If we have a good link with a clean format such as http://www.sqlhtm.com/About.htm, we'll have absolutely no problems but if we're going to depend on google search for links, we might get lucky or unlucky depending on what the SERP[search engine results page] deigns to provide us. Look at the image above and consider the two cases below. First we have some results with a clean web address on the second line of the result below the top title line such as the one below. We can grab this just like it is and use it without changes.

EXAMPLE

Sed Command in Unix and Linux Examples
www.folkstalk.com/2012/01/sed-command-in-unix-examples.html‎


CASE2: MESSY LINK TO CLEAN UP FOR LYNX CLI USAGE

While an address such as the one above is what we want, oftentimes we run out of luck and end up having to rightclick on the title and select the [copy link
address] option to get a really messy google-formatted link with characters that
need to be cleaned up so the link is acceptable to the lynx command line deity.
To do this, we need to delete a portion at the beginning of the messy address,
delete another at the end, and also replace the code [%3A] with colon [:] and [%2F] with forward slash [/] in the reminder of the address.

EXAMPLE

Linux / Unix: sed Command Print Only Matching Lines - nixCraft
www.cyberciti.biz/.../unix-linux-sed-print-only-matching-lines-comman...‎


Do you notice the three dots from hell leading to nowhere in the middle and also at the end? This is no good. So we'll rightclick on the top title link and select the [copy link address] option to get an equally-lousy-if-not-worse address below.

Lynx CLI hates this address, throws an error message! Deciphering this address
is not rocket science but everyone can only have so much patience for things of
this sort, don't you agree? Let us keep moving and finish cleaning up this scary
mess as much as we can.

1.Delete from the beginning all the way upto [URL=].

2.Delete everything from &ei till the end of line.
ie. Delete this at the end of the address:


3.Replace [%3A] with [:] and [%2F] with forward slash[/] in the leftover address.
In this step, we start with address [http%3A%2F%2Fwww.cyberciti.biz%2Ffaq%2Funix-linux-sed-print-only-matching-lines-command%2F] and
end up with the address[http://www.cyberciti.biz/faq/unix-linux-sed-print-only-matching-lines-command]

There, we got what we needed, a really clean lynx CLI-acceptable address!
Using this address in any of the several ways above, we can get what we want - the html or text version of the page with or without the links on the page!

REDIRECTS

If the link we are using is a redirect, then we may not get the text of the
page at once, but only an alternate link in the output of the lynx command. We need to run the same lynx command again with this redirect link and then we will get the required text source of the page.

FINAL WORD

While a picture may be worth a thousand words, it is nice to be able to be content with words for a change, at least once in a while, at least considering the useless, bloated, unhealthy or harmful images on the net! But dude, I know where you are headed right after reading this - right back to your G+s, pinterests and tumblrs, but really can't blame you. Who'd want to stay with text only other than those stuck with dial-up or slow internet?

To search for the string "echo hello >" within the hworld1.log file, type

$grep "echo hello >" hworld1.log

and everything is fine. Check the image above.

Tested on Puppy Lucid 510. Puppy Linux allows both single and double quotes within the grep command. Check with your Unix/Linux platform for details on how things work with regard to single or double quotes.