Mator

Merge Plugins renumbers form IDs. It's necessary when there would be a formID conflict.

A screenshot doesn't help much, unfortunately. If you tell me what mods you're merging I can investigate it at some future point in time. I think you should be able to find some kind of guide somewhere which explains how string files work. My initial thought is that they may be file-specific assets, meaning if your plugin was named MyMergedPlugin.esp, the string file would have to be named MyMergedPlugin_en.strings to be loaded by the game (that would be the english strings file, replace _en with whatever language suffix is used on the string file currently).

What game are you merging plugins for? I don't honestly know exactly how string files work. How does the game decide to load a string file, and how are IDs assigned to items? If you can figure this out you should be able to answer your question on your own. A merged plugin is just a plugin, so if you know how strings files work for plugins in general, getting them to work with a merged plugin should be trivial.

You don't need to keep the ESPs you merged active or installed. I'd recommend keeping them available to some extent so you can reactivate them when you need to rebuild your merge, but you certainly don't need to include them in the merged mod archive.

Nick just let me know in Discord that DynDOLOD.esp is the 0xFF (255th) ESP in his load order. I'm 99% certain that's the issue. Said he's going to try merging some stuff.

If it was a local piece of malware the browser you use wouldn't matter. I'm fairly convinced it's not a local piece of malware though, and that it's something that inserted itself into the IPB assets via some kind of vulnerability in IPB per drudkh's previous posts.

Glad it worked!

Unfortunately Merge Plugins doesn't work properly with BA2 files. It's a bug. It should extract them for you when you check that, but it doesn't.

Extract the BA2s associated with the mods you merged.

I don't have server access either, so it's not something I can address at the moment.

Great find. I actually saw some base 64 encoded stuff in the STEP code when I was looking around, though I think what I saw was probably just part of IPB. I think this is quite likely to be the exact issue here though. EDIT: This article which is linked from there is much more in-depth, definitely something we should do to investigate this Tech & Z: https://peter.upfold.org.uk/blog/2013/01/15/cleaning-up-the-ip-board-url4short-mess/

Not sure how your DLL would get corrupted like that, you should scan your machine for malicious software to be safe. Generally speaking, if a file is getting corrupted like that it means you have something funky afoot. There's the unlikely situation where atmospheric noise flipped a bit on your hard drive platter, but that's not the kind of thing that happens repeatedly. DLLs don't get rewritten by any normal system processes, but are common vectors of attack by malicious software (which can binary-patch inject themselves into the DLL so their code executes when the DLL is loaded). In this case it sounds like some malicious software is trying to binary patch into the DLL but was built for a different version of the DLL.

Funny, that's exactly what he used/uses! I personally use Avast and Spybot S&D, but I haven't done much research lately into detection rates. EDIT: Here's the most recent report I could find about detection rates, though I'd like to find more reports. Always good to have multiple sources. But this reflects what I had found previously several years ago. Though MalwareBytes AntiMalware is incredibly popular, its detection rate for malicious software is sub-par. That doesn't mean you have to buy premium antivirus software, there are many other free solutions which provide better protection. I personally use Avast free. I've previously tried several free AV solutions including AVG and Avira. I found AVGs interface and nagware unworkable, so I removed it. Avira bricked my system at one point (potentially partially my fault for choosing to quarantine a file which was in System Volume Information), so I stopped using it as well. Anyways, what I'd really like to do is remote into a machine which is affected by this, crack open browser developer tools, and systematically step through the JavaScript/network actions to find the source of the redirect to the malicious site(s) when opening STEP. From here I can assess whether or not the JavaScript is getting injected into the browser from software on the local machine or is actually coming from STEP itself (hopefully).

It sounds to me like this has to be some kind of resource problem combined with one or more other problems. NickTheDuke: can you run Skyrim Performance Monitor to see if we can get a look at Skyrim's RAM/VRAM usage? It sounds to me like some kind of buffer is overflowing, could be Papyrus string count or something else. I really don't know. But based on what I've read here I don't think there's a single record or even a single cell which is causing the issue. We could perhaps show this to be the case if we can get two stable builds which have opposite mods enabled/disabled. If we can show the game is stable with selection X enabled and Y disabled, and is also stable with X disabled and Y enabled, it will show with certainty that the problem is not in any specific record (or even any specific model). Just my thoughts, anywho. Sheson is of course the true master of all things DynDOLOD. :)

The thing is that multiple users have reported this behavior when visiting the STEP site, making me think there's something afoot. But if we can't recreate the issue and get some diagnostic information we won't be able to fix it (and it is most likely not related to STEP).

I'll make sure to set SSL up, it's a priority, in my opinion. Also, midst*, not mists. :)

Thanks for posting this here. I'm unable to recreate this issue on my machine (testing with multiple browsers and all protection disabled), but I'll test on some other computers to see if I can get different results.

I hadn't thought of this yet, but it's definitely something that could be built.

You don't select the archive from merge plugins, you extract it then select the .flg file in the Scripts\Source folder, per the instructions in the docs and GP's video.

To extend the analogy, the operating system acts like your mother coming over and throwing away all the old cereal boxes. :)

v0.2 released, see the updated OP for details!

what are the full paths of the asset files? if they're loose assets and aren't using file paths that are generated by the engine based on plugin name/record formID, they should "just work".

Most likely an issue with handling BA2 files. Try extracting them prior to merging.

jdsmith reached out to me to make this script, and I threw it together. Here ya go: let npcs = xelib.GetRecords(0, 'NPC_'); let data = []; let shouldSkipNpc = function(npc) { return xelib.HasElement(npc, 'TPLT') || xelib.GetFlag(npc, 'ACBS\\Flags', 'Is CharGen Face Preset'); }; npcs.forEach(function(npc) { let winningNpc = xelib.GetWinningRecord(npc); if (shouldSkipNpc(winningNpc)) return; let npcName = xelib.FullName(winningNpc); if (!npcName) return; let isFemale = xelib.GetFlag(winningNpc, 'ACBS\\Flags', 'Female'); let winningMod = xelib.GetElementFile(winningNpc); data.push({ name: npcName, gender: isFemale ? 'Female' : 'Male', formID: xelib.Hex(xelib.GetFormID(winningNpc, false, true), 8), winningMod: xelib.GetFileName(winningMod), comments: '' }); }); let headers = ['NPC Name', 'Gender', 'FormID', 'Winning Mod', 'Comments']; let output = [headers.join(',')]; data.forEach(function(npc) { output.push(Object.values(npc).join(',')); }); output = output.join('\r\n'); fh.saveTextFile('exported_npcs.csv', output); Appears to be working for me, though there will be some NPCs included which you may not want to include. I tried to exclude as many undesirable NPCs as possible be excluding ones that used templates, didn't have a FULL name, or had the Is CharGen Face Preset flag set. Here's the output of a run with the current zEdit alpha build on just Skyrim.esm + DLCs: https://puu.sh/xHuY7.csv

Who, me? It's Sheson's code, so I don't have much to say. What's on line 305 in the userscript unit? If we can answer that we can start to figure out what the problem is.