@guy038 Sorry for being unclear! Sometimes I read my older posts and wonder how people managed to find any sense in them. That’s why I post so rarely…

thank you

also,

a(\W+\w+){1,6}\W+for

Hello @ph0enix, @Scott-sumner and All,

if the general template of your data is always :

........@..........;......

Here are, below, 3 other possible regex S/R. Similarly to Scott’s post, you must select the Regular expression search mode, tick the Wrap around option and click on the Replace All button. Note that the Replace with: zone is always Empty

First regex :

SEARCH (?-s)@.+

REPLACE Empty

Results :

123.456.78.0:22000 12.3.456.789:10000

Second regex :

SEARCH (?-s).+?@|;.+

REPLACE Empty

Results :

Name\Name123 NAME\Name-1

Third regex :

SEARCH (?-s).+;

REPLACE Empty

Results :

name Name2017

Best Regards,

guy038

Scott, Thank you. I knew there would be a way and I almost knew it would be something in a regular expression, but only gods like yourself would know what to do. I really like the idea of a delimeter and then using the search to match any above and below the line.

I did not get the chance to try this (although I can’t see why it wouldn’t work) as I thought of an alternative, which was to use Excel and a formula. I managed to open each of my 5 files in one workbook with seperate tabs and use a formula to look at a control tab to find any matches. I think it’s pretty much the same thing as your suggestion, just using vba and not regex.

I thank you for your efforts

@Allan-Biggs

Try pressing Alt+Home in your macro where you are pressing Home.

@C-Ananda-Rao

Well, when I asked for data, I guess I should have asked for an explanation of that data to accompany it… I suppose I assumed that the phone number was the last part of every line where digits were present. Anyway, hopefully @guy038’s response provides the help you needed.

Well @John-Drachenberg I’d try and grab the records in the group of 5 lines from the date added to the url line, combining them all into 1 line (so replacing CR/LF) with some other delimiter. I’d then create a "key at the start of the line, possibly the main part of the url, excluding any /folder names. Sort all the lines so that it would easily match up possible duplicate urls. To my mind a duplicate is any url where even if the /folder names/ portion was different would warrant further inspection.

At this point either eyeball the duplicates, or another regex could mark possible duplicates for further inspection.

Not sure if you actually want a regex to just remove duplicates in the original file, or would be happy just getting a list of possible duplicates which you could then check against the original and remove manually.

Terry

@Alain-Torrens

Settings (menu) -> Preferences… -> Editing (box on left) -> Caret Settings -> Width can get you a much more visible caret/cursor…but not everything you’ve requested.

@Rasmus-van-moortel

can someone please help me?

Nope, not without more detail than that…hint: show some before and (desired) after data.

@Terry-R said:

Note that I have used the “” character before some,

I think this is supposed to read:

Note that I have used the “\” character before some,

https://github.com/pnedev/compare-plugin/releases

Thanks for the heads up. I hadn’t thought of that.

Maybe a “Recent News” thread? :)

Ok, I see you found it by yourself. Note that \d is equivalent to [0-9]. But you need to escape the "." to "\." or else it will match any character e.g. “0n234” would also match.

@Jose-Emilio-Osorio said:

@Scott-Sumner Use the mouse is good. But, I have more than 85000 lines in a file and it´s hard to drag the line to the last line. Any method to do this ?. Thanks for your help.

If it’s a one-shot deal then use two search/replaces in regular expression mode (Ctrl-G in the search/replace box)

Search for: ^
Replace with: 00000000
If you have leading spaces then use search for: ^ *

That’ll put eight zeroes at the front of all lines

The second search/replace is:
Search for: ^[0]+([0-9]{8})$
Replace with: \1

That skips over excess leading zeroes, saves the last eight digits, and the replace part is the saved value.

Now do one more pass which is to search for ^0{8}
This searches for lines that start with eight zeroes.

If you find any then either

There were blank lines in the file and we now have 00000000 there. That may or may not be desirable One of the lines had a number with 9 or more digits and will now have something like 00000000123456789 One of the lines had something other than a number and now has 00000000~whatever was on the line~

If you need to do this a lot then use sed. Google for GNU sed. It’s free. In sed the search/replaces would look like this

sed -r -e "s/^/00000000/" -e "s/^0+([0-9]{8})$/\1/" input-file.txt >output-file.txt

Then to check for errors

sed -r -n -e "/^0{8}/ { = ; p }" output-file.txt

This scans and for every line found that starts with eight zeroes it outputs the line number and then the offending line.

Ok, good to know! Thanks again guys for all the help!