Announcement

Collapse

Please use the Hentai ID thread for all hentai ID requests. Click me for link!

The Identification Thread is Here:

http://www.hongfire.com/forum/showthread.php/447081
See more
See less

Translation Aggregator

Collapse
This is a sticky topic.
X
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Translation Aggregator

    I'm no longer working on Translation Aggregator, but Setx has released an updated version, here. The files attached directly to this post are now outdated

    Translation Aggregator basically works like ATLAS, with support for using a number of website translators and ATLAS simultaneously. It was designed to replace ATLAS's interface as well as add support for getting translations from a few additional sources. Currently, it has support for getting translations from Atlas V13 or V14 (Don't need to have Atlas running), Google, Honyaku, Babel Fish, FreeTranslations.com, Excite, OCN, a word-by-word breakdown from WWWJDIC, MeCab, which converts Kanji to Katakana, and its own built-in Japanese parser (JParser). I picked websites based primarily on what I use and how easy it was to figure out their translation request format. I'm open to adding more, but some of the other sites (Like Word Lingo) seem to go to some effort to make this difficult.

    JParser requires edict2 (Or edict) in the dictionaries directory, and supports multiple dictionaries in there at once. It does not support jmdict. You can also stick enamdict in the directory and it'll detect some names as well, though the name list will be heavily filtered to avoid swamping out other hits. If you have MeCab installed, JParser can use it to significantly improve its results. TA can also look up definitions for MeCab output as well, if a dictionary is installed. In general, MeCab makes fewer mistakes, but JParser handles compound words better, and groups verb conjugations with the verb rather than treating them as separate words.

    TA also includes the ability to launch Japanese apps with Japanese locale settings, automatically inject AGTH into them, and inject its own dll into Japanese apps. Its dll can also translate their menus and dialogs using the ATLAS module (Requires you have ATLAS installed, of course). Versions 0.4.0 and later also include a text hooking engine modeled after AGTH. The menu translation option attempts to translate Windows-managed in-game menus, and is AGTH compatible. The AGTH exe and dlls must be in the Translation Aggregator directory for it to be able to inject AGTH into a process. AGTH is included with the most recent versions of TA.

    The interface is pretty simple, much like ATLAS: Just paste text into the upper left window, and either press the double arrow button to run it through all translators, or press the arrow buttons for individual translation apps. Each algorithm is only run once at a time, so if a window is busy when you tell it to translate something, it'll queue it up if it's a remote request, or stop and rerun it for local algorithms. If you have clipboard monitoring enabled (The untranslated text clipboard button disables it altogether), it'll run any clipboard text with Japanese characters copied from any other app through all translators with clipboard monitoring enabled. I won't automatically submit text with over 500 characters to any of the translation websites, so you can skip forward in agth without flooding servers, in theory. I still don't recommend automatic clipboard translation for the website translators, however.

    To assign a hotkey to the current window layout, press shift-alt-#. Press alt-# to restore the layout. Bound hotkeys will automatically include the current transparency, window frame, and toobar states. If you don't want a bound hotkey to affect one or more of those states, then you can remove the first 1 to 3 entries in the associated line in the ini file. Only modify the ini yourself when the program isn't running. All other values in those lines are mandatory.

    Pre-translation substitutions modify input text before it's sent to any translator. Currently applies to websites, ATLAS, Mecab, and JParser. There's a list of universal replacements ("*") and replacements for every launch profile you've created. I pick which set(s) of substitutions to use based on currently running apps. Note that you do not need to be running AGTH or even have launched a game through TA's launch interface for the game to be detected, but you do need to create a launch profile. May allow you to just drag and drop exes onto the dialog in the future.

    MeCab is a free program that separates words and gives their pronunciation and part of speech. I use it to get the information needed to parse words and display furigana. If you have MeCab installed but I report I'm having trouble initializing it, you can try copying libmecab.dll to the same directory as this program. Do not install MeCab using a UTF16 dictionary, as I have no idea how to talk to it (UTF16 strings don't seem to work). Instead, configure MeCab to use UTF8, Shift-JIS, or EUC-JP. If you have both MeCab and edict/edict2 installed, you can view a word's translation in MeCab by hovering the mouse over it. Also, JParser can use MeCab to help in parsing sentences.

    JParser tends to be a better choice for those who know almost no Japanese - it tells you how verbs are conjugated, handles some expressions, etc. MeCab may well be the better choice for those who know some Japanese, however.

    Source, attached below, is available under the GPL v2.

    Thanks to (In alphabetical order, sorry if I'm leaving anyone out):
    Hongfire Members:
    Freaka for his innumerable feature suggestions and reported issues over the course of development.
    Setsumi for TA Helper and for all his suggested improvements and reported issues, particularly with JParser.
    Setx for AGTH.
    Stomp for fixing the open file dialog not working properly on some systems and adding the tooltip font dialog, and fixing a bug that required admin privileges when certain other software was installed.
    Might sound like minor contributions, but feedback really drives the development of TA.

    Non-members:
    KingMike of KingMike's Translations, who is apparently the creator of the EUC-JP table I used to generate my own conversion table.
    Nasser R. Rowhani for his function hooking code.
    Z0mbie for writing the opcode length detector/disassembler I use for hooking. Apparently was intended for virus-related use, but works fine for other things, too.
    And the creators and maintainers of edict, MeCab, and zlib.

    You might also be interested in:
    *Setsumi's TA Helper and AGTHGrab.
    *errotzol's replacements script.
    *Devocalypse's devOSD.
    *kaosu's ITH (Like AGTH. No direct TA support, due to lack of a command line interface, but definitely worth checking out).

    MeCab
    edict2

    Changelog:
    0.4.9
    * Fixed MeCab/JParser getting stuck when starting a new translation before the last is fixed.
    * Fixed interface lockup while mousing over an item in MeCab while JParser is running.
    * Menu translation will now translate column headings in ListViews (Needed this for the AA launcher)
    * Fixed ATLAS config crash.
    * Global hotkey support. Toggle under "File" menu (Tools is kinda big already). Currently only really supports history navigation. May add more later.

    0.4.8
    * Added history. Logs both original text and translations (For online translators). It logs up to 20 MB of original text, and whatever translations are associated with it. Currently only way to force a retranslation is to toggle one of several options (Autoreplace half-width characters, src/dest language, modify substitutions).
    * Fixed deadlock bug on MaCab mouse over while JParser is running.
    * Fix corrupting built-in text hooker settings when launch failed. Suspect no one uses this, anyways.
    * Drag/dropping an exe onto TA to open up the injection dialog now actives TA.

    0.4.7
    * JParser and MeCab each use their own thread (Mostly).
    * Changed conjugation table format to JSON - plan to do this to a lot of other files (Being careful not to mess up game settings or substitution tables). Currently have way too much file loading code.

    0.4.6
    * Fix WWWJDIC
    * Fix closing injection dialog
    * Updating process list 10+x faster
    * Process list autoupdates
    * Fixed bug that would result in injecting into wrong process when one program is running multiple times.
    * Updated included AGTH version

    0.4.5
    * Added bing support.
    * Updated Honyaku code (They didn't try and block TA, they just modified their HTML)
    * Fixed AGTH command line code.
    * Replaced "/GL" with "/SM" compile option, resulting in faster builds when one has a lot of cores.

    0.4.4
    * Regular expressions are now compiled
    * Injection validation when using addresses relative to dlls (Or function addresses in dlls) should be fixed.
    * Added option to create shortcuts. They'll launch TA (If it's not running) and try to launch the game using the current injection settings (Injection settings that you'd get at the launch screen - the current settings are not saved - it always uses the most recently used ones).
    * Appropriated some of Setsumi's code to make tooltips larger.

    0.4.3
    * Multiple subcontexts now supported. Separate them with semi-colons. AGTH code converter will add two subcontexts, when appropriate.
    * Using aliases for hooks added. Prefix a hook with "[Alias Text]" and that's what will be displayed on the context manager screen as the hook's name. Makes it easier to see context strings.
    * Locale selection added to injection dialog.
    * "Hook delay" added to injection dialog. Actually doesn't delay hooking, delays how long before hooks that use filtering based on calling function's dll are enabled. Generally only the default hooks do this. Increasing this delay may circumvent issues with games that crash when launched with AGTH, but work fine when injected after launching.
    * Added "!" and "~" operators.

    * Stomp's admin privilege fix when using some 3rd party software added.
    * Excite fixed
    * Fixed sanity testing for injection addresses, so when specify a dll or exe name in a text hook, shouldn't erroneously think it's an error when the module isn't loaded in the current address space.
    * Fixed some JParser dicrionary common word parsing, when using versions of edict with entL entries. Also changed treatment of Kanji entries when only their corresponding Hiragana are marked as common.
    * Fixed substitution matching Hiragana with Katakana and vice versa.
    * Fixed a clipboard-related crash bug.
    * Fixed hooks causing crashes when relocating call/jumps (Hopefully...)
    * Fixed AGTH repeat filter length placement (oops).

    0.4.2b
    * Fixed substitution loading/deleting.
    * Fixed << and >>.

    0.4.2
    * AGTH code conversion tool.
    * Injection code checker added.
    * New child process injection handler (Really nifty injection code for that...). Should be a little more robust than before.
    * Option not to inject into child processes added.
    * Auto copy to clipboard added.
    * Both extension filters fixed.
    * Both eternal repeat filters fixed/upgraded.
    * Phrase repeat filter fixed/upgraded.
    * OpenMP/MSVC 2008 SP1 runtime requirement removed
    * char/charBE fixed
    * GetGlyphOutline fixed
    * Copy to clipboard crash when auto translate disabled fixed.
    * Slightly improved dll injection error handling.

    0.4.1
    * More context/filter options.
    * Repeated phrase filter now handles cases where phrase is being extended by a couple characters each time (xxyxyz, etc). Extension filters no longer really needed, unless the repeat starts out too short.
    * Option to handle eternally looping text.
    * Option to ignore text without any Japanese characters.
    * Text which substitution rules reduce to nothing no longer overwrites translated text.
    * Log length limit added.
    * Options to manage default internal text hooks added.
    * Clipboard treated as a context. Its default settings should mirror the old handling.

    0.4.0
    * Added it's own text hooking engine. Probably still buggy.
    * Fixed excessive redrawing when a hidden furigana window had clipboard translation enabled.
    * Works with new, even more poorly formatted edict files.
    * Handles EUC_JP characters that Windows does not (Doesn't use them properly with WWWJDIC at the moment, however). Only really fixes loading edict files with those characters.
    * Fixed right clicking when full screen.
    * Fixed not checking auto Hiragana mode.
    * Less picky when reading MeCab output.
    Attached Files
    Last edited by ScumSuckingPig; 07-11-2015, 11:20 AM. Reason: Change download link, re-upload attachments upon request from Setx

  • You probably just installed in a path without Japanese characters, or are using Japanese region settings so didn't run into the issue I first encountered.

    Anyways, text is only passed to the translators once it's stopped coming, like with AGTH. As the game almost never stops displaying text, that isn't too useful. Need an option to detect endlessly repeating text. Also, my "Extending repeats" option doesn't work because it adds two characters at a time, rather than one. Good game to test on, as it gives me a chance to work on both issues.

    If you reduce the idle delay to say, 20 milliseconds, things get fun. Actually, might be a good idea to not translate if the current string matches the last, and presto - instant hack for that game.

    Comment


    • I'm using a native jp WinXP to minimize any issue.

      The /KF option in agth handles those endless repeats, it does have some issues in practice with finding the correct sentence beginning, but worked pretty well with this game. 20ms is unstable for me, at 40ms it always produces the same result. It does filter out double characters though.

      Comment


      • Curious. I didn't know /KF dealt with eternally repeating text. I apply filters only after I have text to actually filter, so I'm basically going to have to make a separate function to handle it. Advantage of applying filters later is that if you have something occur twice in a row with different numbers of characters, like "GoatGGooaatt", my method handles it. AGTH handles that case by using a simpler text filter than I do (Remove all repeats up to k characters, which can be applied to text immediately upon receiving it. While I do the same if you supply a single value, my "Constant Repeat", "Variable Repeat", and "Custom" (with multiple values) all do something a bit more complicated. AGTH's method can't handle things like "HellHHeellll", in general.

        "Constant repeat" finds the largest single repeat for an entire string, so it handles double characters without any additional configuration. Only fails when the number of repeats varies over a string, or when a string consists entirely of characters that actually are doubled (Or tripled, whatever). Latter case is pretty rare. Even rarer is the latter case when it actually matters. Former case isn't too unusual, unfortunately.

        "Variable repeat" finds the largest repeat > 1 that works for the longest initial region of the string. Then it looks for another subsequent one. It has trouble with transitioning between two values, and cannot switch between values that are multiples of each other (like from 2 to 4, though it can go from 4 to 2) or with values that share a common factor (if given repeats of 4 and 6, it would note that 2 works quite well). First issue is a special case of the latter one, actually.

        "Custom" with multiple values basically does what variable repeat does, only it only checks a couple values. It has issues with the same things, as well.

        With a single value, "Custom" works just like /KS does for AGTH.


        I really want to improve on my variable repeat code to be able to handle most funky cases with minimal configuration, and it can't work without the entire string.
        Last edited by ScumSuckingPig; 05-07-2010, 03:40 PM.

        Comment


        • Endless loops is more or less the only situation where I'd use /KF. If the amount of loops is limited, it's usually possible to find a better hook, although it would be even better to get that handled without hook ofc. I would say "GoatGGooaatt" is pretty unusual in practice. What I believe can be a bigger issue is name + dialogue were only one of them is looping. Which reminds me that this case and the former case mentioned about finding a hook usually rely on using the subcontext split method which I believe you do not support? Another case were this is often needed are Furigana. Without a special split they're often mixed into the normal text, which breaks atlas etc.

          Comment


          • subcontext split? You mean specifying a subcontext via a custom hook? I allow specifying a context, and split on that instead of the context. If you need to split on both, can just add the two values together (_0 + [subcontext]). Still have yet to test my custom hook code, of course, as I've been lazy about getting games with known AGTH hooks to test with.

            In terms of handling name+dialog, where only one is looping: If the looping one is second, the code I'm working on can handle it. If it's dialog:name:dialog:dialog:dialog, etc, you'll always get dialog:name:dialog. If you do split and merge them again, currently there's no guarantee of the order in which they'll be merged. It'll probably be consistent, but it depends on a couple things, including order in which I check them. Really need to think about that. Should probably keep a queue based on when the last text was received, and check merged contexts last.

            Think they can even end up being split (That is, one is moved to the merged context and translated before the other makes it), if I checked the merged context before the individual contexts. This may happen when one of the contexts receives more text before the other one and before the idle time + 200 milliseconds have passed, and the idle time is set to less than 200 milliseconds, such as in the infinite display loop case. Keeping track of the order in which the last data was received will fix this. Handling infinite loops in subcontexts plus this fix should result in handing the text:name:text:text case as well.


            Edit: Also, I've seen "GoatGGooaatt" a few times. I believe the first is generally used to layout the entire string, and the second for display and shadowing. Or maybe it was "GGooaattGoat", and the first was displaying just the shadow.
            Last edited by ScumSuckingPig; 05-07-2010, 08:06 PM.

            Comment


            • Looked again at your docs, yeah you got it. I missed that example: char:_4:_0x48@4180A0 //HAN4:48@4180A0 (Type:String:Context:Length:Charset@address:module : function) What you call context is subcontext in agth, that's probably what got me confused.

              Maybe the best would be to work from game to game regarding the loop handling, at least to me splitting+merging+filtering sounds like quite a task to do it only on theory.

              Comment


              • AGTH sorta uses "threads" and "contexts" interchangeably, which is what got me confused and using different notation. Well, more accurately, it uses "thread" to refer to the text from a context/subcontext pair. I haven't fooled lately much with AGTH /h codes, so was thinking AGTH used thread:context:subcontext, with context being the reference to the calling address, and subcontext being only for custom hooks, rather than subcontext being used for calling address in default hooks, and a customizable value with custom hooks. I'll change my terminology to match, at some point. I'll just ditch "Threads" entirely, most likely, and stick with context/subcontext.
                Last edited by ScumSuckingPig; 05-08-2010, 09:43 AM.

                Comment


                • how to use it?

                  maybe i missed something but dont kow how to launch games with the translator, wanna play badly jiburiru 4 but i cant, dont know too much japanese. im supposed to have ATLAS or something? thks

                  Comment


                  • http://www.google.com/#q=site%3Ahongfire.com+atlas
                    http://www.google.com/#q=site%3Ahongfire.com+agth

                    Comment


                    • Another experimental build. Fixed a couple things in the context manager's interface, added aggressive extension filter and a looping text handler. Also removed text replacement hooks (Menu translation is still in, of course), to shrink the dll a bit. Seems to have fewer issues when the dll is smaller.

                      The basic extension filter only handles 1121231234. The aggressive one handles adding more characters at once (Even if the number varies between repeats), starting with more than one character, adding fewer characters at a time (1212123), and combinations thereof. It almost makes the standard phrase repeat filter unnecessary when enabled.

                      Extension filters are needed for two reasons: First, my repeat phrase filter looks for exactly repeated phrases, and when it finds them, it starts at the end of the final repeat (Starting at the start would cause it to handle much of what the extension filter handles. May make this change in the future), and second, it handles short repeats, as long as they're followed by longer ones. Regardless, best to set games to instantly display all text when possible.

                      <Removed, get 0.4.1 instead>
                      Last edited by ScumSuckingPig; 05-12-2010, 10:55 PM.

                      Comment


                      • In アトラク=ナクア text flashes and disappears replaced by constantly blinking triangle cursor.
                        Spoiler








                        ▽▽▽「…ふふ」



                        さなかに立つ女が、冷たく赤い双眸を細めて微笑った。









                        I'd like very much to have possibility of using text filters with regular clipboard translation too.
                        読んでみた。 読んでみようとした。 読めたらいいな、と思った。

                        Comment


                        • Hmm...Have to think about a general way to fix that.

                          As for applying filters to other text - I've been thinking about treating the clipboard as its own context (With a dummy process). That would allow applying filters to the clipboard, as well as a (relatively) simple way of disabling the clipboard while keeping process translation enabled. Setting the idle time to 0 (Which would be default, for the clipboard) with no filters would result in basically the same behavior as now. Could also add an "Ignore on non-Japanese text" option. Two nice things about that - would fix, to some extent, that repeating cursor (▽▽▽「…ふふ」 would still have the cursor attached, though a single substitution rule would get rid of that - should probably ignore text when substitutions reduce it to nothing, come to think of it), and would allow auto-copying from clipboard when using TA for a source language other than Japanese. As I currently ignore linebreaks from contexts, would also need to make an option to keep them, which is probably a good idea, anyways.

                          Allowing translating to be disabled by process is something else I'm considering.

                          Also be nice to add regexp support to my substitution rules.
                          Last edited by ScumSuckingPig; 05-10-2010, 10:42 PM.

                          Comment


                          • 0.4.1 released. Basically just more enhancements related to the new hooking engine (Or, more accurately, the context manager). Also, the clipboard is now treated as a context. Note that for whatever reason, the latest version is in the middle of the attachment list of the first post. As I don't want to remove older versions, I'm not going to add/remove attachments to sort them. The latest file, however, is also linked at the top of the first post and can also be downloaded from this link.

                            Anyone looking at the source will also notice that I reorganized the files a little, though could still do a better job of it.


                            Anyone wondering about the line break options - some games send all line breaks, so you'll get something like:
                            Name
                            line 1
                            line 2
                            line 3

                            Current options allow keeping just the first line break. Not sure how useful the option really is, as such games don't always have a name before every set of lines, so can just end up with lines 1 to 3, in which case you wouldn't want the first line break. May just switch to a single toggle (Keep/remove line breaks). Or could add the option to "Keep all line breaks in the first x characters", which may handle most cases where there are linebreaks that would be good to keep.

                            Note that my context manager currently doesn't display linebreaks within a block of text. This is (kinda) a new bug with 0.4.1, and will be fixed next release.

                            Edit: Forgot to set the clipboard character/phrase filters to none, by default. Not going to fix that until the next release, however. Not a big issue, since they'll generally find nothing to filter, anyways.

                            Edit 2: Filters are really easy to write, and I'm toying with the idea of letting plugins add new ones. Only problem with this is that then I'd need to switch to drop down menus to select filters, which isn't particularly appealing. A plugin could currently modify a string "pre-substitution" with a filter, but integrating plugin filtering options into my context filters configuration seems a better solution. That having been said, if anyone comes up with an improved fairly general purpose filter, probably be best to add its source directly to TA, anyways.
                            Last edited by ScumSuckingPig; 05-13-2010, 01:19 AM.

                            Comment


                            • I think I know why there are no drives displayed in "My Computer" in the "Launch Program" dialog. I don't know C++, but I did a Google search and found these two links.

                              This one http://www.eggheadcafe.com/software/...-computer.aspx

                              "Make sure you are not calling CoInitializeEx with argument other than
                              COINIT_APARTMENTTHREADED.

                              If COM is initialized on the current thread with COINIT_MULTITHREADED mode,
                              the shell functions (including the file dialog) will not work."
                              And this one: http://www.codeproject.com/KB/dialog...08#xx2582508xx

                              "My problem was described in Microsoft Knowledge base article 28707.

                              Basic issue is that common dialogs dont work in a multithreaded apartment - i.e. if you call:
                              CoInitializeEx(NULL, COINIT_MULTITHREADED)

                              on the same thread as CFileDialog or XFolderDialog you'll see this behaviour...

                              Interestingly enough the problem went away on Vista - but did affect XP and 2000."
                              AGTH wiki

                              Comment


                              • So I tried the 0.4.1 version. Trying to use the internal hook support. Game I tried is a kirikiri engine based, and I noticed it doesn't support autodetection for it yet. Maybe you gonna implement it in the future?
                                Then I tried to convert agth code into TA's format, but don't know how to do it... ^^;;;;
                                The general agth code for kirikiri is /HW-4*14#1@[thread address] or /HW-4*14:-4*0#1@[thread address], what I got is wchar:[_EAX+ESP]@[thread address], but it seems it didn't work....
                                Can someone willing to convert the hook code for me?

                                However, even using the standard output (like GetTextExtendPoint32W), it's actually working if the filter is enabled (all repeats, aggressive, auto). Good job SSP.

                                @Setsumi
                                After TA ver 0.4, the TAHelper's window detacher seems not clearing text before displaying the next text (in JParser or Mecab window), the result is overwritten text on the background, making the text hard to read. Can you fix it ? Thanks

                                Comment

                                Working...
                                X