KanjiTable & KanjiTest v1.0.1 KanjiTable ---------- INTRODUCTION KanjiTable is a small app with a large database, used as a dictionary for the first 1006 or 1945 kanji (Japanese characters). It works with the "Find" button to allow lookups based on "on" and "kun" readings, as well as English definitions. Four different databases are available. They are: kandb1r.prc 1006 kanji, with readings in romaji kandb2r.prc 1945 kanji, with readings in romaji kanjidb1.prc 1006 kanji, with readings in kana (needs J-OS 1.9 or greater) kanjidb2.prc 1945 kanji, with readings in kana (") The kanji can be displayed in rows of 1,2 or 9, sorted according to stroke count, bushu radical number, SKIP entry. Spahn and Hadimitzky, or grade. It can also sort according to frequency of use. I won't bother to explain those here--they come straight from the kanjidic dictionary, from which this was derived. Note that as the database is derived from kanjidic, it has the same restrictions. Please read kanjidic.doc (included) for details. If any entry is pressed, its full dictionary entry is displayed. It is useful to invoke the etched-in button "Find". If you find that some words (ie heart--kokoro) can not be found, you may wish to install "FindHack" by Florent Pillet at . Case is important (the Chinese readings must be searched for using upper case). Because of problems with database locking, "Find" is disabled from within other applications. ROMANIZATION I wasn't sure how to display the 10 or so kanji with readings ending in a small "tsu". Finally, I just decided to display them as "tsu+". (I didn't want to interfere with searching). KNOWN BUGS Okurigana marker (.) interferes with Find--something other than StrStr needs to be used. KanjiTest --------- KanjiTest is a small program which works with the KanjiTable's databases to teach the kanji by using flash cards. It starts with 50 characters, but adds more if a 95% accuracy has been maintained, and the "Merit" is high enough. By using a rough guess as to how many characters are well known, this prevents 100 new characters from being added to the set because of a run of luck. (In this version, the most frequent characters are added first. This may change in a later version). More precise details can be found by pressing the info button at the top of the info page. The top of the display contains something describing the kanji (either the English definition, the readings, or both). Choose the button on the bottom which matches the top. The "?" button can be used if you don't know the answer, or (as is sometimes the case), the description does not match any of the kanji shown. Statistics are kept on each individual character, so it is much more likely that you will see a character you've just gotten wrong than a character you've gotten right 8 times in a row. The fields in the info screen are as follows: %R -- percent right TG -- total guessed #K -- number of kanji in the set NC -- number of kanji to advance (ie, add more kanji). If you get this many kanji in a row, more characters will be added to your working set Merit -- the program's guess about how many characters you know, based on past performance (multiplied by 100). Don't take seriously. AUTHORSHIP and Copyright kanjitable.prc (the KanjiTable program and associated resources), kanjitest.prc (a drill program) copyright (C) 1997 by Jason CJ Neudorf kanjidb.prc (the KanjiTableDB database) is a derivitive of kanjidic, copyright (C) 1996 James William Breen. Please see kanjidic.doc for details regarding licensing LIMITATIONS/LIABILITY This is freeware. You may use it for non-commercial purposes. You may make as many copies as you like, and distribute it freely, but please keep this text file with it. KanjiTable requires up to 12k of extra space, for temporary resources such as indices. If it doesn't have this extra space, it will not be happy. KanjiTable has been tested, but it may yet contain flaws which could cause data corruption or loss. If you have such problems, I assume NO liability. However, please report them to me. ACKNOWLEDGEMENTS jlehett and many (unfortunately forgotten) others on IRC helped with basic concepts bwinton, dagbrown, matthew.chroust --beta testers