Copy/Paste and UTF-8 work both fine on my Windows computer. With the relatively current Pd 0.51.1 and the older 0.47. Thanks, IOhannes!
So it's not the [canvas] that is causing the problem with unicode characters (except for the ascii 35 issue).
The code for the alternative # (8983) in UTF-8 turned out to be "226 140 151" instead of "232 166 131". But this doesn't change the real problem.
On my Debian 9.5 with Pd 0.49 I cannot display anything higher than ascii values. All higher characters are displayed with empty squares or nothing at all.
Copying the patch with the [#( as unicode "8983" results in an empty square even in the message box. So the problem seems rather be an issue of the OS not being able to display higher unicode characters at all. I'll have to figure out how to do this . . .
On my Windows computer I have a addon for displaying unicode. This doesn't seem to be present in my very basic Debian installation that uses fluxbox as a windows manager.
So, there is a bug in [canvas] not displaying ascii 35 correctly and UTF-8 (alternative characters) is not working for me right now. My last solution would be to edit the "Helvetica" or "DejaVu" font for displaying a "#" with a lower number. However, last time I was looking for the location of Helvetica I couldn find it and didn't get any answer here. If anyone knows the location of the font when being used by Pd please let me know.
This means I probably have to wait another few years until someone will fix the [canvas] issue for displaying a sharp "#".
Thanks anyway! I give up for now.
Ingo
-----Original Message----- From: Pd-list [mailto:pd-list-bounces@lists.iem.at] On Behalf Of IOhannes m zmölnig Sent: Sunday, January 17, 2021 9:40 PM To: pd-list@lists.iem.at Subject: Re: [PD] unicode symbols and Pd
On 1/17/21 7:40 PM, oliver wrote:
BTW, I just tried replacing the pound sign [35( with unicode [8983( which is a pound sign for telephones. This works on the current Pd 0.51
how so ?
on my machines (win, debian) ASCII code is still wrapped between 0 and 255, like always.
actually no. ASCII is 0..127 by definition.
otoh, bytes are 8bit numbers, which cover a range of 0..255.
so sending ASCII 8983 to [list tosymbol] outputs the same result as ASCII 23. (I'm on PD 0.51.2)
you need to convert the unicode codepoint to utf8. for the unicodepoint 8983, the utf-8 sequence is "232 166 131". feeding this sequence to [list tosymbol] will give you "?" (which is think is what you want).
And yes, I agree it would be great if canvases would allow this character to be displayed correctly.
but they do. since ages.
i'm honestly baffled by this discussion. afaik we have a working unicode implementation in Pd for more than 10 years. what's the buzz, tell me what's a-happening...
amsdr IOhannes