this defending "approximations", or as you put
it "fuzziness". The problem is, I don't think they're approximations. I
strongly suspect they're often closer to 5 or 10 times out. Think of it this
way and you'll perhaps realise why analog doesn't provide them
ctor. Since the processes that cause the timing to be wrong apply in
> identical ways to pages at the same site [...]
No they don't. Different areas of the site have different audiences.
Different page types have different processes. (Think of an index page vs a
content page).
--
ile it does not work but when I launched it
>directly it
> does?
>
Erm, you didn't say what the error was.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
"The new operating system wi
ations page.
>
I'm a bit late onto this as I've been on holiday, but there is already
something on the Helper Apps page to convert sendmail logs to common log
format.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road,
people
who want ten pictures of the buttons to press for each task, to sysadmins
with ten years of experience. It's very difficult to write docs to satisfy
both. In fact I get about as many comments that the documentation was
unusually easy to follow as complaints that it was unreadable. N
; idea what the overhead would be, but it seems more trouble than it would be
> worth.
>
Oooh, I tried this once, a long time ago. For a week or so the stats on my
site were 'live'. The processing power was negligible, but my sysadmin
didn't like the amount of memory
e and unstable). You can also
> just download the tarball and make it. It build without a hitch on debian.
>
So it should, it's my development platform. :)
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB
On Tue, 5 Sep 2000, Ales Trtnik wrote:
> Hi,
>
> So quotation is the answer. Stupid me. Thanks.
>
> But now another problem:
> analog classifies it as "Unknown Windows" instead of "Windows Me".
>
Windows Me is Windows Millennium? Is it released ye
is was discussed here a few months ago, and Stephen Turner answered
> "Hmm, I thought I'd fixed this. I'll look into it again. Sorry about
> that."
>
> I would just like to checkup on the status concerning this.
Sorry, the status is that I haven't looked at ana
On Fri, 8 Sep 2000, Silviu Dicu wrote:
> hello,
>
>
> is anybody which has been use %q in a log format I used and I had
> strange results.
It's working as far as I know. What problem have you had specifically?
--
Stephen Turner http://www.stats
between Windows 98's first and second editions.
>
But 98 is so passe. "Millennium" is the top word of the moment. Although I
think "Millenium" might be even higher.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, W
xample lines were Corrupt Lines
anyway. Maybe you should check out that possibility.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
"The new operating system will recover more easi
l, that's just the normal analog output. I guess you really need to read
the documentation and find out how to configure the program... If there's
specific things that you can't work out how to do, we might be able to help
you.
--
Stephen Turner http://www.statslab.cam
on the mailing list. -DNODNS will certainly fix it, although you will
lose DNS lookups. It may be that you can also add a library to fix it: see
the LIBS in the Makefile for what has worked on other Unices.
Let me know how you get on.
--
Stephen Turner http://www.stats
equests and bytes, I think you must
have the same logfile in two LOGFILE commands, one with the second parameter
and one without. This would explain the double-counting.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory
On Mon, 18 Sep 2000, Stephen Turner wrote:
>
> Hmm, I've never heard of anyone having trouble with AIX, though you could
> ask on the mailing list.
Whoops, that was already on the mailing list. Too much mail to sort out at
the moment!
--
Stephen Turner
m. Look at the HTML output and it will become obvious.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
"The new operating system will recover more easily from system crashes."
e requests from last week
> are reported. I tried turn LASTWEEK (FALSE) in analhead.h but did not work
> Where i am making mistake ?
>
Did you use FROM?
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3
, contributions are
always gratefully accepted. :)
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
"The new operating system will recover more easily from system crashes."
if so, you might also wonder why you want to waste 8
bytes per logfile line...
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
"The new operating system will recover more easily from system crashes
terface of the sort you're talking about that I can imagine would
simultaneously be insufficiently flexible for the users and much too
complicated to program. But maybe I'm just not imaginative enough. :)
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Lab
gt;
I replied to this in another mail already, but just to get it in the right
thread:
FAILCOLS D (or d)
It's documented too.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
failed referrer report. I read the web archive, but no tip can help me.
> Has anybody an idea?
>
Well, you can't quite do that, but what you can do is restrict the whole
report to only 404's by using STATUSINCLUDE 404 and run it again looking at
just those request
, Netscape thinks it is running on Windows 95.
>
*sigh*
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
"The new operating system will recover more easily from system crashes."
to do it back in
May, but I haven't seen any progress. I'll contact him again. I might put it
back on the "help wanted" list.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
"
lly, Marco, I wasn't referring to you. I had another volunteer.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
"The new operating system will recover more easily from system crashes."
to full
mailboxes. So my total mailbox size was over 2700 messages.
23 out of 271 people unsubscribed themselves from the list.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
"The new operating syst
t his basic assertion that "Reply To all" replies to the
list only; it replies to the author too which is rude. It seems from the
comments page that lots of other people disagree too.
I agree with your other points. This was a very badly configured vacation
program in several ways.
-
quot;GoogleBot",
>for example.
>
Yes, have a look at BROWEXCLUDE.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
"The new operating system will recover m
approve mfjjmand \
subscribe analog-help \
[EMAIL PROTECTED]
end
[The above is broken into multiple lines to avoid mail reader linewrap
problems. Commands can be on one line, or multi-line with '\' escapes.]
If you disapprove, do nothing.
Thanks!
[EMAIL PROTECTED]
-
Moral... never try and read email while listening to the Olympics on the
radio. I changed the password before this hit the list, so don't try
anything...
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, En
On Thu, 21 Sep 2000, mjn wrote:
>
> I reran analog but it doesn't seem to be parsing all of the data. How do
> I get it to reread and make stats for everything anew?
It doesn't save its state. If you rerun it, it will read everything afresh.
--
Stephen Turner
gt; every archive.
Probably the entry is older than the DNSGOODHOURS.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
"The new operating system will recover more easily from system crashes."
nd not ip, so
> it cant be that
>
> Please can someone advise?
>
It looks right. Try running "analog -settings" to see what it thinks it's
meant to be doing.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce R
would be useful. I'll add it to the wishlist, but I'm
afraid it's not likely to happen any time soon. You might have to just
preprocess the logfile...
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce
nipped intentionally]
>
> ...which are great, but I'd really like to see this broken down more, maybe
> showing me 10k-20k, 21-30k, 31k-40k, etc. Is there a configuration option
> to do this that I've missed? Or has anyone else solved this problem?
>
No, there is no
t (twice). Maybe it's just timing out?
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
"The new operating system will recover more easily from system crashes."
http://www.analog.cx/docs/form.html#trouble
point 7.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
"The new operating system will recover more easily from system crashes."
Besides, a lot of browsers let the user configure the UA string.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
"The new operating system will recover more easily from system crashes."
hen configured normally, does show
> the suborg part properly.
>
Possibly you need to set the SUBORGFLOOR lower.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
"The new operating system w
.gz
> DNS NONE
>
> any ideas as to why the discrepancy arises. for the month of july it was
> about 1 million( analog monthy counts more) while for june it is as high as
> 39 million( analog monthly counts more).
Could you post the URL of a sample report?
--
Stephen Turner
You can specify any URL to be -- or not to be -- a page. Of course, the
question is then identifying which ones you do and don't want to count.
But I suppose if your filespace is well organised, a small number of regexps
will sort it out.
--
Stephen Turner
s the percentage of the total number of bytes downloaded from your web
server.
You can have the actual number of bytes if you prefer, using the ORGCOLS
command.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
nd sort
out what's going on.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
"The new operating system will recover more easily from system crashes."
(Micr
>
> When you fix the previous error, this may go away.
>
Actually, this is a new one in 4.12. You need to define -DHAVE_ADDR_T in the
DEFS in the Makefile.
I think I made a mistake here. Too many people need to include this
definition. I'm going to release another new ver
On Tue, 10 Oct 2000, Doug Nelson wrote:
> What is a "bonified request"?
>
I assume it's the same as a "bona fide request". I do like the word
"bonified" though -- "these goods aren't genuine; can you do anything to
bonify them?".
--
St
rn .Z
files back into plain text.
Does anyone know where he might get such a program from?
Alternatively, if you have access to the Solaris machine, you could
uncompress them on that first.
Even better, "gzip" the files rather than "compress" them. gzip compresses
better. Al
On Wed, 11 Oct 2000, Jeremy Wadsack wrote:
>
> Winzip will do this. It also has a command line interface (downloaded separately).
> But it's not free; it's about $35. http://www.winzip.com/
>
Can it decompress to stdout though?
--
Stephen Turner http://
shall pursue, unless
> insanity gets me first...
You mean that analog can't make sense of them (maybe you need to specify a
LOGFORMAT?) or that if you just uncompress the files they are empty, before
you try and analyse them with analog?
--
Stephen Turner http://www.stat
On Fri, 13 Oct 2000, Tom Roper wrote:
> Sorry, yes, analog doesn't find any data in them, though looking at them in
> a text editor finds lots of stuff.
>
Well, probably you've got a lot of Corrupt Lines and you need to change your
LOGFORMAT.
--
Stephen Turner
tput"? Analog always produces something...
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
"The new operating system will recover more easily from system crashes."
(
Apache about this.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
"The new operating system will recover more easily from system crashes."
(Microsoft, aiming h
gt; my config file:
> >
> > rr.com "RoadRunner"
> > aol.com "America Online"
> > wisc.edu "University of Wisconsin, Madison, WI"
>
> This can be done with ALIAS commands, specifically HOSTALIAS.
ORGOUTPUTALIAS, actually.
--
Stephen Turner
On Wed, 18 Oct 2000, Fabien Penso wrote:
>
> Ok found it also, use:
>
> analog +F 001016 +T 001006 on command line.
>
No spaces between +F and 0, and +T and 0.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road,
es the opposite case. I thought people might be interested in it.
http://www.metasystema.org/reply-to-useful.mhtml
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
"The new operating system will rec
ILE? This is all spelled out in gory detail at the top of
anlgform.html.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
"The new operating system will recover more easily from system crashes.&qu
On Sat, 21 Oct 2000, Carlos Santos wrote:
>
> I've tried it. It doesn't work :-<
> I guess i'll have to make lots of LOGFILE commands, one for each file.
>
Only one for each folder, in fact.
--
Stephen Turner http://www.statslab.cam.ac.uk/~s
_log (! -- you can even use Analog for that!)
>
-- Look in the HTML output. You'll see the logo being included at the top.
Make sure it's in the right place.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridg
isplays. Can this
>be done? If so, what would the config file command look like?
>
The easiest way is to restrict the whole report to the time period you want
using FROM and TO. Then you have a list in the Request Report of the number
of requests in that period for all files.
--
Stephen
On Tue, 24 Oct 2000, Dive Centre Manly wrote:
> Just checking to see why the program was started by us 16th October, and
> have not been able to run since that day.
You have to run it again each time you want to generate a report.
--
Stephen Turner http://www.statslab.cam
last command is 1.
>
>
ALIGNMENT is defined in anlghea2.h. Have you edited that file?
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
"The new operating system will recover more easi
gt;
> Could it be that I am looking in the wrong place? Is there a
> different reporting option I need to use which I have not enabled?
>
What status code is returned for /myscript?x=1 ? I'm thinking it could be in
the Redirection Report not the Request Report.
--
Stephen Turner
On Fri, 27 Oct 2000, Tom Messmer wrote:
> Yes, I have both BROWSER and FULLBROWSER reports set
> up
> I've used both default and all the BROWALIASES to no
> avail.
Which version of analog are you using? The Browser Summary improved a lot in
version 3.9.
--
Stephen Turner
On Thu, 26 Oct 2000, Glenn Murray wrote:
>
> I tried hacking anlghea2.h after the make choked, but not before.
And did you get the same error message before that? I can only suggest you
download a clean copy of analog. ALIGNMENT really is defined!
--
Stephen Turner
;
> 450: [not listed: 23 URLs]
>
> Is there something I can do to see what those 450 requests are and what the 23 URLs
>are?
>
REFINCLUDE *
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB,
NMENT is already defined on your OS,
so that this code is not getting included, although no other AIX user has
ever mentioned this, and I don't see how it would give the error you are
seeing.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory,
This all sounds very odd. You might like to put SETTINGS ON in your
configuration file. Then instead of running, analog will say what it thinks
it's supposed to be doing.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Camb
On Mon, 30 Oct 2000, Glenn Murray wrote:
>
> This seemed to do the trick! I received many
> 'anlghea2.h:35: warning: `ALIGNMENT' redefined'
> warning messages, but got an executable.
>
Hmmm. I wonder where ALIGNMENT was defined before then. Has anyone else
x27;s below any other REFINCLUDE or REFEXCLUDE commands.
You can also run with SETTINGS ON to see what analog thinks it's meant to be
doing.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
"The
I guess the thing to do is to look in the logfile directly. Find a specific
host which only requested an image. See if it got included in the Host
Report before the FILEEXCLUDE and excluded from the Host Report afterwards.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1
So, if
you've got a Host Report; or if you've got a Domain Report or Organisation
Report and not HOSTLOWMEM 3.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
"The new operating sys
file included from anlghead.h:87,
> from anlghea3.h:271,
> from analog.c:9:
>
Hmmm. Could you send me that socket.h file (privately to
[EMAIL PROTECTED] ). Thanks,
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical L
r.asp?AREA=homepage
> 21: : 10/Oct/00 20:21: /adbanner/adbanner.asp?AREA=girls
Your first thought was correct, in fact. The top line is the total of the
rest. The reason it doesn't add up is because you are only seeing the lines
with at least 20 request. You need to set the R
the log:
>
> Mozilla/4.0 (compatible; MSIE 5.01; Windows NT 5.0)
>
This doesn't look like what analog produces by default. Are you using some
BROWALIASes? If so, remove them.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberf
have been getting lots of compilation warnings in
previous versions, even if it compiled and ran. Why are people's
expectations so low that no-one mentioned it???
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0
file that gives output
looking like that and I'll have a look? Thanks.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
"The new operating system will recover more e
at it's obvious where it begins and ends. Or alternatively, put the
browser last on the line -- then it will extend all the way to the end of
the line.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, C
On Tue, 31 Oct 2000, John Kelly wrote:
> ---- Stephen Turner <[EMAIL PROTECTED]> wrote:
>
> > I guess the thing to do is to look in the
> > logfile directly. Find a specific host
> > which only requested an image. See if it got
> > included in the Host R
he rpm. Then there's a way of installing it,
but I forget what it is. rpm -i analog-4.13-1.i386.rpm maybe. I use Debian
myself.
If you prefer to grab the source code and compile it yourself, you can, but
it's not necessary.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1
never delimit a field with a character which might occur within
that field.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
"The new operating system will recover more easily from system crashes.&q
On Wed, 1 Nov 2000 [EMAIL PROTECTED] wrote:
> I need help configuring and installing analog. Is this the correct list to
> request help?
>
Yes.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB
gt;
You really need to learn regexps before trying to use them. :)
Replace all instances of \.* by \..* in your examples and it will work.
Explanation left as an exercise to the reader.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilber
On Mon, 6 Nov 2000, Stephen Turner wrote:
> On Sat, 4 Nov 2000, Massimo Mezzini wrote:
> >
> > HOSTALIAS REGEXP:^216\.[0-5]\.*\.* digex.net
> >
> > should be able to catch all the IP in the range
> > 216.0.0.0 - 216.5.255.255 and nothing else.
> >
>
hardly a tutorial. Chapter 7 (I think) of "Learning Perl" is
very good, of course. I don't know about online tutorials, but there must be
lots.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, E
(for help on all errors and warnings, see docs/errors.html)")
>
>
Which server are you using? It looks like the server may be incorrectly
sending error messges (stderr) to the browser.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Labo
18,842: 0.32%: Browne
17,327: 0.29%: Buchanan
4,278: 0.07%: Philips
2,287: 0.04%: Hagelin
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
"The new operating system will recover more
d ones
have already been included under the unindented one which precedes them.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
"The new operating system will recover more easily from system crashes.
w return you to our regularly scheduled programming.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
"The new operating system will recover more easily from system crashes."
gt; 302ISSUCCESS http://my.site/my/script.php*
>
> Now, I have to preprocess log files and replace "302" to "304" or
> "200" for these URLs but I think it is very ugly hack.
>
This is in the FAQ.
--
Stephen Turner http://www.statslab.cam.ac.
On Fri, 10 Nov 2000, Piotr Roszatycki wrote:
> On Fri, 10 Nov 2000, Stephen Turner wrote:
> > > Now, I have to preprocess log files and replace "302" to "304" or
> > > "200" for these URLs but I think it is very ugly hack.
> > >
&
sions 4.11 and later. Paul didn't
say which version he was using. But if it's a later one than that, it may be
that the ASP's returned a redirect or error status code instead of a
success.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laborato
g the logfiles.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
"The new operating system will recover more easily from system crashes."
(Microsoft, aiming high
t;
I can't understand this. I've just tested it and it works fine for me. Are
you saying that you're getting a Browser Report but not an OS Report?
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0W
nparseable". We can't tell if they're successful or
unwanted or what until we've tried to parse them.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
"The new operating system wil
tand it.
>
REQFLOOR 100r
Alternatively REQFLOOR -50r to list the first 50.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB3 0WB, England
"The new operating system will recov
of week 2, 1 copy of week 3 and 1 copy of week 4. And it gets worse
after that. :)
Analog should have warned you that you were analysing overlapping files
though. Anyway, the correct procedure is outlined in docs/cache.html.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sr
ure how to distinguish the two cases though. It's not been a
problem before I introduced regexps, because commas don't normally occur in
names.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge, CB
#x27;s not much I can do. Analog is including
the right things, it's just that somehow your system is fetching broken
versions of them.
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret1/
Statistical Laboratory, Wilberforce Road, Cambridge
On Fri, 10 Nov 2000, Piotr Roszatycki wrote:
>
> 302INCLUDE (*item | "")
> 302EXCLUDE (*item | "")
> or something like that.
>
Well, I'll add it to the wishlist and consider it further. But I won't
promise what I'll decide.
--
Stephen Turner
ILEEXCLUDE *
FILEINCLUDE /dir/*
should work fine. Actually, you don't even need the FILEEXCLUDE * -- the
FILEINCLUDE implies that everything else is excluded (as long as there's no
previous FILEINCLUDE or FILEEXCLUDE).
--
Stephen Turner http://www.statslab.cam.ac.uk/~sret
1 - 100 of 2515 matches
Mail list logo
