mvysny / aedict

Original Aedict 2 source codes

Aedict

Please use the Aedict Issue Tracker to also report bugs for Aedict 3. The files present in this repository contain sources for the old Aedict 2, please see below for more information.

Use Aedict 3 on your desktop

Fire up your browser and head to Aedict Online - an online clone of the Android Aedict. Be warned - it is still in the Alpha phase though :)

Aedict Ultimate

Aedict is now subscription-based, to help fund further development. Please read more at the Aedict Blog, the Aedict Ultimate Blogpost.

Aedict 2 Sources

This is the original repository of the old Aedict 2 code base. Aedict 2 is no longer maintained by me, but I have open-sourced the sources under GPL3. Please feel free to check them out. Beware - the sources are very old, the build system uses an unofficial Maven plugin, so there is no guarantee that you will be able to build the app.

The sources are open-sourced under the GPL3 license.

aedict's Issues

Kanji search broken in 0.9

What steps will reproduce the problem?
1. Look up 'haha' in the Japanese search.
2. Select the top result: mother. You are taken to the entry detail screen.
3. The kanji for 'mother' is in the kanji search box. Click 'Search'.

What is the expected output? What do you see instead?
In v0.8 you would see the hundreds of entries that featured that kanji. But
now you get 'No results' in the default dictionary.

What version of Android are you using?
1.6 on Archos 5 IT.

Keep up the good work, I love the software!

Original issue reported on code.google.com by [email protected] on 4 Mar 2010 at 8:52

Crash selecting examples in Tanaka dict

What steps will reproduce the problem?
1.
2.
3.

What is the expected output? What do you see instead?


What version of Android are you using?


Please provide any additional information below.

Crash Report:

English Search "meeting" Check "Search in example" Click Search Search
results state "Searching in the Tanaka dictionary" First search result is
"The Meeting was Closed" Click "The Meeting was Closed" ERROR: "Sorry! The
application Aedict (process sk.baka.aedict) has stopped unexpectedly.
Please try again. Force Close.

Choosing any of these top search results produces same error.

It would be nice behavior if choosing a search result led to a highlighted
entry. This way, you could show another person which sentence you wish to
express.

Please let me know if you need a stacktrace.

Original issue reported on code.google.com by ken.laberteaux on 9 Mar 2010 at 5:13

Add a notepad functionality

User should be able to save several kanjis or jp words and access them 
easily. Use-cases:
- when performing a lookup of a word consisting of multiple kanjis. User 
should be able to store the kanjis found and should be able to perform a 
lookup of a word composed of such kanjis.
- when learning new set of kanjis, it could help having them available 
quickly as a list, without the need to search for them again and again.

Implementation: a new activity, showing a list of user-defined jp 
kanji/word entries (with appropriate translations, if available) will be 
created. Clicking on a list item should append the kanji/word to an 
EditText. User should be able to perform both search and analyze activities 
for the word currently entered in the EditText. The list must be 
persistent. Items should be deletable (long-clicking on the item will bring 
up the menu).
TODO: how to add an entry to this notepad.

Original issue reported on code.google.com by [email protected] on 18 Dec 2009 at 2:21

Show a verb inflection chart

I have a book of verbs at home. Although Japanese verbs are regular I find
it useful to have all the forms some times. That would be a good feature to
have.

Original issue reported on code.google.com by [email protected] on 6 Mar 2010 at 9:22

osusumu not translated

What steps will reproduce the problem?
1. Try to translate the following sentence: 
2.今週のおすすめﾊﾞｰｹﾞﾝTVｹﾞｰﾑ
3.

What is the expected output? What do you see 
instead?
Osusumu is left untranslated 

Please use labels and text to provide additional 
information.

Original issue reported on code.google.com by [email protected] on 26 Feb 2010 at 7:45

RFE - selection of word and output into textpad or mail

Currently aedict allows lookup.

I'd like to be able to select a word I looked up and insert it into my mail.

An easy way to do this (an also add Japanese support for handsets without
it) would be to use
http://www.adamrocker.com/blog/273/now_mushroom_application.html

The mushroom application above ties into Simeji IME and using it I believe
we could for example type in "cat" using Simeji find the kanji, select the
kanji (this selection functionality needs to be added), and the mushroom
app would output our selected kanji to notepad.  The other way - looking up
by Japanese would work as well.

Is this beyond the scope of your project? 

All the Best,

Shawn

Original issue reported on code.google.com by [email protected] on 12 Sep 2009 at 12:01

Tanaka download not offered

What steps will reproduce the problem?
1. Remove the tanaka dictionary
2. Search for japanese "haha"
3. longclick and pick "Search in examples"

What is the expected output? What do you see instead?
A dictionary download should have been offered. Instead, an error is displayed.

Please use labels and text to provide additional information.

Original issue reported on code.google.com by [email protected] on 2 Mar 2010 at 11:52

RFE - configurable results

When using the simeji input method, can we please choose what results to
enter. 

 For example, when writing e-mail with simeji, enter "くるま"　「ie. kuruma
in hiragana」 and use the aedict integration function.  

Before choosing the desired translation, I'd like to be able to hit menu
(which now shows "show kana table" and "configuration") and in the
configuration menu, I'd like to be able to choose what gets input.

Now if I make a selection by searching via hiragana, I just get the English
gloss input.  There are times when I really want to lookup via hiragana and
input the kanji and other times I want to lookup via hiragana and input all
data -- hiragana -kanji and English gloss (to study).


Shawn

Original issue reported on code.google.com by [email protected] on 7 Nov 2009 at 1:02

Ambiguity with ん

I realise that this is perhaps a feature, not a bug.

What steps will reproduce the problem?
1. In the main search screen, enter `sonna` in the Japanese search box.

What is the expected output? What do you see instead?
I would hope to see the entry for そんな, but I get no results. This is 
because it is searching for みんあ. If you enter `sonnna`, it works.

I know that `nn` is sometimes used to enter ん, but I think this could 
confuse some people (it confused me at first). Would it be possible to 
make `sonna` search for both そんな and みんあ? Failing that, maybe you 
could just do something to bring the issue to people`s attention.

What version of Android are you using?
1.6 on Archos 5 IT

Thanks for the new version, the radical lookup is looking really nice now!

Original issue reported on code.google.com by [email protected] on 26 Mar 2010 at 12:57

Add support for half-width katakana

The half-width katakana should be supported as well, both in search box and 
in the "translate" box.

Original issue reported on code.google.com by [email protected] on 2 Jan 2010 at 8:22

Random IndexOutOfBoundsException popping after a search

Random IndexOutOfBoundsException popping after a search. I have logs on my
phone so I'm hoping to squash this one pretty quickly.

Original issue reported on code.google.com by [email protected] on 4 Mar 2010 at 8:43

Add support for translating sentences

Allow aedict users to be able to translate sentences, e.g. copy-pasted from 
a browser or the like. This should be implemented by adding the "Analyze" 
button to initial activity (the lookup activity), which will optionally 
perform word analysis instead of character analysis, as requested in Issue 
23

Original issue reported on code.google.com by [email protected] on 18 Dec 2009 at 2:10

• Blocked on: #23

Problem with indexer

What steps will reproduce the problem?
1.I'd like to index fj.utf edict file from http://dico.fj.free.fr/dico.php
  (15000 entry-long french-japanese edict file)
2.I try this command line :
  ai -f fj.utf
or
  ai -f fj.utf -e utf-8
3.I can't make indexer work neither with fj.dct nor fj.utf nor else.

What is the expected output? What do you see instead?
C:\aedict-indexer>ai -f fj.dct
org.apache.commons.cli.ParseException: At least one of -u, -d or -f switch 
must
be specified
        at sk.baka.aedict.indexer.Main.<init>(Main.java:122)
        at sk.baka.aedict.indexer.Main.main(Main.java:72)
Indexing failed: org.apache.commons.cli.ParseException: At least one of -
u, -d o
r -f switch must be specified

What version of the product are you using? On what operating system?
Indexer0.4-zip on WinXp SP3 command line with Java Standard Ed. build 
1.6.0_15-b03 installed.

Original issue reported on code.google.com by [email protected] on 2 Nov 2009 at 12:37

Attachments:

• fj.utf
• fj.dct
• fj.asc

Kanjipad warning message reduces available space

I just updated, and the new warning message on the kanjipad is a little 
annoying, since it reduces the available space. What chance of being able to 
disable that in options? It only really needs to be seen once, anyway...

Original issue reported on code.google.com by [email protected] on 6 Mar 2010 at 9:17

Add a simple translate Widget

Add a simple translate Widget: just an EditText with a button, which 
directly transfers user to the Results Activity.

Original issue reported on code.google.com by [email protected] on 28 Oct 2009 at 9:16

'Reading' search issue: "kyou"

What steps will reproduce the problem?

1. Search for 'kyou'. (Either in "Japanese search" or "Reading")

What is the expected output? What do you see instead?

The output should include 'today', beyond others. It won't, even when using
the (kana) reading returned from an "english search" of 'today'.
Exact search will return only 'cooperation' and 'hungary'.

What version of Android are you using?
Android 1.6 w/ Aedict 0.8

Sorry if that's covered already, or somehow was my bad.

Original issue reported on code.google.com by [email protected] on 16 Jan 2010 at 11:31

Implement the kanji lookup based on the SKIP system

The SKIP system: http://www.basic-japanese.com/Hilfsdateien/skipCode.html
The system probably cannot be used without permission from Jack Halpern. I 
sent him a mail and waiting for reply.

Original issue reported on code.google.com by [email protected] on 20 Nov 2009 at 9:20

custom / multiple dict files (enhancement request)

What steps will reproduce the problem?
1.Jim Breen's edict file is converted into Lucene-indexed format
2.it works
3.i assume that ANY .dct file working on Jim Breen's JWPCE or .jli file 
working on JLookUp could be converted and loaded onto Aedict

What is the expected output? What do you see instead?
fj.ct is like french-japanese edict file. I'd like to use it but I even 
don't know how to convert .dct into Lucene-indexed format.
I'd like to be able to choose one or several dic file in aedict just like 
in jwpce or jlookup

What version of the product are you using? On what operating system?
Aedict 0.3 / Android 1.5 / HTC Magic

Please provide any additional information below.
application is already great, I'm just one of those boring french users ;p

Original issue reported on code.google.com by [email protected] on 29 Oct 2009 at 9:37

Attachments:

• fj.dct

Missing the "About" dialog

The "About" dialog is no longer accessible from the application.

Original issue reported on code.google.com by [email protected] on 7 Dec 2009 at 8:53

Nihon-shiki support

I'd like to be able to enter nihon-shiki romaji, not just hepburn. Could
this feature be added? Thanks.

Original issue reported on code.google.com by [email protected] on 13 Oct 2009 at 8:54

No romaji in example search

What steps will reproduce the problem?
1.
2.
3.

What is the expected output? What do you see instead?


What version of Android are you using?


Please provide any additional information below.

Log Collector version: 1.1.0
Device model: Nexus One
Firmware version: 2.1-update1
Kernel version: 2.6.29-bacon1
shade@toxygene )
#1 PREEMPT Thu Jan 7 21:34:22 EST 2010
Build number: ERE27

Click "search in examples" and enter any word in English Search. (e.g.
"meeting").  Several sentence examples are displayed.  Press and hold any
search result.  Select "Show romaji".  Nothing happens.  No romaji is shown
for example sentences.

Original issue reported on code.google.com by ken.laberteaux on 9 Mar 2010 at 5:27

Implement a kanji drawing recognition

There's an open source japanese handwriting recognition engine named TOMOE
http://tomoe.sourceforge.jp/cgi-bin/en/blog/index.rb

after a quick search, i found several links :
a utility i used on palm : http://www.wellscs.com/pilot/index.htm
you already know this one, but readers may not : 
http://ftp.monash.edu.au/pub/nihongo/#wince_r
the well-known kanjipad : http://fishsoup.net/software/kanjipad/
another one, lamerpad : http://www.debian.org.hk/~ypwong/lamerpad.html
http://chasen.org/~taku/software/ajax/hwr/
and else :
http://www.cs.arizona.edu/projects/japan/JavaDict/download/index.html
http://www.akalin.cx/2006/10/21/japanese-handwriting-recognition-systems/

Original issue reported on code.google.com by [email protected] on 7 Nov 2009 at 11:09

Exact matches first

Provide best matches first:
1. when performing a kanji lookup, show kanjis with lowest stroke count 
first;
2. when performing a search, show shortest words (=best matches) first

Original issue reported on code.google.com by [email protected] on 7 Dec 2009 at 8:56

Word Page / Kanji Lookup

I'm fairly sure that the Jim Breen dictionary already supports this (or at
least many sites which use the dictionary support this).  Basically I'm
asking if it would be possible to have a "home page" for each word where
the kanji can be shown in nice big letters for copying down and to show a
short reference as to what each kanji in the word means.

Original issue reported on code.google.com by [email protected] on 2 Nov 2009 at 5:50

Embed to IME (e.g. Simeji)

Allow aedict to be invoked from an IME, to allow quick word translation

Original issue reported on code.google.com by [email protected] on 3 Oct 2009 at 4:32

Add support for word analysis

The "Kanji analysis" activity should be extended with functionality to 
collapse separate kanji characters (where appropriate) to words and perform 
edict lookup on those words.

Original issue reported on code.google.com by [email protected] on 18 Dec 2009 at 2:07

• Blocking: #24

Problem with creating custom dictionary

I have tried to index 3 dictionaries (edict_080320.dic, eijiro v95.dic, 
waeiji v95.dic) using tips about CustomEdictFile on Your site, but they don't 
want to work with aedict (I've tried to make them work with aedict-indexer 
0.6-0.8). The ones which were downloading automatically are working fine.
I was trying new created dicts on Android 1.5, 1.6, 2.0.1 and 2.1 but always 
when I typing "Kanji" or "Flower" or other words, I see nothing, or a lot of 
strange symbols, no kanji at all.
During indexing were no problems. Everything went OK. Indexer created 4 files 
each time in zip folder (_0.cfx; _2.cfs (3, 5); segments.gen; segments_3) and 
I can't find any solution.

What should i do make them working?

Original issue reported on code.google.com by [email protected] on 28 Jan 2010 at 6:47

SKIP finds the kanjis but does not display them

Just perform a SKIP search. The kanjis are displayed in the title of the 
activity but not in the activity itself.

Original issue reported on code.google.com by [email protected] on 15 Jan 2010 at 8:19

(build error) Could not execute: Command = aapt

What steps will reproduce the problem?
1.mvn clean install


What is the expected output? What do you see instead?
I expect apk package in the aedict-apk/target/ directory - the
aedict-apk-VERSION-signed.apk

I get Embedded error: ANDROID-040-002: Could not execute: Command = aapt
package -m -J  /me/workspace/hg/aedict-apk/gen -M
/me/workspace/hg/aedict-apk/AndroidManifest.xml -S
/me/workspace/hg/aedict-apk/res -I
/me/.m2/repository/com/google/android/android/1.5-r2/android-1.5-r2.jar


What version of the product are you using? On what operating system?
mecurial repository
OSX  10.5.8

Please provide any additional information below.

It seems to be an issue with Masa
http://code.google.com/p/masa/issues/detail?id=9

Original issue reported on code.google.com by [email protected] on 11 Sep 2009 at 11:53

Please add example sentence support

It would be very nice to have example sentences available in the search
results. There is an excellent collection here:

ftp://ftp.monash.edu.au/pub/nihongo/00INDEX.html

See example.gz.

I imagine you probably know about it already, but just in case... It is
extremely helpful I find.

Many thanks for this app. I use it a lot!

Original issue reported on code.google.com by [email protected] on 28 Jan 2010 at 1:48

Allow verb deinflection

Allow optional verb deinflection when searching for japanese words.

Original issue reported on code.google.com by [email protected] on 31 Jan 2010 at 2:15

The Word Analysis only offers the first possibility

E.g. when analyzing はな only the "nose" translation is offered, however 
there are more translations (flower). The Analysis Activity should present 
results in an expandable tree, where the user should be able to pick the 
translation.

Original issue reported on code.google.com by [email protected] on 10 Mar 2010 at 10:18

Kanji Lookup

Implement a simple Kanji lookup, using just a stroke count and a radical 
picker. All information is stored in the kanjidic dictionary.

Original issue reported on code.google.com by [email protected] on 7 Nov 2009 at 11:12

Searching for hana yelds only the "nose" result

What steps will reproduce the problem?
1. Search for hana using non-exact match

What is the expected output? What do you see instead?
The list should include also the "flower" translation, like the "exact 
match" search.

Please use labels and text to provide additional information.
The problem is that the non-exact search for hana yields too many results, 
from which only the top 100 results (as provided by Lucene) is displayed.
To solve it, we can display all results retrieved from Lucene (which may be 
an overkill if 10000 or more results are returned). Perhaps better approach 
is to search for an unlimited number of results, but only the common words 
(P). Then the old top-100 search should kick in. Duplicates need to be 
removed.
Also, the user has to be informed that not all words were returned, and 
perhaps an option to display all results should be offered.

Original issue reported on code.google.com by [email protected] on 10 Mar 2010 at 10:25

Sort ordering is better in Tagaini Jisho

First of all, thank you for developing such a great dictionary app. 大好き
です。とっても便利んです。

I have one request, not so much a programmming request as a dictionary data
request. Would it be possible to allow the user to pick which data to use?
 On my computer, I've used edict (jisho.org) and JMDict (tagaini jisho),
and have found the latter much more useful. I know there is a JMDict module
for colordict, but colordict is a far inferior interfce than Aedict.

Any help would be appreciated.

Thanks!

Original issue reported on code.google.com by [email protected] on 4 Mar 2010 at 8:30

show Romaji in the ResultActivity

Really nice, only wished it would show Romaji in the search results list 
directly, instead of having to click on every results individually..

Romaji will be shown right before the english translation.

Original issue reported on code.google.com by [email protected] on 17 Nov 2009 at 2:09

Add stroke order diagrams

http://www.kanjicafe.com/using_soder.htm#download

Original issue reported on code.google.com by [email protected] on 14 Feb 2010 at 1:22

The Configuration dialog should receive the standard Android config dialog look

The Configure Activity should look like the Android 1.6 System Settings 
activity.

Original issue reported on code.google.com by [email protected] on 10 Mar 2010 at 10:27

Add support for KANJIDIC

Add support for KANJIDIC in the Analysis activity, to display information 
such as:
- all readings (including readings in names)
- stroke count
- SKIP code
- translations
- Nelson's radical
- grade

Original issue reported on code.google.com by [email protected] on 9 Nov 2009 at 12:35

Word Page

Basically I'm asking if it would be possible to have a "home page" for each 
word where the kanji can be shown in nice big letters for copying down and 
to show a short reference as to what each kanji in the word means.

Original issue reported on code.google.com by [email protected] on 7 Nov 2009 at 11:03

entering kana づ

The kana づ can't be entered using either du or dzu. Instead it must be
entered using xzu. Perhaps this should be mentioned in the kana table?

Original issue reported on code.google.com by [email protected] on 13 Oct 2009 at 8:57

Radical lookup display problem

I'm using Android 1.6 on an Archos 5 Internet Tablet. On the 'Radical
lookup' screen, the big table of radicals overflows the screen slightly, so
that the last two or three radicals are off the right-hand edge and can't
be seen or selected (as the page will not scroll horizontally). This
happens whether the screen is in landscape or portrait mode (in fact,
changing orientation can be used as a workaround, since the radicals that
are off the edge in one format can usually be found in the other).

Original issue reported on code.google.com by [email protected] on 3 Mar 2010 at 11:25

Add support for multiple edict files

Hi,

Since we saw that any lucene-formated edict file works on Aedict, can we 
imagine it to use several ? e.g edict (en-jp), fj (fr-jp), enamdic 
(names), kanjidic (kanji), compdic (computer and electronic terms)...
2 or 3 will be good.

Thanks

Original issue reported on code.google.com by [email protected] on 3 Nov 2009 at 8:58

Add more downloadable dictionaries

Add support for automatically downloading more dictionaries:
- ENAMDICT
- COMPDIC
- EDICLSD3

http://www.csse.monash.edu.au/~jwb/edict.html

no-go cases: multilingual dictionary (too huge for a phone), searching in 
all dictionaries (you'll have to pick the dictionary to search in, otherwise 
it would be too slow)

Original issue reported on code.google.com by [email protected] on 6 Jan 2010 at 11:16

Crash on Dictionary Download

What steps will reproduce the problem?
1. Open app for the first time on G1
2. Say yes to attempt to download Jim Breen dictionary
3. Force Close

What is the expected output? What do you see instead?
Not crashing

What version of the product are you using? On what operating system?
0.3, T-Mobile G1 on Donut

Please provide any additional information below.

Original issue reported on code.google.com by [email protected] on 22 Oct 2009 at 3:15

available space inaccurate

What steps will reproduce the problem?
1. Download and install
2. run
3.

What is the expected output? What do you see instead?

I don't expect to see a warning about remaining space 


What version of the product are you using? On what operating system?

Andoid ION (from I/O 09 - like a Magic I think)


Please provide any additional information below.

I am using an 8 gb card.  I have 7gb free. Could this the problem?

The download succeeds

Original issue reported on code.google.com by [email protected] on 30 Aug 2009 at 9:42

Tests fails on MacOSX with Java5

What steps will reproduce the problem?
1. mvn clean install
...
tests:

testToRomaji(sk.baka.aedict.JpUtilsTest)  Time elapsed: 0.001 sec  <<< 
FAILURE!
org.junit.ComparisonFailure: expected:<[haha]> but was:<[はは]>::

testToKatakana(sk.baka.aedict.JpUtilsTest)  Time elapsed: 0 sec  <<< 
FAILURE!
org.junit.ComparisonFailure: expected:<[ミニコン]> but was:<[????]>

testToHiragana(sk.baka.aedict.JpUtilsTest)  Time elapsed: 0.003 sec  <<< 
FAILURE!
org.junit.ComparisonFailure: expected:<[はは]> but was:<[??]>

I have:

 $ env
LC_ALL=en-US.UTF-8
LANG=en-US.UTF-8
LC_CTYPE=UTF-8


Any ideas?

Original issue reported on code.google.com by [email protected] on 5 Oct 2009 at 11:23

Pressing the 'search' button should open the Search Activity

Pressing the 'search' button should open the Search Activity (the first
activity opened by Aedict on launch).

Original issue reported on code.google.com by [email protected] on 16 Mar 2010 at 8:30

Feature request: Launching an aedict word search from another application

Note: It is not a bug, but a feature request.

I am one of the developers of AnkiDroid, an application that often 
displays Japanese words. It would be nice if a simple click allowed the 
user to launch aedict and automatically look up the word. Full 
explanation:
http://code.google.com/p/ankidroid/issues/detail?id=51

I am not an expert on how intents work in Android, but I often see an 
application launching another (for instance a messaging application 
launching a search in the address book), so it should be feasible.

Do you already provide such an interface?
If yes, I will research how to use it.
If not, would you be kind enough to offer it?

We seem to share a lot of users, so I guess aedict and AnkiDroid have a 
lot of integration potential. For instance the ability to add a given 
word from aedict to the current AnkiDroid deck would be very useful!

Cheers, keep up the great work!
Nicolas Raoul

Original issue reported on code.google.com by [email protected] on 3 Mar 2010 at 10:58

Show hint on reading

Long click on SearchResult and Analysis activity should allow user to see 
romaji reading of all hiragana/katakana characters. When clicked on the menu 
item the entire list should be converted to romaji (or back to 
hiragana/katakana).

Original issue reported on code.google.com by [email protected] on 6 Jan 2010 at 11:21