nani issueshttps://framagit.org/groups/nani-project/-/issues2024-03-08T15:18:14Zhttps://framagit.org/nani-project/nani-app/-/issues/38Kanji alive audio data2024-03-08T15:18:14ZJulien LepillerKanji alive audio datahttps://github.com/kanjialive/kanji-data-media/tree/master/examples-audiohttps://github.com/kanjialive/kanji-data-media/tree/master/examples-audiohttps://framagit.org/nani-project/nani-app/-/issues/37Add proper names2024-03-09T17:26:25ZJulien LepillerAdd proper nameshttps://www.edrdg.org/enamdict/enamdict_doc.htmlhttps://www.edrdg.org/enamdict/enamdict_doc.htmlhttps://framagit.org/nani-project/nani-app/-/issues/36More kanji information2024-03-07T11:44:38ZJulien LepillerMore kanji informationKanjidic has additional information, which could be shown such as: is it a jouyou, jinmeiyou, JLPT level, and various index numbers in other diccionaries, but that could inflate the size of the file…Kanjidic has additional information, which could be shown such as: is it a jouyou, jinmeiyou, JLPT level, and various index numbers in other diccionaries, but that could inflate the size of the file…https://framagit.org/nani-project/nani-app/-/issues/35No kanji is shown when there are no word results2024-03-09T13:03:19ZJulien LepillerNo kanji is shown when there are no word resultsWhen a search gives no word results, it's considered failed and the app doesn't show any kanji, even if they appear in the search bar. An example would be looking up a rame or a rare kanji. Similarly, no examples are shown (if there were...When a search gives no word results, it's considered failed and the app doesn't show any kanji, even if they appear in the search bar. An example would be looking up a rame or a rare kanji. Similarly, no examples are shown (if there were examples before, they are still shown)https://framagit.org/nani-project/nani-app/-/issues/34Add reset button to radical screen2022-08-22T17:43:41ZDavidAdd reset button to radical screenA clear/reset button in the radical selection screen is a lot more efficient for searching multiple kanji than deselecting each radical manually.A clear/reset button in the radical selection screen is a lot more efficient for searching multiple kanji than deselecting each radical manually.https://framagit.org/nani-project/nani-app/-/issues/33Crash Logs2022-07-29T16:39:08ZgghhkmCrash Logs[crash1.txt](/uploads/0f66188626d5f977f9fd1f1d318d515e/crash1.txt)
[crash2.txt](/uploads/940007c1600262d215dd45b1a8b0c02c/crash2.txt)
[crash3.txt](/uploads/7e72ce15ec1d68b694b6fe85fb531416/crash3.txt)[crash1.txt](/uploads/0f66188626d5f977f9fd1f1d318d515e/crash1.txt)
[crash2.txt](/uploads/940007c1600262d215dd45b1a8b0c02c/crash2.txt)
[crash3.txt](/uploads/7e72ce15ec1d68b694b6fe85fb531416/crash3.txt)https://framagit.org/nani-project/nani-app/-/issues/32Store dictionary data in application data instead of cache2022-07-24T17:33:22ZDavidStore dictionary data in application data instead of cacheI'm not sure if there's a technical reason for it, but Nani downloads and stores dictionary data in the app's cache instead of the app's data. As application cache is usually frequently cleared en masse to save space, this means it's muc...I'm not sure if there's a technical reason for it, but Nani downloads and stores dictionary data in the app's cache instead of the app's data. As application cache is usually frequently cleared en masse to save space, this means it's much easier to lose and thus redownload the dictionaries than if they were part of the app data. In my phone's settings along (Huawei Mate 10 lite running Android 8.0), Clear Data has a confirmation prompt but not Clear Cache which is done instantly.
This is the storage of Nani with the radicals and JMDict_e dictionary installed:
![image](/uploads/090853783cbb6f3a9ae96185e163c12a/image.png)https://framagit.org/nani-project/nani-app/-/issues/31Use Common Voice dataset2022-07-16T11:54:51ZJulien LepillerUse Common Voice datasetThe dataset contains many sentences and pronunciations by native Japanese speakers. Although there are no translations, it could still be very useful.The dataset contains many sentences and pronunciations by native Japanese speakers. Although there are no translations, it could still be very useful.https://framagit.org/nani-project/nani-app/-/issues/30+1 button on results2024-03-07T09:18:05ZJulien Lepiller+1 button on resultsThis would give the possibility to click on a "+1" button (and maybe reset/-1) to "count" the number of times an entry was useful. This would count towards the score of an entry to make it more likely to appear in the future.This would give the possibility to click on a "+1" button (and maybe reset/-1) to "count" the number of times an entry was useful. This would count towards the score of an entry to make it more likely to appear in the future.https://framagit.org/nani-project/nani-app/-/issues/29Regex with Japanese ?2022-07-16T14:10:15ZJulien LepillerRegex with Japanese ?When using a Japanese keyboard, "?" is a different character. Should convert it so the regex system works.When using a Japanese keyboard, "?" is a different character. Should convert it so the regex system works.https://framagit.org/nani-project/nani-app/-/issues/28Crash when attempting to search kanji/words2022-07-16T14:09:32ZDavidCrash when attempting to search kanji/wordsWhen entering a single kanji or a word into the search bar and hitting the search button on the keyboard, Nani instantly crashes. This is true if the kanji was selected from the radical view or written with a keyboard.
Currently install...When entering a single kanji or a word into the search bar and hitting the search button on the keyboard, Nani instantly crashes. This is true if the kanji was selected from the radical view or written with a keyboard.
Currently installed dictionaries include `radicals` and `JMdict_e`. App version 0.5.0 (same as F-Droid). Device used is a Huawei Mate 10 lite running Android 8.0. The logcat files include all log entries with Warn or above levels, as well as more device info. If necessary I can provide more logs.
[logs-2022-07-16-13-06-18.zip](/uploads/e3a1f136ca3d8fba426eb5a3eebfef3a/logs-2022-07-16-13-06-18.zip)https://framagit.org/nani-project/nani-app/-/issues/27no kana for some entries2024-03-07T09:14:54ZJulien Lepillerno kana for some entriesFound an entry with no pronunciation: 亡骸Found an entry with no pronunciation: 亡骸https://framagit.org/nani-project/nani-app/-/issues/26Confusion between same-headword entries2022-07-02T15:42:58ZJulien LepillerConfusion between same-headword entriesWhile doing some tests, I found that a few entries have the same headword, like 女の子 or 女子. The app will always merge them because of the identical headword I think. This was implemented for merging identical headwords from different dict...While doing some tests, I found that a few entries have the same headword, like 女の子 or 女子. The app will always merge them because of the identical headword I think. This was implemented for merging identical headwords from different dictionaries (eg. from different languages). However, the merge result shows the uncommon reading for these two entries, and the app does not show alternate readings either.https://framagit.org/nani-project/nani-app/-/issues/25Handle dictionary deletion and format update2024-03-09T17:27:00ZJulien LepillerHandle dictionary deletion and format updateCurrently, if a dictionary is removed from the server, the file might be kept on the device, but it will no longer be usable, it will not be listed and cannot be removed.
When a format update occurs, the file can be downloaded but is no...Currently, if a dictionary is removed from the server, the file might be kept on the device, but it will no longer be usable, it will not be listed and cannot be removed.
When a format update occurs, the file can be downloaded but is not usable by the app at all.https://framagit.org/nani-project/nani-app/-/issues/24Sometimes, furigana is not shown2021-07-31T00:59:15ZJulien LepillerSometimes, furigana is not shownFor some words, furigana is not shown at all. I think this is when the headword has kanji and katakana, eg. 国際連合カンボジア暫定統治機構【こくさいれんごうかんぼじあざんていとうちきこう】. The issue being that the app can't match the katakana part with the kanji pronunciation...For some words, furigana is not shown at all. I think this is when the headword has kanji and katakana, eg. 国際連合カンボジア暫定統治機構【こくさいれんごうかんぼじあざんていとうちきこう】. The issue being that the app can't match the katakana part with the kanji pronunciation at all. A solution would probably be to convert katakana to kanji on both side before matching parts.https://framagit.org/nani-project/nani-app/-/issues/23Remove duplicate accent patterns2021-07-30T22:50:26ZJulien LepillerRemove duplicate accent patternsSome words have duplicate pitch accent patterns, as in `1,1`.Some words have duplicate pitch accent patterns, as in `1,1`.https://framagit.org/nani-project/nani-app/-/issues/22Empty keyword for an entry in JMdict_dut2022-06-25T17:31:33ZTom SmedingEmpty keyword for an entry in JMdict_dutWhen searching for 国, I get an entry from JMdict_dut that seems to have no keyword: [screenshot](https://tomsmeding.com/f/nani-shot2.jpg) (note: different screenshot than in previous issue, now with kana reading instead of furigana readi...When searching for 国, I get an entry from JMdict_dut that seems to have no keyword: [screenshot](https://tomsmeding.com/f/nani-shot2.jpg) (note: different screenshot than in previous issue, now with kana reading instead of furigana reading). The entry keyword is 国際連合教育科学文化機関 (or at least: my IME gives that for the reading and entering that looks up the entry for me, so I assume it's at least a substring); it has reading こくさいれんごうきょういくかがくぶんかきかん and its definition text is:
> {afk.} unesco; wetenschap en cultuur; vn-organisatie voor internationale samenwerking inzake opvoeding
And it has accent numbers `19, 19`. (Deduplication might be nice?)
The definition text translates to:
> {abbr.} unesco; science and culture; UN organisation for international collaboration regarding child-rearing
Regarding a question on the previous issue:
> The strange thing is that I can't find any "wetenshap en cultuur" in the source file for the dictionary. Does it make sense in Dutch, or is that just random words?
Odd that it seems to not occur; perhaps having the keyword will help. :P
"Wetenschap en cultuur" means "science and culture" and it makes perfect sense, also in the context of unesco (which stands for United Nations Educational, Scientific and Cultural Organization); case in point, we have a Ministerie van Onderwijs, Cultuur en Wetenschap (ministry of education, culture and science) in NL.https://framagit.org/nani-project/nani-app/-/issues/21Search result ordering is suboptimal2021-07-29T14:22:06ZTom SmedingSearch result ordering is suboptimalThanks for the nice, clean app, and thanks for making it open source! :)
When searching for certain more common or shorter words, the most relevant results are quite far down the list, sometimes to the point that the main dictionary ent...Thanks for the nice, clean app, and thanks for making it open source! :)
When searching for certain more common or shorter words, the most relevant results are quite far down the list, sometimes to the point that the main dictionary entry for that word is not even shown at all.
Examples:
- 必要 (shows 5 compounds starting with 必要, then 必要 itself, then 4 more compounds)
- 家族 (same, but 6 before, 4 after)
- 国 (the word itself doesn't even show up -- incidentally, I also get a word here with an empty keyword or something? [screenshot](https://tomsmeding.com/f/nani-shot1.jpg))
- 謎 (same, word itself doesn't show up)
I have nani 0.3.2.1 from fdroid, with the following dictionaries installed: wadoku_pitch, radicals, JMdict_dut, JMdict_e.
Instead of the current behaviour, I would at least like to see an exact match (if any) show up at the top of the list. The ordering of the other results is subject to opinion anyway, and less important than the exact match. (I don't know what you currently use as result ordering.)
Thanks again!https://framagit.org/nani-project/nani-app/-/issues/20Add KanjiDic support2021-07-24T00:59:40ZJulien LepillerAdd KanjiDic supportRelated to #16, this is about adding support for looking kanji up. I am currently working on this, so I expect it to be merged soon. This feature would be a different screen that shows kanji lookup results, by using a tab layout. Current...Related to #16, this is about adding support for looking kanji up. I am currently working on this, so I expect it to be merged soon. This feature would be a different screen that shows kanji lookup results, by using a tab layout. Currently, I have this tab layout and kanji lookup. What needs to be fixed:
* Use only one line for meaning in the same language instead of multiple lines. Kanji can have a lot of meanings that are very short words. This is taking a lot of space, especially if there are multiple languages.
* Related to that, maybe change the layout a little bit.
* Currently it only show kanji that are in the search string. If there is no kanji in the user search, we could get the kanji from the first result, or from all the results. Maybe sort them so the most frequent (in the result or kanji frequency?) appear first.
Current information extracted from kanjidic are number of strokes, meanings and pronunciations.https://framagit.org/nani-project/nani-app/-/issues/19[feature request] Pitch accent line representation2022-06-25T17:30:42ZFrostKiwi[feature request] Pitch accent line representationWould be great to have the pitch accent represented as a drawn line, as seen in other apps.
Though I guess it's time for me to to learn that number system as well^^
![IMG_20210122_033239](/uploads/ad5801bee2fd53dfd4a3f7ed359f629d/IMG_20...Would be great to have the pitch accent represented as a drawn line, as seen in other apps.
Though I guess it's time for me to to learn that number system as well^^
![IMG_20210122_033239](/uploads/ad5801bee2fd53dfd4a3f7ed359f629d/IMG_20210122_033239.jpg)