Dear Alex, I am glad to hear that my input might have been of some help to you. In the meantime, I found the bugzilla entry of the Mozilla Firefox/Thunderbird bug which caused the kinput2-wnn input frontend misbehaved for some time (2+ years). https://bugzilla.mozilla.org/show_bug.cgi?id=1525867 (Inputting Japanese text using kinput2 frontend that uses XIM protocol under linux no longer works well with FF/TB when on-the-spot mode is used.) The problem was only with Mozilla firefox and thunderbird. In the above bugzilla entry, you can find a few screen dumps that show how kinput2-wnn frontend handles input conversion. Like this one https://bugzilla.mozilla.org/show_bug.cgi?id=1525867 and this one https://bugzilla.mozilla.org/show_bug.cgi?id=1525867 BTW, I used to use OS/2 from 1995 (OS/2 Warp) to 2001 or 2002 when I switched to Debian GNU/Linux completely. Good luck with your project to create OS/2 Japanese input front end. Best, Chiaki PS: Hmm, come to think of it, Japanese version of OS/2 used to come with a Japanese input front end developed by IBM Japan, I think. However, I threw away all the OS/2 related magazines, etc. a few years back. The graphical interface worked in a similar manner by highlighting the conversion target somehow. On 2019/03/26 5:42, Alex Taylor wrote: > On Mon, 25 Mar 2019 02:59:34 +0000 石川 千秋 wrote: >> Dear Alex, >> >> The last time you wrote early in 2018, you seem to read English, so I >> am writing in English here. > Yes, thank you. (I can sort of read Japanese, but it takes a long time > and I usually need help from Google or rikaichan.) > > >> it seems that roman input to hiragana or katakana string conversion is >> handled OUTSIDE >> Wnn: That is, the conversion is done BEFORE the string is sent to Wnn. > Yes, I noticed that kinput2 seems not to use the romkan APIs even when > it uses Wnn. > > I did find that 'uum' seems to use romkan, but I don't know any other > program which does. > > >> Let me explain the input behavior of two frontends. Well, actually one >> is the behavior in a dedicated editor. >> The other is a general-purpose input frontend under X. > -- edit -- > > Thank you for this explanation. Actually, it's quite useful for me > because I have been thinking about how to present the interface for > clause (bunsetsu) conversion in my program. You have given me some > good ideas for how to approach it. > > > >> It may be that the romanization feature of Wnn server may not be tested >> extensively since many input frontends find it easier or more flexible >> than to talk to Wnn server. > I think that is likely. I suspect it was written for uum, and not > really designed for graphical IMEs. I found I had trouble using it to > parse through strings with its default functions. (That is how my test > program does, but in my real GUI program it causes problems.) In the > end, my GUI will handle string (buffer) iteration itself, and just call > the romkan function to convert the kana. > > I think I mentioned it before. I'm actually trying to write a graphical > IME (a bit similar to kinput2) for OS/2. Since the OS/2 desktop is very > different from XWindows I an writing the front-end processor all myself. > > Actually, right now it already works well for typing hiragana in any > graphical window! It is using the romkan functions in FreeWnn. > > For katakana, I thought of a new approach this morning. The OS/2 API > has a function to convert hiragana to katakana, so I can take the hiragana > string from romkan and convert it to katakana using this function. That > does not handle special katakana combinations like 「ヴ」 or 「ティ」 but I > found a way to convert those special combinations myself before calling > romkan for the hiragana. > > Supporting halfwidth (半角) katakana will be more difficult, because of the > need to handle decomposed characters, but I might be able to do > something similar. I'll see. > > BTW my project is: https://github.com/altsan/os2-wnnim > There is no release built yet, although it works quite well as I described. > However, if you don't run OS/2 it's not very useful. :) > > I still need to add bunsetsu conversion, which will use the full FreeWnn > server and API. > >> Well, again, I hope someone in the know will chime in, but this romanization >> feature may not be used by many due to the reason I mentioned above. > Understood. Your information was helpful though, thank you. > > >> PS: Of course, the majority of linux users use other Japanese input methods >> according to some statistical data >> gathered by mozilla foundation. But I think kinput2(-wnn) users are still >> there until they perish due to old age :-) > Sometimes I think OS/2 users are the same. :) > > Thanks, > Alex
Fork