Jump to content

BEncode Editor


Ultima

Recommended Posts

Yuck. Thanks for the report. I was cutting corners just slightly when expanding environment variables, which caused the bug. It affects more than just files whose names begin with '%%' -- it affects all paths that contain consecutive '%' characters.

Anyhow, it's fixed, but I think I'll hold off on a release for now (unless it's urgent to you...?).

Link to comment
Share on other sites

Any time's fair game for requesting features... Doesn't necessarily mean I'll implement every feature I see, but I will consider.

How or why people use BEncode Editor? Well, I couldn't imagine there being too many uses other than simple edits to .torrent files, editing µTorrent *.dat files, or the settings files for various other softwares that make use of BEncoded data (like Vuze, for example).

Link to comment
Share on other sites

  • 3 weeks later...

Not really. You'll have to contact the developer of you antivirus software, tell them to analyze the executable, and fix their false positive.

Edit: Hm. Checking something...

Edit: Please redownload the v0.6.0 Unicode build (clear browser cache and all)?

http://www.virustotal.com/analisis/e52d68b5e557c7a89f3b7abe48e0bbc4

That's for the updated executable. I think I should start including the hashes in my release postings to make sure things haven't changed.

That was a rebuild of the exact same source, yet, the results are different. When I scanned the old v0.6.0, it also came up with the same result as you saw. I'm quite certain it was a false positive, but there you have it: a reupload that at least doesn't get scanned as a trojan >_>

3 of the scanning engines still claim it's suspicious, but well, at best, you'll have to take my word that it doesn't do anything even remotely suspicious (however little my words are worth) :/

Link to comment
Share on other sites

I think I'll just post v0.6.1 with the path fix...

Download

[v0.6.1] Unicode, ANSI

Changelog

v0.6.1 (2008-09-18)

* Fix: Paths with consecutive '%' characters fail to open

VirusTotal Results for BEncode Editor v0.6.1 ANSI

File size: 261027 bytes

MD5...: a76ada340a1e507ebc878613bf69b32d

SHA1..: a2663f17fa1d68934d1d01a28ba4decc00b26287

SHA256: 09a6e7302aa384f85f61e61081680ac313fc6dcff578886602cbe07d833e617b

SHA512: ad9dcbe3009e81330ff36b07fe360d51ae226d496aedba7a099336d1f723a433 903342c3c1fe3f1e3e26877e875b84aa14110cf4a449efe79d9471d31f24084

VirusTotal Results for BEncode Editor v0.6.1 Unicode

File size: 288675 bytes

MD5...: 91bc05e96e946c4ced8c4d75b11a9540

SHA1..: 20426db350ea443ecd19719a9dd2d427a4af6072

SHA256: ea9f694edb145582b1e4474a90229863089612af6da1bf15ed915fc43f64cbf5

SHA512: b1481531ea87ce8aaf0443a0175a027475b35a27a9425f250296bbca406bd3d9 7e6a9885d7893119c76b107d9607a196562cc1916186f8022c4671fab6ed5874

Note: I've posted an x64 build, but it has never been tested. I don't have a 64-bit copy of Windows. Play with it at your own risk!

VirusTotal still reports possible malware, but they're false positives. Nothing I can do about that.

Link to comment
Share on other sites

  • 2 weeks later...

Is there any way to use this to combine 2 resume.dat files from seperate computers into one file? I can edit the paths off the old computer ok with it, but obviously I dont want to go reloading 1300 or so torrents on the new one that already has about 900 in its utorrent already.

Link to comment
Share on other sites

Ok, so I am trying to do this,

How do I import one file into the other? I have 2 copies of bencode editor open with the old file in one, and the currently used one in the other, and copy and paste doesnt work between them, and there is no option to import a file in the other.

Also in the main screen of one of the files there is a path to the torrent, the other doesnt have the path infront of the filename.

If I cant copy or paste them, I cant select all of them with a ctrl-A or shift clicking anyway, how am I able to copy the entries over?

Link to comment
Share on other sites

You can "import" another file by adding raw BEncoded data from file. Merging dictionaries/lists isn't easy currently. I do want to get merging in somehow, but right now, I'm trying to refactor my code (it's becoming difficult to add some of the outstanding feature requests the way I currently have things implemented).

Link to comment
Share on other sites

The 64-bit build has a weird bug:

th_bencode_editor_64-bit_bug.png

(Click image for full size)

Looks like it's missing the resource data except dialog resources of course or maybe there are format incompatibilities.

I could probably help you with the 64-bit build if I had the sources available.

Got Visual Studio 2008 installed on Vista Ultimate x64

Link to comment
Share on other sites

I still dont get how I am supposed to get the content of one file into the other file, that post just seems to be showing how to move things around within the file.

I tried using notepad to copy and paste the entire content over but that just broke it, so I am really not sure what the process is to combine these 2 files.

Link to comment
Share on other sites

Cutting/copying/pasting works only on the same file and in the same instance of BEncode Editor. I use an internal clipboard, not the Windows clipboard. Do keep reading this post, though. You'll find something interesting.

@saivert: Ugh, I realized why the x64 version had no resources >_> ResHacker is 32-bit only (duh! xD). At this point, I don't feel like working out x64 issues, so I guess that line of builds has already had its life cut short :P Given a bit more motivation and time, I might get back to it.

Onto something more interesting...

BEncode Editor v0.7.0.0 WIP 1

Download: http://ultima.utorrent.googlepages.com/BEncodeEditorv0.7.0.0wip1.zip

Changes:

v0.7.0.0 (2008-??-??)

+ Feature: Merge binary/dictionary/list data

+ Feature: Undo/redo changes

~ Change: Added F5 for reloading the treeview, Ctrl+R for reloading from disk

~ Change: Allow blank key names when searching using "Value by Key"

~ Change: Binary input/conversion made slightly more lenient/reliable

~ Change: Move up/down keyboard shortcuts changed to Ctrl+Up/Down

~ Change: Show "Find" dialog if never shown before "Find Next" is used

* Fix: Repeated successive conversions into Binary allowed in Find/Replace

* Fix: Switching data types not properly hiding/showing controls in dialogs

* Fix: UI "allows" child to be added to an integer, but crashes on attempt

Notes:

* Dictionary merging does not sort keys or resolve duplicate keys

* Undo/redo only remembers up to 10 changes

Increasing the undo/redo change history size is completely trivial, but I feel like 10 changes should be more than enough.

PLEASE regression test everything (and of course, test the new features as well)! There was a lot of refactoring involved in this version in order to accomodate undo/redo, and while I've tested as best as I can, I can only do so much.

Speed most definitely wasn't a priority in this build, and this will be slower than v0.6.x only because I switched to an older compiler to maintain Wine support. I can make a separate compile just for Wine (and leave the faster build for native Windows 2000+ users), but I don't care enough to decide at this point (trivial to do anyway).

Edit: Oh, uh, I kinda removed commandline loading in this build and forgot to re-add it. I was thinking of how best to implement commandline options. Don't feel like rebuilding/reuploading, so live with it :P

Link to comment
Share on other sites

I know you probably don't want to hear this, but "speed most definitely wasn't a priority"... any numbers to back that up on your end. In any case I know at least one person who'll want to poke at it and give you data.

moogly, WIP = work in progress. As far as Ctrl-A in Post #131 it's mentioned no accelerators work in child dialogs.

Link to comment
Share on other sites

The accelerator+child dialog thing should no longer be an issue. Regarding speed...

v0.6.1.0

[20:16:24] [bDecode] Start: "D:\Desktop\tor\resume.dat_j"

[20:16:38] [bDecode] End (14.633 seconds)

[20:16:38] [TreeView Fill] Start

[20:16:54] [TreeView Fill] End (15.304 seconds)

[20:17:32] [Find/Replace] Made 4226 replacement(s) in 25.443 second(s)

v0.7.0.0 WIP 1 (older compiler)

[20:05:22] [bDecode] Start: "D:\Desktop\tor\resume.dat_j"

[20:05:39] [bDecode] End (16.552 seconds)

[20:05:39] [TreeView Fill] Start

[20:05:54] [TreeView Fill] End (14.604 seconds)

[20:08:29] [Find/Replace] Made 4226 replacement(s) in 27.178 second(s)

v0.7.0.0 WIP 1 (newer compiler)

[20:04:40] [bDecode] Start: "D:\Desktop\tor\resume.dat_j"

[20:04:54] [bDecode] End (14.618 seconds)

[20:04:54] [TreeView Fill] Start

[20:05:09] [TreeView Fill] End (15.149 seconds)

[20:07:04] [Find/Replace] Made 4226 replacement(s) in 25.453 second(s)

v0.6.1.0 used the newer compiler (which broke Wine support). Overall, v0.7.0.0 is actually slightly faster given the same compiler, but I was referring to this build specifically being slightly slower. And that I haven't looked at the possibility for any further speed optimizations at all.

Interestingly, the builds produced by the newer compiler are slower at treeview filling given the same code. Not sure what I can do about that.

Link to comment
Share on other sites

Ahh thank you, but you presume incorrect. Actually, I was referring to get jewelisheaven the news ;) That's quite a respectable speed hit if indeed you did re-factor code to get Undo working. :D

I'm going in circles so I'll ask, I tried sifting through several pages of hits for the launch error I'm getting in WINE 1.1.5 of OS X (after using the guide made for uTorrent). Have you installed it before? It's not loading with "err:module:attach_process_dlls "gdi32.dll" failed to initialize, aborting" I've cleaned ~/.wine twice with both the .mpkg and .dmg . If I can't get this running with WINE, it certainly saves having to boot up VMWare Fusion.

My main attempt is wanting to continue with some standard test files for benchmarking, similar to how uTorrent "works on as weak a system as Windows 95 on a 486 with 14MiB of RAM" I'd like to organize data to show that performance isn't as necessary. Surely http://media.xiph.org/ED/ or the 4200 torrent resume.dat are valid cases but for most people I think seeing an average 32000 piece torrent with 100 files load within X % whether you're running that old 486 or multi-core CPU is a positive commentary of your work.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...