Delete everything after "Text"

steve1040 · Post by **steve1040** » Fri Oct 13, 2006 2:26 am

I'm trying to cleanup 100+ html files

How would I go about searching for
"</BODY></HTML>"
and delete everything starting at next line to the End of each file?

Thanks
Steve

Bob Hansen · Post by **Bob Hansen** » Fri Oct 13, 2006 3:25 am

Since that string is usually near the end of a file, I am not sure you can do that just with RegEx in TextPad.

How does this approach work for you?

Search for "</BODY></HTML>" and Mark All lines above including this line.
Search/Invert All Bookmarks
Edit/Delete/Bookmarked Lines

You might then be able to make a macro to do that for you.

steve1040 · Post by **steve1040** » Fri Oct 13, 2006 12:33 pm

I'm thinking I'll have to get a Unix account and do this there

ben_josephs · Post by **ben_josephs** » Fri Oct 13, 2006 3:42 pm

I don't see why you need a Unix account to do this.

You can readily do it using Helios's own WildEdit (http://www.textpad.com/products/wildedit/).

You can use Perl (for example, from http://www.activestate.com/Products/ActivePerl/), or your favourite scripting language, from a Windows command line.

If you want a Unix shell and Unix utilities, you can get bash and a vast collection of utilities from Cygwin (http://www.cygwin.com/).

steve1040 · Post by **steve1040** » Fri Oct 13, 2006 6:22 pm

ben_josephs wrote:I don't see why you need a Unix account to do this.

You can readily do it using Helios's own WildEdit (http://www.textpad.com/products/wildedit/).

I've downloaded wildedit - How would I perform the task in this application?

ben_josephs · Post by **ben_josephs** » Fri Oct 13, 2006 9:00 pm

Find what: </BODY></HTML>.*
Replace with: </BODY></HTML>\n

[X] Regular expression
[X] Replacement format

Options
[ ] '.' does not match a newline character [i.e., not selected]

Bob Hansen · Post by **Bob Hansen** » Fri Oct 13, 2006 11:30 pm

Hi ben_josephs.

Nice simple solution, well done, and thanks for letting us know that .* in WildEdit also includes \n codes.

If I am understanding you, this means that WildEdit has the ability to scan past multiple \n codes, correct?
I missed that ability somewhere, sure solves a lot of issues.
I limited my solution to TextPad alone.

Thanks again, for the technical kick.....

I will have to spend more time with WildEdit.

ben_josephs · Post by **ben_josephs** » Sat Oct 14, 2006 11:32 am

Yes. The Boost regular expression recogniser that WildEdit uses (http://www.boost.org/libs/regex/doc/index.html) treats the text it is searching as a sequence of characters, not as a sequence of lines. This makes it possible for the regex . (dot) to match any character, including newline (if you want it to). It also makes it possible for \n to have full rights as a regular expression that can be incorporated as a component in other regular expressions without restriction.

The regular expression . when the option '.' does not match a newline character is not selected
is equivalent to
the regular expression (.|\n) when the option '.' does not match a newline character is selected.