New bugs

[Ed]

So I made a new database - then imported 675 new cards (front-to-back vocab) from a text file.

596 of the new cards became classified as "Vocabulary" (298 Production, 298 Recognition); the remaining 79 were classifed as "front-to-back".

Under "Browse cards" I selected the Vocabulary ones only (596 cards), used Ctrl + A to select all...
But I soon discovered Ctrl + A wasn't selecting *all* the cards, only a chunk of them, so I scrolled down to the bottom of the list, selected the bottom card, scrolled to the top of the list, selected the top card, and pressed Shift to select them all.

Then I used the "change card type" option by right-clicking, selecting to convert all the cards to front-to-back.

This still didn't work. Out of the 596 cards, only 377 were converted to front-to-back. The remaining 219 had simply disappeared!!

So I want explanations for:

- Where those 219 cards have gone
- Why most of the cards were erroneously imported as Vocabulary, and not simply front-to-back (I shouldn't have to change the card type every time I import from a uniform tab-separated text file)

Just when I thought I was getting the hang of this program these bugs come up...

* Update: I have just repeated the above process twice and got exactly the same results!

[Peter Bienstman]

Hi,

If your text file contains a line with three columns (e.g. from Mnemosyne's 1.x three-sided cards), Mnemosyne will automatically treat that as data for a Vocabulary CARDTYPE and will create two SISTER CARDS from that data, one dealing with recognition, and the other with production. If you come from Anki, you're probably familiar with the fact that data from a single 'fact' is used to create multiple 'linked cards', so the principle is exactly the same.

If you don't want e.g. recognition cards, you can indeed convert them to Front-to-back only cards (which will delete the recognition card and keep the production card), but another option is 'activate cards' and then simply deactivate the recognition cards.

As for cntrl-a not selecting all cards in the browser, this is a unfortunate side effect/bug of the UI library I use: to speed up display, not all the cards are loaded into the browser at once, but progressively once you scroll down. Cntrl-a selects only the cards that the library had already loaded.

Hope this clarifies things.

Thanks for the feedback, and if you have suggestions on how to improve the documentation, feel free!

Cheers,

Peter

> --
> You received this message because you are subscribed to the Google Groups
> "mnemosyne-proj-users" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to mnemosyne-proj-u...@googlegroups.com.
> To post to this group, send email to mnemosyne-proj-
> us...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/mnemosyne-proj-users/af3df679-8864-
> 4194-9350-08efa25253e2%40googlegroups.com
> <https://groups.google.com/d/msgid/mnemosyne-proj-users/af3df679-
> 8864-4194-9350-
> 08efa25253e2%40googlegroups.com?utm_medium=email&utm_source=foo
> ter> .
> For more options, visit https://groups.google.com/d/optout.

[Ed]

Hi Peter,

Many thanks for your rapid explanation.

I have been through my wordlist and deleted all the answers to the questions. It does, in fact, contain 378 words (so I have lost one somewhere in the import). So I apologize for that. The 675 number confused me and I was sure my wordlist was longer than it actually is.

However, to answer your points:

- The text file I am importing from only contains two columns separated by a tab - the format is *source word - tab - target word(s)*. It is as simple as it could be
- I am not using any cards from earlier versions of Mnemosyne and my text file does not contain 3 columns
- I am not using anything at all from Anki

So the question still remains why this apparently totally arbitrary assignment to card types upon import? As I said above, I am importing from a totally uniform, tab-separated, 2-column text file. So why is the program assigning some cards as front-to-back and some as Vocabulary?

This creates extra work in having to reassign card types by using the "change card type" function and should not be a necessary step. It seems to me there is a significant flaw in the import process from text files?

Thanks for your further thoughts :) Ed

> email to mnemosyne-proj-users+unsub...@googlegroups.com.

[Oisín]

Before jumping to the conclusion that there is a "serious flaw" and bugs in the import process, perhaps it would be a good idea to attach the file you imported from so others can test and investigate properly?

BTW it's generally a useful approach to give software the benefit of the doubt before concluding that a problem you're experiencing is due to a bug in someone else's code, even though it might be. When I work with new programmers I often hear them say things like "... Why isn't this working? It must be a bug in the compiler." -- it almost never is :D

To unsubscribe from this group and stop receiving emails from it, send an email to mnemosyne-proj-u...@googlegroups.com.
To post to this group, send email to mnemosyne-...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/mnemosyne-proj-users/8d827c5b-7a21-4b2c-8d4d-aaf59b01be5e%40googlegroups.com.

[Peter Bienstman]

Please send me the txt file, I'll have a look at it.

Peter

On September 14, 2014 2:34:34 PM CEST, Ed <habene...@gmail.com> wrote:

Hi Peter,

Many thanks for your rapid explanation.

I have been through my wordlist and deleted all the answers to the questions. It does, in fact, contain 378 words (so I have lost one somewhere in the import). So I apologize for that. The 675 number confused me and I was sure my wordlist was longer than it actually is.

However, to answer your points:

- The text file I am importing from only contains two columns separated by a tab - the format is *source word - tab - target word(s)*. It is as simple as it could be
- I am not using any cards from earlier versions of Mnemosyne and my text file does not contain 3 columns
- I am not using anything at all from Anki

So the question still remains why this apparently totally arbitrary assignment to card types upon import? As I said above, I am importing from a totally uniform, tab-separated, 2-column text file. So why is the program assigning some cards a s front-to-back and some as Vocabulary?

[Ed]

Thanks Peter.

Here's the text file.

Regards, Ed

[Peter Bienstman]

Hi,

Plenty of the entries are separated by two tabs as opposed to a single tab, that explains why they get imported as vocabulary card type with an empty pronunciation field.

I guess in a future version I could add a workaround for these malformatted files.

Thanks for the info!

Peter

> -----Original Message-----
> From: mnemosyne-...@googlegroups.com [mailto:mnemosyne-
> proj-...@googlegroups.com] On Behalf Of Ed
> Sent: 14 September 2014 18:47
> To: mnemosyne-...@googlegroups.com
> Subject: Re: [mnemosyne-proj-users] New bugs
>
> Thanks Peter.
>
> Here's the text file.
>
> Regards, Ed
>
>
> On Sunday, 14 September 2014 16:33:37 UTC+1, Peter Bienstman wrote:
>
> Please send me the txt file, I'll have a look at it.
>
> Peter
>
>
> On September 14, 2014 2:34:34 PM CEST, Ed <habene...@gmail.com

> users+un...@googlegroups.com.

> > To post to this group, send email to mnemosyne-
> proj-
> > us...@googlegroups.com.
> > To view this discussion on the web visit
> > https://groups.google.com/d/msgid/mnemosyne-
> proj-users/af3df679-8864-

> <https://groups.google.com/d/msgid/mnemosyne-proj-users/af3df679-
> 8864->
> > 4194-9350-08efa25253e2%40googlegroups.com
> <http://40googlegroups.com>
> > <https://groups.google.com/d/msgid/mnemosyne-
> proj-users/af3df679- <https://groups.google.com/d/msgid/mnemosyne-
> proj-users/af3df679->
> > 8864-4194-9350-
> >
> 08efa25253e2%40googlegroups.com?utm_medium=email&utm_source=foo
> <http://40googlegroups.com?utm_medium=email&utm_source=foo>

> > ter> .
> > For more options, visit
> https://groups.google.com/d/optout

> <https://groups.google.com/d/optout> .

>
>
>
>
>
>
> --
> You received this message because you are subscribed to the Google Groups
> "mnemosyne-proj-users" group.
> To unsubscribe from this group and stop receiving emails from it, send an

> email to mnemosyne-proj-u...@googlegroups.com.

> To post to this group, send email to mnemosyne-proj-
> us...@googlegroups.com.
> To view this discussion on the web visit

> https://groups.google.com/d/msgid/mnemosyne-proj-users/b1ffdeef-a525-
> 4dee-9c89-3988fc44fa2c%40googlegroups.com
> <https://groups.google.com/d/msgid/mnemosyne-proj-users/b1ffdeef-
> a525-4dee-9c89-
> 3988fc44fa2c%40googlegroups.com?utm_medium=email&utm_source=foot

[Ed]

Ah, so that's the reason for it. Thanks Peter. I used two tabs in some cases because the source and target were almost too close to be able to read easily otherwise.

I appreciate this is free software and you have spent - and do spend - many hours of your time developing it.

Thanks for Mnemosyne. The more feedback I get the more I am appreciating it and the more I realize it is actually working fully as you intended.

Hopefully I won't be bothering you anymore for a while, now I appreciate things a bit better and I have got my set-up more or less how I want it.

Thanks again, Ed

> email to mnemosyne-proj-users+unsub...@googlegroups.com.