Bashscripts.org

Hi,

I have one very long text line like this:
<tr bgcolor='#FFFFFF'><td width='50%' class='un_subj'><a href=../result/1237.html>Waitec </a></td><td width='50%' class='un_subj'><a href=../result/455.html>EXP Computers </a></td></tr><tr bgcolor='#FFFFFF'><td width='50%' class='un_subj'><a href=../result/1242.html>Wearnes </a></td><td width='50%' class='un_subj'><a href=../result/475.html>Formosa </a></td></tr><tr bgcolor='#FFFFFF'><td width='50%' class='un_subj'>..........

I need to extract each /result/????.html in a new line into a file.

How do I do it?

Thanks

hi guy, I found a solution with sed and grep. Give it a try:

Suppose the long line is in a file named blah, then

So you can see what you want. If necessary, just pipe it to a text file with ">>file.txt" appended.

thanks a lot, that worked

For something with a few less moving parts, try this!



You may have to escape the slashes, I don't remember... grep -o is one of my favoritest commands ever for scripting stuff like this! Here's what I get for your sample line:



Hope this helps!
-J

wow, I use grep a lot but did not know that you can do this with it

thanks

thx jeo! That's fabulous!

another question, what if I want to extract everything between html> and </a> from the same line?
Example:
...<a href=../result/158.html>Atapi</a></td><td ...
after extraction I should see Atapi

Note: there are names using 2 words, using special characterslike () and '

Thanks in advance

Now you're getting fancy... I know there's a better way to do this, using regex and parentheses, but here's the first thing that came to mind...



If I figure out the right way to do it I'll post it later

EDIT: forgot to add sample output... This is on debian, so your results may vary...

odd, I've tried something similar but it did not want to work

it does not include the special character once but I can do those manully since there're not much

thanks a lot for you help

Doh! I forgot you mentioned special characters... Digging deeper! after work...

Okay, I still think there's a cleaner way to do it, but you can modify the one that I sent to include the characters that you expect. I added the parentheses () here:

[code]
grep -Eo 'l>[a-zA-Z() ]*<\/a' log.txt |sed 's/l>\|<\/a//g'
[code]

You should be able to add other special characters as well. I just hope you're not expecting any "<>" characters or else it will start matching everything between the first "l>" and the last "</a" on each line. Hope this helps!

-Jeo

that did it

thanks