Anlizt6: The GM6.1 Analyzer & Corruption Finder

[IsmAvatar]

This will analyze any GM6.1 file from the beginning to the end. This is not the old version that didn't read the whole file.
You still must have access to the C drive to use this.

users.nni.com/cmagicj/anlizt6.zip (35 KB)

What it doesn't analyze:
Whether Images are the correct Format or not (it'll only take their size and skip over them)
Whether your code has any errors in it, etc (lol)

What it does analyze:
Sprites (but not the image format - just basic settings for them)
Sounds (again, not the format)
Backgrounds (no format)
Paths
Scripts
Fonts
Timelines
Objects
Rooms
Game Information
The Resource Tree
Global Game Settings (with the includes and constants)

Make sure to read anlizt6.txt before using. Otherwise you won't be able to use it (at best, you'll open it up and see a black screen flash)

This was written in Pascal and is the first full release. There are still a few things that I may add in a second release, but I wantto make sure it doesn't have any problems first. You're welcome to try it out on any GM6.1 file, or even try a few GM6.0 files. If you get a corruption even though you know GM can open the file just fine, tell me, and I'll try and fix it. It is also helpful to include the GM6 file that gives the problem, and (optionally) the analyze.txt file (optionally, because I can make one myself quite easily)

This is a precursor to LGM. Who knows, maybe I'll make one to read in your analyze.txt file and turn it into a gm6 file >.< Doubtful, but you never know.

Tell me if you have any problems with it and I'll help you out.

Sorry, I did not archive the old version. I can't imagine you wanting the old version anyways.


Just to give you a clue why I'm working on this:
I was working on a sprite exporter/importer. You can export sprites just fine, but for some random reason, when it imports them it corrupts the file. I made this to find the corruption. So far it hasn't found anything, but that was using the incomplete version of anlizt6.

[Xenith]

sounds cool i've never had a corrupted file before but i'll be sure to test it if i do,

i'm not curious about why you made the program, but how you made it, programing is so confusilizing

[IsmAvatar]

Please test it on files that are not corrupted. At this time I'm more concerned about finding out if it reports a good file as corrupted so that I can fix that.
If you ever come across a corrupted file, please give it to ME, because I need to study them to further the development of this project, and you probably wouldn't have a clue how to use this to examine or fix a corrupted gm file, since I didn't document it thoroughly enough yet.

I made it in Pascal, the source code is included in the zip. I used binary reading/writing functions. all the functions I used can be looked up in the Pascal Manual: ftp.freepascal.org/pub/fpc/docs-pdf/rtl.pdf (8MB!!!) or online at www.freepascal.org/docs-html/rtl/index.html (don't know size - but I doubt it's small). Search in the units that the program has next to "uses" as well as the System unit.

As for understanding what bytes to do what to, please see the byte documentation that I spent a good few months designing (which at this time is pretty much every other post in this forum).

[Xenith]

if i ever get a corrpted file i'll give it to you.

[IsmAvatar]

A bug was noticed that I created when I created a function to take care of a repetative task. I corrected the bug only for the resources that appeared in the corrupted file I was working with, and forgot to correct it for other resources. I've now fixed it so it works for all resources.

If you downloaded before 2005 October 2 @ 1900 GMT, please redownload the file (same link) to get the fixed version

[IsmAvatar]

I've been using a completed version for quite some time now and keep forgetting to release it. It is now released. If you downloaded the file before 2006 August 30th at 17:30 GMT, please re-download the file.

I am considering this a completed version. I will collect any bug reports posted here, and other such things, and release a version 2 some time, which will allow you to choose the file yourself. Version 2 will likely be written in Java.

[Jim]

G'day ISMAvatar,

I've been having quite a bit of trouble with corrupt files lately. I have tried the program from the 1st post to find the corruption but it tells me that there is no corruption that it can find.

Here is the link to a zip file containing my GM6 file and a small text file detailing how to recreate the corruption.

members.westnet.com.au/brick27/Jimgm6.zip

Hopefully you will be able to pinpoint what is wrong and update your program to find this sort of problem in the future.

Thanks for taking the time to look into this and I look forward to hearing from you soon.

Jim

[Jim]

Further to the above, you say in the anlizt6.txt file to look at the source code to understand what the corruption error means. I'm not up to scratch on binary so it makes no sense to me.

The new corruption I am getting now is stating a corruption error of 65, how do I found out what this means? Do I need to refer to the binformatni.txt file that you have mentioned previously? Where so I get this file from?

Cheers

Jim

[IsmAvatar]

The source code of anlizt.
Error code 65 means that one of your Scripts is reporting the wrong version.
Error code 0 means that it was trying to read a string but read past the end of the file because the string reported its size too large. Unfortunately, this error code is somewhat ambiguous, which is why Error codes should also provide the location of the corruption and the value that it received at that position. These might seem like nonsense to you, but they are very important to me.

The error code tells me where in the source code it went wrong, which in turn allows me to know what kind of resource it was, and what data it was trying to determine about that resource. This allows me to know Why it's corrupted, and how to fix it. The location tells me where in the gm6 file the corruption occurs, and the byte value tells me what the corruption looks like.

I will look into this further.