open the results of a find command in a browser

[GrouchyGaijin]

Hi Guys,

OK - that is a pretty bad title for a post.
I don't really have the proper vocabulary to explain what I mean, but here it goes.

I have all of the man files on my system converted to html files in a directory in my home.

If I run:

#!/bin/bash
read -p "Searching for: " search_string; find /home/grouchygaijin/man-pages-html -iname "$search_string" 2>/dev/null

It works.
The script brings up the path to the html file(s) that matches the search.
Now, I could just be happy and copy the path to the file then paste it into the browser, but I thought it would be much cooler to be able to have the browser open up the file when I click on it.

The question is:
How can I tell the script to let me choose to open the search results (path to the file found) in a browser.

I really appreciate the help!

[Vaphell]

xdg-open "$file" to open with program associated with the file type or simply browser_name "$file"


#!/bin/bash

dir="$HOME/man-pages-html"
read -p "Searching for: " search_string
readarray -t results < <( find "${dir}" -iname "*${search_string}*.html" )

if (( ${#results[@]} == 1 ))
then
xdg-open "${results[@]}" # open the only result immediately
elif (( ${#results[@]} > 1 ))
then
select s in "${results[@]/#$dir\//}"; do break; done # allow user to select file from the list of results
xdg-open "$dir/$s"
else
echo "no files found, sorry"
fi

[GrouchyGaijin]

Wow!
THANK YOU!
Now I need to figure out what you did.

[Vaphell]

readarray line captures the output of find and stores it in the results array, each line is a separate element, ${results[0]} - 1st element of the array, ${results[1]} = 2nd, etc.
${array[@]} = all elements.
if you want to see the contents of the array you can use something like this:

printf "'%s'\n" "${results[@]}"

if the array length ( ${#results[@]} ) is 1 then simply run xdg-open <results>

if it's greater than one: create menu populated with elements of results[] using select. It's a potentially infinite loop in case one wants to put repetitive tasks in it, so break is needed to jump out right after the choice. Whatever you have chosen is opened with xdg-open.
Array elements store full paths but in menu i trimmed them a bit for aestetic purposes: ${results[@]/#$dir\//} = elements of results[] with $dir path removed. $dir is glued back with selected value in the next line.

to illustrate:

$ prefix=ABC
$ x=ABCsomething
$ echo $x:${x/#$prefix/}
ABCsomething:something
$ array=( "ABC111" "ABC222" "ABC333" )
$ echo ${#array[@]}
3
$ printf "'%s'\n" "${array[@]}"
'ABC111'
'ABC222'
'ABC333'
$ printf "'%s'\n" "${array[@]/#$prefix/}"
'111'
'222'
'333'
$ select s in "A" "B" "C"; do echo "you selected '$s'"; break; done
1) A
2) B
3) C
#? 3
you selected 'C'
$ select s in "${array[@]/#$prefix/}"; do echo "you selected '$s'"; break; done
1) 111
2) 222
3) 333
#? 2
you selected '222'

[schragge]

@OP
Just wondering why using http://manpages.ubuntu.com or http://manpages.debian.net is not an option?

[GrouchyGaijin]

@OP
Just wondering why using http://manpages.ubuntu.com or http://manpages.debian.net is not an option?

Well, I suppose they are options. They sound like better options actually, I just didn't think of it.
(I was thinking to myself that I'd probably need to update my directory of man pages if I install a new program.)
How can I modify this script to use the manpages.ubuntu.com link instead of my local directory.

[schragge]

The fourth link under Navigating the Repository on manpages.ubuntu.com is Download the 'dman' shell script (http://manpages.ubuntu.com/dman) to run on a command line. The link before that is Install the search engine plugin for your browser.

The manpages on manpages.ubuntu.com are just static HTML files, so e.g. dpkg(1) (http://manpages.ubuntu.com/manpages/precise/en/man1/dpkg.1.html) in English for precise is

http://manpages.ubuntu.com/manpages/precise/en/man1/dpkg.1.html

[GrouchyGaijin]

Cool - thanks man!