I’ve been really getting into developing my KickTweet project and set up a Twitter account for it at http://twitter.com/KickTweet. I wanted to feed in my subversion commits so I did some searching.

What I found was twitvn a monstrous (50-ish ‘real lines) Python script that sends commits to Twitter. I find that ridiculous. So here is my version, it drops right into the post-commit script and could be reduced to 2-3 active lines.

1
2
3
4
5
6

#!/bin/bash
REPOS="$1"
REV="$2"
 
TWEET="SVN Log (r$REV): $(svnlook log $REPOS -r $REV)"
curl -u KickTweet:mySecretPassword -d status="${TWEET:0:139}" http://twitter.com/statuses/update.xml

Note that the use of the ‘${TWEET:0:139}’ is a definite bashism, and not portable.

1
2
3
4
5
6
7
8
9
10
11

#!/bin/bash
REPOS="$1"
REV="$2"
 
TWEET="SVN Log (r$REV): $(svnlook log $REPOS -r $REV)"
 
if [ "${#TWEET}" -gt 140 ]; then
  TWEET="${TWEET:0:137}..."
fi
 
curl -u KickTweet:mySecretPassword -d status="${TWEET:0:139}" http://twitter.com/statuses/update.xml

I had a bad morning today. I was working on a script that parses out log files and ran it from my desktop. It’s supposed to change to a specified directory and check for, then delete, any logs that are over a month old. Unfortunately I didn’t have that directory on my machine, and I didn’t exit the script after a failed directory change. Poof, there go my desktop documents.

As an impulse reaction I looked for something to replace ‘rm’, and found safeRM. I was not impressed. Why have a ‘dustbin’ folder if we already have that built into the desktop environment, i.e. the Trash can?

Here is my own brief script that (on openSuSE) moves things to the trash. Note that it is not set up for moving directories and does not emulate ‘rm’ entirely.

1
2
3
4
5
6
7
8
9
10
11
12
13

#!/bin/bash
 
TRASHDIR=~/.local/share/Trash
X=1
NAME=$(basename $1)
while [ -f $TRASHDIR/$NAME ]; do
        NAME=${NAME}_$X
        X=$(($X + 1))
done
mv $NAME $TRASHDIR/files/
echo "[Trash Info]" > $TRASHDIR/info/$NAME.trashinfo
echo "Path=$PWD/$1" >> $TRASHDIR/info/$NAME.trashinfo
echo "DeletionDate=$(date +%Y-%m-%dT%H:%M:%S)" >> $TRASHDIR/info/$NAME.trashinfo

By the time I got this far I decided that this wasn’t worth pursuing. I’ll just try to be more careful and will absolutely run backups. I thought I’d publish it anyway just in case someone wants a safeRM alternative or to be able to put things in the trash from the command line.

Here’s a modification of a bash script I made a while back to convert anything that mplayer can play into an mp3 file. Change bitrates and the like to taste.

1
2
3
4
5
6
7
8
9
10
11

#!/bin/bash
#
# any2mp3
#
for i in *.$1; do
mplayer -vc null -vo null -ao pcm:fast "$i" -ao pcm:file="${i%.$1}.wav"
done
for i in *.wav; do
lame -h -V2 --vbr-new "$i" "${i%.wav}.mp3"
done
rm *.wav

Say you are in a directory of m4a files, just run any2mp3.sh m4a and wait.

Do you ever spend 20 minutes figuring out the perfect way to do some arcane task on the command line, only to forget it and need it again in a month? I do. A lot.

I started keeping a directory with little plain text documents that had the command line I wanted inside them. I recently uploaded them to my static site and wrote a script to use them easier.

It’s real simple, just download the howto bash script in the following directory http://static.velvetcache.org/howtos/ and place it in your path. Now make a directory for your HowTo’s somewhere and put the whole path into a file in your home directory called “.howto”.

You should be able to call $ howto –sync and it will pull down a tar of the most current files, then expand it into your HowTo directory.

If you download the howto.tar.gz by itself, be warned that it is a tar bomb and will rudely scatter its files all over your directory.

Lastly, onceyou have some HowTo’s installed, just use $ howto -l to list out the possible choices, then $ howto print-unix-timestamp, replacing “print-unix-timestamp” with the name or number of the HowTo you want.

Below is the howto shell script, because I like including code in my posts whenever I can.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78

#!/bin/bash
 
function printUsage {
	echo "Usage: howto [-l | --list] file-name"
	exit 1
}
 
if [ $# -ne 1 ]; then
	printUsage
fi
 
if [ ! -f ~/.howto ]; then
	echo "No .howto file found in your home directory!"
	exit 1
fi
 
HOWTODIR=$(cat ~/.howto)
 
if [ "$1" == "--sync" ]; then
	LRECENT=$(cat $HOWTODIR/CURRENT)
	SRECENT=$(wget -O - http://static.velvetcache.org/howtos/CURRENT 2> /dev/null)
	if [ $? -ne 0 ]; then
		echo "Can't contact download server."
		exit 1
	fi
	if [ "$LRECENT" != "$SRECENT" ]; then
		cd $HOWTODIR
		if [ $? -ne 0 ]; then
			echo "Can't change to the HowTo directory."
			exit 1
		fi
		wget -O howtos.tar.gz http://static.velvetcache.org/howtos/howtos.tar.gz
		if [ $? -ne 0 ]; then
			echo "Can't download new files."
			exit 1
		fi
		tar -zxf howtos.tar.gz
		if [ $? -ne 0 ] ; then
			echo "Can't untar the updates."
			exit 1
		fi
		rm -f howtos.tar.gz
		exit 0
	else
		echo 'Nothing new to download.'
		exit 0
	fi
fi
 
if [ "$1" == "--list" ] || [ "$1" == "-l" ]; then
	COUNTER=0
	for howto in `ls $HOWTODIR`; do 
		echo "$COUNTER $howto"
		COUNTER=$(expr $COUNTER + 1)
	done	
	exit 0
fi
 
echo $1 | grep '^[0-9][0-9]*$' > /dev/null 2>&1
if [ $? -eq 0 ]; then
	COUNTER=0
	for howto in `ls $HOWTODIR`; do
		if [ $COUNTER -eq $1 ]; then
			cat $HOWTODIR/$howto
			exit 0
		fi
		COUNTER=$(expr $COUNTER + 1)
	done
	echo 'That HowTo was not found.'
	exit 1
elif [ ! -f "$HOWTODIR/$1" ]; then
	echo 'That HowTo was not found.'
	exit 1
fi
 
cat $HOWTODIR/$1
 
exit 0

I’m releasing version 0.6 of Nix Monitor today. It’s gotten a bit bigger and harder to configure, so I wrote a kickin configurator script using Zenity for all my prompts. Check it out at http://kde-look.org/content/show.php/Nix+Monitor?content=67399