Remove references to pages that don't exist (i. e. have no .prp file) #1366
Conversation
|
I'm reasonably sure that these entries weren't in the .age files for PotS or MOUL... my guess is a bunch of development files were used when open-sourcing moul-scripts, or something to that effect |
| @@ -55,7 +55,7 @@ void plClientMsg::AddRoomLoc(const plLocation& loc) | |||
| if (loc.IsValid()) | |||
| fRoomLocs.emplace_back(loc); | |||
| else | |||
| hsStatusMessage("Trying to load an invalid room, ignoring"); | |||
| hsAssert(false, "Trying to load an invalid room"); | |||
There was a problem hiding this comment.
The challenge with this is that it only fires in debug builds.
There was a problem hiding this comment.
Well, we could make it a hard error everywhere. I can't think of a good reason why you'd want to silently ignore invalid page loads, but perhaps there are weird cases that I don't know about.
Or just leave the original log message in place, in addition to the assert. Though I feel like the log message on its own isn't a good tool for uncovering these issues, considering that it has appeared on every game start for years without anybody paying attention to it...
|
I've been sitting on this for awhile, thinking that something isn't quite right with this idea. I finally remembered what's not good about this today. Artists who use the Max plugin have to write their .age files by hand, so it's possible that the .age file that an artist hands off has PRPs, potentially for future content, listed that they don't actually distribute yet. I'm not sure how I feel about requiring cutting down the .age file manually in that case. I'd like to hear from others before proceeding. |
|
I know that prpshop wont load the prps if the first prp listed in the age file doesnt exist. |
|
That sounds like a good bug to file on H-uru/PlasmaShop 😉 |
|
Looking at this again, perhaps I'll revert my .age file changes here. The nonexistant entries don't cause any problems, so if they're potentially useful during development, we can just leave them in. The Python changes are the important bit, because they fix all the cases where the game tried to actually load nonexistant pages. |
dgelessus
force-pushed
the
remove_nonexistant_pages
branch
from
99dd47d
to
9a57e84
Compare
And add an
hsAssertinstead of just a log message when trying to load a nonexistant page, so that such issues will be easier to notice in the future, at least in debug builds.The Python changes are the most relevant bit - they fix all scripts that tried to actively load nonexistant pages. The .age changes are less important in practice, because as far as I can tell, the engine doesn't do anything with pages that are just listed in the .age file and never explicitly loaded somehow.
I'm also a bit suspicious that so many nonexistant pages were listed in the .age files, including pages that never existed in any public version of Uru (as far as I can tell). I'm assuming that they are safe to remove, but perhaps these entries have some purpose that I'm not aware of...