I believe that's a case of 'Functioning as Designed' - it's not supposed to play before the reveal. I believe Mike has it that way to prevent self-cheating, though I can certainly see the point of using it as a hint. Though this has cropped up enough times that it might be worth adding an option for in 2.0.1... I know I sometimes hit the button before the reveal, having forgotten that it doesn't work...
2.0.0 Bug Report Thread
- Thread starter mikelove
- Start date
stephanhodges said:
mikelove said:
just remember now, that when i was setting up my profiles/scorefiles i also had some strange effects. went through all the settings for several times, always pushing "OK" on each option page, until it was consistent, because some settings it did not remember when not "OK"ed on the option page... specially the score file was not taken over without that, if i remember correctly...
i had no possibility at that time to report a bug, and later on forgot there was a problem
Mike-
We are indeed using a hacked ROM in our Axims X50vs, to get Windows Mobile 6 running on them. Sorry for not mentioning that earlier - it had slipped my mind that this was a non-standard configuration. These machines were configured for us by a vendor of a Japanese-language dictionary system, which is co-habiting with Pleco.
Being non-standard, it's reasonable to suppose that all bets are off regarding how Pleco behaves in this environment. Nevertheless, in case it's of any interest, here is the setup for the reveal button: Before clicking on Reveal, we show the definition only. When we click the Reveal button, we show the character, the headword and the definition. The audio is not set to autoplay when the card is revealed.
Peter
We are indeed using a hacked ROM in our Axims X50vs, to get Windows Mobile 6 running on them. Sorry for not mentioning that earlier - it had slipped my mind that this was a non-standard configuration. These machines were configured for us by a vendor of a Japanese-language dictionary system, which is co-habiting with Pleco.
Being non-standard, it's reasonable to suppose that all bets are off regarding how Pleco behaves in this environment. Nevertheless, in case it's of any interest, here is the setup for the reveal button: Before clicking on Reveal, we show the definition only. When we click the Reveal button, we show the character, the headword and the definition. The audio is not set to autoplay when the card is revealed.
Peter
I've converted the file to Unicode but the problem is exactly the same. The reader freezes and turning off and on the display seems to be the only way to trigger the next event to happen!
There is another bug in the reader: Occassionally it looses the sound, and when exiting the reader, the dictionary also doesn't have the ability any more. If I exit pleco and launch it again the sound is back!
Actually I don't know if it's Pleco or my PDA that is behaving strangely, because my PDA is quite old, but judging from the fact that Pleco 1.0 had never crashed, maybe these are really bugs... I don't know...
There is another bug in the reader: Occassionally it looses the sound, and when exiting the reader, the dictionary also doesn't have the ability any more. If I exit pleco and launch it again the sound is back!
Actually I don't know if it's Pleco or my PDA that is behaving strangely, because my PDA is quite old, but judging from the fact that Pleco 1.0 had never crashed, maybe these are really bugs... I don't know...
Magnus - to play audio before the reveal, set the "Show" field in "Test Setup" to "Head + Audio" or "Pron + Audio" or whatever you want + Audio - that'll make it available right away. (bit confusing to access it two different ways, but there are some good design reasons for it)
mfcb - there was a bug on saving profile settings for a while which was fixed late in the beta-test; was this with 2.0 final or an earlier release?
PeterP - well we can still take a look at this at least - sounds like the most likely problem is some kind of memory leak / instability with the font rendering on that revealed entry. There are a couple of other bugs that might relate to this, actually, so if we put them all together perhaps something will start to emerge.
F_Kal - strange that it would do that even with Unicode - certainly could be a faulty device, but the audio failing sounds like it might be some other kind of problem; perhaps you're just running low on heap memory, how much Program memory does the System / Memory control panel say you have while Pleco is running:
mfcb - there was a bug on saving profile settings for a while which was fixed late in the beta-test; was this with 2.0 final or an earlier release?
PeterP - well we can still take a look at this at least - sounds like the most likely problem is some kind of memory leak / instability with the font rendering on that revealed entry. There are a couple of other bugs that might relate to this, actually, so if we put them all together perhaps something will start to emerge.
F_Kal - strange that it would do that even with Unicode - certainly could be a faulty device, but the audio failing sounds like it might be some other kind of problem; perhaps you're just running low on heap memory, how much Program memory does the System / Memory control panel say you have while Pleco is running:
BrianTung
秀才
I didn't see this in a cursory search; apologies if this is a duplicate...
There might be a problem with some of the stroke order animations. Specifically, in characters with the chuò ("halt") radical where there are separate traditional/simplified forms (e.g., 這/这), the simplified radical is shown in the standard animation as having three strokes, while the alternate animation shows the traditional one having four. All well and good, I believe. But when there is only one form (e.g., 近), only the standard animation is used, which shows the three-stroke version. The authoritative Taiwanese handbook for stroke order indicates it should have four. Comments?
EDIT: The character 過 has a separate problem. Since there is a simplified variant (过), there is no issue with the number of strokes in the radical in either form. However, in the traditional character, the rest of the character is incorrectly shown, I think, as having eight strokes, when it should have nine: the thing at the top is a T (crossbar first, then upright second), not a 7 (written in one stroke in the animation). At least that's what the Taiwanese handbook seems to show.
There might be a problem with some of the stroke order animations. Specifically, in characters with the chuò ("halt") radical where there are separate traditional/simplified forms (e.g., 這/这), the simplified radical is shown in the standard animation as having three strokes, while the alternate animation shows the traditional one having four. All well and good, I believe. But when there is only one form (e.g., 近), only the standard animation is used, which shows the three-stroke version. The authoritative Taiwanese handbook for stroke order indicates it should have four. Comments?
EDIT: The character 過 has a separate problem. Since there is a simplified variant (过), there is no issue with the number of strokes in the radical in either form. However, in the traditional character, the rest of the character is incorrectly shown, I think, as having eight strokes, when it should have nine: the thing at the top is a T (crossbar first, then upright second), not a 7 (written in one stroke in the animation). At least that's what the Taiwanese handbook seems to show.
System
-HTC Touch
-Windows Mobile 6 Professional
Handwriting Recognition:
If I write my characters directly on the dictionary screen or in the box located at the bottom of the
dictionary screen, the handwriting recognition works so slow that its impossible to enter the strokes
in the correct order and shape. Again, in the dedicated window which i can enter over the "ru"-button,
everything works perfectly and in a natural flow.
It seems that displaying all those characters plus the handwriting recognition at the same time
are too much for the 200mhz processor in my device.
Any suggestions?
(did reinstall of software)
-HTC Touch
-Windows Mobile 6 Professional
Handwriting Recognition:
If I write my characters directly on the dictionary screen or in the box located at the bottom of the
dictionary screen, the handwriting recognition works so slow that its impossible to enter the strokes
in the correct order and shape. Again, in the dedicated window which i can enter over the "ru"-button,
everything works perfectly and in a natural flow.
It seems that displaying all those characters plus the handwriting recognition at the same time
are too much for the 200mhz processor in my device.
Any suggestions?
(did reinstall of software)
This is a minor thing that maybe doesn't qualify as a bug, but in the self-scored test mode, if I have the tall drawing area turned on and magnified headwords, there is only room for one line of text when the definition is revealed. But it looks like there's *almost* enough room; there is blank space between the first line of text and the top of the drawing area that looks like it's just about big enough to fit a second line of text. I bet if the top of the drawing area were lowered by just a pixel or two, you'd be able to fit another line of text in there, which would save me much scrolling back and forth during my flashcard sessions.
To be clear, this is only a problem when large-size Chinese characters are shown along with pinyin and/or definitions. I do get two lines of text when it's just, e.g., pinyin and definitions.
To be clear, this is only a problem when large-size Chinese characters are shown along with pinyin and/or definitions. I do get two lines of text when it's just, e.g., pinyin and definitions.
Minor minor glitch that I don't care all that much if you ever fix: The first time you suspend a flashcard session, you are taken back to the PD2 main UI page (the desired result). If you suspend the same session again after resuming it, you're taken to the flashcard dialog and you have to hit its "Done" button to get back to the main UI.
When this problem happens the free memory is at 4MBs, but I guess that's normal for my device(without anything started free memory is at 11MBs). Actually with pleco 1 I was always running lower on memory(usually 2MBs free), but never had anything like that....when I find the time, I'll try doing a hard reset to my pda, and see how it goes...
BrianTung - it'd be tough to support that change in 近 because the Unicode standard only defines one code point for that character - the extra dot is considered a font variation rather than a different character. So there'd be a ton of work needed on the encoding / character-handling end to even allow Pleco to differentiate between those, along with of course the effort to update the database to include those extra variants. With 過, checking Wenlin at least it seems like that's supposed to be drawn with one stroke - they show a total of 8, so perhaps it's a regional variant (or just one of those annoying little areas where there's no universal agreement on how characters should be drawn).
The fact of the matter is that Mainland China outputs an order of magnitude more dictionaries / linguistic data / etc than Taiwan/HK do, so most of the data that's available for us to license tends to be skewed towards that - we've done all we can to add support for traditional characters, even paid the company we licensed the current stroke data set from to add traditional character support (which it didn't have at all originally), but it's just not possible for us to support traditional characters 100% as well as we do simplified.
LarsMoire - yeah, per ldolse this is a side effect of TouchFlo; do a Google search for "disable touchflo scrolling" and that'll tell you how to turn it off.
koreth - borderline bug with the drawing box, but also a 10-second fix so we'll take a look. With the saved sessions, is this happening even in 2.0 final? I thought we fixed it a few betas ago.
F_Kal - that's odd, not sure why 1.0 would have been using more memory - hard reset certainly might help when you get around to it, perhaps 1.0's screwed up your filesystem in some way.
The fact of the matter is that Mainland China outputs an order of magnitude more dictionaries / linguistic data / etc than Taiwan/HK do, so most of the data that's available for us to license tends to be skewed towards that - we've done all we can to add support for traditional characters, even paid the company we licensed the current stroke data set from to add traditional character support (which it didn't have at all originally), but it's just not possible for us to support traditional characters 100% as well as we do simplified.
LarsMoire - yeah, per ldolse this is a side effect of TouchFlo; do a Google search for "disable touchflo scrolling" and that'll tell you how to turn it off.
koreth - borderline bug with the drawing box, but also a 10-second fix so we'll take a look. With the saved sessions, is this happening even in 2.0 final? I thought we fixed it a few betas ago.
F_Kal - that's odd, not sure why 1.0 would have been using more memory - hard reset certainly might help when you get around to it, perhaps 1.0's screwed up your filesystem in some way.
mikelove said:
yes, that was during beta, not in the final...
another thing is starting to become a problem for me. recently i started to suspect, that the flashcards i am tested with are more frequent than they should be. lets say, if the score is 800 and i have a 100/day setting, i would expect to see the card again just after at least a week...
yesterday i had about 500 cards to review, and after about 250 i stopped the session (arrived at work
). everything looked normal, i got my statistics. when i started the next session i was tested approx 900 cards, and soon after i recognized, that many of these cards i was tested already in the morning and the scores would not permit the cards to show up again... then i recalled that between these two sessions i had to quit pleco and start it again, because when i tried to look up something it did not show any entries from any dictionary as a match. in fact it almost looked like it was hanging, but it was still responsive to menues and input, just no entries from the dicts coming up...
i remember, that you said that during a flash session the scores of cards are updated immediately. i assume, that at the end of a session there will be also a commit of the whole transaction... so i suspect, that there must have been for whatever reason an open transaction from before, which prevented the databases to be commit'ed, and therefore (at least) the morning session was rolled back when i started pleco again...
(just a guess, but i have seen errors like that in my own software
) i had the feeling that i am tested cards too often already several times, but i never could be sure like yesterday... also the fact that i had to restart pleco between the 2 sessions was a little bit more graved into, as i was thinking already about how to describe the error i experienced with the dicts... unfortunately i have no idea how long pleco was running before the "first" session, could be 2 or 3 days...
additionally i have occational hickups with the audio, but the problem is very minor, as its only 1 card, the following cards will play the audio fine again. moving back in the session and trying to play the audio for the failing card wont help, the audio for this card is lost... but it plays again after a restart of pleco. i mention this only to tell you, that the work-around is almost perfect, but the problem is still around...
mfcb - the flashcard score changes are definitely saved immediately during sessions, but I suppose it's possible that there might be some other transaction that's accidentally getting wrapped around them - SQLite doesn't allow nested transactions, though, once you COMMIT once you've committed everything, so I'd think that this would only affect the first card - every other one would be COMMITted as soon as its score was updated. Very odd... perhaps there's some disconnect between SQLite and WM's more general file-caching system that's causing all of SQLite's file changes to be lost. (we've had one or two other reports of this sort of thing, actually)
PeterP - Pleco will find / load flashcard files quite happily from SD cards (just put them in with your other Pleco databases), but I'd strongly advise against putting them on one because of how maddeningly inconsistent the Windows Mobile SD card drivers are - the card could disappear / become inaccessible right in the middle of a big flashcard transaction, in which case you could lose a lot of recent data changes. (we've got some ideas for how to work around this but they all tend to be extremely complicated)
PeterP - Pleco will find / load flashcard files quite happily from SD cards (just put them in with your other Pleco databases), but I'd strongly advise against putting them on one because of how maddeningly inconsistent the Windows Mobile SD card drivers are - the card could disappear / become inaccessible right in the middle of a big flashcard transaction, in which case you could lose a lot of recent data changes. (we've got some ideas for how to work around this but they all tend to be extremely complicated)
BrianTung said:
Hi!
For this I had a seperate thread under "Chinese language" - there seems to be a lack of agreement on how this character is to be written.
regards
Daniel
HTC Touch (ELF), WM Prof. 6
Am I such an idiot or why cant i switch my flashcards to English - Chinese?
Ill get an error message, when I want to start a session after I switched the settings like "no such cards" or something.
Plus, where do i activate the character writing then?
Right now, this input fleld i can turn on is nothing but for handwriting the english translation of the chinese symbol i see
at the moment ?!
Cant be the final solution as everyone is using ones old paper flashcards the other way round, i guess.
Am I such an idiot or why cant i switch my flashcards to English - Chinese?
Ill get an error message, when I want to start a session after I switched the settings like "no such cards" or something.
Plus, where do i activate the character writing then?
Right now, this input fleld i can turn on is nothing but for handwriting the english translation of the chinese symbol i see
at the moment
?!Cant be the final solution as everyone is using ones old paper flashcards the other way round, i guess.
i tried to remap a flashcard without a definition (was assigned to a non-existing user dict entry) to a different dict definition.
the card had headword + pron, but definition was empty !
switching to other dicts did show the definition of the first entry of the selected dict, but not the entry for the right headword.
it only switched between the dicts, that had the headword, but always showed wrong definition.
after several tries to input the headword again, to provide some definition (by selecting the non-dict option) and so on i accidentially switched the card to english-chinese and of course back, after entering the (automatically erased) headword again it worked like expected.
the card had headword + pron, but definition was empty !
switching to other dicts did show the definition of the first entry of the selected dict, but not the entry for the right headword.
it only switched between the dicts, that had the headword, but always showed wrong definition.
after several tries to input the headword again, to provide some definition (by selecting the non-dict option) and so on i accidentially switched the card to english-chinese and of course back, after entering the (automatically erased) headword again it worked like expected.
LarsMoire - setting the card language to E-C will cause Pleco to test you on flashcards based on English-Chinese dictionary entries; if you don't have any of those you'll get that no-cards message. If you just want cards to show the English and hide the Chinese parts of entries, you'd configure that through Test Setup - set "Show" to Definition on that screen. If you want to see English and write in characters, you'd do that through that same screen - set the Test Type to "Free-answer," "Show" to Definition (or Pron + Defn to show the Pinyin too), and "Ask" to "Headword."
mfcb - how did you remap this card? Using the Batch command, or by going into Edit Card and changing which entry it linked to? The definition in Edit Card will only update after you tap on the Search button (next to the Input Field on that screen) so that could be the problem here.
mfcb - how did you remap this card? Using the Batch command, or by going into Edit Card and changing which entry it linked to? The definition in Edit Card will only update after you tap on the Search button (next to the Input Field on that screen) so that could be the problem here.