Linux-Misc Digest #52, Volume #26 Mon, 16 Oct 00 16:13:04 EDT
Contents:
chmod ("Johannes G�tz")
Howto empty log files (Tobias Schenk)
Re: Is there a MS Word (or substitute) for Linux? (Roberto Teixeira)
Re: Is there a MS Word (or substitute) for Linux? (John Hasler)
Re: Is there a MS Word (or substitute) for Linux? (Brian Langenberger)
rdate; adjusting bios. (Thaddeus L Olczyk)
Re: rdate; adjusting bios. (Lew Pitcher)
Re: Is there a MS Word (or substitute) for Linux? (Bartek Kostrzewa)
Re: Is there a MS Word (or substitute) for Linux? (Jean-David Beyer)
Re: Is there a MS Word (or substitute) for Linux? (The Ghost In The Machine)
Re: Is there a MS Word (or substitute) for Linux? (The Ghost In The Machine)
Re: Is there a MS Word (or substitute) for Linux? (Roberto Teixeira)
I/O in application programs. (Fred)
----------------------------------------------------------------------------
From: "Johannes G�tz" <[EMAIL PROTECTED]>
Subject: chmod
Date: Mon, 16 Oct 2000 21:10:30 +0200
Hi!
Is it possible to change the default attributes of files generated by a
user?
Every file that is written should have the a+wX attributes.
Is it possible to change this behaviour wth in.ftpd?
Thanks for your answers,
Johannes
[EMAIL PROTECTED]
------------------------------
From: Tobias Schenk <[EMAIL PROTECTED]>
Subject: Howto empty log files
Date: Mon, 16 Oct 2000 21:14:07 +0200
Hi,
How can I archive log files?
When I rename an existing log file then still messages will be written
in it.
I exspected that a new file would be created and filled.
How does this mechanism work?
How to archive correctly?
I have seen this effect in vbox log file and in messages.
Thank you,
Tobias
------------------------------
From: Roberto Teixeira <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.advocacy
Subject: Re: Is there a MS Word (or substitute) for Linux?
Date: 16 Oct 2000 17:15:08 -0400
>>>>> "Jan" == Jan Schaumann <[EMAIL PROTECTED]> writes:
Jan> The most portable document format is PDF (Portable Document
Jan> FOrmat - D'uh). RTF is not half as portable.
Not to start a document format war, but isn't PDF a proprietary
format? What about PS? I don't know if PS is proprietary, but it sure
is *very* portable
--
Roberto Teixeira
------------------------------
From: John Hasler <[EMAIL PROTECTED]>
Subject: Re: Is there a MS Word (or substitute) for Linux?
Date: Mon, 16 Oct 2000 18:03:58 GMT
Jean-David Beyer writes:
> It had the required space the instant before I sent it, but I see someone
> has deleted it. It is there now, and I will now press "Send".
It isn't there.
--
John Hasler
[EMAIL PROTECTED]
Dancing Horse Hill
Elmwood, Wisconsin
------------------------------
From: Brian Langenberger <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.advocacy
Subject: Re: Is there a MS Word (or substitute) for Linux?
Date: 16 Oct 2000 19:23:12 GMT
In comp.os.linux.advocacy Roberto Teixeira <[EMAIL PROTECTED]> wrote:
:>>>>> "Jan" == Jan Schaumann <[EMAIL PROTECTED]> writes:
: Jan> The most portable document format is PDF (Portable Document
: Jan> FOrmat - D'uh). RTF is not half as portable.
: Not to start a document format war, but isn't PDF a proprietary
: format? What about PS? I don't know if PS is proprietary, but it sure
: is *very* portable
I believe both Postscript and PDF are Adobe proprietary, but are
well documented to the point where anyone can make a reader
(Ghostview, xpdf) or writer (ps2pdf, dvipdfm) for them,
effectively making an open standard. And since Adobe hasn't
shown much desire to make us participate in the
"File Standard Hoop-Jumping Olympics", there's little to fear
that either will be severely mucked-up anytime soon.
------------------------------
From: [EMAIL PROTECTED] (Thaddeus L Olczyk)
Subject: rdate; adjusting bios.
Date: Mon, 16 Oct 2000 19:33:25 GMT
Reply-To: [EMAIL PROTECTED]
I use rdate to adjust the system clock.
The problem seems to be when I reboot.
The time becomes very wrong.
I suspect this is because the bios is wrong.
Is there some way ( other than by hand )
of adjusting the system bios.
------------------------------
From: [EMAIL PROTECTED] (Lew Pitcher)
Subject: Re: rdate; adjusting bios.
Reply-To: [EMAIL PROTECTED]
Date: Mon, 16 Oct 2000 19:40:08 GMT
On Mon, 16 Oct 2000 19:33:25 GMT, [EMAIL PROTECTED] (Thaddeus L
Olczyk) wrote:
>I use rdate to adjust the system clock.
>The problem seems to be when I reboot.
>The time becomes very wrong.
>I suspect this is because the bios is wrong.
>Is there some way ( other than by hand )
>of adjusting the system bios.
>
There are a number of utilities to adjust the hardware clock.
Slackware Linux comes with hwclock and /sbin/clock, either of which
can be used to set the hw clock.
Check your distro, and/or check freshmeat (http://www.freshmeat.net/)
for clock utilities.
Lew Pitcher
Information Technology Consultant
Toronto Dominion Bank Financial Group
([EMAIL PROTECTED])
(Opinions expressed are my own, not my employer's.)
------------------------------
Date: Mon, 16 Oct 2000 21:47:25 +0200
From: Bartek Kostrzewa <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.advocacy
Subject: Re: Is there a MS Word (or substitute) for Linux?
jazz wrote:
>
> I really need a powerful word processor with templates, styles, etc.
>
> What is available for Linux? How about for Powerpoint and Excel?
>
> Thanks ---
> Jazz
go for staroffice (respectively openoffice from now on)
www.openoffice.org
Then there's abiword (it's a little unstable still) www.abisource.com
Many others to cite still.
--
Best regards,
Bartek Kostrzewa - [EMAIL PROTECTED]
<<< http://technoage.web.lu >>>
------------------------------
From: Jean-David Beyer <[EMAIL PROTECTED]>
Subject: Re: Is there a MS Word (or substitute) for Linux?
Date: Mon, 16 Oct 2000 15:46:55 -0400
John Hasler wrote:
> Jean-David Beyer writes:
> > It had the required space the instant before I sent it, but I see someone
> > has deleted it. It is there now, and I will now press "Send".
>
> It isn't there.
I can see that. In my e-mails that I send (in the "sent" folder), the needed
spaces are missing, too, and have been since at least last May (as far back as
I have retained anything), except once.
I examined some of the stuff in my inbox. They have all kinds of signature
separators:
"-- ", "--", "---", "=====", "" (i.e., nothing). They also have some with
"business cards". Let me try an experiment: Netscape formats the separator and
everything after it as "Formatted" (which gets me a different type style and
constand-width format type face. I manually changed it to "Normal" for the
separator. I checked and the space is still there.
--
.~. Jean-David Beyer Registered Linux User 85642.
/V\ Registered Machine 73926.
/( )\ Shrewsbury, New Jersey
^^-^^ 3:35pm up 6 days, 21:13, 4 users, load average: 3.64, 3.74, 3.69
------------------------------
From: [EMAIL PROTECTED] (The Ghost In The Machine)
Crossposted-To: comp.os.linux.advocacy
Subject: Re: Is there a MS Word (or substitute) for Linux?
Date: Mon, 16 Oct 2000 19:48:35 GMT
In comp.os.linux.advocacy, Matthias Warkus
<[EMAIL PROTECTED]>
wrote
on Mon, 16 Oct 2000 15:23:25 +0200
<[EMAIL PROTECTED]>:
>It was the Sun, 15 Oct 2000 19:50:23 -0400...
>...and jazz <[EMAIL PROTECTED]> wrote:
>> > claire
>>
>>
>> Are you this offensive in real life too?
>
>Claire does not exist in real life. There has been more than a dozen
>of trolls with similar stances and a similar idiom and vocabulary
>around on this group over the last months. They all had pseudonyms,
>they all posted from New York City IIRC, etc.
>
>The troll has changed its stance from blunt and total denial of any
>quality of Linux whatsoever to a slightly differentiated one, where he
>pretends to know its advantages as a server operating system. He has
>also become a bit, but not much, less offensive. Furthermore, he has
>adopted a female nickname, which is a clever move -- he's wagering on
>the hormones and/or instincts of the overwhelming male majority in
>this group.
>
>Of course the troll ("Claire") is not a woman. No woman would be that
>silly, only men are stubborn and playful enough to play a game to the
>point of total absurdity and beyond (e.g. the Cold War).
>
>In the de.* hierarchy, we tell everyone who doesn't post under their
>realname to fuck off, which is a useful policy IMO.
Gee, thanks. :-P
Mind you, at least I try. :-) I'm not quite sure what to make
of claire at times; at times s/he is almost reasonable, but at
times s/he comes on as a very argumentative individual.
Of course, I also don't change my moniker every three months,
either. I also try to admit my shortcomings -- I'm not an expert
on high-volume multi-CPU servers, for example. (And I doubt NT
can handle such, even were the CPUs ix86-based. However, I'm
not an expert on benchmarks, either. :-) )
>
>mawa
>--
>Bl�mchenpfl�cker!
>Bonsaig�rtner!
>Beinrasierer!
>Beischlafbettler!
--
[EMAIL PROTECTED] -- who is the Ghost, not the Troll In The Machine :-)
(and who doesn't frequent de.* anyway)
------------------------------
From: [EMAIL PROTECTED] (The Ghost In The Machine)
Crossposted-To: comp.os.linux.advocacy
Subject: Re: Is there a MS Word (or substitute) for Linux?
Date: Mon, 16 Oct 2000 19:50:51 GMT
In comp.os.linux.advocacy, Matthias Warkus
<[EMAIL PROTECTED]>
wrote
on Mon, 16 Oct 2000 15:17:34 +0200
<[EMAIL PROTECTED]>:
>It was the Mon, 16 Oct 2000 18:18:14 +0800...
>...and Todd <[EMAIL PROTECTED]> wrote:
>> This is a time that will be long remembered. It has seen the end of UNIX's
>> hold on design software, it will soon see the end of its hold on publishing
>> software.
>
>Unix? A hold on publishing software?
>
>Hallooooo? <waves hand>
Well, there was nroff/troff in the early 80's, perhaps. I'm not
sure if that counts or not. :-)
>
>Publishing software has, if any, been an Apple domain... for DTP to be
>a real success on Unix workstations, the font handling has always been
>too primitive.
>
>The only notable exception is Framemaker, AFAIk. This may change in
>the next few years.
One hopes it changes to TeX (or Lyx or Kylix) and not to Word... :-)
[.sigsnip]
--
[EMAIL PROTECTED] -- insert random misquote here
------------------------------
From: Roberto Teixeira <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.advocacy
Subject: Re: Is there a MS Word (or substitute) for Linux?
Date: 16 Oct 2000 17:51:33 -0400
>>>>> "Bartek" == Bartek Kostrzewa <[EMAIL PROTECTED]> writes:
Bartek> go for staroffice (respectively openoffice from now on)
Bartek> www.openoffice.org
Bartek> Then there's abiword (it's a little unstable still)
Bartek> www.abisource.com
Bartek> Many others to cite still.
What about KOffice? I've been using it for about a month or two and it
seems to be nice.
--
Roberto Teixeira
------------------------------
From: Fred <[EMAIL PROTECTED]>
Subject: I/O in application programs.
Date: Mon, 16 Oct 2000 15:58:54 +0000
I am learning Unix. In the past I have written many
programs, most of which I coded in IBM mainframe assembly
language. Also I have programmed other machines, usually in
assembly. But assembly is not the only programming notation
I have used, and I (like nearly everyone else) use C when I
work with Unix.
I find that some of my past experience is transferable
to Unix. For example, the Unix handling of signals (and
especially signal masks) is much like the way hardware
handles interrupts. But a 100% transfer is not possible, nor
do I expect that. Therefore I am not surprised that a
difficulty has arisen. It pertains to I/O. There are a
number of ways in which I could try to explain this
difficulty, but the best way is by example, using assembly
language as a point of reference.
Suppose that we wish to write a simple application
program that repeatedly reads a record from a file,
computationally modifies the data in the record, and then
writes the changed record to another file. The program uses
an input buffer to read the data, a computational buffer for
modifying the data, and an output buffer to write the data.
In its basic schema, this program may be described in the
following sequence of steps. (For simplicity, I omit error
handling).
Step 1: Open the files.
Step 2: Commence an input operation on the input file.
Step 3: Tell the operating system to suspend this program's
execution until such moment as the most recently-initiated
input operation is finished.
Step 4: Test the most recently-completed input operation, to
determine whether EOF has been encountered. If so, go to
step 12.
Step 5: Move all data from the input buffer to the
computational buffer.
Step 6: Commence an input operation on the input file.
Step 7: Modify the data in the computational buffer.
Step 8: Tell the operating system to suspend this program's
execution until such moment as the most recently-initiated
output operation (if any) is finished.
Step 9: Move all data from the computational buffer to the
output buffer.
Step 10: Commence an output operation on the output file.
Step 11: Go to step 3.
Step 12: Tell the operating system to suspend this program's
execution until such moment as the most recently-initiated
output operation (if any) is finished.
Step 13: Close both files.
Step 14: Stop.
An IBM systems programmer might do I/O using the X'9C'
instruction, but the typical IBM application programmer uses
the EXCP macro to accomplish steps 2, 6, and 10. He uses the
WAIT macro to accomplish steps 3, 8, and 12. However, no one
need know IBM mainframe language in order to answer my Unix
question. He need only understand what I attempt to
accomplish by the fourteen steps listed above. I run three
kinds of procedure, which are input, output, and compute. I
attempt to run them concurrently, and to coordinate them so
that they do not interfere with one another. For example, I
do not want to alter the contents of the output buffer during
the time interval in which the output driver is obtaining
data from it. And I do not want to move data from the input
buffer to the computational buffer at a moment when the most
recent input operation has not finished filling the buffer.
Note that the above model does not require the
application program to handle any interrupts or to process
any signals. All such work is left in the hands of the
operating system. As far as I know, the above model is
consistent with standard techniques employed by application
programmers around the world. Can it be used by a C
programmer who works in the Unix environment? Probably so.
But I do not know how.
I assume that steps 2 and 5 can be implemented using the
Unix read call. I assume that step 10 can be implemented
using the Unix write call. I assume that my program can use
the O_NDELAY flag or the O_NONBLOCK flag (perhaps in
combination with other flags) to avoid "getting stuck" in
those steps. It appears, therefore, that steps 2, 5, and 10
present no real obstacle to implementation of the above
scheme.
But what is the Unix analog of IBM's WAIT macro? In
other words, how can I implement steps 3, 8, and 12? At
first glance it appears that I can do so via the Unix select
function, or perhaps poll. But can I? For guidance on this
point, let us turn to W. Richard Stevens's book entitled
Advanced Programming in the Unix Environment. (Addison-
Wesley, 1993.) This is not the most recent book in the
subject, but perhaps that's just as well. Not everyone has
the most recent release of Unix. If I want my program to be
portable, I should make it compatible with systems of
moderate age.
In section 12.5 of his book, Stevens treats the topic of
"I/O Multiplexing." Therein Stevens carefully cautions his
reader that "I/O multiplexing is not yet part of POSIX." In
other words, reference to the POSIX specification will not
answer my questions. I must determine how things work in the
real world. And because I wish to maximize the portability
of my program, I must not limit my inquiry to the workings of
my specific Unix platform. Rather, I must try to learn how
most Unix platforms do this kind of work.
At first glance, Stevens appears to offer real guidance.
The first sentence of section 12.5.1 says, "The select
function lets us do I/O multiplexing under both SVR4 and
4.3+BSD." That seems promising. Stevens goes on to explain
how file descriptors can be grouped into sets, so that one
call to the select function can be used on more than one
descriptor. Also he explains that each call can specify
three sets, one of which corresponds to each of three
conditions. Those explanations are quite clear. Then he
says, "A positive return value specifies the number of
descriptors that are ready. In this case the only bits left
on in the three descriptor sets are the bits corresponding to
the descriptors that are ready." From his remarks, I gather
that the select function will (optionally) suspend execution
of the application program until such moment as there is a
change in one or more of these bits.
So far, so good. I say this because, although the
above-quoted passages may appear cryptic when lifted out of
context (as I have done), they are clear enough when read in
combination with other things Stevens says about I/O
multiplexing. The problem comes a couple of paragraphs
later, where he says, "We now need to be more specific about
what 'ready' means." Stevens then proceeds to define "ready"
so that the word has no genuine meaning. He says, for
instance, that "A descriptor in the write set is considered
ready if a write to that descriptor won't block."
In passing, I wonder how the kernel can know for sure
that a future I/O operation won't block, without knowing the
details of the operation. What about an attempt to write a
truly humongous number of bytes? It seems to me that such an
attempt might block, unless the driver has a truly humongous
buffer. However, that is merely an aside. My real problem
is as follows.
No doubt it is nice to know that the next call to write
will not block, but that's not the same as knowing that the
previous call to write is finished. Similarly it is nice to
know that the next call to read will not block, but that's
not the same as knowing that the previous call to read is
finished, and much less is it the same as knowing that the
read did not encounter an EOF condition. In other words, the
select function offers a prediction about the future, but
that is not what I need at steps 3, 8, and 12. What I need
is not a prediction about the future, but rather a statement
about the present. For example, when my program arrives at
step 3, either the input operation is finished, or it's not.
If it's finished, I can safely copy data from the input
buffer to the computational buffer. If it's not finished, I
need to go to sleep until such time as it is finished.
Apparently, neither select nor poll meets this need.
Yes, Stevens says that the select function enables I/O
multiplexing under both SVR4 and 4.3+BSD. But he includes in
section 12.5 no example program showing how this can be
accomplished.
Does anyone know?
If so, please drop me a line at [EMAIL PROTECTED]
Thank you.
------------------------------
** 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
******************************
