prideslayer Posted July 19, 2012 Author Posted July 19, 2012 Oh bleh, that makes sense in the context of a mod script. Well if the other system works (load with no name/id/whatever) then that will work fine as well.
sen4mi Posted July 20, 2012 Posted July 20, 2012 If you want to detect game load, you might try this: Set up two values, one which is preserved in save game, and one which is not. Then, in game mode, check if they are different. If they are not different, put the same current time in both of them.
Chase Roxand Posted July 20, 2012 Posted July 20, 2012 Since v7 I've had the GECK crash when trying to save scripts. Removing NVSEE makes it stable. The source of the problem seems to be: playerREF.NX_GetEVFl "Sexout:Core:Sextype:Oral" It was working before, but today it's being derpy.
DoctaSax Posted July 20, 2012 Posted July 20, 2012 Since v7 I've had the GECK crash when trying to save scripts. Removing NVSEE makes it stable. The source of the problem seems to be: playerREF.NX_GetEVFl "Sexout:Core:Sextype:Oral" It was working before' date=' but today it's being derpy. [/quote'] I'm having a similar issue. A script (any script; no NVSEE code in it) saves fine with the save button, no powerup warnings either. But when I simply want to exit the script editor, I'm asked to save again. Clicking on yes makes my geck crash, clicking on no means a save w/o compile (at least some of the time: I did an sqv in game to check whether a new quest var showed & it didn't). Only had this since yesterday, and the only thing that changed is switching to v7. Removing v7 removes the problem.
Chase Roxand Posted July 20, 2012 Posted July 20, 2012 Anyone have v6, so we can mod until this bug is fixed?
DoctaSax Posted July 20, 2012 Posted July 20, 2012 Anyone have v6' date=' so we can mod until this bug is fixed? [/quote'] Here ya go:
prideslayer Posted July 20, 2012 Author Posted July 20, 2012 hmmmm I had a similar problem today and I was getting really confused, nothing seemed to solve it, could not edit *any* esms or esps. For no reason at all, restarting my PC fixed it, and I am working fine in v7 again. I don't know if steam being resident was forcing it to get cached despite being updated or what...
Halstrom Posted July 20, 2012 Posted July 20, 2012 hmmmm I had a similar problem today and I was getting really confused' date=' nothing seemed to solve it, could not edit *any* esms or esps. For no reason at all, restarting my PC fixed it, and I am working fine in v7 again. I don't know if steam being resident was forcing it to get cached despite being updated or what...[/quote'] I've been getting a error just opening anything non vanilla in GECK, it may be related to V7 I suspect. ---------------------------Warning - PowerUp - REPORTS --------------------------- Loading a user created or misnamed BSA Origin: 008A3942 "Yes to all" will disable all warnings of this origin. --------------------------- Yes No Cancel ---------------------------
prideslayer Posted July 20, 2012 Author Posted July 20, 2012 Can't say I've seen that one before. My problem was editing anything in the sexout esp caused the failed save issue.. where you click save on a script, then close the script window and the 'wanna save?' dialog box popped up, which never worked, almost like the powerup wasn't installed. Just opening an esp and clicking save would crash the geck. Again I have no idea why (believe me, I tend to want to solve this kind of stuff) but simply rebooting fixed the problem and I'm able to edit all the esp's and esm's again. I am of course using v7, same version as you guys, no 'dev version' here.
prideslayer Posted July 20, 2012 Author Posted July 20, 2012 I wonder if you guys got a corrupt v7 when the download (.7z) was messed up? Is your v7 a .7z or a .zip? I can't see that honestly being the problem since if the .7z was corrupt, the .dll wouldn't extract in the first place, just tossing it out there.
Chase Roxand Posted July 20, 2012 Posted July 20, 2012 7z. Has no problems extracting. Should I try the zip?
prideslayer Posted July 20, 2012 Author Posted July 20, 2012 It couldn't hurt, but I really doubt it will help. Might want to just try closing down the geck, fomm/wrye, and steam. Check the process list to make sure they're really gone, especially if you use FONV4GB which never actually seems to exit for me and, ending up with dozens of the damn things running if I forget.
Chase Roxand Posted July 20, 2012 Posted July 20, 2012 Will do in a few. (On a side note, I wish I could use FONV4GB. It doesn't work on W8.)
prideslayer Posted July 20, 2012 Author Posted July 20, 2012 Doesn't work for me in Win7 either, or at least doesn't seem to. I gave up on it month or so ago after fighting with it for about a week.
prideslayer Posted July 21, 2012 Author Posted July 21, 2012 This is really strange.. any suggestions from those having the problem, I'm all ears. I can't reproduce it here.
prideslayer Posted July 21, 2012 Author Posted July 21, 2012 Chase if you want to look in (or just attach for me to look in) the NVSE logfiles, maybe they'll have a clue.. nvse_editor.log, nvse.log, nvse_loader.log, and nvse_extender.log. All to be found in FNV directory once you've run the geck or the game once. They're cleared out and recreated each run. If you have any ending in '0' in there, that means the engine and geck were running at the same time, so just close everything, delete all of them, and run the geck again and try to make an edit.
prideslayer Posted July 21, 2012 Author Posted July 21, 2012 Should have taken my own advice.. something is amiss.. argh. plugin C:\Program Files (x86)\steam\steamapps\common\fallout new vegas\Data\NVSE\Plugins\\nvse_extender.dll (00000001 Extensions for NVSE 00000007) disabled, fatal error occurred while loading plugin
prideslayer Posted July 21, 2012 Author Posted July 21, 2012 FEH how did I miss this and why is it working here anyway? New version coming in a moment.
prideslayer Posted July 21, 2012 Author Posted July 21, 2012 v8 in the OP Fixed GECK issue. This one is a .7z again, and test download worked OK for me. Let me know if it goes haywire with the mime type again (gibberish in browser).
prideslayer Posted July 21, 2012 Author Posted July 21, 2012 This is kind of depressing.. I was saving V8 for a big powerful musclecar kind of release. Guess that will have to wait for V10 or V12.. and *then what*?
prideslayer Posted July 21, 2012 Author Posted July 21, 2012 9Beta is in the OP. This is not for general usage, just for testing. Confirmed it does work in the GECK before uploading. No new functions (betas will never have new functions, for now). 9Beta replaces the standard map class with the google sparse map class, which supports serialization and file save/load internally. Testing it out to prep for save/load in the extender, need to make sure all the map stuff still works right -- all the NX_Get/Set business uses this.
prideslayer Posted July 21, 2012 Author Posted July 21, 2012 I may be jumping the gun, but since I did once already, that's only fair. The save/load thing may be a little more problematic than I thought. MAY be. I forgot that the serialization code needs to understand how to fix up reference IDs if a game is saved, the mod load order rearranged, and then that save is loaded. FOSE does this automatically. The serialization interface may actually be intact in NVSE, I'm not sure, but I don't think it's there. This means I'm going to have to fix up the reference IDs myself when I load. Now there's a chance I can use some code from FOMM or something to poke around in the savegames. If so, this shouldn't be too difficult, since mod load order is stored in the savegames anyway so that the engine can do this itself. If not, I'm going to have to store the load order myself with "my" savegame, and then compare the two on a load. This too should be possible, but it's a bit more work than I'd anticipated.
prideslayer Posted July 22, 2012 Author Posted July 22, 2012 v9beta2 available in OP This is a beta and is not for general play or modding, just testing and evaluation. I've tested the new functions myself in game via the console, and made sure to open the geck, modify an NX using script, and save this time before uploading. All seems OK on my end, but due to the hiccup with the last release, I think betas are in order for NX prior to release, just like with sexout itself. Changes in v9b2 - The google unordered map code was removed, I won't be able to use it as I'd hoped anyway, so back to the STL. This means the code from the previous beta has been removed. - Two new functions: (ref.)NX_ClrEVFl "key" (opt), and (ref.)NX_ClrEVFo "key" (opt). Flip sides of the same coin, same as Get and Set. These functions will clear an entry in NX. You should call this if you're unsetting a value, rather than just setting it to zero, as this will free up memory. How to use: If the key is blank, ALL keys on the reference are removed. Doing this *will* break other mods, including sexout. Use only in extreme circumstances, and only from the console. Don't do this in a mod!! The "opt" is an optional integer argument that currently can take one of two values: opt == 1 : Removes the value only if the key match is exact. opt == 2 : Removes all values whos keys substring match the given key. Example: ref.NX_SetEVFl "test:a:a" 1 ref.NX_SetEVFl "test:a:b" 2 ref.NX_SetEVFl "test:a:c" 3 ref.NX_SetEVFl "test:b:a" 4 ; given the above vars, here is what the Clr function can do.. ; option 1, clear an exact key ref.NX_ClrEVFl "test:a:b" ; OR ref.NX_ClrEVFl "test:a:b" 1 ; option 2, clear all the keys starting with "test:a" ref.NX_ClrEVFl "test:a" 2 ; option 3, clear all keys on a reference. opt is ignored in this case. ref.NX_ClrEVFl "" - Modified the NX_Get* functions to do a count() first. This is just a c++ STLism to do with maps, and prevents memory from growing if an NX_Get* is called on an unset value. Both Get* routines return 0 if the key does not exist, just like before. To avoid breaking any mods that do use this in testing/beta, I will not be removing these functions, so they're safe to use -- but since this is a beta, if there are issues, *how* they work has an infinitesimal chance of changing.
Recommended Posts