I don't remember what I did .. I just opened one bin file, then another and then change the palette around.. I think the error was in relation to that, but I can't recreate it. It may very well have just been from the way I was trying to run it trying to get 2.7 to work.

I was just trying to see if I could figure out anything that would cause the problem you were having with text alignment in the campaign menu ...

When I open the glucmpgn, I see:


If I open the properties window and close it (making no changes), the green rectangle is changed to:


I was just suggesting the text alignment issue may be related to a miscalculation of the white textbox size (or perhaps a missing variable or property or something), but that's just a guess.

PyTBL (Well, I guess all of the tools) happily places its window outside the desktop area if its settings file has saved such coordinates, which can happen when switching monitor setup. Of course I can just edit the settings file every time I disconnect my second monitor, but it could become somewhat annoying when there are 10+ files to edit. Is there an easy way to detect when window would appear entirely out of desktop area and force it inside?

Could be because I work on cleanup related stuff after I finish things still in dev. While its still in dev I would open one file per launch.

Interesting. I don't think this is affecting the text alignment but its a bug either way.
Edit: This was a result of the automatic calculations that happen when you type also being run when the fields were set programmatically while loading the dialog. Fixed now.

I've rewritten the window size loading so this should no longer happen.

Thanks for the reports!

Post has been edited 1 time(s), last time on Apr 2 2015, 1:34 am by poiuy_qwert.

Since commit 67d739b the string editing field in PyTBL is created minimally thin and has to be resized.
Also could there be a way to jump to a string by its index?
Aaalso when pressing Ctrl-F to search for a string, it would make sense to have the text field selected so I could just Ctrl-F => Write text => Enter without using mouse.
And esc as a hotkey to close search window?
And maybe there could be checkbox for wrapping around when searching?

Fixed

I can add that.

Fixed

Added

Added


Thanks for the reports!

I still have to click with mouse after having done one search.

Also, I complained a week or two ago in shoutbox how PyTBL can't handle characters outside of the 128 ascii characters properly, but it didn't happen on your system.. I came up with this patch to fix it though I have no idea if it breaks something on other systems.

Some PyDAT issues:
-Open directory does not actually open units.dat? The filename on bottom changes but unit stats don't.
-Ctrl-D and Ctrl-Alt-O give errors
-Ctrl-F selects se search field but could it also select the search string there so it gets deleted after writing something?
-This seems to be more of python ui limitation, but is it possible to allow typing letters when dropdown listbox is selected to search for a specific entry?

Edit: Is there any reason why the placement box is located at staredit tab instead of being next to unit dimensions? Sure, Datedit does it too (And Arsenal 3 as well?), but it is really misleading place for it as it is used for the building placement size and such in-game. Also it's existence is easy to forget when changing unit dimensionbox <.<

Edit2: The dimension box preview seems to be off by one pixel. Even if the dimension box is (0, 0, 0, 0) the unit will use 1x1 pixel space, unit with (3, 2, 2, 2) box will use 6x5 pixels and so on. Most of the units have same sized dimension and placement boxes, but the previewer shows dimension box smaller.

Edit3: The sound tab is completely incorrect.

Thanks for the continuous updates <3

Post has been edited 4 time(s), last time on Apr 5 2015, 7:36 am by Neiv.

PyMS icons were bugging me (No offense, poiuy). I didn't like them and it disturbed my workflow

And being that I am Corbo and like to change stuff in other people's tools I decided to make some new icons and decided to share them with you guys. I didn't make ALL of them, I am missing the saving ones, but it is pleasant enough for me to work now. My problem was that they were too... "off".


I could make saving ones too if anyone is interested, otherwise, here you go:
http://www.mediafire.com/download/l7c2041b875vd87/PyMS_Icons.zip


A few screenshots:

I forked the PyMS repo and added in your images, Corbo. If this look good, I can issue a pull request back to the original repo with your images.

Here's the repo.

I'll look into these soon, thanks!

Yeah I stopped caring much about the icons somewhere along the way But thats one reason PyMS is open source, edit them to your liking!

I wouldn't merge this into the main PyMS repo unless it was a complete and cohesive icon set. It also doesn't fit the style that I am looking for.

They are a little gray, aren't they?

Also, am I supposed to use a specific versions of Python? Most of the applications work in Python 2.7, but pyDAT crashes on me when I try to run it.

You only need Python 2.6+ for most programs, you need PIL for any program that display images (other than icons), and to work with MPQ's you need to be running in 32-bit mode. Best bet to fix the issue is to get me the crash log. Check PyMS/Libs/stdeo.txt if there are no error dialogs or tracebacks on the command line.

Fixed.

In certain circumstances it would happen on my system so I did change some stuff and it was working for me. I made the changes you did but with the latin-1 encoding for cross platform, hopefully it works for you.

Fixed

Ctrl+D is fixed, wasn't getting any issues with Ctrl+Alt+O though

Done

Yeah.. I'll see what I can do

Don't know, but for consistency I'm not going to change it.

I tweaked the preview, I believe its better.

Fixed.

Thanks for the reports.

This is the error I get with Ctrl-Alt-O:
Exception in Tkinter callback
Traceback (most recent call last):
File "C:\Ohjelmia\Python27\lib\lib-tk\Tkinter.py", line 1532, in __call__
return self.func(*args)
TypeError: openmpq() takes exactly 1 argument (2 given)
Exception in Tkinter callback
Traceback (most recent call last):
File "C:\Ohjelmia\Python27\lib\lib-tk\Tkinter.py", line 1532, in __call__
return self.func(*args)
TypeError: openmpq() takes exactly 1 argument (2 given)

Fixed thx

Some pygrp issues:
1) Opening grp and importing vertical bmp containing more frames causes the following error:

(Deleting all frames before importing avoids it)

2) Saving grp causes another crash:

Code

Exception in Tkinter callback
Traceback (most recent call last):
 File "C:\Ohjelmia\Python27\lib\lib-tk\Tkinter.py", line 1532, in __call__
   return self.func(*args)
 File "C:\Pelei\Scbw\Ohjelmia\PyMS\PyGRP.pyw", line 683, in save
   self.grp.save_file(self.file)
 File "C:\Pelei\Scbw\Ohjelmia\PyMS\Libs\GRP.py", line 457, in save_file
   f.close()
 File "C:\Pelei\Scbw\Ohjelmia\PyMS\Libs\utils.py", line 1447, in close
   raise e
WindowsError: [Error 183] Cannot create file, already exists

Nvm, another program didn't let it save

3) Grps are encoded inefficiently: Every non-transparent color is encoded as [repeat this color for n pixels], even if there are no vertical lines of repeated colors. Not too huge deal but happened to notice it

Edit: 4) Opening and saving again loses the last row/column of each frame

Post has been edited 2 time(s), last time on May 17 2015, 3:32 pm by Neiv.

I'll look into it

Hmm this sounds like it may be due to the large changes i've made to the grp handling recently. Previously I worked hard to make sure PyGRP was the best at compressing GRP's, its sad to hear that I broke that


Thanks, hopefully I'll have time to look into these soon.

PyGRP: Having problems with importing Single BMP Vertical framesets:



Additionally, while I was working with a *.grp:

1) I opened the grp file ORIGINAL. (ORIGINAL happens to be Kakaru, neutral critter in Twilight, the bird thing.)
2) I deleted every frame except for its death animation, from top to bottom in the frame list using keyboard shortcut DELETE.
3) I used SAVE AS to save to file OUT.
4) Expected output: file OUT retains remaining frames (death animation) and frames are complete. Actual output: file OUT retains remaining frames but frames are not complete: frames have been cropped according to the limits drawn in the file ORIGINAL and by the "Frame Outline (Red)" feature.

These were all a result of the same issues, and they have been fixed. Thanks!



I've made the error more graceful, thanks!



I re-wrote the compression so it should be back to having the highest compression. Thanks!



I believe this is fixed, Thanks!

The new PyGRP seems to corrupt files. The corruption is visible on inspection in PyGRP.

Can you provide sample bmp's or a grp to open+resave that reproduces the corruption?