FenPhoenix on 21/8/2019 at 18:22
That's sounding pretty doable in my head. I'll dig into the code and see what it'll take. Thanks for the suggestion!
Twist on 21/8/2019 at 18:47
Maybe this is self-evident, but I would hope this would also eventually work for, say, DarkLoader -> NewDarkLoader -> AngelLoader, or other similar variations. Just a thought.
goldwell on 9/9/2019 at 01:47
Hey Fen,
Thank you for working on this!
I just gave it a spin but i'm still running into the same issue as before however because for example if I wanted to import the finished tag from darkloader, I can't seem to import it also from FMsel.
So say if you have the following setup
FM 1 - Finished in darkloader
FM 2 - Finished in darkloader
FM 3 - Finished in FMSel
I was hoping that it would import all of the data from darkloader for example, and then import all of the data from FMSel and AngelLoader would represent the data combination of the two. This import window you've introduced is very robust but it has the same issue as before because it's only importing from one source, just now I can choose the specifics of where it comes from.
Does that make sense?
FenPhoenix on 9/9/2019 at 02:13
Okay, I'm sick and my head's not working fully so bear with me. The way it works now is like:
You choose to import finished status from FMSel.
FM 1:
-It imports it from DarkLoader -> the FM is marked "finished on expert" but it ignores it
-It imports it from FMSel -> the FM is not marked finished -> it imports this blank value because you chose to import from FMSel
So really what it should do is more like:
-If an FM has different values for finished (say) in the different loaders, it takes the highest priority one, unless the highest priority one is blank, then it takes the next lowest until it a finds a non-blank value
Is that about right?
goldwell on 9/9/2019 at 02:47
Quote Posted by FenPhoenix
Okay, I'm sick and my head's not working fully so bear with me. The way it works now is like:
You choose to import finished status from FMSel.
FM 1:
-It imports it from DarkLoader -> the FM is marked "finished on expert" but it ignores it
-It imports it from FMSel -> the FM is not marked finished -> it imports this blank value because you chose to import from FMSel
So really what it should do is more like:
-If an FM has different values for finished (say) in the different loaders, it takes the highest priority one, unless the highest priority one is blank, then it takes the next lowest until it a finds a non-blank value
Is that about right?
Yes that is exactly what I was hoping for!
Sorry for not being clearer earlier.
FenPhoenix on 9/9/2019 at 03:35
No problem! Sorry for taking so long. Now I know what the exact goal is, I'll get started. Thanks for the feedback; it's really way more helpful than you know. Without anyone telling me anything, I can only guess "everything's fine" or "everything's broken" ;)
FenPhoenix on 12/10/2019 at 11:40
(
http://fenphoenix.com/apps/AngelLoader/AngelLoader_v1.1.5.zip)
AngelLoader v1.1.5 is out. Fixed a few visual bugs and a couple others, so releasing this one for polish.
Unfortunately, no multi-loader priority-based import as of yet. It broke my brain trying to figure out how to implement it. The entire architecture of the import code (and the way FM data is structured) is basically at a 90-degree angle from how it would need to be in order to easily support this. Not saying never, but it's not in for now. Sorry :(
Other issues are cropping up with future plans, to wit, AngelLoader's .NET Core 3 performance on my machine - as measured as well as eyeballed - seems to be running between 2x and 10x slower than the current .NET Framework-based version (and yes, I double- and octuple-checked I'm running in Release profile and every other thing I could think of). As you can imagine this makes me extremely unhappy, as I was planning to switch to .NET Core 3 for comprehensive high-DPI support among other things, but now I'm considering what other options I have. Maybe it's because I'm still using Windows 7. That would be annoying because it is still an officially supported OS for .NET Core 3, but if that's the problem then fine, I'll switch to Core 3 cause everyone but me is prolly on Win 10 by now anyway and I will be too before the end of the year. So that's the state of things. Meh.
Additions:
-Added ability to specify which fields to import in the Import dialogs (but no multi-loader priority-based import, sorry)
Display and rendering fixes:
-Fixed: Glitchy rendering could occur in some situations with the following controls: "+", "-" and "x" buttons; the Settings window tab page buttons.
-Fixed: When the FMs list contained more FMs than could fit on the screen, it hadn't been scrolled manually since startup, and you scrolled the list to the bottom using the keyboard, there could be a glitchily-drawn last row.
-Fixed: The progress box could in some situations have an empty "%" character shown when no percentage should be visible.
Other fixes:
-Fixed: When working with .7z archives, a whole load of false-positive errors would be logged, ballooning up the log file.
-Fixed: When ctrl+up or ctrl+down was pressed at the start or end of the FMs list respectively, the selected FM would be reloaded unnecessarily. This goes with a similar previous fix for other key combos that do the same thing.
caqo on 13/10/2019 at 02:23
Hey Fen....
Did you added some new text to the language file?
FenPhoenix on 13/10/2019 at 02:49
Just these:
Code:
ImportData_Title=Title
ImportData_ReleaseDate=Release date
ImportData_LastPlayed=Last played
ImportData_Finished=Finished
ImportData_Comment=Comment
ImportData_Rating=Rating
ImportData_DisabledMods=Disabled mods
ImportData_Tags=Tags
ImportData_SelectedReadme=Selected readme
ImportData_Size=Size
This is mostly the same as the column header text, except for that "Selected readme" one. I have all these duplicate lines because using the same text for completely different places in the UI makes me a little nervous. Might end up with an FF6 "conductor / impresario" type situation. Probably not, but well, yeah.
Also, I realize now I should make a note specifying what, if any, new text appears in each new release. I'll do that from now on.