Truncation of monster names

[half]

In my proto-variant, I have had a longstanding bug with monsters' names being truncated. For example, at one point

White worm mass

appeared as:

Wh

and now the white worm masses were fine but

Metallic green centipede

appears as

Me

I'm pretty sure it has also happened with more than two characters in the string before getting cut off.

Has anyone had this happen to them before? It strikes me that somehow a NULL character is getting written into the text descriptions or something like that. It mainly happens on my Windows build (the same creature getting mangled for each playtester in a given development version), but I have had the problem once on my Mac build.

Any ideas?

[zaimoni]

Yes, there's a fairly awful hack in data loading dating back to the initial conversion to data files: all of the strings for monster names are stuffed into a single long C array containing the NUL-terminated representation of the strings in sequence. There's a few other instances. Look for ?_name and ?_text global variables for the large arrays, and name and text fields within *_type structs for offsets into these arrays.

It sounds like the construction and/or offsets into these arrays isn't going perfectly.

[Twilight]

Originally Posted by half

In my proto-variant, I have had a longstanding bug with monsters' names being truncated. For example, at one point

White worm mass

appeared as:

Wh

and now the white worm masses were fine but

Metallic green centipede

appears as

Me

I'm pretty sure it has also happened with more than two characters in the string before getting cut off.

Has anyone had this happen to them before? It strikes me that somehow a NULL character is getting written into the text descriptions or something like that. It mainly happens on my Windows build (the same creature getting mangled for each playtester in a given development version), but I have had the problem once on my Mac build.

Any ideas?

This seems like it might be the old "Blu" bug. The thread:
http://forum.nppangband.org/viewtopic.php?id=151
gives a fix - at least for NPP. Hopefully it's the same bug and this helps.

[half]

Originally Posted by Twilight

This seems like it might be the old "Blu" bug. The thread:
http://forum.nppangband.org/viewtopic.php?id=151
gives a fix - at least for NPP. Hopefully it's the same bug and this helps.

Thanks Twilight. You have saved me a lot of array out of bounds debugging. That is almost certainly the same bug. It looks like my code forked off from NPP between the fix to load.c and the fix to birth.c. I've now removed the troublesome line, and all should be fine (I can't check without a windows box, but will do so in a week or so).

[Twilight]

Originally Posted by half

Thanks Twilight. You have saved me a lot of array out of bounds debugging. That is almost certainly the same bug. It looks like my code forked off from NPP between the fix to load.c and the fix to birth.c. I've now removed the troublesome line, and all should be fine (I can't check without a windows box, but will do so in a week or so).

Luckily I just started playing *bands again and checking the angband forums/r.g.r.a etc after years away in time to recognize this one. Glad to have been of help.