this. > If you could take a look at the tab-widget/ folder I created and reason if the changes I'm doing are good, that would be awesome already. what I need is a "looks fine, continue" - because in my current timeframe I don't wanna work in something that you think it's unecessary. Tomaz > /D > >
_______________________________________________ subsurface mailing list [email protected] http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface
