Console character sets - SCO

This is a discussion on Console character sets - SCO ; I have been testing character sets on the character-based console of OS6, and finding problems. In a nutshell, it looks like half of the Unixware solution was moved to OS6, leaving an unusable setup. I'm working with a default installation, ...

+ Reply to Thread
Results 1 to 4 of 4

Thread: Console character sets

  1. Console character sets

    I have been testing character sets on the character-based console of OS6,
    and finding problems. In a nutshell, it looks like half of the Unixware
    solution was moved to OS6, leaving an unusable setup.

    I'm working with a default installation, in what should be a US locale.
    Here is what I have found:

    1. The console's ROM seems to be loaded with an ISO-8859-1 character set,
    with some linedraw characters in hex-8B through hex-99. (In previous Open
    Servers, it was left in its default state which was generally codepage
    437, at least in the US).

    2. The virtual terminal sessions log in with mapchan set to "ibm" which
    maps output from ISO to CP-437.

    a) If I display a chart of characters from hex-80 to hex-ff, it does
    show the linedraw characters mentioned above, but most of the accented
    characters do not appear correctly.

    b) If I remove the mapchan ("mapchan -n"), the chart shows correctly.

    3. The virtual terminal sessions also log in with a 'mapscrn' set to do
    some translation. I'd like to know where this gets done in the bootup
    sequence. If I change mapscrn to a straight-through mapping, it screws it
    up again. I am still trying to figure out the interaction between
    mapchan and mapscrn, if any.

    4. Some of the above changes based on whether I login as root or not, but
    I don't have a handle on that yet.

    Of course, none of this affects computing with US characters, but...

    Can somebody help me out?

    Regards,
    .....Bob Rasmussen, President, Rasmussen Software, Inc.

    personal e-mail: ras@anzio.com
    company e-mail: rsi@anzio.com
    voice: (US) 503-624-0360 (9:00-6:00 Pacific Time)
    fax: (US) 503-624-0760
    web: http://www.anzio.com

  2. Re: Console character sets

    Bob,

    The console driver in OpenServer 6 is the SVR5 kernel's console driver.
    While this driver is not the same driver as the OpenServer 5 console
    driver, well behaved applications should no experience any problems
    when running on the console.

    If you have an applicaiton that is experiencing problems then can you
    describe to
    me what it is attempting to do and how it is failing?

    Thanks,

    John


  3. Re: Console character sets

    I'm having a very similar situation. My legacy app sends a sequence of
    hex values to the console to draw rudimentary boxes and charts. These
    characters are not working at all on the OSR6 console.

    The hex characters are 0xda, 0xd9, 0xbf, 0xc0, 0xb3, 0xc4 (essentially
    the corners and sides of a box).

    I've tried every possible font in /usr/lib/vidi and mapchan and none of
    them work properly so I'm stuck and would really appreciate some help
    on getting this app to display properly on OSR6

    Thanks
    -santo
    santo71@gmail.com


  4. Re: Console character sets

    On Tue, 9 Aug 2005 santo71@gmail.com wrote:

    > I'm having a very similar situation. My legacy app sends a sequence of
    > hex values to the console to draw rudimentary boxes and charts. These
    > characters are not working at all on the OSR6 console.
    >
    > The hex characters are 0xda, 0xd9, 0xbf, 0xc0, 0xb3, 0xc4 (essentially
    > the corners and sides of a box).
    >
    > I've tried every possible font in /usr/lib/vidi and mapchan and none of
    > them work properly so I'm stuck and would really appreciate some help
    > on getting this app to display properly on OSR6


    If your app is hard-coded to those values, "bad form", and you're in for a
    bit of work, I think. Here's what I know about OSR6 character mode
    console:

    1) It comes up as an at386-ie terminal type.

    2) It comes up with a mapchan in place for IBM437, which is WRONG. Null
    out the mapchan (in each session).

    3) the console session(s) will now operate as at386-ie. Presumably, the
    "ie" is for "international edition".

    4) You can see this charset by catting out a character chart. If you don't
    have one, find one at
    http://www.anzio.com/~ras/ascii80.txt
    You will see that this chart has many international and accented
    characters in the range of A0 to FF. In fact this is the ISO-8859-1 set,
    which is identical to Windows codepage1252 except that the latter has
    characters assigned to hex-80 through hex-9F. By contrast, the at386-ie
    chart has the linedraw characters in this area (there they are!).

    5) You may be able to create a mapchan file to translate your expected
    charset (437) to the at386-ie set, but I don't think you'll find one ready
    to go.

    6) to make your app more device-independent, if you have access to the
    source, you could output the 'smacs' sequence from the terminfo (or
    equivalent from termcap), then issue the appropriate ascii characters for
    the graphics characters you want, then issue the rmacs sequence.

    7. Note that there may be other subtle differences between the behavior of
    the OSR6 console and the 5.0.x console. Pay particular attention to the
    interaction between color attributes and monochrome attributes such as
    bold, reverse, and normal.

    8) Finally, note that your legacy app may in fact be more
    device-independent, and all you need to do is change your TERM variable.


    Regards,
    .....Bob Rasmussen, President, Rasmussen Software, Inc.

    personal e-mail: ras@anzio.com
    company e-mail: rsi@anzio.com
    voice: (US) 503-624-0360 (9:00-6:00 Pacific Time)
    fax: (US) 503-624-0760
    web: http://www.anzio.com

+ Reply to Thread