-
-
Notifications
You must be signed in to change notification settings - Fork 417
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Unicode Japanese Characters #57
Comments
Yes, and to be more precise the characters are "flipped" on the X axis |
Adds classic mode which can be invoked with "c", which uses Japanese characters instead of the regular English characters that cmatrix outputed earlier. Done to match up with the original matrix. The starting, ending, and range for randum number (corresponding to the unicode character) is modified to suite the Japanese character range. Requires appropriate Japanese fonts to make it work. Fixes #57 Signed-off-by: Abishek V Ashok <[email protected]>
I managed to get the Japanese characters easily but could'nt flip them. |
Hey! I'd be happy to know which font is required for having the japanese characters show up? |
Any Japanese font complying with utf-8 works fine |
I have font installed and they work (tested japanese characters in browser & terminal) |
Strange. Will look into it |
im having the same issue as @ask6155 |
Hey guys, did anyone manage to get this working? I made sure to install a japenese font but at the moment cmatrix -c leaves a blank screen. Im using ubuntu server btw. |
Nope still researching on it |
same here on archlinux |
See also: https://help.accusoft.com/PrizmDoc/v12.1/HTML/Installing_Asian_Fonts_on_Ubuntu_and_Debian.html It might be nice to include a list of fonts that are known to work with this. |
Package-Manager: Portage-2.3.103, Repoman-2.3.23 Bug: abishekvashok/cmatrix#57 Signed-off-by: Jeroen Roovers <[email protected]>
Issue is from 2018, tried running |
Thanks for this pointer |
Hello, I have made the thing working. I don't have the time to make a correctly formatted pull request, but here is the trick. A little workaround to print it is to replace all calls to
The other thing is to link to ncursesw instead of ncurses.
Also note that all the characters in the range (12288; 12351 => 0x3000; 0x303f) are not displayed correctly (maybe a missing font, my browser doesn't print all characters correclty too). You can display all the kanji correctly with the range [0x4e00;0x4db5]. |
I was working on trying to fix this issue yesterday. The current japanese range is incorrect, as 0x3000-0x303f are Japanese punctuation characters, whereas the movie uses hiragana and katakana (0x3041-0x30ff), and kanji (scattered variously from 0x3400-0x4db0 and 0x4e00-0x9fa0) source I'll work on a pr if no one else gets to it before I have time later this month. |
Getting some weird overlap, which will require a more intricate digging into the code than I have bothered to do so far. The film frame in OP seems to also have some numbers, which would be impossible to include given the current method of random character generation, so I may work on revamping the system to work on any arbitrary character set while I'm at it. |
Just noticed #112 which seems to implement a fix already... However, after cloning the repo linked in 112 and running it, I get something that looks like this: Compared to my current progress of: Given that, I think I'll keep working on it, unless @abishekvashok wants to merge the existing PR (in which case please let me know so I don't waste my time) |
this being what this app is - the overlap is actually kinda cool IMHO |
@PensiveCynic I was thinking that as well. Would tie in well with #93 if I can get it working in a controlled manner |
@space-pagan if you figure out what's causing it exactly maybe make it a cli option? ;-) |
I know what was causing it - the use of japanese characters which took up more than one unicode char cell (not sure what the proper term for this is, double-wide seems to refer to the fact that their representation is two bytes of data), which caused column alignment to be off. I just need to figure out how to control it. But yes, agreed, this can be a cli option. |
For the record, #112 is technically what is done in the movie, because it uses half-width kana. By looking at your solution, I guess you are using the regular katakana. The problem with those is that they occupy the space of two characters (note that the spaces between columns in my solution is what cmatrix always does, not something I added, and it looks like your characters occupy what was supposed to be whitespace). So while I believe that your solution looks nicer as it is now, I hope in the future to create a mode that mixes katakana, latin script & numerals to create something that is much closer to the original movie style, and I think I need half-width kana for that. Maybe it's worth it to merge both solutions (as different flags) so that the users can choose what they prefer. Or maybe we should use your solution for japanese-only output since it looks nicer and we'll only use half-width kana if/when I get around to implementing mixed-script (which needs some discussion & planning as it's not really compatible with the current way of choosing random characters to print as we use a single contiguous character range). |
@space-pagan Nope, I was wrong. I looked at your code and you also use the half-width forms. The difference in appearance is because you explicitly changed the spacing I guess. So your solution is basically a superset of mine because the only thing I did was enable wchar support and change the character range. You should open a PR to get this merged and we should close mine if you do that. |
I had originally used full-width when taking the screenshot, but then switched to half-width later on. I think that adding custom charset support at the same time as fixing this would make the most sense so that people can chose between full width, half width, etc... I've just been busy with school and work and haven't had a chance to do anything since my last comment. I'll get on that soon :) |
Just to clarify @space-pagan,
If the output is same, we could merge #112 of @patatahooligan and then we could add an additional flag for the full width characters -- just as suggested :D What are your takes on this guys? |
That should be fine. I don't remember what other changes I had made since I was working on fixing many things at once. If there's anything missing I'll add it in a pr after you merge. |
I don't know if you tried out both our versions so I'll briefly describe the situation to make sure we're on the same page. #112 is a small fix that does the absolute minimum to get the japanese characters to print. It doesn't look great, probably because the character set is not well-suited for cmatrix's default behavior of leaving whitespace columns between the characters. I also haven't fixed problems beyond the printing of the characters, eg the rain drop's head is not white for some reason. space-pagan's version contains my fix but also includes adjustments/fixes to make the output look nice, plus various other changes (code comments, docs, etc). These changes affect other modes of cmatrix beyond the japanese characters, eg try simply So the crucial question to decide what you want to do is: do you like space-pagan's other changes? If yes, merge directly from them and skip #112 as it offers nothing more. If you don't like their changes then it makes sense to merge #112 for now just to have something that works and discuss with them what you would want changed to merge their version. That's how I see it, if you guys have differing opinions let me know. |
Another choice, I hope. |
So, to be clear, how do i make this happen in termux |
@JanzenJohn For #112 you obviously need a font that has the desired unicode characters. I guess you could quickly check that by copying and pasting these characters in your terminal and seeing if they display properly. If they don't display in your browser either, then you probably don't have a unicode font for that as well.
|
This might be the easiest option: |
Fixed by #137 |
@space-pagan ´s fork works for me on macOS when it comes to japanese characters. However the window does not resize and it draws on black background which renders my blurry/vibrancy background of my hyper terminal window just opaque. The official 2.0 -c option does not work on macOS. |
To be more precise, according to this scifi.stackexchange.com post they're not only katakana, but just normal Latin alphabet digits, letters, and symbols; the rest are Japanese characters (mostly half-width katakana, though there's at least one kanji in there as well). In total, that's around 67 characters. Unfortunately it seems not possible to mirror characters with just Unicode, maybe could be used a specific font. |
@Fufu-S Warp also fails. |
@unrealapex: Please read this: Any updates? |
v2.0 fails to render anything on MacOS 11.6.5 and iTerm 3.3.12. In the default terminal, I get some boxes showing, but no font. Do I need to actually install a Japanese font? If so, how do I do that exactly? |
Meanwhile, people subscribed to this issue might be interested on UniMatrix project. |
Me no coder, but Ms Bing (Precise) suggested this, which solved two probs with one (AI) stone :
Ref:
|
The original matrix animation uses Japanese characters.
https://static.independent.co.uk/s3fs-public/thumbnails/image/2017/10/27/09/matrix-code.jpg?w968
It would be nice if unicode support was added so cmatrix could display Japanese characters.
The text was updated successfully, but these errors were encountered: