radioman - I'm going to Best Buy today to replace our broken-for-several-weeks-now E2 since it seems like we're not going to be able to get rid of these E2 crashing bugs without actually doing some more E2-specific testing. (the E2 has a version of Palm OS, 5.4.7, that's unique to it and was never shipped on any other device, annoyingly enough) Hopefully it'll all turn out to be related to that now-fixed DBCache leak... but glad to hear the session ran correctly at least.
Sarevok - per your later e-mail I think that session screwup may have had something to do with the flashcard database being on your SD card, rather than your dictionaries being there; our file access code does a generally better job of handling files on memory cards going missing than SQLite's does, so if the file's on the card I'm more inclined to blame SQLite. (though since SQLite is open-source we still can / ought to fix it - given the many file corruption problems in 1.0 on WM, though, I'd rather not even touch SQLite's code at all until we've gotten at least one clean release out the door).
Not sure why the clipboard commands aren't working - might be one of the changes we've made to the way Pleco handles keyboard focus (so that key commands are always directed to the Input Field), that could prevent cut/copy/paste from seeing the correct field to cut/copy/paste from.
marsch - strange that #1 is still happening... are you sure that file's 0 bytes? If you, say, move it to the My Documents folder does the file size change? Is your \\Program Files\Pleco folder named in English or does your system localization put it in another language? (some localizations still call it "Program Files" but others don't) These are both running Windows Mobile 2003, right?
#2/3 I'm seeing here so those should be easy to fix. #4 that is indeed the intended behavior - once we added the looping option to review-incorrect it seemed silly to keep having them affect history (since some people like to loop incorrect cards 10 or 20 times and that would really screw up the card history if they kept being counted). #5 has now been fixed, thanks for catching that. With #6, do you do anything after pressing that iTask button or do you just press it and switch right back into Pleco? Do other parts of the flashcard screen still draw correctly or is the whole thing blank? #7 is indeed pretty minor but takes all of 10 seconds to fix (just calculate the second percentage by subtracting the first one from 100
.
#8 that's actually the intended behavior - with built-in dictionaries we assume that anything in parentheses should be stripped out (since most of the time they're (r)s and (zi)s and other unimportant things). Are there a lot of useful characters in parentheses in HanDeDict? Because we could certainly flag that to be treated like a user-created dictionary (in which characters in ()s are retained in flashcards) if most of the content in ()s is useful.
The Duelist - not sure if the reorganization would make a difference but we'll try it out anyway. (haven't managed to reproduce the problems here so far)
daniel123 - #1 and #2 have been found / fixed - #1 was a glitch from some changes we made to custom dictionary searches (which share a lot of code with flashcard searches), #2 we were deleting the file from a different place than we were creating it - the saved session file is in \\Program Files\Pleco now so if you go into there you can delete it and get rid of that message. Not sure what's going on with the sound playback issue in #3, but since we have a 110 series iPAQ here it should hopefully be easy to reproduce - do you have a rough idea of how many flashcards it takes to set this off? 100? 500?
