Messages

211

AGI Development Tools / Re: WinAGI: feature request - lower case GameID

« on: February 12, 2021, 10:44:41 PM »

Hmm, I guess I could. Every Sierra game I know of uses all upper case for GameID (even though the check in the set.game.id function is in fact case sensitive). It's just one of those long standing traditions I guess.

I assume the ScummVM check is also case-sensitive? Meaning, if you use 'PQ1', ScummVM will think it's the original Sierra version? Why not just use 'PQ1H' or something like that? The GameID property isn't actually linked to the AGI game - you can set it to anything you want despite what the game logics actually use for the set.game.id command.

In the meantime, let look into what it will take to implement lower case GameIDs.

Oh, and I did drop a comment on your pull request. Hope it helps.

212

AGI Development Tools / Re: WinAGI: feature request - batch mode

« on: February 12, 2021, 10:00:20 PM »

Probably can't make it completely GUI free, as I'd have a lot of things to unhook from the compile functions.

... suddenly I understand what's the big frikkin deal with recent, more "modern" UI development paradigms. I need a drink.

I don't know how to read this. All I was saying is that the main functions for opening and compiling games currently provide feedback in the form of progress windows, status bars, and message boxes. I'd have to deal with all of these to go sans-GUI. It's not difficult, but it will take some time that I'd rather spend on something else.

Is that an example of a "modern" UI development paradigm? Why is that bad?

213

AGI Development Tools / Re: WinAGI: feature request - batch mode

« on: February 12, 2021, 03:48:24 PM »

This is probably the best place to make feature requests! Separate threads works for me, unless the requests are closely related.

Regarding batch mode, I can easily add command line switch options ( '/c' for a full compile; '/r' for rebuild VOL, '/d' for compile dirty logics) along with a wag file name that will run the compiler when WinAGI starts. Probably can't make it completely GUI free, as I'd have a lot of things to unhook from the compile functions. But I could make the GUI disappear immediately after running.

Would that work for you? Are there any other switches/functions you think you might need?

214

AGI Development Tools / Re: WinAGI 2.1.5 - fails to compile PQ1

« on: February 11, 2021, 02:24:57 PM »

Short answer - your files are marked read-only. If you remove that mark, everything works just fine.

Longer answer - not all my error messages are helpful; I should have provided more useful information to the user about why the files weren't writing. I've tweaked that section of code so it points you in the right direction if read-only files are encountered. Update will be available sometime this weekend.

I think a better long term solution will be to verify all files can be accessed when the game is loaded, and provide immediate feedback to the user at that time. I'll add that to my ToDo list for next version.

thx for finding this for me.

215

AGI Development Tools / Re: WinAGI 2.1.5 - fails to compile PQ1

« on: February 11, 2021, 11:57:03 AM »

That's not good. It compiles fine for me though. But now I'm wondering if I have the right DOS version of the game. My game files are a bit of a mess, as I'm constantly tweaking things as I do testing.

The first error indicates one of the game resources needed by the compiler is not loading. But the only resources that the compiler needs loaded are the object file and words.tok file. Location 66662 in vol.0 is in the middle of View 188. The compiler does not need to load views, and View 188 isn't even referenced in Logic 6 anyway. That makes me think my version is not the same as what you have.

I have game version 2.0G, running on AGI interpreter version 2.917. Can you check your game and interpreter version? Also, if possible, can you zip up all your game files, including source directory, and make them available to me so I can try to duplicate the problem?

Once I find out what's going on, I'll get a fix out for you.

Again, sorry for the inconvenience.

216

AGI Development Tools / Re: WinAGI: retain "Show message arguments by number" settings

« on: February 11, 2021, 11:07:09 AM »

Dang it. That's on me. When I was adding that setting, I changed the name that is used in the game's config file at one point. Since my dev file had both the old and new name in it, i thought it was working correctly. But what's happening is that it uses the old property name to read the setting, and the new property name to save the setting.

It's easy to fix though. In a text editor (WinAGI has a text editor built in!), open up your winagi.config file, which can be found in your install directory, usually in 'Program Files (x86)\WinAGI GDS'. If it's not there, try 'Users\{user}\AppData\Local\VirtualStore\Program Files\WinAGI GDS'*** - Windows 10 sometimes moves program configuration files automatically.

Find the [Decompiler] section, and add this line:
 

Code: [Select]

  MsgsByNum = True
then save the file. If the line is already there, just change the value to 'True'. Ignore the line that says 'MsgsByNumber = True'. If you want to change the setting later, you'll have to manually edit the config file again.

I'll get a fix out that addresses this permanently by the end of the week.

Sorry for the inconvenience.

***The AppData folder is a hidden folder; if you don't see it, make sure you configure File Explorer to show hidden files and folders.

217

AGI Development Tools / Re: WinAGI Character map

« on: February 10, 2021, 05:01:37 PM »

Yeah, I kinda broke my own rule about providing multiple input methods - I tried to make sure there were keyboard/toolbar/menu options available for all program functions. But this feature was a late add on, and I didn't include the menu/toolbar options partly because i reasoned it was likely to be very infrequently used (and partly because I was lazy ). Sorry about that.

I guess the good news is that the Help file was actually helpful! 

If this is a deal-breaker for anyone, I can probably get a revision out that adds both a toolbar button and menu item. Otherwise, hopefully by the time you want to use it a year from now, there may be a C# version out that has all the features you need.

218

AGI Development Tools / WinAGI Version 2.1.5

« on: February 09, 2021, 10:12:13 PM »

This version adds an option to decompile logics with message argument tokens (m1, m2, etc.) instead of literal strings. Also addresses two minor bugs found by pierrelapin (when opening words.tok files with no group zero words and views with bad description pointers).

Link is here: http://agiwiki.sierrahelp.com/images/6/69/WinAGI_2_1_5.zip

219

AGI Development Tools / Re: WinAGI 2.1.3 & Original Games: Errors and Warnings

« on: February 03, 2021, 07:09:47 PM »

Thank you for the bug reports! I wish you were around for the earlier beta tests!

Manhunter SanFrancisco, 3.02    7/26/89
03-févr-21: WARNING: Sound215 was skipped because it's VOL file (MH2VOL.6) is missing.
03-févr-21: WARNING: Sound216 was skipped because it's VOL file (MH2VOL.6) is missing.
Compile!

Warnings are informational only; they don't mean the game is broken or that WinAGI can't work with them. The warnings you see here are fairly common in Sierra Games. When Sierra built the DIR files, they often left old index information in for resources that they were no longer using. It's usually not a problem because none of the logics ever try to load them (at least not in any of the games I've examined). But the pointers are bad, and WinAGI is just informing you that they are there. WinAGI also deletes the bad pointers from the DIR files.

Mixed Up Mother Goose
WINAGI Crash when clicking on Words Editor icon ! (No Words in this game)
Compile!

This is a good catch. The WORDS.TOK file for this game has no words in group 0 (the words that get ignored). Which is fine; they are not required. But the editor expects them there, and when it tries to set up the form with the list of groups, it gets stuck when it can't find group 0. I'll get a fix out for this shortly.

Police Quest 1, 2.0G     12/3/87 (AGI STUDIO has no problem and the resource is readable)
View.036, error 513 in preview: can't read past end of resource
Compile!

Another good catch. This view has an invalid pointer for the View Description; it is set to a value that is past the end of the file.  WinAGI has always treated that as a fatal error, as attempting to load a view description with an invalid pointer can create problems. But for this view, the view description property is never used, so it's not really a problem. I will get a fix out that ignores this error and repairs the pointer automatically.

Space Quest 1, 1.1a 1986-11-1 (ok with other downloads but same version, AGI STUDIO can't read the bad version at all)
Logic.099, Error 688 at position 10 in preview mode:  Unable to decode logic (invalid goto position or if/then block
Logic.100, idem at position 197
=> refuse to compile... certainly a corrupted archive (another one works well)

If there is a bad logic, due to corrupted source VOL files, then WinAGI will correctly refuse to compile it. Maybe I am misunderstanding your comment?

Space Quest 2, 2.0f (2.0A OK and never contains Logic141 & Picture147)
03-févr-21: WARNING: Logic141 was skipped because it's location (131071) in the VOL file(VOL.0) is invalid.
03-févr-21: WARNING: Picture147 was skipped because it's location (196607) in the VOL file(VOL.0) is invalid.
compile!

This is the same as the Manhunter comment. WinAGI is just informing you that the DIR files have bad pointers in them. Most likely they existed at one time during development, but were dropped at some point. Sierra never bothered to update the DIR files. WinAGI automatically removes the invalid pointers when it imports the game.

For each entry, you add "Compile!" at the end. I think you mean that after importing, the game can be successfully recompiled? If the games can be imported and successfully recompiled, I'm not sure what the issue is. Other than the Space Quest corrupted game, are there any that don't compile? That'd be useful information.

If you know of any other significant issues, I'd love to get that information as well. While I don't want to spend a lot of time on the VB6 code, I will make fixes for significant errors.

Thanks again for the information,

220

AGI Development Tools / Re: Original AGI code?

« on: February 02, 2021, 10:57:10 AM »

I have also partially disassembled Sierra's original Logic compiler (cg.exe), which has helped me validate their syntax.

221

AGI Development Tools / Re: WinAGI Version 2.1.3 Is Available

« on: February 02, 2021, 03:11:12 AM »

Hello Pierre,

Welcome to our AGI community. It's a great place to discuss AGI related issues, and you can always count on the users here to help answer any questions you might have.

I'm sorry to hear you are having problems with WinAGI. All I can say is that I've done extensive testing on it over the last ten years or so, and I have successfully decompiled and recompiled every original Sierra DOS game with no errors. But there were a lot of different versions of each game released, and it's possible you have a version that I don't have that does cause errors when loaded in WinAGI. I don't think that's too likely though.

And yes, you do see a  lot of warnings when you decompile/recompile original Sierra games, but those are all there because of bugs or bad coding practices used in the original games. WinAGI is able to do deep analysis of the resources and logic code and exposes those flaws that other AGI tools gloss over or ignore. If you don't like the warnings, you can disable them by setting the error detection level to low. But you miss out on one of the best features of WinAGI (in my opinion) because it will help you find those subtle errors and bugs that are otherwise really hard to detect.

Regarding unrecognized resources, many of the original Sierra games had bits of old code, old resources and invalid directory pointers that got left in the game, even though they are never accessed. WinAGI is able to identify those 'orphaned' invalid resources so they can be corrected if doing a recompile. The errors are not caused by WinAGI- on the contrary, WinAGI points out what other tools miss.

I am not sure I understand your comment about message variables. The terms 'message' and 'variable' have distinct meaning in AGI syntax, so 'message variables' isn't a thing (translation error I assume?) I *think* you are referring to a decompiling option to hide inline display of messages as text strings; so when decompiled, you want to see this:

Code: [Select]

  if (said("look"))
    {
    print(m1);
    }
instead of this:

Code: [Select]

  if (said("look"))
    {
    print("You see nothing of interest.");
    }
If that's the case, then you are correct; WinAGI currently doesn't offer that as a decompile option. But it took me less than 10 minutes to add it. Unfortunately, I saw this post literally one minute AFTER I just posted a new version (2.1.4)!!! So I'll have to make another version, including installer, assuming this is the fix you want. If so, I can get that out in the next couple days.

 Support for delimited comments (/* ... */) was removed intentionally from WinAGI several years ago. There were two reasons for this. First, the original syntax used by Sierra did not have this feature; keeping the syntax as close to the original for nostalgia purposes is something that many in the AGI fan community feel very strongly about. Secondly, removing them greatly simplified the logic coding functions, and also provided huge performance gains when using syntax highlighting in large source files.  It's possible that support for them will be added back in when I move to C#, but most likely not. Time will tell...

Your comment about logics, views and sounds not being recognized, if true, certainly is a concern. But as I said, I have never had any issues like that in years of testing with original DOS versions of AGI games.

I encourage you to start another thread that includes the specific issues you are seeing. If you give me details, I may be able to tell you if it really is a bug that needs addressing, or if you are perhaps not using WinAGI correctly.

222

AGI Development Tools / Re: WinAGI Version 2.1.4 Is Available

« on: February 02, 2021, 01:11:06 AM »

Buggers. The toolbar handler uses the Key property of the pressed button to set the draw mode; I don't know how, but the Key for the absolute line button got deleted, so the button just gets ignored now. I have no idea what I did that caused that, because I haven't worked on that code in a very long time. It's an easy fix. I will try to post an update shortly, since the picture editor is not very functional without it. While it's not very efficient, you could use the trapezoid option to draw a temporary set of lines, then edit the coordinates to be the line you actually want. (use Ctrl+Ins to add new coordinates as needed).

 If you find any other major bugs like this, please let me know.

FWIW, I'm working in C# now, so other than fixes for things that are severely broken, there won't be any more updates to version 2. Maybe by some time this summer, I'll be ready to release a more modern version... with a lot of new bugs...

thx,

No problem! I'll be sure to share if I find anything else like that in my fiddlings. I'll be looking forward to both the Pic Editor fix and whatever more you got in store later down the line, makes me happy to see this program still being maintained after so long. 

I posted an updated install file. Link here: http://agiwiki.sierrahelp.com/images/3/37/WinAGI_2_1_4.zip

This takes care of the picture editor bug. Thanks for your help in finding this.

223

AGI Development Tools / Re: WinAGI Version 2.1.3 Is Available

« on: January 27, 2021, 10:47:58 AM »

I was fiddling with this a bit, and I noticed that in the Picture Editor, the button for Absolute Line does not function properly. You can click it and it will show Absolute Line selected but the tool will continue to be whatever it was previously. I'm on Windows 7, for what it's worth.

Buggers. The toolbar handler uses the Key property of the pressed button to set the draw mode; I don't know how, but the Key for the absolute line button got deleted, so the button just gets ignored now. I have no idea what I did that caused that, because I haven't worked on that code in a very long time. It's an easy fix. I will try to post an update shortly, since the picture editor is not very functional without it. While it's not very efficient, you could use the trapezoid option to draw a temporary set of lines, then edit the coordinates to be the line you actually want. (use Ctrl+Ins to add new coordinates as needed).

 If you find any other major bugs like this, please let me know.

FWIW, I'm working in C# now, so other than fixes for things that are severely broken, there won't be any more updates to version 2. Maybe by some time this summer, I'll be ready to release a more modern version... with a lot of new bugs...

thx,

224

SCI Development Tools / Re: SCI01 interp decompiling

« on: December 10, 2020, 10:05:20 PM »

You might have better luck with Ida Pro. it's a highly capable disassembler. It's a commercial tool, but they have made one of their earlier versions available for free, and it works great on MSDOS programs.

I took a quick look at Reko- it couldn't even figure out the entry point for simple DOS programs. Not sure how well it would do with a program as complex as SCI.

Decompiling (disassembling) is a lot of work, and a bit of an art. I've spent years disassembling AGI. I have Version 2.917 100% disassembled, and other version mostly done (I only go to them when I'm looking for differences).  But it took A LOT of work.

If someone is serious about trying to decompile/disassemble SCI, it can be done, but just be sure you understand what your getting yourself into!

225

AGI Development Tools / Re: WinAGI Version 2.1.3 Is Available

« on: December 07, 2020, 02:43:02 PM »

OK, OK, OK!!!

I know I need to switch. But I'm an old curmudgeon, and frankly it scares me a bit that I have such a steep learning curve awaiting. I have Visual Studio (and have had it for years) installed, but every time I open it up, I just get discouraged at how complicated even simple tasks seem. I know it's just a matter of buckling down and learning it, but like I said, I'm old and stuck in my ways. It's not rational, but I can't help it. Some day...

Regarding the problem with the text editor - it's not Windows; it was me. Again. Stupid versioning- the installer had version 3.0.30 of the OCX, but it was supposed to be 3.0.31. So I was installing the wrong one (When I tweaked the OCX, I forgot to copy the new version to the install folder). I fixed the installer, and tested it on a clean Windows 10 system, and it works great - no errors.

You may notice that the first post changed a bit - after getting a special request, I added support for Apple IIgs sound. More fun stuff!

Thx for your support (and for your 'gentle' nudging toward a modern dev platform...)