Pleco for Android 2.4.5 Bug Report Thread

[skripp]

Since no one has mentioned this before, this is maybe something wrong with my installation. On my android phone I can't see the price of any of the new dictionaries (Longman, TCM, gu dian). On my iPad I can see the price.

 

[mikelove]

Since no one has mentioned this before, this is maybe something wrong with my installation. On my android phone I can't see the price of any of the new dictionaries (Longman, TCM, gu dian). On my iPad I can see the price.

Hmm, yeah, looks like the catalog file was glitched by a recent update - fixed now, thank you very much for bringing this to our attention.

 

[grahambond]

I discovered a bug this morning in the new (Android 2.4.5) Flashcard system. When you go back to a card that has previously been reviewed within the same session (something I sometimes do when I realise that I have been overly generous/mean in the way that I had marked myself in my self-assessment) and try and change the card score, rather than REPLACING the old score, it seems that the new mark you give to yourself is merely ADDED to the flashcard history. Moreover, no change is made to the flashcard score.

Are you doing this through the "undo" command, or by some other means? You need to use "undo" in order for it to reverse the old score - "change score" works differently.

If you are using "undo," though, is there any chance this card might have appeared more than once in the same session? Are you "undo"ing after advancing to a future card and then backtracking, or are you "undo"ing on the same card after you've marked it correct / incorrect but not yet advanced to the next card?

Yes, I am doing this through the 'undo' command, not the 'change score' command. Having seen this bug in action now a few times, I have a few additional observations.

The problem is inconsistent. There are times where the 'undo' button works normally, and times where it does not. I will probably need to see the problem occur a few more times, in a few more contexts, to have a clearer idea of what kind of variables are changing. However, my initial instinct is to say that it happening in those unstable Eng-Ch 'Fill in the blanks' writing flashcard tests that I mentioned in my last post. Every single time I do a writing test, the system crashes at some point between 30 cards and 45 cards, or so. I've noticed the 'Undo' problem at least three times in this specific context. By contrast, in a (more stable) Eng-Ch 'Self Assessed' Writing flashcard test, I can review hundreds of cards with no problem, and the 'undo' button SEEMS TO work normally.

As I said, I'll need to play around a bit more to clarify this. However, the constant crashing during the Eng-Ch 'Fill in the blanks' writing flashcard tests is a daily and highly predictable phenomenon. I'll try and do as you say and send a Error file report.

 

[mikelove]

The problem is inconsistent. There are times where the 'undo' button works normally, and times where it does not. I will probably need to see the problem occur a few more times, in a few more contexts, to have a clearer idea of what kind of variables are changing. However, my initial instinct is to say that it happening in those unstable Eng-Ch 'Fill in the blanks' writing flashcard tests that I mentioned in my last post. Every single time I do a writing test, the system crashes at some point between 30 cards and 45 cards, or so. I've noticed the 'Undo' problem at least three times in this specific context. By contrast, in a (more stable) Eng-Ch 'Self Assessed' Writing flashcard test, I can review hundreds of cards with no problem, and the 'undo' button SEEMS TO work normally.

Hmm... are you frequently exiting and reopening the test by any chance? Going to a separate screen in Pleco, say?

As I said, I'll need to play around a bit more to clarify this. However, the constant crashing during the Eng-Ch 'Fill in the blanks' writing flashcard tests is a daily and highly predictable phenomenon. I'll try and do as you say and send a Error file report.

OK, thanks.

 

[grahambond]

The problem is inconsistent. There are times where the 'undo' button works normally, and times where it does not. I will probably need to see the problem occur a few more times, in a few more contexts, to have a clearer idea of what kind of variables are changing. However, my initial instinct is to say that it happening in those unstable Eng-Ch 'Fill in the blanks' writing flashcard tests that I mentioned in my last post. Every single time I do a writing test, the system crashes at some point between 30 cards and 45 cards, or so. I've noticed the 'Undo' problem at least three times in this specific context. By contrast, in a (more stable) Eng-Ch 'Self Assessed' Writing flashcard test, I can review hundreds of cards with no problem, and the 'undo' button SEEMS TO work normally.

Hmm... are you frequently exiting and reopening the test by any chance? Going to a separate screen in Pleco, say?

I hardly ever open and close the Pleco application. However, I do often move between different screens within Pleco to, for example, check on a dictionary definition of a word that has cropped up in my Flashcard session in order to see a few example sentences of how it is used etc.etc. Is this likely to be a factor? If so, I'll try and pay attention to whether the problem occurs in a session in which I have done a lot of this kind of flipping.

As a general comment, a system crash becomes very predictable on the basis of the speed it takes to move to the next card in a session. I did a session just this morning where I noticed that, after cards 20 or so, it was taking increasingly long to open the next card (the difference between say 0.05 secs [normal speed] and 0.5 secs [warning territory]). It seems to take longer and longer for the card to open until, inevitably, after 30-40 cards, I get the crash message. All that said, I'll try and produce a (more helpful) error report and send it across shortly.

 

[mikelove]

I hardly ever open and close the Pleco application. However, I do often move between different screens within Pleco to, for example, check on a dictionary definition of a word that has cropped up in my Flashcard session in order to see a few example sentences of how it is used etc.etc. Is this likely to be a factor? If so, I'll try and pay attention to whether the problem occurs in a session in which I have done a lot of this kind of flipping.

It doesn't seem particularly likely, but the biggest single change in flashcard architecture in 2.4 was that we modified the format we use to save / restore flashcard sessions so that it's now a native Android one (which actually gets rid of a lot of crashes when returning to Pleco after not using it for a while), so because of those changes we're naturally inclined to look there to explain bugs.

As a general comment, a system crash becomes very predictable on the basis of the speed it takes to move to the next card in a session. I did a session just this morning where I noticed that, after cards 20 or so, it was taking increasingly long to open the next card (the difference between say 0.05 secs [normal speed] and 0.5 secs [warning territory]). It seems to take longer and longer for the card to open until, inevitably, after 30-40 cards, I get the crash message. All that said, I'll try and produce a (more helpful) error report and send it across shortly.

This makes it seem more like a memory leak - an error report would (probably) confirm that but we'll investigate it in the meantime, thanks.

 

[Lijie]

On the galaxy Note 10.1(ice cream sandwich, will upgrade to jelly bean and check again). Reader mode.
Portrait mode shows blank, Landscape is normal.
edit:fixed in jellybean