Match between two strings of which one may occur often in python

Question

I get a large swab of output from pacman -Qi. This output is piped using subprocess and converted to a string using str.decode. This all works.
But then I would like to be able to cut a a block related to one package out of that large swap of output, like this:
Name : steam
Version : 1.0.0.25-3
URL : http://steampowered.com/
Licenses : custom
Groups : None
Provides : None
Depends On : bash desktop-file-utils hicolor-icon-theme curl dbus
freetype2 gdk-pixbuf2 ttf-font zenity lib32-libgl
lib32-gcc-libs lib32-libx11
Optional Deps : lib32-ati-dri: for open source ATI driver users
lib32-catalyst-utils: for AMD Catalyst users
lib32-intel-dri: for open source Intel driver users
lib32-nouveau-dri: for Nouveau users
lib32-nvidia-utils: for NVIDIA proprietary blob users
lib32-flashplugin: for flash video, copy to
$XDG_DATA_HOME/Steam/ubuntu12_32/plugins/
Required By : None
Conflicts With : None
Replaces : None
Installed Size : 6424.00 KiB
Packager : Daniel Wallace <danielwallace at gtmanfred dot com>
Architecture : x86_64
Build Date : Fri Feb 8 21:51:42 2013
Install Date : Mon Feb 11 11:35:09 2013
Install Reason : Explicitly installed
Install Script : Yes
Description : Digital distribution client - open beta - bootstrap package
The goal is to search the large swab of output for a package installed on a specific install date.
I've tried many kinds of pattern matching but I cant seem to wrap my head around it. Does anyone know a short, sweet, fast, clean way to do this?
Greetings
Last edited by welpert (2013-02-11 16:59:24)

Answer 1

#!/usr/bin/env python
# -*- coding: UTF-8 -*-
import re, sys
pacman_data = sys.stdin.readlines()
for line in pacman_data:
match = re.search( "^install date[ \t]*: (.*)", line, re.I )
if match: print( match.group( 1 ))