Also, I really don't like where the "New Text" button is. Happy to entertain suggestions on where to put it. Same goes for the title field - if you're going quickly, it might be where you accidentally dump in text to analyze.
3. おしまい - it cannot yet handle "this is the word, but the spelling is different". Will consider for future upgrade.
4. こと fixed
5. 叫んでるの fixed
6. 死にたがってる <-- weird, still working on it
7. Names - this has nothing to do with the name dictionary that jisho uses. It's being split up by the layer below renshuu, which is something I cannot control. There might be a way to do name searches in the future, but that's a significantly large project.
8. うわあ not fixable at present.
9. 言わなくちゃ fixed
10. そりゃ - so, that's not going to be fixable at the moment - there is not a way to replace it with different kana, then hold onto the original - the data would still come up very strange on the end of the analysis.
11. 涙してた - unfixable at the moment due to kuromoji (doesn't have the dict entry for it)
12. 輝ける fixed
13. 彷徨ったの fixed
14. じゃん fixed
While I'm grateful that members of the discord are trying it, I *need* to have bugs reported here - there is additional information I need (see first post), and I can't keep track of those in the discord.
Okay, to report this „officially“ and so others know it has been reported: Numbers and acronyms etc. seem to be counted as words and marked unknown which might give an incorrect result.
They are not affecting the numbers, they were just colored wrong (and the numbers did not have the correct kana applied to them). Both of these should be fixed!
FYI - anyone reporting issues, please try and paste the original text (or part of text) in here (in addition to the screenshot) - that will make it easier for me to look into it :)
Issue not related to parsing, but I tried doing a large chunk of text ( https://www.renshuu.org/index.... ) and eventually had to give up waiting for it. Then I got 4 notifications. The first three were for small chunks I did a long time ago, and finally the fourth one was for this text. I might guess those first three were also the ones I closed before they completed, so now I got all of the notifications at once.
First of all I wanna say that it's a really cool tool
I'm not sure if maybe I didn't understand the usage, but I tried to create a schedule of the words I didn't know, and it would be cool if you could set a jlpt level, because otherwise words you know that you haven't studied before on renshuu, will be included.
Also I didn't quite understand the color code in the reader, because some words I didn't know were red and others weren't.
That's a good suggestion as a filter, I'll definitely add that when I get a chance!
The red words are ones that could not be found in the dictionary (or at least, the program couldn't find them) - those are definitely ones I'd like to hear about if you definitely think they exist in renshuu.
Dave - it'll help if I can get snippets of the original text, as the parser acts differently depending on the text around it. Specifically, that first one with いわば - can you copy the line that it appeared in?
The red words are ones that could not be found in the dictionary (or at least, the program couldn't find them) - those are definitely ones I'd like to hear about if you definitely think they exist in renshuu.
Ah, okay, that makes sense. It was a pretty old book, so it's possible they were kanjis that aren't in use anymore. It was always just one random kanji inside a word, so it's likely that that was the reason.
Ah, the initial one contains spaces, new lines, all that stuff. It's a "best guess" until it actually parses up the file. I'll adjust it to fix that number once parsing is complete.
Yay, that's great! It's really helpful! Although I noticed that you can't choose two options from the same category without no results remaining (for example N1 and not jlpt).
