* study vocab and sentences containing only known kanji
* review kanji using japanese prompts created from known vocab
* suspends and unsuspends new cards in vocab and sentence decks based on seen kanji in kanji decks
* adds prompts to kanji decks based on seen and new words in vocab decks
* lists kanji used in vocab and sentence decks missing from kanji decks
* lists kanji in kanji decks missing from vocab and sentence decks

https://sites.google.com/site/ankinihongo/

Is there a way to get this to work for kanji compounds or words, like, have the program compare a certain number of characters (kanji and/or kana) in a series, and suspend/unsuspend cards that way, in any decks you care to compare?

In other words, to suspend/unsuspend cards in a sentence deck if those cards have words/compounds that match the cards in another deck.

That's really awesome.

Do people really have that many suspended cards though? I don't know it might just be me, but I have been SRSing nearly a year and in that total time Anki has only auto-suspended like 10 cards or less. Which I just manually unsuspended. But it is cool that there is so much support for so many different features of Anki.

Yonosa Wrote:Do people really have that many suspended cards though? I don't know it might just be me, but I have been SRSing nearly a year and in that total time Anki has only auto-suspended like 10 cards or less. Which I just manually unsuspended. But it is cool that there is so much support for so many different features of Anki.

I guess this is for people who use pre-made decks and only want stuff they know unsuspended in the beginning.

Tobberoth Wrote:I guess this is for people who use pre-made decks and only want stuff they know unsuspended in the beginning.

yeah, sorry, I forgot to say what it's actually for...

so it's meant as an alternative to sorting pre-made decks by kanji, or to using the suspend by character plugin, the advantage being that it's dynamic and will adapt automatically as you lean new kanji in any order

also it can be used as an alternative to kanji-vocab for japanese prompts, being dynamic and based off your learnt vocab

nest0r Wrote:Is there a way to get this to work for kanji compounds or words, like, have the program compare a certain number of characters (kanji and/or kana) in a series, and suspend/unsuspend cards that way, in any decks you care to compare?

In other words, to suspend/unsuspend cards in a sentence deck if those cards have words/compounds that match the cards in another deck.

I'd like to add support for activating sentences based on vocab, but it's a bit trickier than doing it just on kanji -- you'd need to parse the sentence and obtain all the words used, converted to dictionary form, to be able to compare them you your vocab cards

I see. By the way, this script bombpersons worked on might be relevant in terms of logic behind the comparisons: http://forum.koohii.com/showthread.php?p...2#pid70022 - I never could get that stuff to work.

nest0r Wrote:I see. By the way, this script bombpersons worked on might be relevant in terms of logic behind the comparisons: http://forum.koohii.com/showthread.php?p...2#pid70022 - I never could get that stuff to work.

thanks, everything in that thread seems to have evaporated now, but radical_tyro's script might be useful...

cangy Wrote:

nest0r Wrote:I see. By the way, this script bombpersons worked on might be relevant in terms of logic behind the comparisons: http://forum.koohii.com/showthread.php?p...2#pid70022 - I never could get that stuff to work.

thanks, everything in that thread seems to have evaporated now, but radical_tyro's script might be useful...

I think these are the scripts bombpersons wrote in that thread: http://www.mediafire.com/?ywrok2zwmdw

nest0r Wrote:I think these are the scripts bombpersons wrote in that thread: http://www.mediafire.com/?ywrok2zwmdw

thanks, but they don't do any parsing, I think that was just in dawhite's

cangy Wrote:

nest0r Wrote:I think these are the scripts bombpersons wrote in that thread: http://www.mediafire.com/?ywrok2zwmdw

thanks, but they don't do any parsing, I think that was just in dawhite's

Well damn. The worst part is, even if you eventually get, say, Mecab working as a parser for such a plugin or something, I wouldn't be able to use it because Mecab doesn't work for me (though Mecapi, the Web API for Mecab, does work, as does JParser from the Translation Aggregator at Hongfire).

cangy Wrote:I'd like to add support for activating sentences based on vocab, but it's a bit trickier than doing it just on kanji -- you'd need to parse the sentence and obtain all the words used, converted to dictionary form, to be able to compare them you your vocab cards

you don't need to do this, do you cangy? Can't you just relegate responsibility to the user?

e.g. make the user input "食べ” To match all the conjugations of　食べる

I understand it's far from perfect as the base form of　見る、　”見”　ｗill match many more things than just conjugations of　見る、　But I think it is a reasonable trade-off.

：）

omg cangy, i'd never noticed it before, but all this time i've been doing it 123578946.

juniperpansy Wrote:e.g. make the user input "食べ” To match all the conjugations of　食べる

I understand it's far from perfect as the base form of　見る、　”見”　ｗill match many more things than just conjugations of　見る、　But I think it is a reasonable trade-off.

yeah, as you say, when you get to something like 見る, it degenerates into just being based on kanji only anyway...

I have thought of manually entering the words, but I'd do it for the sentences -- enter the words in the sentence you don't yet know and it'll be unsuspended later when you've covered that vocab

chamois Wrote:omg cangy, i'd never noticed it before, but all this time i've been doing it 123578946.

sorry, I've got no idea what that means!

your sig: my stroke order has been wrong for ages...

chamois Wrote:your sig: my stroke order has been wrong for ages...

By 'sig' do you mean 'signature', and by 'signature' do you mean icon/avatar? ;p

Also, I always write that kanji with 4/5 reversed (as it's shown in the icon).

Edit: No wait, nevermind, I write it how it is in the icon. Muscle memory remembers better than I do. ;p

Edit 2: No wait, I write the 6 last.

Edit 3: But for some kanji, for the primitive on the right, I often write the 5 first.

Edit 4: None of these edits reflect my thoughts on stroke order, which is that it's only important insofar as when you're learning the kanji, you have a consistent framework to develop an internalized familiarity/motor program for the characters, but in practice, studies show that stroke order is only marginally important (minor contribution to recall of radicals), it's the radicals that are important for recall.

Edit 5: No wait, I write it how it is in the icon. Definitely. I think. It depends on how long I make the horizontal stroke. If I make it too short, I write the 6 last.

the latest version can put the keyword into the seen prompt as a fallback when there are no seen words

This is another plugin that is very useful to me. I am currently studying RTK1, but simultaneously I work through kore. I sorted kore by frequency and use swanki to unsuspend/suspend vocab cards of which I've seen all kanji already in RTK1.
To quote the creator:

cangy Wrote:so it's meant as an alternative to sorting pre-made decks by kanji, or to using the suspend by character plugin, the advantage being that it's dynamic and will adapt automatically as you lean new kanji in any order

It works seamlessly with multiple decks: my RTK1-3 deck and Custom_kanji deck (with f.i. kanji from the RTK supplement) are parsed for seen kanji, and sentence and vocab cards are unsuspended of KO2001 and kore.

Fyi, with only 500 kanji you can already study ~1600 kore words, most of them kana-only, but certainly not all.

Some kanji were missing for which there were words in my kore deck, so I added those. In the end I decided adding all kanji from that "Beyond RTK3 thread" (for now suspended of course) but that is beside the point. All kanji for which there are no words (atm quite large an amount you can imagine with all the kanken kanji in there...) can be looked up in a text file as well. Very friendly.

I've disabled the function that updates the kanji decks for now, since I did not yet look at its workings and don't need the Japanese keywords yet. I guess I'll start using that as well once I've got some vocab down the belt.

I would still like to see something like suspend/unsuspend (across decks, based on seen) by string, personally. ;p