@Endocore:
Oh, your post certainly took me back... and not in a good way! You have just reminded me of a number of headaches I was having back in the day. Thanks a lot!
Anyway, I have seen what you described... various erratic errors due to dialog files being very large. (Most of the time the problem was just that the dialog would say "error" instead of pulling the actual dialog, just as the problem you reported.) And, to be more specific, I don't think it was related directly to the size of the file so much, but to the number of dialog options. The reason Kladwtwn.msg was the size it was because, one more dialog file inside it, a character from Colly, refused to work consistently. So, I started combining dialog in other files, like aicrshdr, for example. But, I don't think a character who just likes to say a lot will necessarily break the dialog. I really do think it is the number of {xyz}{}{} in the file and not the actual amount of dialog that causes the file to be broken.
But, yeah, I sometimes still combine message files, but not as often anymore. And there haven't been any problems with doing so. But, I think I did make another odd 'discovery' back in the day. Or else I am remembering it wrong. Part of me says that there wasn't a limit on the message files. But rather, there was a "bad block" of message files to use. I'm sorry, I can't remember the number where I think things went bad. Looking at my scripts.lst file, though, it had to be somewhere around script 1775 or so, because it was the Colly scripts that were originally giving me that headache. But, by skipping ahead 30 lines or so (put a1.int, a2.int, etc. in scripts a whole bunch of times, then adding the new real script), I never encountered the problem again. At least, that is what I think happened. Please give that a shot.
Thanks for the info on the decompiling errors. When I do encounter any errors like that, I just try to revert to an earlier version of one of my scripts or I will take Killap's latest version and add in my changes. But, I guess I'd better update to the latest version of sfall and its int2ssl, too! And that "internal error"! Ugh! Yeah, I remember the first time I found that while compiling. Going through a script over and over and over again to find the problem for a new variable I added but forgot to set at the top of the script. Now, it is the first thing I check.
And Dahlia was indeed one of the NPCs Pogue4 had recommended recruiting. You're right: it seemed like she had the greatest potential to just get up and go. There were several we discussed, but the only other one I can remember him suggesting was Chrissy from the squatters after you saved her, even though I thought she'd be too young.
Oh, your post certainly took me back... and not in a good way! You have just reminded me of a number of headaches I was having back in the day. Thanks a lot!
Anyway, I have seen what you described... various erratic errors due to dialog files being very large. (Most of the time the problem was just that the dialog would say "error" instead of pulling the actual dialog, just as the problem you reported.) And, to be more specific, I don't think it was related directly to the size of the file so much, but to the number of dialog options. The reason Kladwtwn.msg was the size it was because, one more dialog file inside it, a character from Colly, refused to work consistently. So, I started combining dialog in other files, like aicrshdr, for example. But, I don't think a character who just likes to say a lot will necessarily break the dialog. I really do think it is the number of {xyz}{}{} in the file and not the actual amount of dialog that causes the file to be broken.
But, yeah, I sometimes still combine message files, but not as often anymore. And there haven't been any problems with doing so. But, I think I did make another odd 'discovery' back in the day. Or else I am remembering it wrong. Part of me says that there wasn't a limit on the message files. But rather, there was a "bad block" of message files to use. I'm sorry, I can't remember the number where I think things went bad. Looking at my scripts.lst file, though, it had to be somewhere around script 1775 or so, because it was the Colly scripts that were originally giving me that headache. But, by skipping ahead 30 lines or so (put a1.int, a2.int, etc. in scripts a whole bunch of times, then adding the new real script), I never encountered the problem again. At least, that is what I think happened. Please give that a shot.
Thanks for the info on the decompiling errors. When I do encounter any errors like that, I just try to revert to an earlier version of one of my scripts or I will take Killap's latest version and add in my changes. But, I guess I'd better update to the latest version of sfall and its int2ssl, too! And that "internal error"! Ugh! Yeah, I remember the first time I found that while compiling. Going through a script over and over and over again to find the problem for a new variable I added but forgot to set at the top of the script. Now, it is the first thing I check.
And Dahlia was indeed one of the NPCs Pogue4 had recommended recruiting. You're right: it seemed like she had the greatest potential to just get up and go. There were several we discussed, but the only other one I can remember him suggesting was Chrissy from the squatters after you saved her, even though I thought she'd be too young.
Last edited: