2.0.1 RC 1 ready for testing

[Peter Bienstman]

http://users.ugent.be/~pbienst/pub/mnemosyne-2.0.1-RC1-setup.exe

I'd appreciate if some people could check this version. If there are
no unexpected major bugs, this will become the official 2.0.1 release
in a few days or so.

Changelog:

-Fix occasional hang during review (Thanks to Scott Youngman and Chris Shanks
for helping to debug this).
-In cloze card types, you can now provide hints, e.g. [cloze:hint] will show
[hint] in the question as opposed to [...]
-Added Sentence card type plugin. If you put this in the 'Sentence' field:
"La [casa:house] es [grande:big]", you'll get three sister cards with the
following questions: "La casa es grande", "La [house] es grande",
"La casa es [big]"
-Take size of media player widget (Linux only) into account when calculating
the optimal split between question and answer.
-Fixed bug where card appearance settings would not be synced between
different
machines.
-Fixed issue in Activate cards, where a set without a name and a saved set
would interfere.
-The dialog where you name saved sets is now a combobox which also list the
current saved sets, to allow for easier updating of a saved set.
-Message boxes are now centered over the current window (card browser, ...) as
opposed to the main window.
-Longer initial interval if the first grade you gave a card is 5.
-Fixed spurious firing of missing media detection (reported by Geir).
-Fixed a few bugs when displaying the edit or preview windows from the card
browser and using PgUp and PgDown to move to the previous/next card (reported
by Chris Schmidt).
-When adding a duplicate card, rather than forbidding the user to add it, ask
the user if he really wants to do this (requested by Laurent Humblet).
-The webserver now supports the -d option to specify the data directory
(requested by Sirius533).
-Fixed crashes in card browser with non-latin clone names (reported by Gabriel
Benamouzig).
-Prevent uninstalling a card type plugin if a clone of that card type has been
created, even if it does not yet have any cards.
-Further fixes for Hebrew (reported by Eliyahu Switzer and Chris).
-Upgraded many of the libraries on which Mnemosyne depends (PyQt, numpy,
matplotlib)
-Updated user contributed translations

[jacen...@gmail.com]

Just ran through 27 new cards and did not have one freeze. This fix works for me. Of course, my cards are text only. So I may not experience some of the issues others experience. Thanks again.
p.s. Love the new default look and font.

[Chris]

I discovered that hint trick totally by accident on one of the test versions.  And I was considering suggesting it too!

   - Chris Shanks

[Stephan Maseizik]

Peter Bienstman wrote:

> I'd appreciate if some people could check this version. If there are
> no unexpected major bugs, this will become the official 2.0.1 release
> in a few days or so.

I tested 2.0.1 on kubuntu 12.04 with ~160 revisions and about 25 new added
cards. I encountered no bugs except one that I think is a design-bug:

When adding a duplicate card that I want to combine with the old card,
mnemosyne is ignoring the old answer. Example:

Already in the spanish-german database:

Q: completar
Aussprache: <audio src="path/to/soundfile/completar.ogg>
A: vervollständigen, ergänzen

Now I want to add a new meaning to completar (not remembering that it is in
the database including a soundfile already and the best way would be to use
the cardbrowser). So I add completar as a new card:

Q: completar
A: ausfüllen

If I then choose to combine the two cards, all I get presented as a
combined card is:

Q: completar
Aussprache: <audio src="path/to/soundfile/completar.ogg>
A: ausfüllen

If I would accept this, as a result I would lose the meaning
"vervollständigen, ergänzen"! What I want as a result is the following:

Q: completar
Aussprache: <audio src="path/to/soundfile/completar.ogg>
A:vervollständigen, ergänzen;
ausfüllen

I think this is a bug und I should be given the opportunity to create the
described combined new card. So I really would like to see my old card and
my new card at the same time and be given the chance to edit a combined new
card. When entering the new combined card to the database the old and
formerly new card can be deleted from the database.

Stephan

[Peter Bienstman]

Quoting "Stephan Maseizik" <st....@web.de>:

> I think this is a bug und I should be given the opportunity to create the
> described combined new card.

I tried the following, with front-to-back only cards.

Added a card:

Q:question
A:1

Added a second card:

Q:question
A:2

When I choose 'merge and edit', I get to edit a card which looks like:

Q:question
A:2/1

So I cannot reproduce your problem. Can you give me a detailed series
of steps which reproduce the bug?

Peter

[Stephan Maseizik]

Peter Bienstman wrote:

> I tried the following, with front-to-back only cards.
>
> Added a card:
>
> Q:question
> A:1
>
> Added a second card:
>
> Q:question
> A:2
>
> When I choose 'merge and edit', I get to edit a card which looks like:
>
> Q:question
> A:2/1

I could reproduce your steps. It is working correctly for two-sided cards!
For my Spanish database I use a vocabulary-clone, putting the question in
the first, the soundfile into the second field and the answer into the
third. Sometimes I put a remark in the last field. You should be able to
reproduce the described behavior using a the vocabulary card type.

Stephan

[Peter Bienstman]

Fixed, thanks!

Peter

> --
> You received this message because you are subscribed to the Google
> Groups "mnemosyne-proj-users" group.
> To post to this group, send email to mnemosyne-...@googlegroups.com.
> To unsubscribe from this group, send email to
> mnemosyne-proj-u...@googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.
>
>
>

[Stephan Maseizik]

Peter Bienstman wrote:

> Fixed, thanks!
Well, I think it is me who has to say THANK YOU!

Stephan

[dangr...@gmail.com]

Hello,
I am a recent adopter of mnemosyne. After switching my deck from front-to-back to front-to-back and back-to-front I began experiencing the hang issue when trying to show answer as described in another thread. Hangs would occur within 5 to 15 cards or so. (Windows 7)

I read through that thread about 1/2 an hour ago, then downloaded and installed the 2.0.1-RC1 revision. I just finished going through my scheduled cards (about 60 or so) without a hang.

Thought I'd post to let you know of another user with good results.

Regards,
Dan

[Scott Youngman]

I've been through my deck four times and did not experience one hang (including three successive runs on a test database of 276 cards). That is such a complete contrast to the regularity of hangs in previous versions that I am confident the timezone bug has been fixed. Thank you, Peter!

[Scott Youngman]

Correction: the test database had 276 scheduled + Not memorized cards.

[Stephan Maseizik]

One more suggestion to card merging: I have to choose whether or not to
merge two cards before I am presented the old card that already is in the
database. So if I opt to merge the cards, I can only "preview" the result or
accept it with "OK". I think a "cancel merging"-option is missing to go
back to start. There could be reasons for changing your mind about merging
when you have been presented to old card from the database.

Regards
Stephan

[Peter Bienstman]

Normally, you are shown the data of the fields that need to be unique
in the dialog that you are presented with.

Peter

Quoting "Stephan Maseizik" <st....@web.de>:

[Stephan Maseizik]

Peter Bienstman wrote:

> Normally, you are shown the data of the fields that need to be unique
> in the dialog that you are presented with.

I attached two screenshots to show what I mean. In the first picture, where
I have to choose whether or not to merge, I don't know the answer of the old
card that already is in the database (which is "A1"). So if I opt to merge,
there is no way back. That's why I think a "cancel" or "cancel-merging" -
button to go back one step, where I can choose to "add anyway" or "don't
add" would be helpful.

For me that has no high priority, but I think a user should always be given
the opportunity to undo his last step, if he thinks it was wrong. Just my
opinion.

Regards
Stephan

[Gnome]

I haven't had the timezone error. Just installed this release, reviewed a few cards and everything seems to be working fine.

[m...@unm.edu]

I'm using Linux and get a non-fatal error that brings up the following pop-up message after every card is graded:

Uncaught exception!
Traceback (innermost last):
File "/usr/lib/python2.7/site-packages/Mnemosyne-2.0.1_RC1-py2.7.egg/mnemosyne/pyqt_ui/review_wdgt.py", line 279, in grade_answer
self.review_controller().grade_answer(grade)
File "/usr/lib/python2.7/site-packages/Mnemosyne-2.0.1_RC1-py2.7.egg/mnemosyne/libmnemosyne/review_controllers/SM2_controller.py", line 135, in grade_answer
if self.rep_count % self.config()["save_after_n_reps"] == 0:
ZeroDivisionError: integer division or modulo by zero

I click OK and go to the next card.

[Peter Bienstman]

Your 'save after x reps' setting is set to zero. I've changed the
dialog so that it's no longer possible to put zero there, but if you
put it to zero, you'll need to change it manually.

Cheers,

Peter

> --
> You received this message because you are subscribed to the Google
> Groups "mnemosyne-proj-users" group.
> To post to this group, send email to mnemosyne-...@googlegroups.com.
> To unsubscribe from this group, send email to
> mnemosyne-proj-u...@googlegroups.com.

> To view this discussion on the web visit
> https://groups.google.com/d/msg/mnemosyne-proj-users/-/Te_iJ1V9R7MJ.

[Peter Bienstman]

Valid point. The dialog now shows the info from all the field so as to
help you make up your mind if you want to merge.

Peter

Quoting "Stephan Maseizik" <st....@web.de>:

[Stephan Maseizik]

Peter Bienstman wrote:

> Valid point. The dialog now shows the info from all the field so as to
> help you make up your mind if you want to merge.

Thanks Peter!
I installed the new revision, tested it and: I do like it! It is clear and
gives all the needed information without bloating the program.

Best regards
Stephan

[m...@unm.edu]

2.0.1 appears to be working on my two Linux boxes (2.0 had been crashing). This new version became available in the ArchLinux user repository today.