A new script...

[Adam W.]

...to aid in quickly deleting spams encountered on groups. Useful to me,
after some modifications might be also useful to someone else.

I invoke with an email address of a spammer (optionally only part before
@gmail.com). It greps the overview database for posts from that spammer
(only tradindexed is supported), shows a summary (with subjects, MsgIDs,
and newsgroups) and, if not cancelled at this point, copies these articles
to some location (in my case it's a location from which reposts to local
spam groups and NoCeM reports are generated), cancels them locally, and
adds e-mail to the text file that's used by the inn filter to kill these
articles on entry.

One drawback: if an email of a spammer appears in other headers in the
overview (for example in the Subject line), such article will be cancelled
as well. That's one reason to review the summary before proceeding.

#!/bin/bash

[ $# = 1 ] || { echo "Syntax: spam.sh <email>"; exit 1; }

if [[ "$1" = *@* ]]; then
email="$1"
else
email="$1"@gmail.com
fi

# Some basic paths
base=/usr/local/news
overview="$base"/spool/overview
gh="$base"/bin/grephistory
sm="$base"/bin/sm
ci="$base"/bin/ctlinnd

# Directory to copy cancelled articles to. They're processed later to copy
# articles to chmurka.spam.* groups and create NoCeMs
copydir="$base"/spool/pyother

# File with list of e-mail messages belonging to known spammers
# (inspected by the filter)
spammers="$base"/etc/spammers.txt

tmpfile=$(mktemp)
trap 'rm -f "$tmpfile" "$tmpfile.overview" "$tmpfile.mids"' exit

# Takes some time, might think of something faster
echo "Scanning overview database..."
grep -r "$email" --include '*.DAT' "$overview" > "$tmpfile.overview"

echo
echo "List of subjects follows"
awk -F'\t' '{print $2}' < "$tmpfile.overview" | sort | uniq

echo
echo "List of Message-IDs follows"
awk -F'\t' '{print $5}' < "$tmpfile.overview" | sort | uniq > "$tmpfile.mids"
cat "$tmpfile.mids"

echo
echo "List of newsgroups follows"
sed -e 's/:.*$//' -e 's/^.*\///g' -e 's/\.DAT$//' "$tmpfile.overview" | sort | uniq

echo
echo "Press Enter to continue, Ctrl-C to abort"
read x

echo "$email" >> "$spammers"

for mid in $(cat "$tmpfile.mids"); do
# Let's get the token for the given Message-ID and validate if:
# - it's known (should be)
# - is not already cancelled

token=$("$gh" "$mid")
[ $? -eq 0 ] || { echo "$mid: token not found"; continue; }
[ "$token" = "/dev/null" ] && { echo "$mid: already cancelled"; continue; }

# Create a random filename to copy the article to, and copy it
filename="$copydir"/$(hexdump -C /dev/urandom | head -n 1 | cut -b 11-58 | sed 's/ //g')
"$sm" "$token" > "$filename.tmp" || { echo "$mid: could not retrieve"; rm -f "$filename.tmp"; continue; }
mv "$filename.tmp" "$filename.to-rewrite"

# Finally, let's cancel
"$ci" cancel "$mid"
done

[yamo']

Hi,
Adam W. a tapoté le 29/01/2024 03:54:

> ...to aid in quickly deleting spams encountered on groups. Useful to me,
> after some modifications might be also useful to someone else.

Could you generate a list like NoceM Notices?

<mid1> group1
<mid2> group1 group2
<mid3> group3
...

It will be also useful, and could be checked before sending it (and
NoceM may be faster has ctlinnd cancel).


--
Stéphane
UTILISATEURS de GOOGLE GROUPS, vous n'aurez bientôt plus accès à Usenet.
<https://support.google.com/groups/answer/11036538>
Des serveurs gratuits de remplacement : <http://usenet-fr.yakakwatik.org>
Des logiciels : <http://usenet-fr.yakakwatik.org/lecteurs-de-news.html>

[yamo']

Hi,
Adam W. a tapoté :

>
> # Finally, let's cancel
> "$ci" cancel "$mid"

Thanks a lot!
Very useful!

I've modified here by those lines to have a NoceM Notice :

NEWSGROUPS=$(egrep '^Newsgroups: ' "$filename.to-rewrite" | head -1 | awk '{print $2;}' | sed 's/,/ /g')

# Finally, let's cancel

#"$ci" cancel "$mid"
echo "$mid $NEWSGROUPS"

--
Stéphane