Vocabulary dictionary

Kanji dictionary

Grammar dictionary

Sentence lookup

test
 

Forums - Fall 2024 Focus: Text Analyzer Improvements

Top > renshuu.org > Feature Requests/Improvements > Finished/Rejected Requests

Page: 2 of 4



avatar
ハシュミナ
Level: 756
you can already do that by adding all words to a list and then using the Advanced search to show the overlap between the words in the list you created and the 10k word list or any other list you made (just not at the same time, you can only compare two lists, other wise it only shows vocabs that are in all lists you chose)

Yes, that's how I'm doing it at the moment but it's always a hassle to go back and forth and create like temporary lists and rename everything properly.

The community list navigation is not really user-friendly, like it always resets and uncollapses.

1
4 months ago
Report Content
avatar
Gibolt
Level: 558

An awesome feature would be using other's public / shared parsed documents, and being able to find those that have a high overlap with what you know.

I really want to find books / writings / lyrics / shows at my level, but it is slow to do so by uploading one at a time. Discovering new content based on existing known documents paired with my known words would be a killer feature for practicing.

1
4 months ago
Report Content
avatar
Gibolt
Level: 558

It would be great to remember where a document was last scrolled to for texts past a certain length (or number of lines).

When reading a chapter of a book, it is easy to lose place and annoying to find that place again the next day. This could be as simple as an indicator at the start of a line or a "scroll" to button (when available). Quite a few other reading tools I've used have this, and it makes jumping back in that much easier.

Keeping track of it could be done during scroll events or every so often. It should track which character / line was in the top left, not the scroll position, so that a different window size on later view wouldn't affect it.

0
4 months ago
Report Content
avatar
マイコー
Level: 283

An awesome feature would be using other's public / shared parsed documents, and being able to find those that have a high overlap with what you know.

I really want to find books / writings / lyrics / shows at my level, but it is slow to do so by uploading one at a time. Discovering new content based on existing known documents paired with my known words would be a killer feature for practicing.

I can say with fairly high confidence that this will not happen - it's more work than I want to consider to deal with copyright issues if submitted texts are made public.

1
4 months ago
Report Content
avatar
Gibolt
Level: 558

I can say with fairly high confidence that this will not happen - it's more work than I want to consider to deal with copyright issues if submitted texts are made public.

I figured that might be the case... Maybe grow a service to rival Kindle and Netflix in size, just to get rights access :D

Another option would be allowing pooling of stats + word lists for content, similar to how lists operate now. Instead, allow grouping by content source / chapter / episode... This should be ok since the original content wouldn't be maintained or accessible, and should be considered a Fair Use derivation for education purposes.

This would provide the same value, allowing filtering by source, to decide what to read/watch next. The actual content could then be found by user (or even linked to with affiliate links) through an official source.

0
4 months ago
Report Content
avatar
Gibolt
Level: 558

Currently status indicator while parsing is always 0. It would be nice if this actually started updating, and even better if it suggested total time/time to completion, if taking >~10 sec.

1
4 months ago
Report Content
avatar
Gibolt
Level: 558

Assuming you can track reading position of longer documents, it would be really awesome to track how long we spent reading a given document, and optimally what/how many/what percentage of words were clicked.

For multi-file documents (book chapters), you could chart out relative progression of reading speed / words clicked, plus chart out both reading speed and even predicted speed of remaining sections, to highlight how much someone is improving.

0
4 months ago
Report Content
avatar
Gibolt
Level: 558

Once file lists are added, it would be super handy to be able to paste one document + title, and then add a specific character to break the larger file into smaller items in a new list. For example, putting <br> or a rare unicode character between chapters, to end up with a list of all the files split apart.

The title could also use a custom string, which would be replaced by an incrementing file number on split.

0
4 months ago
Report Content
avatar
Gibolt
Level: 558

Currently, opening a file just opens a large modal on the main /breakit url. The browser back button goes back to previous url but doesn't update the content, instead of the file list since it is still the same url. When it doesn't react, I then click back again, going back 2 pages. Overall, it feels very unexpected.

Fixing this would be a nice QOL improvement, preventing a mistake I've made >100 times.

1
4 months ago
Report Content
avatar
すねだ
Level: 1628

How about adding some documents that are already out there open source? Ordering those by difficulty and parsing them might make them more accessible. (It would of course still be a lot of work, but maybe it can be outsourced to beta testers/community)

3
4 months ago
Report Content
avatar
gillianfaith
Level: 1073

Small QOL thing, but for the Pro features it would be nice to have the list option work the same as the schedule option and allow you to add to an existing list, rather than having to make a new list for each text.

2
4 months ago
Report Content
avatar
Gibolt
Level: 558

More QOL feature / setting ideas:

1. Render document as a single page, instead of paged.

2. Break pages at end of final line, rather than in the middle.

3. Allow adjusting vertical spacing between lines. Many other readers have more gap between paragraphs, to make the distinction slightly more obvious and improve readability.

4. Allow defining/ignoring red (missing) terms, and/or allow hiding the red until hover. Documents with names are spotted with red words, which is highly distracting from reading. Reparse doesn't allow defining words that aren't in the dictionary.

5. Enable auto-speak on word hover (if audio available).

6. Default to the Reader tab, instead of Analysis.

7. Once groups exist for documents, a Next button at the end, and maybe a Previous button at the beginning (when a next/prev file exists) would make navigation simpler.

8. Ability to mark a document as Read (small dot indicator?) in list

9. Ability to favorite a document

10. Notes attached to a document, and maybe even its own tab. Example: Adding notes on a grammar item. Tab could show all my notes on that document next to the line the note is about, or even all notes on files in a group.

11. Highlight words / lines that you find valuable. Similar to 10, but could indicate without interaction.

12. Inline / temporary parsing. Would parse the content immediately on pasting, update once parsed, and disappear once tab is closed. Even better if the content could be updated without having to open the separate edit document modal and find where to change the original source.

13. Load text from txt file / PDF. May not be worth the work, but would be nice not to have to open the file separately and then paste the content over. Could then default to use the file's title as well

14. Building on 13, bulk upload of files

15. Ability to cancel analyzing a file while it is processing. Currently have to wait until it is done.

16. Option to bias rendering dialog above/below hovered word

17. Adjust / invert font + background color (only for text analyzer)

18. Adjust font / font size (only for text analyzer)

0
4 months ago
Report Content
avatar
VoidWinter
Level: 57

1. Filter by frequency/JLPT. The current filter only allows "Nx and up". If there are lower JLPT/more frequently used words I don't yet know, I want to learn those first. Rather than both "Nx and up" and "Nx and down" allow to just multi-select the individual JLPT (or frequency) category in the filter.

2. Filter by frequency of words found in the analyzed text. If I'm reading a book, it makes more sense to learn a word used 100 times over a word used once. E.g. "only include words used more than XX times" where XX is a whole number input, defaulted to 1 (everything). Might be useful to extend this to Kanji and Grammar as well.

3. List what file types the tool allows to import. I spent an inordinate amount of time trying to figure out what I could use. Epub, then pdf, then docx and none of them worked. I'm guessing it only takes txt files? I ended up just copy/paste from the docx into the tool manually.

And yes, my ideal use case would be using it to make vocab lists for books I'm about to/currently reading.



1
4 months ago
Report Content
avatar
マイコー
Level: 283

3. I'm probably not going to be able to take the time to break apart anything other than txt files, sorry :(

1
4 months ago
Report Content
avatar
VoidWinter
Level: 57

3. I'm probably not going to be able to take the time to break apart anything other than txt files, sorry :(

No worries conversion in calibre is quick to do!

However even changing the wording to "Or send a txt file" would help clarify what file types are accepted. The choose file box currently allows any kind of file to be selected, so it's not intuitive what the tool allows.


3
4 months ago
Report Content
avatar
Václav89
Level: 133

Romaji support would be appreciated.

0
4 months ago
Report Content
avatar
マイコー
Level: 283

What do you mean by "romaji support"?

1
4 months ago
Report Content
avatar
マイコー
Level: 283

Alright, although there was a lot more than this, I got an initial batch of 16 changes/additions for the TA, and will be starting soon. I won't be able to spend so much time on it that I can cover EVERYTHING, but I think this will help a lot :). Excited to get started!

4
4 months ago
Report Content
avatar
マイコー
Level: 283


**[Improvement]** "Untitled" removed from Text Analyzer

**[Fix]** Text analyzer correctly updates loading percentage for larger texts

**[Improvement]** Text Analyzer uses file name when choosing title for text.

**[Improvement]** Can delete texts in Text Analyzer that are in the middle of analysis

**[Improvement]** Put Reader tab as default to be opened

**[Improvement]** Groups added to Text Analyzer


In particular, I'd love for everyone to play with the group names, and see if you have any issues.

3
3 months ago
Report Content
avatar
Gibolt
Level: 558

Notes on groups:

Super excited to have this feature! Initial thoughts:

- Minor: Don't show Ungrouped header / Dropdown when no other group exists

- Minor: Something is off about the header; feels a little weird rendered as a link with same emphasis as other links (minus the size). It is good enough, and I don't have a suggestion other than a chevron showing open state.

- Header vertical centering with text count would be nice.

- Got an undefined group when I tried to make my first one

- Group list isn't shown in the Adjust groups section. No option to edit / delete (I assume these are coming)

0
3 months ago
Report Content
Getting the posts


Page: 2 of 4



Top > renshuu.org > Feature Requests/Improvements > Finished/Rejected Requests


Loading the list
Lv.

Sorry, there was an error on renshuu! If it's OK, please describe what you were doing. This will help us fix the issue.

Characters to show:





Use your mouse or finger to write characters in the box.
■ Katakana ■ Hiragana