2.0.2 Beta 1 Bug Report / Feedback Thread
- Thread starter mikelove
- Start date
Pleco retains full screen after switching on/off - now you just need to retain it between executions 
(e.g. minimize Pleco, close it via the task manager, start it again and it's not in full screen). Would also like a way of accessing the menus without having to hop out of full screen (Treo Pro doesn't have hard L/R keys, so if it's not on the screen I can't use it).
(e.g. minimize Pleco, close it via the task manager, start it again and it's not in full screen). Would also like a way of accessing the menus without having to hop out of full screen (Treo Pro doesn't have hard L/R keys, so if it's not on the screen I can't use it).I get crashes (reset) when I try and compress my database. I'm going to send you the files by email, since I backed them up just before installing the new version, and the compression was almost the first thing I tried, other than just looking at things.
Another full text concern anomaly. In the CC dict entry for 此起彼伏 ci3qi3bi3fu2 I notice there is a notation that this is a chengyu and it even says 成语 in Hanzi. I was following another post about people wanting a chengyu dictionary, and one of the reasons was to easily search for them. I thought, wow if the CC dict automatically tags the chengyu than I can just to a full-text search for 成语。 Well, I tried that and the full text search does not find 成语 anywhere in the CC dictionary.
Why can't #成语 find anything in the CC dict?
Why can't #成语 find anything in the CC dict?
FYI, I am overjoyed with this product and the support I am receiving. I think almost all of my concerns about the product have been addressed.
On that positive note, there is one issue that I think I brought up before and would like to submit it again for consideration because it is mildly annoying.
Is it possible to make the "Copy to Input" feature follow the dictionary search order?
In particular, I am often in the Guifan dictionary looking things up and run across some term that I don't recognize. When using copy to input it looks it up in the Guifan. Well my Chinese is still at the level that I'm looking up this word because I don't know it at all and would prefer to first search the Tuttle, than the ABC dictionary to find an English explanation. Then it is a simple matter of pressing my assigned "Back" key and I'm back where I left off. If I want to go back to that explanation I simply press forward once.
As it is now, I highlight the work, press copy to input, then have to drop down the dictionary list and select the Tuttle dictionary. Then to return to where I left off I have to press back at least twice. Then if I want to see the definition again to see how it fits in with the Guifan entry I have to press forward at least twice.
Thanks.
On that positive note, there is one issue that I think I brought up before and would like to submit it again for consideration because it is mildly annoying.
Is it possible to make the "Copy to Input" feature follow the dictionary search order?
In particular, I am often in the Guifan dictionary looking things up and run across some term that I don't recognize. When using copy to input it looks it up in the Guifan. Well my Chinese is still at the level that I'm looking up this word because I don't know it at all and would prefer to first search the Tuttle, than the ABC dictionary to find an English explanation. Then it is a simple matter of pressing my assigned "Back" key and I'm back where I left off. If I want to go back to that explanation I simply press forward once.
As it is now, I highlight the work, press copy to input, then have to drop down the dictionary list and select the Tuttle dictionary. Then to return to where I left off I have to press back at least twice. Then if I want to see the definition again to see how it fits in with the Guifan entry I have to press forward at least twice.
Thanks.
sthubbar - 成语 is too common a term for the full-text indexer, at least in that dictionary - it appears about 3000 times.
That Copy to Input change you mention is actually kind of a tricky thing programming wise, so while I agree it's a good idea it may be a while before we can put it in - basically there's not a good mechanism right now for the search system to tell the difference between a Copy to Input and any other dictionary query.
That Copy to Input change you mention is actually kind of a tricky thing programming wise, so while I agree it's a good idea it may be a while before we can put it in - basically there's not a good mechanism right now for the search system to tell the difference between a Copy to Input and any other dictionary query.
OK. Got another one for you.
After a failed full-text search, the back button will return to the next to last found entry rather than the last entry.
I guess the full-text search is something I don't understand. So I guess the program has to pre-index these full text items? Well, here's another question. Does the full-text search look at headwords? I searched for aidewen which produced an entry 爱德问 in the HDD dictionary. When I then did a search for #爱德问, it comes up as no entry found. I guess that is also by design. Right?
After a failed full-text search, the back button will return to the next to last found entry rather than the last entry.
I guess the full-text search is something I don't understand. So I guess the program has to pre-index these full text items? Well, here's another question. Does the full-text search look at headwords? I searched for aidewen which produced an entry 爱德问 in the HDD dictionary. When I then did a search for #爱德问, it comes up as no entry found. I guess that is also by design. Right?
Yeah, that looks like a bug - if the display's blank due to a failed search then back should get you back to the last entry you were viewing.
The full-text indexes are generated when we make the dictionary database files, so the limit on the number of indexed entries isn't something built into Pleco, it's something we apply when creating the databases for performance / stability reasons.
And yes, headwords aren't included in the full-text index by design.
The full-text indexes are generated when we make the dictionary database files, so the limit on the number of indexed entries isn't something built into Pleco, it's something we apply when creating the databases for performance / stability reasons.
And yes, headwords aren't included in the full-text index by design.
Found two problems:
1) The audio dropping-out bug is back. The first time I saw this with 2.0.1 Beta 2. You suggusted to move some Pleco files from card to internal memory. Then I moved most files to internal memory (about 100 MB). About 24 files in internal memory 12 rest on sd card. After moving it seemed that the problem has gone.
Until 2.0.2 Beta 1. Now the same problem came back. I cannot move more files to internal memory. What else can I do? Is there a new problem in 2.0.2 Beta 1? Fortunately it is not so often but it is a little pity that in 2.0.0 the audio worked without any mistake and now it is worse.
2) Import of a unicode flashcard file has a new effect. I always use my own definitions in a unicode text file (three columns separated by a space followed by a tab). Some of the entries are wrongly mapped to the ABC dict instead of using my given defintion. I know this behaviour from before. Sometimes the importing makes tiny mistakes. This was the same in 1.03 and in 2.0x. But this time it is different. I have about 20% mistakes and when I edit the entry and change from dictionary mapped to custom entry my origin definition will be restored as if it was saved anywhere. This is new. Don't know if this a feature or a bug because of the false mapping.
Maybe I have to take an eye on it when importing new vocabulary after next lesson.
Apart from that I like the new features. Good job Mike!
Daniel
1) The audio dropping-out bug is back. The first time I saw this with 2.0.1 Beta 2. You suggusted to move some Pleco files from card to internal memory. Then I moved most files to internal memory (about 100 MB). About 24 files in internal memory 12 rest on sd card. After moving it seemed that the problem has gone.
Until 2.0.2 Beta 1. Now the same problem came back. I cannot move more files to internal memory. What else can I do? Is there a new problem in 2.0.2 Beta 1? Fortunately it is not so often but it is a little pity that in 2.0.0 the audio worked without any mistake and now it is worse.
2) Import of a unicode flashcard file has a new effect. I always use my own definitions in a unicode text file (three columns separated by a space followed by a tab). Some of the entries are wrongly mapped to the ABC dict instead of using my given defintion. I know this behaviour from before. Sometimes the importing makes tiny mistakes. This was the same in 1.03 and in 2.0x. But this time it is different. I have about 20% mistakes and when I edit the entry and change from dictionary mapped to custom entry my origin definition will be restored as if it was saved anywhere. This is new. Don't know if this a feature or a bug because of the false mapping.
Maybe I have to take an eye on it when importing new vocabulary after next lesson.
Apart from that I like the new features. Good job Mike!
Daniel
1) This is still a head-scratcher, the audio system really hasn't changed at all in several releases - I suppose there could be some sort of severe memory leak somewhere else in the code that's eventually causing the system to have too little memory to keep playing audio, but we haven't seen any signs of that so far. We'll keep looking though.
2) That's actually the new default behavior, though you can change it back to the old way if you like - tap on the "Dicts..." button in the "Import" screen and select "Prefer File" for "Definition Source."
2) That's actually the new default behavior, though you can change it back to the old way if you like - tap on the "Dicts..." button in the "Import" screen and select "Prefer File" for "Definition Source."
I haven't had any issues, but I did notice that ADS entry for cheng (orange) is missing the "g" in the pinyin.
I really like the way you handled moving through related words (on the details screen). Very intuitive and easy to use.
Feedback. selecting categories takes a lot of time and scrolling at the Manage Categories -> Set Default, and on the main "add to flashcard" button when selecting an alternate category, because the list is completely expanded. I noticed the option to not expand lists, but the way it reads, I assume that it doesn't show the sublists at all.
You have a copy of my flashcards, so you can see what I mean.
How about changing the "Default" so that it works just like "Group"? I.E. Select the Default button, then quickly browse to the one you want to make a default.
Not sure best way to improve the "Add to flashcards" category selection, but that would save a lot of time now, as I have to scroll through 4-6 windows to switch from one book to another (during different classes, etc).
Last, I accidentally "discovered" a feature in the reader that brought up a "go to line/character position" dialog, but can't get it to come up now with current version. I used a finger swipe from right to left across bottom of Palm screen. I cannot find a menu for that dialog anywhere. Seems it could be on the regular menu, at least?
*** EDIT *** Ha, found that go to dialog. Clicking on the bottom (where it shows 5.63% of text already read, etc) brings it up.
I really like the way you handled moving through related words (on the details screen). Very intuitive and easy to use.
Feedback. selecting categories takes a lot of time and scrolling at the Manage Categories -> Set Default, and on the main "add to flashcard" button when selecting an alternate category, because the list is completely expanded. I noticed the option to not expand lists, but the way it reads, I assume that it doesn't show the sublists at all.
You have a copy of my flashcards, so you can see what I mean.
How about changing the "Default" so that it works just like "Group"? I.E. Select the Default button, then quickly browse to the one you want to make a default.
Not sure best way to improve the "Add to flashcards" category selection, but that would save a lot of time now, as I have to scroll through 4-6 windows to switch from one book to another (during different classes, etc).
Last, I accidentally "discovered" a feature in the reader that brought up a "go to line/character position" dialog, but can't get it to come up now with current version. I used a finger swipe from right to left across bottom of Palm screen. I cannot find a menu for that dialog anywhere. Seems it could be on the regular menu, at least?
*** EDIT *** Ha, found that go to dialog. Clicking on the bottom (where it shows 5.63% of text already read, etc) brings it up.
Tacpis Palm with 2.02 Beta
Repeatable crash when I select Statistics screen, which first always selects "all".
I have a 3 level category system under Z - 18 - 1800,1810,1820, etc.
When I select "Z" (top level), it crashes, causing a RESET screen every time. I can select several other categories without problem. It should be reproducible with the file you already have.
BTW, selecting "ALL" results in several seconds waiting before being able to change. Wouldn't it be "better" (except in crashes!) to leave the selection at the last selected item? Selecting ALL is "optimizing" for the worst case situation, and I still have most of the semester to go with adding words, etc. I've already deleted a couple thousand words to just reduce the size of the category selections list.
Repeatable crash when I select Statistics screen, which first always selects "all".
I have a 3 level category system under Z - 18 - 1800,1810,1820, etc.
When I select "Z" (top level), it crashes, causing a RESET screen every time. I can select several other categories without problem. It should be reproducible with the file you already have.
BTW, selecting "ALL" results in several seconds waiting before being able to change. Wouldn't it be "better" (except in crashes!) to leave the selection at the last selected item? Selecting ALL is "optimizing" for the worst case situation, and I still have most of the semester to go with adding words, etc. I've already deleted a couple thousand words to just reduce the size of the category selections list.
Please consider adding a "per card" timer instead of a single session timer. I would like to try a method of study varying the time per card between 10-30 seconds, as I have some peculiar (I think) memory problems. For example, I have to memorize E-C and C-E P-C, E-P-C all separately, with seemingly little cross-over. I.E. I can study just one modality and get it, and not have a clue if asked via different mode. I can do writing tests, and have no idea how to read the characters afterward, etc.
The time would reset after each card. At this point, I couldn't say if keeping any statistics on these times would be of any interest or value.
Very Minor: If you add or edit a score file name that is too long, it will overwrite the "Manage" button afterward. The next time you enter the screen, it then puts the "Manage" button on top [correctly].
The time would reset after each card. At this point, I couldn't say if keeping any statistics on these times would be of any interest or value.
Very Minor: If you add or edit a score file name that is too long, it will overwrite the "Manage" button afterward. The next time you enter the screen, it then puts the "Manage" button on top [correctly].
Thanks! Good idea about the "Default" button, particularly on Palm where the name often gets cut off by the list anyway. And about the "per card" timer, though getting that to trigger something once it reached a certain time limit would be a lot more complicated than just getting it to update / reset that timer display for each card.
For choosing a category in the popup menu, why not just temporarily move the books you're using in class to the top of your flashcard category list? That way they'll be right at the top of that menu.
I'm not getting a crash pulling up statistics for "Z" - is there any specific error message associated with this? The statistics screen actually does save / restore the last selected item - works fine on a Treo I just tested it on at least - so I'm not sure why that isn't happening in your case.
For choosing a category in the popup menu, why not just temporarily move the books you're using in class to the top of your flashcard category list? That way they'll be right at the top of that menu.
I'm not getting a crash pulling up statistics for "Z" - is there any specific error message associated with this? The statistics screen actually does save / restore the last selected item - works fine on a Treo I just tested it on at least - so I'm not sure why that isn't happening in your case.
radioman
状元
With regard to answer quality on repetition spaced flashcard sessions, I am not seeing how the various numbers affect the score. For instance, I believe that if I started with a new card and number of points = 100, and mark it wrong (i.e., 1,2,3, it appear to not move the specific score of the card. and if I mark it right (4,5,6 it can move it up to 400). If it is my first time evaluating a card, I would expect to see it every day for a few days based on repetition based algorithms as I understand them. But with the 400, does that not mean I will only see it 4 days after the last time I saw it (assuming 100 is the value set per day?).
I figure there is more going on here than I am seeing. Sorry if this is a repeat topic.
I figure there is more going on here than I am seeing. Sorry if this is a repeat topic.
That only matters after the first answer, though it's factored into the difficulty from the get-go - the score is always 400 (or higher / lower with a different difficulty setting) after the first correct answer, regardless of whether it's a 4, 5, or 6. (there's an option in Advanced / Scoring to scale that initial score by the difficulty factor, though) And yes, 400 means you won't see it for 4 days - we're considering tweaking a lot of this stuff in 2.1, though (we're sticking a bit too strictly to the SuperMemo approach in our current version, I think).
radioman
状元
I was also going to suggest this because of my interest in studying sentences. Part of my exercises is to go through a sentence in my head and get the rhythm of it down. Right now I use an external timer for 1 to 3 minutes depending on the difficulty of the sentence its annoying and one more thing batteries. So it would be great if there was a way to have some sort of notification set that highlights that you have been on a particular card for a certain time period. Even if the current "timer" (really a session timer) notification on the flashcards could have another option as "current card timer".
stephanhodges said:
I had a small issue using the category split feature. I had the items sorted by create time, so that they would retain the ordering in the book. I took my list of 90 words and split it into 10 words per group. The words in each of the groups to not even come close to matching the initial ordering. I think this function should use whatever the current sort is set in the
Second, please add this to your list of enhancements. Of course, I don't know what is so simple that it might make the "next" release, but hopefully they are all improvements and worth considering...
When splitting, I would like to be able to:
a) Yes/No prepend the parent group name to the beginning of the newly split subgroup. Example: I have parent group "Z/Z20" (for chapter 20 of Zonghe). Would want "Z20" prepended. This is because current list handling does not show parents if the subgroups are too numerous. So, many subgroups with the same name are hard to distinguish. Perhaps a change to those dialogs would be sufficient also. But, I personally like this naming convention, so still think it's useful on the approach that not everyone uses the same viewpoint or usage model.
b) able to specify generated pattern. I want a numeric 2 digit pattern, for example. So: (with #a) I would get "Z2000", "Z2010", "Z2020"... when split by groups of 10. If split by groups of 100 or more, then would use a 3 digit pattern.
Second, please add this to your list of enhancements. Of course, I don't know what is so simple that it might make the "next" release, but hopefully they are all improvements and worth considering...
When splitting, I would like to be able to:
a) Yes/No prepend the parent group name to the beginning of the newly split subgroup. Example: I have parent group "Z/Z20" (for chapter 20 of Zonghe). Would want "Z20" prepended. This is because current list handling does not show parents if the subgroups are too numerous. So, many subgroups with the same name are hard to distinguish. Perhaps a change to those dialogs would be sufficient also. But, I personally like this naming convention, so still think it's useful on the approach that not everyone uses the same viewpoint or usage model.
b) able to specify generated pattern. I want a numeric 2 digit pattern, for example. So: (with #a) I would get "Z2000", "Z2010", "Z2020"... when split by groups of 10. If split by groups of 100 or more, then would use a 3 digit pattern.
Currently, card scores are kept with the card. I'd like to be able to keep them instead with the profile. I really want to be able to study and score independently for hanzi->english, writing, english-hanzi, etc. My ability to recognize far exceeds my ability to write, and as a result, I cannot use the spaced repetition system, afaik.