Posts: 185
Threads: 0
Joined: Aug 2018
They are unavailable until 1406
Posts: 55
Threads: 0
Joined: Mar 2012
Is there an easy way to contribute small fixes to help text? It feels like a waste of time to report single small spelling/grammar errors. I try to batch them up to make a worthwhile bug report, but it's easy to forget them before I've seen enough to report. Do you have the code up on github? Searching for "seravy" didn't pick up anything relevant.
p.s. in the Mind Storm help text "effect" -> "affect"
Posts: 10,463
Threads: 394
Joined: Aug 2015
Does github even work for binary/hex edited files?
If forgetting is the problem, you should make a text file to store them until you make a post.
June 20th, 2019, 12:28
(This post was last modified: June 20th, 2019, 12:32 by myk002.)
Posts: 55
Threads: 0
Joined: Mar 2012
Sure, github works for any kind of file, though a text file (maybe yaml) that gets compiled into a patch against the binary would make more sense. It would be easier for the community to edit, the commit history would be useful, and pull requests would have sensible diffs. What's the binary format? Maybe I could write a tool that does the compilation.
For the record, I do accumulate help text fixes in a text file, but this kind of community contribution is exactly the use case for github. The question is whether the introduction of a new tool to your workflow would help the development of the project or just get in your way. (I'm hoping it's the former, but I know any change to development workflow is disruptive)
Posts: 10,463
Threads: 394
Joined: Aug 2015
Well, unless you can point me towards a simple-to-use hex editor that has inbuilt git support, it would be quite inconvenient.
More importantly, how many people are out there who would want to contribute code in hex?
I mean, sure there are plenty of people contributing ideas and bug reports, but I only really know of one person who actually contributed patch data.
To make matters worse, hex patches aren't convenient for people to work with, you need to apply it to your files then disassemble to be able read it. So you can't even click on the commit and read the contents.
Besides, does github even allow illegal projects? Remember, we are hacking a commercial game without permission.
Posts: 55
Threads: 0
Joined: Mar 2012
I don't think the original MoM binaries (or derivatives) should be put on github, just the parts that we can factor out and represent as text files. As you point out, it wouldn't be very useful to upload any binary code. What parts could the community contribute to? Help text could be one part. Maybe unit stats. Probably not game logic.
For example, we could start a github repository (say seravy/MoM) with just the extracted help text in human-editable yaml format and a small script that error checks and compiles the text into a format that could be inserted into the MAGIC.EXE binary. When you want to change some help text, you'd edit the yaml file in a text editor instead of the binary directly in a hex editor. Then you'd run the script to generate the modified binary.
Just an idea. I'm not sure how difficult it would be to write the compilation script. I see the text in the binaries, but I don't know yet how it is ordered/indexed.
Posts: 10,463
Threads: 394
Joined: Aug 2015
I still fail to comprehend how that would help me work less when it already only takes a few seconds to change files that do work with text. Assuming someone else does it for me and submits a pull request, I have to read it, approve it, makes changes to parts I like in a different way, download the updated file, apply it to the executable, and finally split between the default and optional versions. That takes longer than reading the forum post and changing it directly myself. It's not like typing in 2-3 characters worth of text is such a hard work I need someone else to do it. As far as unit stats go, the decision takes the long time. Changing the number in the text file doesn't.
Posts: 55
Threads: 0
Joined: Mar 2012
Many of the steps could be automated away (e.g. 1. edit/approve the pull request, 2. build binary), but making development difficult for you is the last thing I want to do. I really appreciate the work you've done on this game. I'm just trying to help in any way that I can.
June 23rd, 2019, 10:09
(This post was last modified: June 23rd, 2019, 10:10 by Seravy.)
Posts: 10,463
Threads: 394
Joined: Aug 2015
Found a bug during my test game :
Quote:5.6c
-Fixed 5.6 bug : When a failed spell or book treasure is replaced by mana, game behavior is indeterminate. (possible memory corruption, crashes, etc)
In my case the bug caused an already conquered node to show as unconquered the next turn (although there were no monsters or treasure left).
Posts: 10
Threads: 0
Joined: Jun 2019
I'm having some issues with a tower not giving me a reward. Every time I step onto the tower, it gives me the message "You have found a tower."'
There was some weirdness when one of my heroes died as well, it wasn't displaying the name of her dropped weapon after I won. I've had quite a few issues with text strings.
Also have had an issue where players could cast an infinite number of Holy Weapons in battle.
save1.gam is the save file for the first slot, right?
|
