[arch-general] script to reformat 'pacman -Ss' output into 2 readable columns (prevents blindness)

[David C. Rankin]

Guys,

	I developed a script to help me read the output of pacman -Ss in 2 nicely 
formatted columns. The output of 'pacman -Ss srchterm' drives me nuts trying to 
read down the package names and descriptions with all the tab indented and 
wrapped text -- so I fixed it. Download the script here:

http://www.3111skyline.com/dl/Archlinux/scripts/srch2list.sh

	The script takes this:

extra/ttf-khmer 5.0-2
     TTFont collection for Khmer (Cambodia)
extra/ttf-linux-libertine 4.4.1-3
     Serif (Libertine) and Sans Serif (Biolinum) OpenType fonts with large 
Unicode coverage
extra/ttf-mph-2b-damase 001.000.4.dfsg.2-2
     Super-Unicode TTFont covering full Plane 1, and the following scripts: 
Armenian, Buginese, Cherokee, Cypriot Syllabary, Cyrillic, Deseret, Georgian, 
Asomtavruli, Nuskhuri but no Mkhedruli, Glagolitic, Gothic, Greek, Hanunoo, 
Hebrew, Latin, Limbu, Linear B, Old Italic, Old Persian cuneiform, Osmanya, 
Shavian, Syloti Nagri, Tai Le, Thaana, Tifinagh, Ugaritic, Vietnamese
extra/ttf-ms-fonts 2.0-3
     Core TTF Fonts from Microsoft


	and turn it into this:

ttf-khmer                             TTFont collection for Khmer (Cambodia)
ttf-linux-libertine                   Serif (Libertine) and Sans Serif
                                       (Biolinum) OpenType fonts with large
                                       Unicode coverage
ttf-mph-2b-damase                     Super-Unicode TTFont covering full
                                       Plane 1, and the following scripts:
                                       Armenian, Buginese, Cherokee, Cypriot
                                       Syllabary, Cyrillic, Deseret, Georgian,
                                       Asomtavruli, Nuskhuri but no Mkhedruli,
                                       Glagolitic, Gothic, Greek, Hanunoo,
                                       Hebrew, Latin, Limbu, Linear B, Old
                                       Italic, Old Persian cuneiform, Osmanya,
                                       Shavian, Syloti Nagri, Tai Le, Thaana,
                                       Tifinagh, Ugaritic, Vietnamese
ttf-ms-fonts                          Core TTF Fonts from Microsoft


	or with the '-d' option for semi-doublespace:

ttf-khmer                             TTFont collection for Khmer (Cambodia)

ttf-linux-libertine                   Serif (Libertine) and Sans Serif
                                       (Biolinum) OpenType fonts with large
                                       Unicode coverage

ttf-mph-2b-damase                     Super-Unicode TTFont covering full
                                       Plane 1, and the following scripts:
                                       Armenian, Buginese, Cherokee, Cypriot
                                       Syllabary, Cyrillic, Deseret, Georgian,
                                       Asomtavruli, Nuskhuri but no Mkhedruli,
                                       Glagolitic, Gothic, Greek, Hanunoo,
                                       Hebrew, Latin, Limbu, Linear B, Old
                                       Italic, Old Persian cuneiform, Osmanya,
                                       Shavian, Syloti Nagri, Tai Le, Thaana,
                                       Tifinagh, Ugaritic, Vietnamese

ttf-ms-fonts                          Core TTF Fonts from Microsoft



	The script needs 1 input. That is a filename 'srchfile' that contains the 
output from 'pacman -Ss srchterm > srchfile' There are 2 options, '-h or --help' 
gives a brief description and '-d' which causes the output to be essentially 
double-spaced. (each package description remains single spaced if it requires 
multiple lines, but a blank line is included between package descriptions) On 
the todo list is having the script take the input from stdin and eliminate the 
srchfile.

NOTE: the '-d' option goes AFTER the filename.
(I'll make the cli input position independent in the next version)

	The script does some pretty cool stuff and has some of my favorite funtions for 
the bash addicts out there. It reads 'srchfile' and formats the output in two 
columns 'pkgname' and 'description'. The pkgname column width is adjusted to the 
longest filename and the description column (in multiple lines, if necessary) 
spans the remainder of the xterm width leaving a one-char margin on the right. 
To make reading easier the description is parsed before being output so that 
descriptions that span multiple lines only break on spaces and not words while 
staying withing the column width. The column widths are dynamic based on the 
terminal window size. (via obtaining the terminal width and height from stty 
output). (for some reason, $COLUMNS and $LINES are not available in my scripts?? 
environment? exports?)

	The script was fun to write as it incorporated a good many of the finer points 
of bash. It is also well commented so if you are learning bash, or want to 
learn, this script can be a useful tool for some of the cooler things that can 
be done with arrays and strings using index manipulation and substring 
parsing/parameter substitution.

	The ultimate goal was to make the pacman search output readable, and I think 
this makes reading the search results *much* easier, at least for me.

	I guess what I should do for the next version is just incorporate the pacman 
-Ss call into the script itself and simply read the output of the search 
directly into the input array instead of reading the input file into an array. 
Simple enough, but I'll have to put that out as the next revision.

	Give this a whirl and see what you think. Let me know what works and what 
doesn't and also how I do things better/cleaner than I have currently done them. 
Who knows, maybe we will look at 'zenity --list' and provide a gui output next 
time :p

	Enjoy.

-- 
David C. Rankin, J.D.,P.E.
Rankin Law Firm, PLLC
510 Ochiltree Street
Nacogdoches, Texas 75961
Telephone: (936) 715-9333
Facsimile: (936) 715-9339
www.rankinlawfirm.com