Jimineep

I'm trying to create a fairly simple bash script which will search a given site for all of the jpg images on the site...its not working though...any ideas?? Am I even using the right commands for this job!!!!? Or is it the piping?

# a friendly greeting

echo hello $USER, hope all is well

echo What is the website you wish to download your images from? Remember to give the full url including http://

read SITE

echo And what compression ratio would you like? from 0-100 I would recommend a value of 65 to get a balance of file size and quality

echo A lower number gives less quality but a smaller file size, A higher number gives greater quality but makes a bigger file.

# get the jpegs from the site

wget -O - $SITE

#filter out all of the jpeg files

grep -O - -f "\.jpg$" |

#compress the jpeg file to make it smaller (although lower quality) and put it in a directory

| cjpeg - -quality $COMPR > ~/compressed/

[/code]

cheers x

mackenga

One problem is that wget -O - $SITE doesn't do what you're expecting it to. It would download the index page of the site and write the code to stdout.

Secondly, your pipelines are wrong. Some examples of how you can use pipelines:


1 uses cat to type out the password file, grep to find the line with my details, gzip to compress these (pointlessly; I was groping for a decent example!) and writes them to a file called foo in the current directory.

2 gets a list of all my running processes and displays on the terminal only those lines that match "netscape".

3 reads a file called words in the current directory and pipes them to sort (which orders them ascibetically I think by default) then to uniq (which removes duplicate lines, but assumes ordered input) and writes the result to /usr/dict/words.

Downloading all the JPG files from a site is also not really that simple a job I'm afraid. Even if you just want the images from the main page, you need to do some complicated things to parse out the URLs then fetch each of those (though I do think wget may be able to do that for you; investigate the -r option. I don't have the manpage in front of me but I have a good feeling about that).

Just to add: ah yes, http://www.lns.cornell.edu/public/CO...et_3.html#SEC7 should be useful for doing recursive retrievals using wget.

Jimineep

OK, scrap that, it is too complex for my first ever script!!

So i'm gonna try a simpler one... ure use of uniq inspired me

So could I use wgrep to search a specific website, and then use sort to take all of the words from this website, then arrange them in order and uniq write them to a specific dictionary:

something like this


I am a little unsure of the piping...do I not need to define STDOUT/STDIN?

I wont get to test anythign until tomorrow so if anyone is running linux I would be grateful if u could help me out and tell me if it works or if there are any glaring errors

I've gotta get my head round all this by september when I begin my masters in bioinformatics! I am supposed to be competent with the CLI :eek: