Linux-Misc Digest #489, Volume #25 Fri, 18 Aug 00 17:13:06 EDT
Contents:
Re: dosemu (Ronald J Roy)
Re: Help! Urgent problem with gzip (Jan Johansson)
Re: newsgroup reader??? (Don Grbac)
Best Free E-Mail for Linux? (Ron J Theriault)
Re: STTY and ERASE (Johan Kullstam)
Re: Booting to /dev/sda2 (steve)
Win4Lin: anyone use it with Micro$oft Outlook on Exchange? (Adam Finkelstein)
Re: How do you determine the job number from lpr? ([EMAIL PROTECTED])
Re: Hard Drive is Lieing. (Richard Lyon)
AutoMagic Linux e-mail -- How do I set it up? (softrat`)
Re: STTY and ERASE (Floyd Davidson)
Re: STTY and ERASE (Floyd Davidson)
Re: STTY and ERASE (NF Stevens)
Re: STTY and ERASE (NF Stevens)
Re: STTY and ERASE (NF Stevens)
Re: STTY and ERASE (NF Stevens)
Re: regex example (Tony Lawrence)
bash ([EMAIL PROTECTED])
Re: How do you determine the job number from lpr? (Arlan Lucas de Souza)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Ronald J Roy)
Subject: Re: dosemu
Reply-To: [EMAIL PROTECTED]
Date: Fri, 18 Aug 2000 19:13:53 GMT
On 18 Aug 2000 01:39:00 GMT, Jonathan M Hill <[EMAIL PROTECTED]> wrote:
>Ronald J Roy ([EMAIL PROTECTED]) wrote:
>: I have dosemu working, but how do you add programs? I installed from source
>: and used the freedos image off the redhat6.1 cdrom.
>
>
>Hello Ronald;
>
> I'm not sure of what you are asking....I installed dosemu some time
>ago on redhat 5.1 and was glad to see that it accesses the floppy disk
>directly from DOS as a: which helps.
>
> I have a DOS partition on the hard drive and access that from DOS
>or Linux. I am not a big DOS user so have not bothered to configure the
>CD-ROM drive to be accessible from dosemu. My system is dual boot so
>the one time DOS needed the CD-ROM drive I justed booted DOS.
>
> A friend installed dosemu on a machine that did not have a DOS
>partition on the hard drive. The install script apparently created a
>large file and formatted the contents with a DOS file system. There is
>a trick involving loopback that you can use to mount the file as a
>file system under Linux. Sorry, I cannot remember the details.
>
> Jonathan Hill
> [EMAIL PROTECTED]
thanks for the info. I replaced freedos with drdos and used lredir to mount
the file system: lredir c: linux\fs\home. After that it was a simple matter of
changing directories and running a dos program directly.
------------------------------
From: Jan Johansson <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.help
Subject: Re: Help! Urgent problem with gzip
Date: Fri, 18 Aug 2000 19:22:37 GMT
>All he needs is a couple of million computers that can try
>each possible bit pattern in turn until the file unzips.
>Conceptually like the SETI project. Seriously: if it was
>important enough, it could be done. Do you think the CIA
>would let a little thing like that stop them if they needed
>the data?
BAFAB could do it, however, it would cost you a few thousand dollars
(about $1500 just to look at the device and give you a
estimated-chance-of-recovery) but we have used them for crucial data.
------------------------------
From: Don Grbac <[EMAIL PROTECTED]>
Subject: Re: newsgroup reader???
Date: Fri, 18 Aug 2000 15:28:01 -0400
[EMAIL PROTECTED] wrote:
> Patrick M Geahan <[EMAIL PROTECTED]> did eloquently scribble:
> Nutscrape news is a very substandard one.
>
> --
Gee, I kind of like Netscape Messenger! I am using it right now. I have used
all the others mentioned on UNIX and I prefer Netscape.
The only complaint I have about it is the font for the message list is too small
and can't be changed. It should follow the font selection preferences, but does
not. The messages themselves do use the preference.
Netscape makes it a breeze to watch threads of interest.
Don
--
[Continued Democracy requires Quality Public Education]
------------------------------
From: [EMAIL PROTECTED] (Ron J Theriault)
Subject: Best Free E-Mail for Linux?
Date: 18 Aug 2000 14:05:27 -0500
What's the best free e-mail service to use for someone
who runs Linux?
What's the easiest free Web hosting site to use for
someone familiar with Unix and Apache?
--
Ron Theriault : CS Department, Texas A&M Univ.
[EMAIL PROTECTED] http://www.cs.tamu.edu/staff/ron
"Victimless crime" is a euphemism for "political crime".
--
Ron Theriault : CS Department, Texas A&M Univ.
[EMAIL PROTECTED] http://www.cs.tamu.edu/staff/ron
"Victimless crime" is a euphemism for "political crime".
------------------------------
Subject: Re: STTY and ERASE
From: Johan Kullstam <[EMAIL PROTECTED]>
Date: 18 Aug 2000 15:33:59 -0400
Tony Lawrence <[EMAIL PROTECTED]> writes:
> Floyd Davidson wrote:
>
> > The fact that MS-Windows happens to do it that way does NOT make
> > it standard,
>
> Don't you love it? Microsoft wasn't ANYTHING 25 years ago,
> but now we get dweebs who ask (whine, more typically) "Why
> couldn't they have used a backslash like Microsoft does?"
that's only a veneer. back in the day (of ms-dos 2), you could
configure ms-dos to accept forward slashes. in fact, it would if you
picked anything else but / as the option marker. so if you picked, -,
it would look a lot like unix. the magic files, nul, prn and friends
could also be set to /dev/nul, /dev/prn &c.
in fact, in its guts, ms-dos 2 and continuing at least through the 3
series would actually take the path, convert backslashes into forward
slash and then proceed to break up the path into its parts. (i know
since it was witnessed by me personally by tracing int 21h calls using
soft-ice)
to this day, microsoft operating systems will accept forward as well
as backward slashes anywhere it won't be confused with an option
char. this is one reason why an NT port of emacs works. it's
probably also demanded by posix but i'm not sure.
> (assuming they don't think "/" is a backslash, which a lot
> of them do!) or "Geez! You'd think they could have named it
> 'tracert'" and of course my favorite "Why do I have to
> login?" Drives me nuts!
--
J o h a n K u l l s t a m
[[EMAIL PROTECTED]]
sysengr
------------------------------
From: steve <[EMAIL PROTECTED]>
Subject: Re: Booting to /dev/sda2
Date: Fri, 18 Aug 2000 13:36:50 -0600
[EMAIL PROTECTED] wrote:
> Guys,
>
> I've seen many posts regarding booting with LILO and only getting 'LI'.
>
> Well, I've tried all the suggestions I've seen with no luck. Here's my
> setup:
>
> /dev/hda1 = NTFS
> /dev/hda2 = WIN98
> /dev/sda2 = Linux
>
> /dev/hda1 and /dev/sda2 both have the boot flag set. I am
> running bootpart for the NT OS Loader (to get Win98 bootable) and am
> also using it to boot to Linux. This is where the weird stuff comes in.
> First, let me say that I have /etc/lilo.conf set up very simply. No
> linear/disk/bios options. boot and root are both set to /dev/sda2.
>
"both boot and root MUST be on the first SCSI device (/dev/sda1 id 0)
(pg. 477 RH 6.1 Reference Guide)", even if you use the IDE MBR. This
apparently some limitation of LILO and using an IDE disk(s) alongside
a SCSI disk(s). You must have some other SCSI device, can it be moved?
------------------------------
From: [EMAIL PROTECTED] (Adam Finkelstein)
Crossposted-To: linux.redhat.misc,alt.os.linux,linux.redhat
Subject: Win4Lin: anyone use it with Micro$oft Outlook on Exchange?
Date: 18 Aug 2000 15:41:05 -0400
Reply-To: [EMAIL PROTECTED]
Good day folks in Linux-land.
Win4Lin has received good reviews--a good way to run Micro$oft apps on a
Linux box. Has anyone used it on a network where they then accessed Outlook?
Specifically, the mail and calendar functions in Outlook?
I'd love to hear your experiences--I'v got to run Outlook. I'd much rather
run it from within my Linux box.
Thanks!
Adam
--
Adam Finkelstein
[EMAIL PROTECTED]
http://metalab.unc.edu/bees/adamf
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: How do you determine the job number from lpr?
Date: Fri, 18 Aug 2000 19:46:01 GMT
Thanks, but lpq -P seems to list all jobs to that printer. In my case
where you have one user "nobody" submitting a lot of jobs, how can you
know which job in the queue was yours? It seems to me that you have to
determine this at the moment the job is queued. Once it is released to
the queue, you can make a good guess, but it seems very difficult to
know for certain which job was yours.
I thought another answer was using the -J option, but that merely
affects the banner page. And yes, I might still be able to leverage
this in grepping the control file looking for my string and then
matching up the control file with the print file, but surely there has
to be a simpler way.
Maybe there is a way to manage this through peeking and managing
the .seq file?
Mike
In article <Pine.LNX.4.21.0008181527340.3452-
[EMAIL PROTECTED]>,
Arlan Lucas de Souza <[EMAIL PROTECTED]> wrote:
> On Fri, 18 Aug 2000 [EMAIL PROTECTED] wrote:
>
> > I need to know the job number that was assigned when lpr spools the
> > print job. I can't seem to find a way to get the job number with
> > certainty.
> [...]
>
> Try lpq -P<printername>
>
>
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: Richard Lyon <[EMAIL PROTECTED]>
Subject: Re: Hard Drive is Lieing.
Date: Fri, 18 Aug 2000 20:00:20 GMT
In article <[EMAIL PROTECTED]>,
N/A <[EMAIL PROTECTED]> wrote:
> i have deleted linux completly and re-installed windows, i dont have
> access to or to create a windows startup disk for some reason but
anyway
> Windows says i now have a smaller hard drive than i actually have,
linux
> is respsonsible for this, i even used partition magic on Mandrake
7.0. to
> clear it and it stills says i have a smaller disk space. why and how
can i
> fix it?
>
> --
> Posted via CNET Help.com
> http://www.help.com/
>
Rerun Partition Magic and delete the old linux partition. Then do a
resize on the DOS partition and this will recover the drive space.
This is best done with Partition Magic version 4 or 5.
--
Richard Lyon
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: softrat` <[EMAIL PROTECTED]>
Subject: AutoMagic Linux e-mail -- How do I set it up?
Date: Fri, 18 Aug 2000 13:11:34 -0700
I would like to automagically send and receive e-mail from my Linux box
to my E-Mail provider on the Internet such that my users do everything
from the BSD (or whatever) 'mail' command if possible.
Sending Mail: Out-going mail is identified with the '@'.
Receiving Mail: Periodically query my E-Mail provider.
Always: Check to see if my ISP is there. If it isn't try again later. If
it is, do the send/receive stuff.
Apparently neither fetchmail nor sendmail is set up properly to do this
magic, if that is even possible.
Thanks for the help.
--
the softrat
mailto:[EMAIL PROTECTED]
PS. I got lost in the 'man' pages for 'mail', 'sendmail', and
'fetchmail'.
---
A husband is what is left of the lover after the nerve has been
extracted.
-- Helen Rowland
------------------------------
From: Floyd Davidson <[EMAIL PROTECTED]>
Subject: Re: STTY and ERASE
Date: 18 Aug 2000 10:21:34 -0800
Thomas Dickey <[EMAIL PROTECTED]> wrote:
>Floyd Davidson <[EMAIL PROTECTED]> wrote:
>
>> Probably true on hardcopy terminals, but he is going back even
>> farther than that, to tape punches. The BS key would backspace
>> the tape, the DEL key would punch all holes, thus deleting
>> whatever had been punched previously.
>
>well, when I used paper tape, we didn't use rubouts anyway (I recall
>a comment from someone who said a tape with a lot of rubouts had
>caused one of our tty's to fail ;-). DEC used to distribute some
>de-commented source that had rubouts where the comments used to
>be -- but that's a different protocol.
Could be that DEC didn't make a very good tape machines??? But
Teletype equipment (model 15's and 28's for example) certainly
didn't have any problems with it. Neither did 32's and 33's.
But there was also another use for "rubouts" with paper tape:
repairing and splicing tapes!
For those with punches that poke a complete hole (and filled up
a little box with "chad") this was not nearly as easy, though it
could be done. But with the "chadless" tape punch that punched
only about 320 degrees of the circle and thus did not produce
chad, it was easy.
A broken tape could be mended, or a new tail could be added onto
the end of a tape. Just start the new tail section with a
series of deletes so that all the holes were punched, and splice
that tape to the end of the first tape by placing the first tape
into the tape reader and putting the start of the second tape,
where all of the holes are punched, directly over it, for say 4
to 6 inches. The tape-reader is then activated and the hole sensors
punch the attached chad from the bottom tape up into the hole in
the second tape, and the two are relatively well stuck together.
Sometimes one might need to inspect it carefully and push a few
recalcitrant holes through with the point of a pencil or a
paperclip.
A torn tape could be mended by taking a short length with
all of the holes punched and placing that directly over the
torn part of the original tape while running it through a
reader.
>> (Regardless, on my terminal the key labeled "Backspace" most
>
>mine does:
>
>ymmv.
Does it really? You must be actually using a real terminal, as
opposed to a PC keyboard then, eh? Given that you probably work
with curses more than anybody alive... I guess that comes as no
surprise!
Floyd
--
Floyd L. Davidson [EMAIL PROTECTED]
Ukpeagvik (Barrow, Alaska)
------------------------------
From: Floyd Davidson <[EMAIL PROTECTED]>
Subject: Re: STTY and ERASE
Date: 18 Aug 2000 11:33:36 -0800
Tony Lawrence <[EMAIL PROTECTED]> wrote:
>Floyd Davidson wrote:
>
>> The fact that MS-Windows happens to do it that way does NOT make
>> it standard,
>
>Don't you love it? Microsoft wasn't ANYTHING 25 years ago,
>but now we get dweebs who ask (whine, more typically) "Why
>couldn't they have used a backslash like Microsoft does?"
>(assuming they don't think "/" is a backslash, which a lot
>of them do!) or "Geez! You'd think they could have named it
>'tracert'" and of course my favorite "Why do I have to
>login?" Drives me nuts!
What burns me about it is that usually the reason UNIX does
it this way or that is because somebody with a lot of smarts
figured out what the *best* way to do it is. And if it took
three or four tries to get it right, we changed 3 or 4 times!
MicroSoft then comes along and, literally, does it different
just to be different. That means they chose a way that is not
the best just to be not the best. And almost any given mistake
like that which was in DOS 1.0 is *still* in WindowsNT, or at
least it seems like it. Their totally idiotic "more" command,
for example.
--
Floyd L. Davidson [EMAIL PROTECTED]
Ukpeagvik (Barrow, Alaska)
------------------------------
From: [EMAIL PROTECTED] (NF Stevens)
Subject: Re: STTY and ERASE
Date: Fri, 18 Aug 2000 20:30:30 GMT
f"Andrew N. McGuire " <[EMAIL PROTECTED]> wrote:
[snip]
>
>~~ >
>~~ >as for X, rxvt can be configured at compile time. i have this in my
>~~ >.xdefaults, but i try to use ^? everywhere (since emacs uses ^H as the
>~~ >help function).
>~~
>~~ Then emacs is broken. ^H is in the ascii character set as backspace.
>~~ If a piece of software cannot even adhere to the most basic standards
>~~ then it should fixed.
>
>I am curious, which standard would that be?
>
ascii
HTH
Norman
------------------------------
From: [EMAIL PROTECTED] (NF Stevens)
Subject: Re: STTY and ERASE
Date: Fri, 18 Aug 2000 20:30:40 GMT
Floyd Davidson <[EMAIL PROTECTED]> wrote:
>Charles H. Chapman <[EMAIL PROTECTED]> wrote:
>>Andrew N. McGuire wrote:
>>> On Thu, 17 Aug 2000, NF Stevens quoth:
>>> ~~ Then emacs is broken. ^H is in the ascii character set as backspace.
>>> ~~ If a piece of software cannot even adhere to the most basic standards
>>> ~~ then it should fixed.
>>>
>>> I am curious, which standard would that be?
>>
>>That "backspace" (0x8) always deletes the character before the cursor and
>>delete (0x7f) always deletes the character the cursor is on. For all its
>>other faults, at least every single Windows program adheres to this
>>standard and there's no reason why Linux/Unix programs couldn't do the
>>same.
>>
>>Chuck
>
>The fact that MS-Windows happens to do it that way does NOT make
>it standard, and it not true that BS _always_ deletes the
>character before the cursor.
The fact that this behavious seems to be universal on
every version of unix apart from linux is, to me,
an indicator of a de facto standard. As part of my
job I have to remotely access a large number of systems,
include aix, hpux, dynix, solaris, winnt. I have no
problems with unless the connection starts or terminates
on a linux box. In my experience it is only linux which
does not, by default, generate a backspace character from
the key which, from its labelling, is the backspace key.
In this respect linux is broken.
> A destructive BS is _not_
>universal, and indeed at one time was non-existant! Unix was
>doing it the right way long before Windows was a gleem in the
>eye of Bill Gates. If Windows had followed the appropriate
>standards you would not be so confused too.
>
>For all its faults, why would we take another faulty "standard"
>from MS-Windows and use it to screw up the rest of the world?
What is faulty about having the backspace key generate a
backspace character?
Norman
------------------------------
From: [EMAIL PROTECTED] (NF Stevens)
Subject: Re: STTY and ERASE
Date: Fri, 18 Aug 2000 20:30:59 GMT
Johan Kullstam <[EMAIL PROTECTED]> wrote:
[snip]
>i think this stuff comes from the card puncher days. a lot of the
>weird ascii control codes seem to stem from card whalloping.
This is now totally irrelevant. We aren't using punched cards
these days. IME ^H as the default erase character is the
de facto unix standard. If you are using your linux box
standalone then you can work round its brokenness but if
you remotely access other unix or windows boxes then it
is obvious that linux is out of step.
Norman
------------------------------
From: [EMAIL PROTECTED] (NF Stevens)
Subject: Re: STTY and ERASE
Date: Fri, 18 Aug 2000 20:30:49 GMT
Johan Kullstam <[EMAIL PROTECTED]> wrote:
>[EMAIL PROTECTED] (NF Stevens) writes:
>
>> Johan Kullstam <[EMAIL PROTECTED]> wrote:
[snip]
>>
>> Then emacs is broken. ^H is in the ascii character set as backspace.
>> If a piece of software cannot even adhere to the most basic standards
>> then it should fixed.
>
>ironically, that was my first tack. the vt100 sends ^? <del> as
>opposed to ^H <bs> when you press that wide key somewhat above return.
>since a vt100 sends <del>, i figured sun's attempt at emulation was
>broken in sending <bs>. perhaps your idea of a standard should be
>fixed as it is that which is broken.
In linux the TERM environment variable is usually set to "linux"
not "vt100" so what a vt100 does is irrelevant. Most pc keyboards
have a backward pointing arrow on the key above the return key.
Pc documentation invariably refers to this as the "backspace key".
Given this I consider that any system which produces any ascii
character other than backspace from this key is broken.
The other problem with linux is that, by default, it uses
a character other than ^H as the default for erase. Every
other unix system that I encounter seems to use ^H as the
default erase character. One can, of course, easily change
the default but 1) this doesn't work _before_ one has
logged in and 2) you cannot always automate it because
it depends on the system from which the keystroke originates.
If, as I do, you login in remotely to many different systems
then it is painfully obvious that linux is out of step with
the rest of the unix world.
Norman
------------------------------
From: Tony Lawrence <[EMAIL PROTECTED]>
Subject: Re: regex example
Date: Fri, 18 Aug 2000 16:48:28 -0400
john slimick wrote:
>
> I'm trying to include the use of
> regex into some course material
> and the man page is very terse.
>
> Could someone provide an example
> of the regcomp input string
> that matches the usual style
> phone number of
>
> (xxx)xxx-xxxx
>
> where I can capture the
> the starting and ending indices
> of the three
> numeric fields.
>
> I am baffled at the use of
> parentheses here -- if \( \)
> delimits a sub-expression
> then how do I quote the parens
> around the area code?
I don't have to quote them :
sed 's/\((...\)/\1/' (simplified of course)
works for me..
--
Tony Lawrence ([EMAIL PROTECTED])
Linux articles, help, book reviews, tests,
job listings and more : http://www.pcunix.com/Linux
------------------------------
From: [EMAIL PROTECTED]
Subject: bash
Date: Fri, 18 Aug 2000 20:53:57 GMT
Hi,
is there a way to access command line parameters in alias definition ?
Thanx
James
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: Arlan Lucas de Souza <[EMAIL PROTECTED]>
Subject: Re: How do you determine the job number from lpr?
Date: Fri, 18 Aug 2000 17:49:20 -0400
On Fri, 18 Aug 2000 [EMAIL PROTECTED] wrote:
> Thanks, but lpq -P seems to list all jobs to that printer. In my case
> where you have one user "nobody" submitting a lot of jobs, how can you
> know which job in the queue was yours? It seems to me that you have to
> determine this at the moment the job is queued. Once it is released to
> the queue, you can make a good guess, but it seems very difficult to
> know for certain which job was yours.
[...]
Ok, I think I understanding your problem this time. What about this:
1) Create a temporary directory, for example, ~/JobsToPrint
2) Copy the files to be print to ~/JobsToPrint
3) Print ~/JobsToPrint/* using 'lpr' command with '-r' option
(remove the file upon completion of spooling)
Now, you'd know the printed and non-printed files simple listing the
~/JobsToPrint content.
------------------------------
** 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 (and comp.os.linux.misc) via:
Internet: [EMAIL PROTECTED]
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
******************************
