ray@vantage.UUCP (Ray Liere) (03/27/89)
We are considering starting an archive for comp.sources.unix and
comp.sources.misc postings.
I have been very interested in the recent discussions in this group
about various archiving approaches.
As a first cut, I plan on writing a script that will break apart the BODY.TEXT
file into the individual postings.
A question comes to mind: how typically does one organize subsequent
related bug reports, patches, comments on experiences, discussions, etc.
that are posted in the comp.sources.bugs and comp.sources.d groups?
It would be nice to be able to "mate" applicable postings in these
groups to the originally posted source in some automated manner ...
Any pointers would be very much appreciated.
Thanks.
Ray Liere
mail: Vantage Consulting and Research Corporation
1017 Molalla Avenue, Suite 3
Oregon City, Oregon 97045, USA
voice: (503)657-7294
uucp: hplabs!hpfcla!hpubvwa!vantage!raywyle@inf.ethz.ch (Mitchell Wyle) (03/31/89)
In article <17900001@vantage.UUCP> ray@vantage.UUCP (Ray Liere) writes: >We are considering starting an archive for comp.sources.unix and >comp.sources.misc postings. Other people have been doing it for a long time. >As a first cut, I plan on writing a script that will break apart the BODY.TEXT >file into the individual postings. Please post your script. >related bug reports, patches, comments on experiences, discussions, etc. >that are posted in the comp.sources.bugs and comp.sources.d groups? Very often. I apply patches directly from rn and mush. I ignore bug reports unless we've already been bitten by them. >It would be nice to be able to "mate" applicable postings in these >groups to the originally posted source in some automated manner ... I'm not sure you want to take the human out of the loop to separate the appropriate bugs/patches from meaningless ones. Do you need postscript stuff for A4 paper? Do you use MMDF as your mail transport? Do you care about code that extracts text from X.400 p2 format? You need a human filter to separate the men from the chaff, the wheat from the boys... >Any pointers would be very much appreciated. Check out narc; simple shell scripts based on find(1)ing files in a news source spool directory which arrived after a touch(1) file was accessed are as efficient as C code. Also look at sending expire(8)ed source articles into your script. BTW: A neat way to have a script re-schedule itself (when you don't have access to crontab or Sun OS 4.0) is: case `date +%a` in Mon) day=Tue ;; Tue) day=Wed ;; Wed) day=Thu ;; Thu) day=Fri ;; Fri) day=Sat ;; Sat) day=Sun ;; Sun) day=Mon ;; esac; at 0200 $day <your script> Why extract out the files from those carefully packed cshars R$ puts together? I find them much more convenient. -- -Mitchell F. Wyle wyle@ethz.uucp Institut fuer Informationssysteme wyle@inf.ethz.ch ETH Zentrum / 8092 Zurich, Switzerland +41 1 256 5237
guy@auspex.auspex.com (Guy Harris) (03/31/89)
>BTW: A neat way to have a script re-schedule itself (when you don't >have access to crontab or Sun OS 4.0) is: > >case `date +%a` in If you don't have access to "crontab" or to a system that provides per-user "crontabs" (it's not a Sun invention; it comes from the S5R2 "cron", so most S5 systems have it), you may well not have a "date" command that accepts format strings, either; for example, 4.xBSD, as distributed by Berkeley, has neither, at present. It is, of course, trivial to write a "day of the week" program that coughs up the current day of the week; the trick in question can be used once you've written such a program.