Linux-Misc Digest #209, Volume #27 Sat, 24 Feb 01 03:13:01 EST
Contents:
Re: Does JDK 1.3 work with kernel 2.4? (Tom Waterhouse)
Where's the physical address of kernel function after linked... (gu weining)
Re: MS Windows XP vs Linux (Paul Lew)
Re: Web page publishing (David)
Re: Where's the physical address of any kernel source code? (gu weining)
Re: how to print man pages in good quality (James Silverton)
Re: Virtual/software UART? (Byron A Jeff)
Re: why /dev/sdx# is a truly unsafe way to refer to physical devices
([EMAIL PROTECTED])
Re: Web page publishing ([EMAIL PROTECTED])
Re: surviving in a Windows only environment? ([EMAIL PROTECTED])
Re: MS Windows XP vs Linux ("Chaz")
Re: Substitute for Microsoft Outlook? (Georg Schwarz)
Re: surviving in a Windows only environment? (Georg Schwarz)
kernel traffic page not updated since Feb 9 ([EMAIL PROTECTED])
Re: MS Windows XP vs Linux (Matt Haley)
Avance Logic 4000 ("Nikola")
Avance Logic 4000 ("Nikola")
Avance Logic 4000 ("Nikola")
Re: /bin/sh ("Peter T. Breuer")
Re: how to print man pages in good quality (Floyd Davidson)
----------------------------------------------------------------------------
From: Tom Waterhouse <[EMAIL PROTECTED]>
Crossposted-To: comp.lang.java.misc
Subject: Re: Does JDK 1.3 work with kernel 2.4?
Date: Sat, 24 Feb 2001 04:18:50 GMT
I am getting the following message:
[tomw@linuxserver /]$ java
/usr/bin/java: /usr/bin/cut: No such file or directory
/usr/java/jdk1.3/bin/i386/native_threads/java: error while loading shared
libraries: libjvm.so: cannot load shared object file: No such file or
directory
[tomw@linuxserver /]$ uname -r
2.4.0-0.99.11
This is with the Fisher release of Redhat 7. Is this error a Redhat error
or a Java problem, or am I supposed to set a shared libarary path?
[EMAIL PROTECTED] wrote:
> Sorry for such an obvious question, but I can't seem to find a
> definitive answer digging around DejaNews, so I'll just ask it
> outright:
>
> Does the Sun JDK 1.3 work with Kernel 2.4? Anyone out there using it?
------------------------------
From: gu weining <[EMAIL PROTECTED]>
Subject: Where's the physical address of kernel function after linked...
Date: Fri, 23 Feb 2001 22:19:22 -0600
I mean I need to find the physical address of some function,
say rt6_lookup() of /usr/src/linux/net/route.c, after it is compiled
and linked to the memory. My work forces me to do this.
I copied my previous mail as follows. Need help.
===========
Where is the physical address of any kernel source code?
I need to access the physical address of kernel through driver,
but I don't know enough about it. I only knew these:
1. In the /proc/ksyms file, there are some relationship about
functions and address. How to put any one of function inside
kernel source code to ksyms? What's the address here? Physical?
2. I perhaps know, I should write loadable driver to access the
physical address, then insmod ..., then through ioctl to get
the data from kernel. Is it a correct and easy way to do this?
Any suggestion is appreciated.
Thank you so much.
Have a great weekend!
Weining Gu
University of Illinois
at Urbana-Champaign
> [EMAIL PROTECTED] wrote:
>
> gu weining <[EMAIL PROTECTED]> writes:
>> Where is the physical address of any kernel source code?
>
> There is none. The kernel source code sits in files on your hard
> drive. MAYBE. That assumes that you've got kernel source code on
> your system, which is not a necessary precondition for running
Linux.
>
> The most likely place to look for the kernel source code is
> /usr/src/linux, which is where it is often installed, by default
------------------------------
From: [EMAIL PROTECTED] (Paul Lew)
Subject: Re: MS Windows XP vs Linux
Date: Sat, 24 Feb 2001 04:43:40 GMT
On Sat, 24 Feb 2001, Mark Bratcher <[EMAIL PROTECTED]> wrote:
>In article <OZDl6.273$[EMAIL PROTECTED]>, Arctic Storm wrote:
>>MS Windows XP vs Linux
>>Microsoft's new Windows XP is touted as the easiest Windows ever, and is
>>expected to unify Windows OS's. MS developed XP by getting feedbacks from
>>average users and computer neophytes, to become easy/intuitive enough for
>>your Grandmother to use.
>
>MicroSoft says this about _every_ release of Windows. What do you think
>so far? The only thing I see is for $90 you get more bloat without any of
>the old bugs fixed. And an increase in "innovation" which usually translates
>into the incorporation of another embedded technology to make Windows
>compatible only with itself (and it's not even good at that).
>
>An aside: my favorite interpretation as to what "XP" means is that it is
>an emoticon.
>
"XP" = exasperation.... as in that's what m$ claimed for win2k.
also the "xp" ver doesn't exist as not available in stores.
------------------------------
From: David <[EMAIL PROTECTED]>
Subject: Re: Web page publishing
Date: Sat, 24 Feb 2001 04:58:23 GMT
Grant Edwards wrote:
>
> In article <[EMAIL PROTECTED]>, Brian Goodyear wrote:
>
> >What are you people using to create and maintain your web pages. I've
> >searched and so far found nothing which is powerful and easy.
>
> A text editor with an HTML mode (emacs/jed/vim/nedit) and a
> brain. ;)
>
> Oh, and a good HTML checker.
Even AbiWord can write and/or edit HTML.
--
Confucius say: He who play in root, eventually kill tree.
Registered with the Linux Counter. http://counter.li.org
ID # 123538
Completed more W/U's than 99.082% of seti users. +/- 0.01%
------------------------------
From: gu weining <[EMAIL PROTECTED]>
Subject: Re: Where's the physical address of any kernel source code?
Date: Fri, 23 Feb 2001 22:59:16 -0600
Thank you.
I do need to know.
-Weining
------------------------------
From: James Silverton <[EMAIL PROTECTED]>
Subject: Re: how to print man pages in good quality
Date: Sat, 24 Feb 2001 00:17:28 -0500
Jean-David Beyer wrote:
>
> James Silverton wrote:
> >
> > Jean-David Beyer wrote:
> > >
> > > Bob Tennent wrote:
> > > >
> > > > On Thu, 22 Feb 2001 17:40:49 +0100, bv wrote:
> > > > >
> > > > >Can anyone tell me how to print man pages in really good quality?
> > > >
> > > > man -t <command> | lpr
> > > >
> > > That works pretty well for me for almost all commands, but some are
> > > screwed up the same way as if you do man command | lpr instead of
> > > man command | col -b | lpr . I.e., some stuff obviously meant to be
> > > in bold is sloppily overtyped on my HP 660Cse printer that otherwise
> > > does very well.
> > >
> > >>>>>>>>>>>Deletions<<<<<<<<<<<<<<<<
> >
> > The method that seems to work for me is to go to the originals and thus
> > give, for example:
> >
> > zcat /usr/man/man1/foo.1.gz | groff -mandoc | lpr
> >
> > This works even tho' I am using a non-PostScript printer with apsfilter.
> >
> Well, I have no vacation.1.gz anywhere. I do have
>
> /usr/man/man1/vacation.1 and I tried piping that through | groff
> -mandoc | lpr
>
> and it is just as lousy as ever. (Recall that most commands come out
> just fine; there are just a few that are loused up.)
I'm sorry that groff -mandoc did not work. Looking at my notes, it
almost always seems to work but I did recently have some trouble with
the program tpipe that I extracted from the Unix Power Tools CD. There,
"man -t | lpr" works fine but the "zcat | groff" method seems to fail
giving doubled letters where they should be bold. I don't have the time
or enough knowledge to investigate the problem. Perhaps, someone more
expert can say what's happening.
Best wishes,
Jim.
--
James V. Silverton
Potomac, Maryland.
------------------------------
From: [EMAIL PROTECTED] (Byron A Jeff)
Subject: Re: Virtual/software UART?
Date: 24 Feb 2001 00:17:51 -0500
In article <X_ul6.1128$[EMAIL PROTECTED]>,
Tauno Voipio <[EMAIL PROTECTED]> wrote:
-
-"SLee000" <[EMAIL PROTECTED]> wrote in message
-news:[EMAIL PROTECTED]...
-> I am developing a 'virtual/software' UART (Universal Asynchronous Receiver
-> Transmitter).
-> That is, using software to simulate the RS232 Serial port (COM0, COM1,
-COM2
-> etc.).
-> Instead of writing a byte to a hardware register and letting the hardware
-UART
-> ( like an 8250 or
-> 16550) shift the bits out, I will be doing the shifting in software,
-> writing/reading only to
-> a generic GPIO pin (as in an Intel Northbridge chipset) or to one pin of
-the
-> parallel port.
->
-> Has anyone developed something like this for Linux/UNIX before?
->
-
-The first advice is: Don't.
-
-It was a popular way to handle 16 serial lines at the time of the
-minicomputers (1970's). At least Data General Nova and several Digital PDP's
-used this way of running several 110 bit/s lines together.
-
-For the current line speeds the interrupt overhead is prohibitive. For
-decent reception the bit interrupt rate should be at least 8 times the bit
-rate. The usual UART's are using sampling at 16 times the bit rate.
-
-Been there and done that.
It's doable with a bit of help. Think about using a Real-Time Linux program
to manage the timer interrupt.
As for sampling the input line, sampling at a 3X rate gives good results.
The original poster didn't specify a speed. Maybe 19.2K or 38.4K might be
good enough.
No clue on the PC timers. Just be aware that Linux uses those timers so
if you mess with them it's at your own peril.
BAJ
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: why /dev/sdx# is a truly unsafe way to refer to physical devices
Date: Sat, 24 Feb 2001 05:46:17 -0000
On 23 Feb 2001 23:11:31 GMT David Mathog <[EMAIL PROTECTED]> wrote:
| I never much cared for the Linux convention of /dev/sdx# for referring to
| scsi disks. Now I've got a reason to really hate it. On one linux system
| I have 5 scsi disks and /dev/sdb had been acting up so after rebooting (for
| unrelated reasons) I went to run badblock on it. With -w. I had rebooted
| into a mode that doesn't mount anything but the root disk. Well guess what
| folks, on this reboot that disk did not come up. Because I didn't try to
| mount it no messages appeared to that effect, but the disks were all
| renamed with c->b, d->c, e->d. Consequently the badblock command munched
| the disk that had been, for every other boot, /dev/sdc. You guessed it -
| the users' files. I had backups, but it's taking a while to restore
| everything.
There has been a long going debate over whether devices should be
named by their physical address, or their available sequence. So
far the only available reasons I've seen for sequential naming is
to give you the flexibility of shifting your hard drive target IDs
to accomodate some other device being added like a tape, or to be
able to specify that the root filesystem to be mounted as / be the
first drive found.
If these arguments were strong, then why were they not applied to
the IDE drive naming? Why isn't the first IDE drive named "hda"
even if it is on the secondary IDE bus.
I do find the IDE fixed naming scheme to be a LOT easier to work
with. And I'd really like to have this work for SCSI as well.
However, SCSI device addressing is quite extensive when you consider
the potential for 8, maybe more, SCSI adapters, 8 logical units,
16 targets (potentially the adapter can be moved to a different
target number), and of course many partitions (I have one drive
on one machine right now with 28 partitions due to it being a
triple boot between Linux, FreeBSD, and OpenBSD). Right now the
partition limit in Linux for SCSI is 15. This means we need at
least 8*8*16*16 = 16384 device codes. That's 25% of the whole
device code space. To get as many partitions on SCSI as IDE has,
there would be no device code space left for IDE to have any.
Since Linux is going in the direction of devfs, maybe this can be
the ultimate solution. If we can get away from the notion that
devfs has to be mounted, and have it always present once the kernel
is up, then we can actually have as vast a space of devices as we
would ever imagine available even for mounting / on.
Were it not for the need for 25% of device code space needed to do
SCSI, I'd suggest actually having both sequential and fixed device
naming. But still, one fair compromise would be to do so for non-LUN
devices on say, 2 SCSI controllers. That would be just 2 majors.
--
=================================================================
| Phil Howard - KA9WGN | Dallas | http://linuxhomepage.com/ |
| [EMAIL PROTECTED] | Texas, USA | http://phil.ipal.org/ |
=================================================================
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Web page publishing
Date: Sat, 24 Feb 2001 05:49:31 -0000
On Sat, 24 Feb 2001 03:27:55 GMT Grant Edwards <[EMAIL PROTECTED]> wrote:
> In article <[EMAIL PROTECTED]>, Brian Goodyear wrote:
>
>>What are you people using to create and maintain your web pages. I've
>>searched and so far found nothing which is powerful and easy.
>
> A text editor with an HTML mode (emacs/jed/vim/nedit) and a
> brain. ;)
I use emacs and brain.
> Oh, and a good HTML checker.
I use brain and Netscape.
Also, to uploaded I use rsync via ssh.
--
=================================================================
| Phil Howard - KA9WGN | Dallas | http://linuxhomepage.com/ |
| [EMAIL PROTECTED] | Texas, USA | http://phil.ipal.org/ |
=================================================================
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: surviving in a Windows only environment?
Date: Sat, 24 Feb 2001 05:55:21 -0000
On Fri, 23 Feb 2001 22:12:19 GMT Bob Hauck <[EMAIL PROTECTED]> wrote:
| Exchange Server _can_ speak both of those if it is enabled. I use Pine
| and IMAP with Exchange just fine, as well as Netscape Mail. You can see
| shared folders this way, but not the Calendar.
Try TradeClient from Bynari ... http://www.bynari.net/Products/products.html
--
=================================================================
| Phil Howard - KA9WGN | Dallas | http://linuxhomepage.com/ |
| [EMAIL PROTECTED] | Texas, USA | http://phil.ipal.org/ |
=================================================================
------------------------------
From: "Chaz" <[EMAIL PROTECTED]>
Subject: Re: MS Windows XP vs Linux
Date: Sat, 24 Feb 2001 06:54:36 -0000
Im afraid i prefer freebsd to linux.
xp stands for 'extra pants'
"Arctic Storm" <[EMAIL PROTECTED]> wrote in
message news:OZDl6.273$[EMAIL PROTECTED]...
> MS Windows XP vs Linux
> Microsoft's new Windows XP is touted as the easiest Windows ever, and is
> expected to unify Windows OS's. MS developed XP by getting feedbacks from
> average users and computer neophytes, to become easy/intuitive enough for
> your Grandmother to use.
> Linux/GNU caters to people who have a reasonable background in computers,
if
> not experts. This is why Linux/GNU is not as prevalent as MS Windows.
> Listening to feedbacks from computer experts for improvements is
important,
> but expending efforts into designing a system that can be easily used by
> "non-computer" public is just as important.
>
>
------------------------------
From: [EMAIL PROTECTED] (Georg Schwarz)
Subject: Re: Substitute for Microsoft Outlook?
Date: 24 Feb 2001 06:58:46 GMT
[EMAIL PROTECTED] (Grant Edwards) writes:
>>Is there any Linux mail client that can work with Microsoft Exchange
>>mail servers?
>Depends.
>Most will work if POP/IMAP/SMTP is enabled on Exchange server.
I don't thibk POP3 and IMAP are enabled, and I don't think they will be.
>Use IMAP rather than POP if given a choice. If they aren't
>enabled, then you're pretty much screwed. Bynari.com has a sort
>of a solution where you run have the a server-app on a Windows
>machine and a client-app under Linux.
I see. That's not a feasable option.
>Nobody has a MAPI client under anything but MS-Windows, and
>nobody ever will.
do you happen to know whether the Mac version of any MS app does support
it, BTW?
--
Georg Schwarz http://home.pages.de/~schwarz/
[EMAIL PROTECTED] +49 177 2437545
------------------------------
From: [EMAIL PROTECTED] (Georg Schwarz)
Subject: Re: surviving in a Windows only environment?
Date: 24 Feb 2001 07:01:23 GMT
[EMAIL PROTECTED] (Lee Webb) writes:
>MS Exchange Server *does* support POP3: it just has to be explicitly enabled.
and I don't think this is going to happen in that particular case,
especially since the mail server sits behind a firewall, which would also
have to be reconfigured.
BTW, you might check your newsreader's address config; it seems to be
incorrect.
--
Georg Schwarz http://home.pages.de/~schwarz/
[EMAIL PROTECTED] +49 177 2437545
------------------------------
From: [EMAIL PROTECTED]
Subject: kernel traffic page not updated since Feb 9
Date: Sat, 24 Feb 2001 02:06:41 -0500
I always like to read the kernel traffic page at
http://kt.linuxcare.com/kernel-traffic/latest.epl
which usually updates every week. It's currently still
back on the Feb 9 issue. Is there something wrong
or did every one just get too busy with the new
kernel being out to write stuff? I'm hoping it's
not going away.
Mark
------------------------------
From: [EMAIL PROTECTED] (Matt Haley)
Subject: Re: MS Windows XP vs Linux
Date: Sat, 24 Feb 2001 07:34:57 -0000
On Sat, 24 Feb 2001 01:18:38 GMT,
Arctic Storm <[EMAIL PROTECTED]> wrote:
[ blah, blah, blah ]
....and your point?
--
Matt Haley <[EMAIL PROTECTED]>
Mandrake 7.2 / RedHat 6.1 / Windows 98 SE / FreeBSD 4.2 / Windows NT 4
------------------------------
From: "Nikola" <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux,comp.os.linux.hardware,comp.os.linux.help,comp.os.linux.redhat,comp.os.linux.setup,linux.redhat.misc
Subject: Avance Logic 4000
Date: Sat, 24 Feb 2001 08:05:36 +0100
I have problem with setting my Avance Logic 4000 sound card.
I try with Alsa drivers, but I don't know how to set config files.
Can someone help me?
Thanks!
------------------------------
From: "Nikola" <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux,comp.os.linux.hardware,comp.os.linux.help,comp.os.linux.redhat,comp.os.linux.setup,linux.redhat.misc
Subject: Avance Logic 4000
Date: Sat, 24 Feb 2001 08:05:36 +0100
I have problem with setting my Avance Logic 4000 sound card.
I try with Alsa drivers, but I don't know how to set config files.
Can someone help me?
Thanks!
------------------------------
From: "Nikola" <[EMAIL PROTECTED]>
Crossposted-To:
comp.os.linux,comp.os.linux.hardware,comp.os.linux.help,comp.os.linux.redhat,comp.os.linux.setup,linux.redhat.misc
Subject: Avance Logic 4000
Date: Sat, 24 Feb 2001 08:06:25 +0100
I have problem with setting my Avance Logic 4000 sound card.
I try with Alsa drivers, but I don't know how to set config files.
Can someone help me?
Thanks!
------------------------------
From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Subject: Re: /bin/sh
Date: Sat, 24 Feb 2001 08:32:32 +0100
Francis Litterio <[EMAIL PROTECTED]> wrote:
> "Peter T. Breuer" <[EMAIL PROTECTED]> writes:
>> Bash when called as sh behaves as sh.
> I've heard that too, but then this behavior is confusing (blank lines
Nothing to hear. It's what the man page SAYS.
If bash is invoked with the name sh, it tries to mimic the
startup behavior of historical versions of sh as closely
as possible, while conforming to the POSIX standard as
well. When .....
But they are relying on the fact that bash is a sh-compatible. They're
just talking about the startup sequence here.
> inserted before prompts for clarity):
> $ ls -l /bin/sh
> lrwxrwxrwx 1 root root 4 Oct 21 1999 /bin/sh -> bash
> $ cat > zzz
> #!/bin/sh
> echo ~
> echo $PPID
> echo $(date)
> VAR=99
> let "VAR = VAR + 1"
> echo $VAR
> ((VAR = VAR + 1))
wuzzat? Oh ...
((expression))
The expression is evaluated according to the rules
described below under ARITHMETIC EVALUATION. If
the value of the expression is non-zero, the return
status is 0; otherwise the return status is 1.
This is exactly equivalent to let "expression".
> echo $VAR
> ^D
> $ chmod 755 zzz
> $ ./zzz
> /home/franl
> 844
> Fri Feb 23 21:34:55 EST 2001
> 100
> ./zzz: VAR: command not found
> 100
> $ bash zzz
> /home/franl
> 844
> Fri Feb 23 21:36:44 EST 2001
> 100
> 101
> So when bash is invoked as sh, it allows tilde expansion, $(...) command
> substitution, bash-specific variables such as PPID, the let "..."
> syntax, but not the ((...)) syntax (which it seems to read as
> doubly-nested subshells).
Yes.
> That's some Bourne-shell compatibility mode!
I don't understand. Is it or is it not compatible with Bourne-shell?
If sh insists that () is for subshells, then it's right to
parse it that way. But I think the problem is that
(( VAR = VAR + 1 ))
is an expression, while
((VAR = VAR + 1))
is a subshell (pair). Anyway, talk to the authoor about it.
Peter
------------------------------
From: Floyd Davidson <[EMAIL PROTECTED]>
Subject: Re: how to print man pages in good quality
Date: 23 Feb 2001 21:54:24 -0900
James Silverton <[EMAIL PROTECTED]> wrote:
>Jean-David Beyer wrote:
>> James Silverton wrote:
>> > Jean-David Beyer wrote:
>> > > Bob Tennent wrote:
>> > > > On Thu, 22 Feb 2001 17:40:49 +0100, bv wrote:
>> > > > >Can anyone tell me how to print man pages in really
>> > > > >good quality?
>> > > >
>> > > > man -t <command> | lpr
>> > > >
>> > > That works pretty well for me for almost all commands,
>> > > but some are screwed up the same way as if you do man
>> > > command | lpr instead of man command | col -b | lpr
>> > > . I.e., some stuff obviously meant to be in bold is
>> > > sloppily overtyped on my HP 660Cse printer that otherwise
>> > > does very well.
By any chance are these particular man pages pre-formatted
instead of coming in [nt]roff source?
>> > >>>>>>>>>>>Deletions<<<<<<<<<<<<<<<<
>> >
>> > The method that seems to work for me is to go to the
>> > originals and thus give, for example:
>> >
>> > zcat /usr/man/man1/foo.1.gz | groff -mandoc | lpr
This is almost essentially the same as using the -t option with
the man command. (A slight difference is discussed below.)
>> > This works even tho' I am using a non-PostScript printer
>> > with apsfilter.
>> >
>> Well, I have no vacation.1.gz anywhere. I do have
>>
>> /usr/man/man1/vacation.1 and I tried piping that through |
>> groff -mandoc | lpr
>>
>> and it is just as lousy as ever. (Recall that most commands
>> come out just fine; there are just a few that are loused up.)
>
>I'm sorry that groff -mandoc did not work. Looking at my notes,
>it almost always seems to work but I did recently have some
>trouble with the program tpipe that I extracted from the Unix
>Power Tools CD. There, "man -t | lpr" works fine but the "zcat
>| groff" method seems to fail giving doubled letters where they
>should be bold. I don't have the time or enough knowledge to
>investigate the problem. Perhaps, someone more expert can say
>what's happening.
There are several things to look at here, because there is a
long chain of events which leads to a man page being displayed
or printed, and problems at different points result in different
behavior.
First, man pages can be (rarely) in pre-formatted form, but they
are usually source code for the formatter. They can also be
compressed or uncompressed. The man command will figure out
either of the above, and properly display the results to the
screen. (The "pre-formatted" version is probably formatted for
screen presentation, and it may look relatively ugly no matter
how it is massaged before printing. Of course if your version
of "screen" isn't what it was pre-formatted for, it may look
ugly there too!)
If the man page is not source code, the -t option to man will
not produce a useful output file. (And that appears to be the
problem described above, though I don't have that man page and
cannot verify that to be the cause.)
FACT: Pre-formatted man pages are evil.
But there are a couple of other likely problems too. Just zcat'ing
a file directly to groff is usually valid, but sometimes not. The
-t option to man causes this command to be executed (note that I've
wrapped the line with '\'):
(cd /usr/man ; /usr/bin/gunzip -c /usr/man/man1/dvi3812.1.gz | \
/usr/bin/gtbl | /usr/bin/groff -Tps -mandoc)
The first part, cd'ing to /usr/man is not necessary for manually
invocation. And "/usr/bin/gunzip -c" is the same as "zcat".
But the man command also piped the source code through
"/usr/bin/gtbl", and once in awhile somebody will have gotten
carried away and put enough effort into a man page that it
actually needs to be filtered by gtbl (rare, but nice when it
happens).
Also, if your printer is not a PostScript printer then most
likely the printing system is using Ghostscript to convert from
PostScript to whatever the printer can understand. Clearly that
works better in some cases than in others. Common problems are
fonts... But there are some possible variations which may or
may not produce better output.
For printing a straight ascii file, one can pipe to groff
directly, and use either "-Tascii" or "-Tlatin1", which may or
may not produce different results.
To produce a different PostScript variation, or to directly
produce HPCL for HP printers, "-Tdvi" can be used with groff and
then another post-processing stage can use, for example dvi2lj
or dvips, to produce a final output. And some interesting
variations can be found using different options for the dvi*
programs too, such as using 600 dpi options with a 300 dpi
printer.
Do note that while trying to find the best way to print any
given man page sounds a little complicated, consider trying to
_write_ a complex man page and making sure that it will produce
reasonable output in each of those different printing methods!
That can be tedious, because sure enough whatever looks *best*
using the preferred method, won't work at all on some required
method.
--
Floyd L. Davidson <http://www.ptialaska.net/~floyd>
Ukpeagvik (Barrow, Alaska) [EMAIL PROTECTED]
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list by posting to comp.os.linux.misc.
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Misc Digest
******************************
