Active Topics
-
The N900 is now 15 years old! (8)
to Nokia N900 by whayong - 15 hrs, 1 min ago -
Sailfish OS for the Motorola Moto G7 Power (XT1955-5) - (ocean) (4)
to SailfishOS by Maemish - 5 days, 11 hrs ago - more...
|
2009-12-02
, 17:00
|
|
Posts: 21 |
Thanked: 1 time |
Joined on Nov 2009
@ Sweden
|
#12
|
Originally Posted by aspidites
I've sent an email with an anki deck attached. Now, I ask any Mnemosyne user to send one as well =)
Sure. I can't seem to PM, but my email is aspidites at inbox dot com.
All I need is a sample deck from each program and I can look into writing a converter.
|
|
2009-12-02
, 17:18
|
|
Posts: 402 |
Thanked: 229 times |
Joined on Nov 2009
@ Missouri, USA
|
#13
|
hmm... sqlite file. Not going to be as easy as converting plain text, but still more than doable. :-) Thank goodness for python's native sqlite support!
Edit:
I'll try and google again for a mnemosyne file in the mean time.
Edit 2:
I'v found an example file,but I'm not yet sure what kind of file it is. It appears to be a database file, but not in sqlite format. It appears to be a pickled binary, which pythhon can handle. Now I just need to find out which fields I need from anki file and convnert.
Also, the anki database seems to be more complicated than I though. As such, I will have to take a minute to study its structure.
BTW: I tried running anki in scratchbox and get errors about not being able to import QtWebkit, even though the deb packages is clearly installed. Anyone else?
Edit 3:
The file I downloaded seems to be corrupted so i still need a valid mnemosyn file in order to try and work this.
It does seem that mnemosyn saves less data than anki, so converting to anki might not be so bad.
Last edited by aspidites; 2009-12-02 at 18:00.
Edit:
Edit 2:
I'v found an example file,
Also, the anki database seems to be more complicated than I though. As such, I will have to take a minute to study its structure.
BTW: I tried running anki in scratchbox and get errors about not being able to import QtWebkit, even though the deb packages is clearly installed. Anyone else?
Edit 3:
The file I downloaded seems to be corrupted so i still need a valid mnemosyn file in order to try and work this.
It does seem that mnemosyn saves less data than anki, so converting to anki might not be so bad.
Last edited by aspidites; 2009-12-02 at 18:00.
|
|
2009-12-02
, 23:14
|
|
Posts: 21 |
Thanked: 1 time |
Joined on Nov 2009
@ Sweden
|
#14
|
Originally Posted by aspidites
I found some Mnemosyne decks from the official web page that seem to be working; one of them is attached to this post and at your disposal.
hmm... sqlite file. Not going to be as easy as converting plain text, but still more than doable. :-) Thank goodness for python's native sqlite support!
Edit:
I'll try and google again for a mnemosyne file in the mean time.
Edit 2:
I'v found an example file,but I'm not yet sure what kind of file it is. It appears to be a database file, but not in sqlite format.It appears to be a pickled binary, which pythhon can handle. Now I just need to find out which fields I need from anki file and convnert.
Also, the anki database seems to be more complicated than I though. As such, I will have to take a minute to study its structure.
BTW: I tried running anki in scratchbox and get errors about not being able to import QtWebkit, even though the deb packages is clearly installed. Anyone else?
Edit 3:
The file I downloaded seems to be corrupted so i still need a valid mnemosyn file in order to try and work this.
It does seem that mnemosyn saves less data than anki, so converting to anki might not be so bad.
As for running Anki in scratchbox I can't say, since I don't have a phone to try it on yet
Dying to get it though T_TI had a brief look at the Mnemosyne program as well and the possible settings for a deck, and I can understand why Anki would be more advanced. In Anki for example you can have more than just two fields on a card, whilst this is not possible in Mnemosyne. This most likely means that while a conversion from Anki to Mnemosyne might be possible, you might lose quite a bit of data from the Anki deck in the process. Converting Mnemosyne to Anki should work better and without losses, though I'm only speculating. If I'm right, then too bad it's the Anki->Mnemosyne conversion we want to do =P Might be better off trying to get Anki to run properly. Damn I want that N900 and start playing around with stuff like this!
|
|
2009-12-02
, 23:31
|
|
Posts: 70 |
Thanked: 23 times |
Joined on Sep 2009
|
#15
|
I tested anki and it had a import from mnemosyne function which seemed to work, although I probably will use anki when adding new material next time.
What I found good about Anki was the buttons for bold, italics etc., online sync, copy-paste images directly. Apparently with AnkiOnline you could review them in the browser. I will try this when I get the device.
What I found good about Anki was the buttons for bold, italics etc., online sync, copy-paste images directly. Apparently with AnkiOnline you could review them in the browser. I will try this when I get the device.
|
|
2009-12-03
, 00:48
|
|
Posts: 402 |
Thanked: 229 times |
Joined on Nov 2009
@ Missouri, USA
|
#16
|
You are right Koeus. It's getting late, but I'll definately give those files a looking over and figure out what data in anki correlates to mnemosyne.
I wish I had an n900 as well ;-(
Anyways, I'll try to bang out something after work tomorrow and report my successes/failures, unless someone beats me to it.
arono, unfortunately we are trying to go the other way around > import anki file into mnemosyne.
Edit:
I played around a bit just to see how much work I had ahead of me tomorrow. Turns out the data in mnemosyne is picked (using cPickle module) and when trying to unpickle the data mnemosyne-specific data is imported, which I of course don't have access to from a regular python script.
In short, I'll have to track down all the modules from the mnemosyne source before I can extract the data, unless there is some easier method to getting to the data?
Edit 2:
I just realized that both programs can export/import from tab separated text files.
Koeus, is it sufficient to open up anki, export file as tab separated text file, then open up mnemosyne and import from tab separated text file?
I tried it on your file and it seemed to work, but I'm not familiar with either program so don't know how much info was lost...
Last edited by aspidites; 2009-12-03 at 01:11.
I wish I had an n900 as well ;-(
Anyways, I'll try to bang out something after work tomorrow and report my successes/failures, unless someone beats me to it.
arono, unfortunately we are trying to go the other way around > import anki file into mnemosyne.
Edit:
I played around a bit just to see how much work I had ahead of me tomorrow. Turns out the data in mnemosyne is picked (using cPickle module) and when trying to unpickle the data mnemosyne-specific data is imported, which I of course don't have access to from a regular python script.
In short, I'll have to track down all the modules from the mnemosyne source before I can extract the data, unless there is some easier method to getting to the data?
Edit 2:
I just realized that both programs can export/import from tab separated text files.
Koeus, is it sufficient to open up anki, export file as tab separated text file, then open up mnemosyne and import from tab separated text file?
I tried it on your file and it seemed to work, but I'm not familiar with either program so don't know how much info was lost...
Last edited by aspidites; 2009-12-03 at 01:11.
|
|
2009-12-03
, 11:28
|
|
Posts: 21 |
Thanked: 1 time |
Joined on Nov 2009
@ Sweden
|
#17
|
Originally Posted by aspidites
I unfortunately have no idea what picked/pickled data is och means, nor what cPickle module is, so I can't really help you there =/ (You're welcome to explain it though ^^)
Edit:
I played around a bit just to see how much work I had ahead of me tomorrow. Turns out the data in mnemosyne is picked (using cPickle module) and when trying to unpickle the data mnemosyne-specific data is imported, which I of course don't have access to from a regular python script.
In short, I'll have to track down all the modules from the mnemosyne source before I can extract the data, unless there is some easier method to getting to the data?
Originally Posted by aspidites
I just looked into it, and you're quite right, it is indeed possible! While the fact that Anki utilises more than two fields to a card turned out to be no problem (it exports both fields on the backside of the card so that when Mnemosyne imports it, all backside data is shown in a single field), the most serious thing that was lost was all the scheduling data; i.e. all cards are imported as if they were brand new and have never been reviewed. This means that for us avid Anki users it's pretty much pointless to import it since it's the scheduling data that's the important part. Rereviewing 2000+ cards from scratch is not something that's done in the wink of an eye, and I personally would never do it if I didn't absolutely have to.
Edit 2:
I just realized that both programs can export/import from tab separated text files.
Koeus, is it sufficient to open up anki, export file as tab separated text file, then open up mnemosyne and import from tab separated text file?
I tried it on your file and it seemed to work, but I'm not familiar with either program so don't know how much info was lost...
In other words, we have a way of importing decks from Anki to Mnemosyne, but if this is going to be of any use to Anki users that want to use Mnemosyne on the N900, then we still need to solve the issue of exporting/importing the scheduling data.
Edit:
Also, Mnemosyne seems to lack the feature of displaying furigana, which Anki does very nicely. A minor aesthetical drawback which just further convinces me that Anki is the better piece of software among the two of them ^^ The best solution would still probably be to just get Anki functioning properly on the N900
(Although digging around in the possibilities of exporting/importing files inbetween the two of them is quite interesting.)
Last edited by Koeus; 2009-12-03 at 11:32.
|
|
2009-12-03
, 12:35
|
|
Posts: 402 |
Thanked: 229 times |
Joined on Nov 2009
@ Missouri, USA
|
#18
|
Agreed Koeus! Really, I was thinking Anki looks much more polished (not to mention easier to spell) than mnemosyne. Perhaps I'll look into trying to get an initial port of Anki started today (no promises). Of course, Even if I did get it ported, i'm not so sure about making it finger friendly :-( -- may be a task for someone else.
|
|
2009-12-03
, 13:12
|
|
Posts: 21 |
Thanked: 1 time |
Joined on Nov 2009
@ Sweden
|
#19
|
Originally Posted by aspidites
*laughs* I've been cursing to myself every time I've had to write "Mnemosyne", so I totally agree with you on that XD
Agreed Koeus! Really, I was thinking Anki looks much more polished (not to mention easier to spell) than mnemosyne. Perhaps I'll look into trying to get an initial port of Anki started today (no promises). Of course, Even if I did get it ported, i'm not so sure about making it finger friendly :-( -- may be a task for someone else.
If you feel like looking into porting Anki, then by all means, feel free (and encouraged ^^)! I'll help with whatever I can, but I suspect my mediocre Java skills won't do much good in this case. Nonetheless, I might be able to be of some use (I'm seriously considering looking into C++ and/or Python in the near future). And I dare say, what we need the most is probably just someone getting the project started, and then people can pick up on the way. Hopefully someone else might pick up on the finger touchy part
(And now, back to my Java examination (yet again); somehow this discussion is a lot more appealing =P)
|
|
2009-12-04
, 02:08
|
|
Posts: 402 |
Thanked: 229 times |
Joined on Nov 2009
@ Missouri, USA
|
#20
|
Well in it's current state, we wouldn't be able to port it because PyQt and PySide are both lacking QtWebKit modules in fremantle. I guess that leaves us with the option of either a) waiting until it becomes available or b) gutting QtWebkit portions out and replacing with something functionally equivalent (assuming it was only used for formatting).
Again, I don't really use Anki, so I'm not familiar with its interface. Does it load any external web pages, or would it be save to say that webkit is just for formatting?
Again, I don't really use Anki, so I'm not familiar with its interface. Does it load any external web pages, or would it be save to say that webkit is just for formatting?
All I need is a sample deck from each program and I can look into writing a converter.