-
Notifications
You must be signed in to change notification settings - Fork 81
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Collaborators #400
Comments
@nycz I rebased your top branch on current dev: https://github.com/wrye-bash/wrye-bash/compare/nycz-wx-begone-DEV. I would like the linux stable support in, as well as the iewin work. Please have a look and let me know when we could start. I need to refactor balt a bit too and this will cause nasty conflicts with your branch, so this is rather urgent. An immediate refactoring is change the |
@wrye-bash/devs @Sharlikran - the fallout3/nv branch is once again out of sync with the dev branch - needs be rebased onto dev and RecordHeader and canEditMasters changes be applied there (in clear commits we will later squash). I do hope this is the last time we have to do this, as I would like to see it merged for 307 proper. Code quality in there is inferior to main tree - warnings count increases by couple hundreds, long lines and ravioli code (I try to deraviolize in some commits in there the patcher but more work to do). Plus I personally can't offer any support and as it's well known we do not offer all the functionality @valda 's fork offers. However it's two major games, the code is there and since we have to refactor game handling and eventually the patcher, we need to merge. Remaining issues, forum reports as detailed in posts above (the bug ones not design/feature ones) and merging the fo3 branch is what we need to concentrate to. Branch: https://github.com/wrye-bash/wrye-bash/commits/feature-sharlikran-valda-fallout @Sharlikran I think this is for you :) Please reply to me at #150 |
Sorry I have permanently retired from anything computer related to work on my marriage and care for my mom in her last days. |
So, the 'resizing status bar icons' part of this came up again. We no longer crash, but resizing icons does not work correctly. I've asked on Discord if this has ever worked, in any previous version, and no one knew of any... maybe it worked due to some undocumented behavior in an ancient wx version? |
The StatusBar mixin is a hack - IIRC it does not work on phoenix at all - maybe we should bug the wx guys directly for this |
@Utumno / @Infernio (since you two seem to be the most active) 2 questions for you:
Main reason I ask this is I couldn't find a good related open issue to ask the following question, and forum posts feel like they mix users' posts with dev talk.
I'm not really getting back into active development, just want somethings to toy around with to scratch that coding itch, without much pressure. Feel free to delete this comment after you answer my Q's. |
Here is the place :) Lo you have all the rights in the world, hash up a branch over dev - usually Infernio handles nightly but I will too (I messed a regex up) and I can cherry pick. You can in theory push to nightly but needs some little coordination |
There's also a dev chat on discord, even though Utumno hates it :P But any actual discussion should stay on github |
Let's use this issue for now and will have a look at gitter |
If an exception occurs when importing `bolt`, the code used to give a bugdump errors when it tries to access `bolt.scandir`, resulting in the following traceback, which hides the original traceback. Launching Wrye Bash 307 in debug mode fail bash\bolt.py Traceback (most recent call last): File "Wrye Bash Launcher.pyw", line 89, in <module> bash.main(opts) File "bash\bash.py", line 181, in main _import_bolt(opts) File "bash\bash.py", line 57, in _import_bolt msg = u'\n'.join([dump_environment(), u'', u'Unable to load bolt:', File "bash\bash.py", line 162, in dump_environment if bolt.scandir is not None: AttributeError: 'NoneType' object has no attribute 'scandir'
… EEE - ugly but can't use MI EEE SSS SpecialPatcher(CBash_Patcher): EEE FIXME: @@ -1301,3 +1301,3 @@ def scan(self,modFile,record,bashTags): if u'R.ChangeSpells' in tags: - spells = curSpells + spells = curSpells # FIXME noop elif u'R.AddSpells' in tags: Revisit isActive/isEnabled for this patcher. allowUnloaded == False?? SpecialPatcher erosion only CBash patchers use scan_more - RacePatcher just used those overrides?? More initDataPruning - tweaks not done yet SSS RACE
Posting an archive of images and documentation from this post from the old forums here so it doesn't get lost when they shut down in March: This is about CBash/PBash discrepancies, as the name implies :) |
Following from Utumno comment at #380 regarding my question of where to put the rest of the issues/requests I wanted to report, I'll paste in here my notes of all the things I thought during these last 5 months using Bash for SSE. Raw unfiltered notes :)All these reported issues apply to v307.202003151850, where it matters. General issues:
UI related
For mod list:
Usability requests:
Tweak Settings:
Screenshots:
Character Save/Profile Management:
Extra Stupid requests, one can dream, right? :)
New Bashed Patch Tweaks [for Skyrim SE only?]*
These are all my notes, in a somewhat easy to read format, I hope! 😕 |
Adding some responses and links as I run through: General issues:
UI related:
For mod list:
Usability requests:
Tweak Settings:
Screenshots:
Character Save/Profile Management:
Extra Stupid requests, one can dream, right? :)
New Bashed Patch Tweaks [for Skyrim SE only?]*:
|
Went and opened a bunch of issues from all the parts that I think sound interesting and would require more discussion. |
Guys you re too fast for me :P Re: one instance - this should have broken when daedalos switched it to wx.SingleInstanceChecker - before we used a lock file - that was per install - this seems a better behavior after all. But as Infernio said this is #250. However it is true that this sounds like a regression Pushed a commit for Has extra Dirs -> 52402e6 - is that what you mean? Needs work but then squash to random BAIN fixups |
It looks like you can give a mutex name to |
As for #250, gonna read it carefully later! Gonna provide clarification for the things you mentioned above:
Hope I haven't missed anything. |
Seems yes - comes from bc1d743 a bit difficult to untangle that one |
diff_dir in newest nightly added some loose ends from previous merge |
ut-336-drop-table is nightly ready - some of it is already on nightly (that I believe is bug free), all of it looks like bba78b3...ut-336-drop-table. It all but finishes off #336 all of sudden (well actually the main commit is dated 1/1/22) but could use some more testing - in particular swapping Oblivion.esm (or not), save profiles and delete. If any of you are using git could you give it a run? |
Traceback
The source is 49b8f49. |
Thanks sibir was running it in Oblivion that does not have that issue -
will try and post a fix when near a computer again
…On Sat, Feb 24, 2024, 8:22 AM sibir ***@***.***> wrote:
ut-336-drop-table crashes & throws the following on launch:
Traceback
Traceback (most recent call last):
File "C:\Users\*****\Documents\GitHub\wrye-bash\Mopy\bash\bash.py", line 426, in main
_main(opts, wx_locale, wxver)
File "C:\Users\*****\Documents\GitHub\wrye-bash\Mopy\bash\bash.py", line 573, in _main
frame = bapp.Init() # Link.Frame is set here !
^^^^^^^^^^^
File "C:\Users\*****\Documents\GitHub\wrye-bash\Mopy\bash\basher\__init__.py", line 4276, in Init
self.InitData(progress)
File "C:\Users\*****\Documents\GitHub\wrye-bash\Mopy\bash\basher\__init__.py", line 4302, in InitData
bosh.modInfos = bosh.ModInfos()
^^^^^^^^^^^^^^^
File "C:\Users\*****\Documents\GitHub\wrye-bash\Mopy\bash\bosh\__init__.py", line 2175, in __init__
super().__init__(ModInfo)
File "C:\Users\*****\Documents\GitHub\wrye-bash\Mopy\bash\bosh\__init__.py", line 1611, in __init__
if do_refresh: self.refresh(**self._boot_refresh_args)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\*****\Documents\GitHub\wrye-bash\Mopy\bash\bosh\__init__.py", line 2359, in refresh
ldiff = self.refreshLoadOrder(forceRefresh=mods_changes or
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\*****\Documents\GitHub\wrye-bash\Mopy\bash\bosh\__init__.py", line 2101, in _modinfos_cache_wrapper
ldiff.affected |= self._refresh_mod_inis_and_strings()
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\*****\Documents\GitHub\wrye-bash\Mopy\bash\bosh\__init__.py", line 2439, in _refresh_mod_inis_and_strings
k for k, v in self.items() if v.isMissingStrings(
^^^^^^^^^^^^^^^^^^^
File "C:\Users\*****\Documents\GitHub\wrye-bash\Mopy\bash\bosh\__init__.py", line 973, in isMissingStrings
bsa_infos = self._find_string_bsas(cached_ini_info)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\*****\Documents\GitHub\wrye-bash\Mopy\bash\bosh\__init__.py", line 939, in _find_string_bsas
modInfos.get_bsa_lo(cached_ini_info=cached_ini_info,
^^^^^^^^^^^^^^^^^^^
AttributeError: 'NoneType' object has no attribute 'get_bsa_lo'
The source is 49b8f49
<49b8f49>
.
—
Reply to this email directly, view it on GitHub
<#400 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AAKNIV73NW3KY4MEIAG6FTTYVGPMPAVCNFSM4EGPAMWKU5DIOJSWCZC7NNSXTN2JONZXKZKDN5WW2ZLOOQ5TCOJWGIZDSNRTGI4Q>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
Aaand repushed |
I'm tackling #26. Early boot is crystallizing more and more. ini_files filed for divorce from bass/bush/env as it became apparent that it is vital to early boot (cough #679 cough). Edit: don't mind the branch yet, I'm just paranoid about pushing commits :P |
I pushed some fixups to Oblivion.esm swapping (mostly ghosting related - introduced in 4dfa03e, that setGhost was needed). I also disallowed deactivating the master_esm (related to swapping as allowing "inactive" master esm complicated swapping even more). |
Thanks Inf - will gonna have to merge a 336-table-prelude to dev and add table removal to nightly - if anything stands out there let me know. From the rest, if you can have a look at 90b9f90 (and 9364e3f) - turns out whole line comments would be parsed as settings - I patched it but looking at cb9de8b there might be other places to do the same and/or a better way? |
@Utumno https://www.afkmods.com/index.php?/topic/4966-wrye-bash-all-games/&do=findComment&comment=190580 - You refactored empty dir cleanup semi-recently, right? Looks like a regression from that. A reproducer, if I understand Arthmoor's post correctly, should just be a package like this:
Which, when uninstalling, will leave behind an empty dir contained in an otherwise empty dir, and we try |
I went ahead and merged the bsa changes (in particular ffff1af...eea41fa) - apart from the ! seems safe and fixes more bugs than (maybe) adds - actually the ! are for a new feature (Morrowind bsa lo) and pointing out existing limits in str detection refresh. I also merged in nightly the rest of the _game_lo refactoring (well half of it) that started in ff78cce - the latter's been in nightly for a while. It contains a rewrite of IniGameLo so it's again Morrowind testing. These changes would be what I would consider needed for an RC. It would be nice to include the remaining half of _game_lo refactoring but the first part must be included as it already started conflicting (per game overrides were moved out of _game_lo into GameInfo in bcf2c93). In particular - if you use Bash for Morrowind please give a whirl to new nightly - build in https://github.com/wrye-bash/wrye-bash/actions/runs/9329992303 |
Pushed a WIP/POC of replacing all those checkbox pngs with just a few SVG
|
Wow, that looks really good! I might find some time this week, in which case I'd love to try my hand at making a GUI for that (if you don't preempt me on it of course 😅). |
Really nice Lo - it would be nice to have only svgs been read by the code - most of what pngs remain are used in the readme so are not loaded - remains this dreaded image/tools directory and image/games. Wow I pushed some nit mainly addressing the TODOs on the exceptions you left and some random wrapping in f24a970 |
Go for it! I was thinking we maybe show a preview of all of the possible colored checkboxes with some mark (probably a check) in them in the GUI for customizing - for colorblind people to be able to see them all next to each other). Then this part is where we'll have to read in from settings. I did notice that
Thanks :) Going through all the rest of the images (thankfully none of those require svg variables or combining svgs) and replacing those that are used in program is a goal. Hopefully online converters can handle all of those nicely. Then more cleanup (maybe most of |
This comment was marked as off-topic.
This comment was marked as off-topic.
This comment was marked as off-topic.
This comment was marked as off-topic.
I pushed a squashed version of ut-336-578-wip on nightly - I have fixed all bugs I could find but could use some more testing from someone else. Among others it contains a rewrite of pinned mods handling which should come handy in relation to the Starfield lo and ccc file lo handling - see #681 and #682 . The rest is some ongoing #336 and #353 refactoring (entangled with the lo changes) - in particular in renaming and copying operations that opens the way for the last part (in ut-lab) that removes the best part of the refactoring scaffolding from DataStore (new_info, add_info, delete_refresh and so on). This could also use some testing so could be left out - we are in RC state I'd say (minus the lo fixups mentioned and maybe some of the triage/high priority items from the project board) Edit: I went ahead and moved most open issues in the 313 milestone to the 314 one (314 moved to 315). Feel free to move back and forth as needed. I also dusted up the project board 'triage' column - moved some to 'Lower priority recent'. |
Thanks for the fixup @sibir-ine ! Meanwhile I have pushed some changes that address #682 - I think I 'pinned' it :P If this survives a nightly round then merging time - we should also include a squashed version of the svg changes by @lojack5 I think |
Sorry for the delay (phone issues and didn't want to remove authenticators while I waited for a new one). Yeah the SVG stuff is pretty good to go, only thing I could do is run them through an optimizer to cut down on a few lines. And maybe a nicer wizard overlay if anyone has ideas, not really an artist myself. The GUI for customizing is separate and can always be added later. |
Thanks Lo - we can even add the opts later (and the wand) - I will squash-add the svg branch on my next nightly round (I am working on smoothing out the 336-578 branch for merging). |
wxPython finally has a maintainer again 🚀 Sorry for my radio silence btw, I started playing FF14 and that has just eaten my free time 😅 |
Well Wrye Bash still lacks maintainers :P |
I've fully given up on my PR for wx type-hints ever being even looked at, even with new releases. I'll keep lurking here though ;) |
I went ahead and merged some commits that seemed ok after a couple months - will merge more of it - if you have any time https://github.com/wrye-bash/wrye-bash/milestone/16 and the high priority items on the board (https://github.com/orgs/wrye-bash/projects/2/views/1 in particular https://github.com/orgs/wrye-bash/projects/2/views/1?pane=issue&itemId=66626565 I think we should not fix) is all we need done for a release I was thinking also of adding a nightly section in the Nexus files page where we can post stable nightlies |
Wow, my PR to wx got merged in two hours 😮 Granted, it was a simple fix, but the new maintainer seems to actually merge stuff. |
Is the PR GIF related? Meanwhile I would suggest that some items of the project board be transferred to #200 - especially the internal todos and low priority ones. Let's keep the project board for user reports (thanks @sibir-ine for keeping it updated!) and have #200 for our todo lists |
No, unfortunately not (I'll get around to fixing up the GIF branch soon(TM)). Anyway: note to self, the XDG user dir support we have in linux.py is wrong - we need to read $XDG_CONFIG_DIR/user-dirs.dirs to implement that correctly. |
Opening this up to try and centralize the various bits of info that are spread in forums and github repos/commit comments/issues - kind of a 308 wip thread - technical thread. Messages may be edited or deleted - this thread is meant to have condensed info on ongoing refactoring and bugfixing.
The text was updated successfully, but these errors were encountered: