[arch-general] What is FONT_MAP for?
Hi, I've looked in the wiki, but I cannot find what FONT_MAP is for. This wiki page seems to confuse it with the keymap: https://wiki.archlinux.org/index.php/Fonts#Console_fonts since it says "Now, set the proper keymap, [...]: FONT_MAP=...". However, I thought the keymap was configured with "KEYMAP=...". My system appears to work with: KEYMAP="fr-latin1" FONT="ter-122b" FONT_MAP="" I can't type accented characters in the console (neither the text consoles F1-F6, nor the X terminals like urxvt or konsole) but those characters are displayed fine when I cat a UTF-8 file. Not a big issue, just curious. So what is FONT_MAP for? Thanks, Jerome -- mailto:jeberger@free.fr http://jeberger.free.fr Jabber: jeberger@jabber.fr
On Thu, Nov 1, 2012 at 10:33 AM, "Jérôme M. Berger" <jeberger@free.fr> wrote:
So what is FONT_MAP for?
Check the setfont(8) manpage. -t
Tom Gundersen wrote:
On Thu, Nov 1, 2012 at 10:33 AM, "Jérôme M. Berger" <jeberger@free.fr> wrote:
So what is FONT_MAP for?
Check the setfont(8) manpage.
Thanks. So if I understand correctly, it is useful for programs that output 8 bit characters that are not valid UTF-8 sequences and serves to convert their output into valid unicode for display, right? Jerome -- mailto:jeberger@free.fr http://jeberger.free.fr Jabber: jeberger@jabber.fr
On чет, 01 ное 2012 14:28:43 CET, "Jérôme M. Berger" wrote:
Tom Gundersen wrote:
On Thu, Nov 1, 2012 at 10:33 AM, "Jérôme M. Berger" <jeberger@free.fr> wrote:
So what is FONT_MAP for?
Check the setfont(8) manpage.
Thanks. So if I understand correctly, it is useful for programs that output 8 bit characters that are not valid UTF-8 sequences and serves to convert their output into valid unicode for display, right?
No, some console fonts don't have a Unicode map, so they're essentially an index -> glyph file where index is from 0-255. A unicode map adds a "Unicode Code Point -> index" mapping. Most fonts in /usr/share/kbd/consolefonts should have bult-in maps (haven't checked though). For those that don't have it, there's the -m option in setfont or FONT_MAP. Without an unicode map, you must make sure the loaded font has the same layout as the charset you're using. Without the unicode map you can't use utf8 -- дамјан
Damjan wrote:
On чет, 01 ное 2012 14:28:43 CET, "Jérôme M. Berger" wrote:
Tom Gundersen wrote:
On Thu, Nov 1, 2012 at 10:33 AM, "Jérôme M. Berger" <jeberger@free.fr> wrote:
So what is FONT_MAP for?
Check the setfont(8) manpage.
Thanks. So if I understand correctly, it is useful for programs that output 8 bit characters that are not valid UTF-8 sequences and serves to convert their output into valid unicode for display, right?
No, some console fonts don't have a Unicode map, so they're essentially an index -> glyph file where index is from 0-255. A unicode map adds a "Unicode Code Point -> index" mapping.
Most fonts in /usr/share/kbd/consolefonts should have bult-in maps (haven't checked though). For those that don't have it, there's the -m option in setfont or FONT_MAP.
Without an unicode map, you must make sure the loaded font has the same layout as the charset you're using. Without the unicode map you can't use utf8
Now I'm confused. According to setfont(8), there are two kinds of maps (not counting the keymap): the console map (option -m) and the unicode font map (option -u). What you describes appears to be the unicode font map but that still leaves the other one. BTW, which does FONT_MAP refer to? I tried to grep through /etc to see how it is used in the initscripts, but this only shows the vconsole.conf entry... Jerome -- mailto:jeberger@free.fr http://jeberger.free.fr Jabber: jeberger@jabber.fr
On Thu, Nov 1, 2012 at 7:22 PM, "Jérôme M. Berger" <jeberger@free.fr> wrote:
Damjan wrote:
On чет, 01 ное 2012 14:28:43 CET, "Jérôme M. Berger" wrote:
Tom Gundersen wrote:
On Thu, Nov 1, 2012 at 10:33 AM, "Jérôme M. Berger" <jeberger@free.fr> wrote:
So what is FONT_MAP for?
Check the setfont(8) manpage.
Thanks. So if I understand correctly, it is useful for programs that output 8 bit characters that are not valid UTF-8 sequences and serves to convert their output into valid unicode for display, right?
No, some console fonts don't have a Unicode map, so they're essentially an index -> glyph file where index is from 0-255. A unicode map adds a "Unicode Code Point -> index" mapping.
Most fonts in /usr/share/kbd/consolefonts should have bult-in maps (haven't checked though). For those that don't have it, there's the -m option in setfont or FONT_MAP.
Without an unicode map, you must make sure the loaded font has the same layout as the charset you're using. Without the unicode map you can't use utf8
Now I'm confused. According to setfont(8), there are two kinds of maps (not counting the keymap): the console map (option -m) and the unicode font map (option -u). What you describes appears to be the unicode font map but that still leaves the other one.
BTW, which does FONT_MAP refer to? I tried to grep through /etc to see how it is used in the initscripts, but this only shows the vconsole.conf entry...
Jerome -- mailto:jeberger@free.fr http://jeberger.free.fr Jabber: jeberger@jabber.fr
If you look at "man vconsole.conf", you see there's both FONT_MAP and FONT_UNIMAP. FONT_MAP refers to the console map, not the unicode font map.
Jan Steffens wrote:
On Thu, Nov 1, 2012 at 7:22 PM, "Jérôme M. Berger" <jeberger@free.fr> wrote:
Damjan wrote:
On чет, 01 ное 2012 14:28:43 CET, "Jérôme M. Berger" wrote:
Tom Gundersen wrote:
On Thu, Nov 1, 2012 at 10:33 AM, "Jérôme M. Berger" <jeberger@free.fr> wrote:
So what is FONT_MAP for?
Check the setfont(8) manpage.
Thanks. So if I understand correctly, it is useful for programs that output 8 bit characters that are not valid UTF-8 sequences and serves to convert their output into valid unicode for display, right? No, some console fonts don't have a Unicode map, so they're essentially an index -> glyph file where index is from 0-255. A unicode map adds a "Unicode Code Point -> index" mapping.
...
Now I'm confused. According to setfont(8), there are two kinds of maps (not counting the keymap): the console map (option -m) and the unicode font map (option -u). What you describes appears to be the unicode font map but that still leaves the other one.
BTW, which does FONT_MAP refer to? I tried to grep through /etc to see how it is used in the initscripts, but this only shows the vconsole.conf entry...
If you look at "man vconsole.conf", you see there's both FONT_MAP and FONT_UNIMAP.
FONT_MAP refers to the console map, not the unicode font map.
Thanks Jan. So the unicode font map is for when the font file does not provide the association between unicode character and glyph index, like Damjan wrote. What about the console map? Is it what I described: a mapping from 8 bits to Unicode for programs that output "extended ASCII" characters? This is what I understood from the setfont man page, but given Damjan's earlier post I prefer to ask for confirmation. Jerome -- mailto:jeberger@free.fr http://jeberger.free.fr Jabber: jeberger@jabber.fr
participants (4)
-
"Jérôme M. Berger"
-
Damjan
-
Jan Steffens
-
Tom Gundersen