I am trying to fetch contents of table from a wepage. I jsut need the contents but not the tags <tr></tr>. I don't even need "tr" or "td" just the content. for eg:
<td> I want only this </td>
<tr> and also this </tr>
<TABLE> only texts/numbers in between tags and not the tags. </TABLE>
also I would like to put the first column output like this in a new csv file column1,info1,info2,info3 coumn2,info1,info2,info3
I tried sed to deleted patters <tr> <td> but when I fetch table there are also other tags like <color> <span> etc. so I want is to delete all the tags; in short everything with < and > .
sed 's/<[^>]\+>//g' will strip all tags out, but you might want to replace them with a space so tags that are next to each other don't run together: <td>one</td><td>two</td> becoming: onetwo. So you could do sed 's/<[^>]\+>/ /g' so it would output one two (well, actually one two).
That said unless you need just the raw text, and it sounds like you are trying to do some transformations to the data after stripping the tags, a scripting language like Perl might be a more fitting tool to do this stuff with.
As mu is too short mentioned scraping HTML can be a bit dicey, using something that actually parses the HTML for you would be the best way to do this. PHPs DOM API is pretty good for these kinds of things.