I want to remove certain range. Please help me

[Johnnychangprogrammer]

Hi Everyone!

I want to remove certain range.

I have a lot of files.

The examples are below.

Line 146 of tommy.html: <td width=“150” id=“content_1” style=“background-color:#F0FCFF;”>123456-7890123</td>
Line 146 of jimmy.html: <td width=“150” id=“content_1” style=“background-color:#F0FCFF;”>163456-7875865</td>
Line 146 of michael.html: <td width=“150” id=“content_1” style=“background-color:#F0FCFF;”>164689-8965698</td>
Line 146 of kimberly.html: <td width=“150” id=“content_1” style=“background-color:#F0FCFF;”>748956-8971457</td>
.
.
.
.
.

So I want to change 13 digit numbers to 7 digit numbers.

Line 146 of tommy.html: <td width=“150” id=“content_1” style=“background-color:#F0FCFF;”>123456-7******</td>
Line 146 of jimmy.html: <td width=“150” id=“content_1” style=“background-color:#F0FCFF;”>163456-7******</td>
Line 146 of michael.html: <td width=“150” id=“content_1” style=“background-color:#F0FCFF;”>164689-8******</td>
Line 146 of kimberly.html: <td width=“150” id=“content_1” style=“background-color:#F0FCFF;”>748956-8******</td>
.
.
.
.
.

Have a good day!

[Claudia Frank]

@Johnnychangprogrammer

Some questions to clarify the real data.
Is it always line 146? In each of the files?
Is always <td width=“150” id=“content_1” style=“background-color:#F0FCFF;”> in front of the 13 digit number?
You wrote you want to have 7 digits but your example shows 7 digits and 6 times asteriks.
Which version do you want to have?

If the layout is not consistent, is the 13 digit number unique?

Cheers
Claudia

[Scott Sumner]

@Johnnychangprogrammer

One way might be:

Find what zone: (?<=\d{6}-\d)\d{6}(?=</td>$)
Replace with zone: ******
Search mode: Regular expression

[Johnnychangprogrammer]

@Claudia-Frank

Yes. It is always in line 146 each of the files.

[Johnnychangprogrammer]

@Scott-Sumner

Thanks. That’s great !

But other lines are also replaced.

I want to change the exact lines.

[Scott Sumner]

@Johnnychangprogrammer

Ah, I missed the “line 146” requirement the first time around…I thought it was part of the data, I suppose!

Maybe this:

FInd what zone: (?-s)((?:^.*\R){145}.*?-\d)\d{6}(?s)(</td>$.*)
Replace with zone: \1******\2

[guy038]

Hi, @johnnychangprogrammer, @scott-sumner, @claudia-frank and All,

UPDATE :

Do not trust in the two regex S/R, related in this post and refer to my second post, and to Scott explanations, below !

---

Scott, I think that we can, also, use the shortened regex S/R, below :

SEARCH (?-s)((?:^.*\R){145}.*?-\d)\d{6}

REPLACE \1******

Because, as the remaining of file, after line 146, must be preserved, the best way to do is to ignore all that part, isn’t it ?

---

@johnnychangprogrammer, as you will use the Find in Files dialog, which scan every file from beginning to end, performing a Replace All operation, you could, as well, use the third version, below :

SEARCH (?-s)(?:^.*\R){145}.*?-\d\K\d{6}

REPLACE ******

In case, you would use the Replace dialog, instead, with that third version, you should :

• Tick the Wrap around option

• Click, exclusively, on the Replace All button ( not the step by step Replace button )

Cheers,

guy038

[Scott Sumner]

@guy038 said:

the remaining of file, after line 146, must be preserved, the best way to do is to ignore all that part

Hi Guy,

The reason I used .* near the end of my regex was to protect against the possible case where the file is a lot longer than 146 lines and has matches farther on down in the file. Remember that when N++ finds/replaces a match, it starts looking again at the character position following the match/replacement–for more matches. Your first SEARCH suggestion fails in that regard–meaning that matches past line 146 will be changed, and this didn’t seem to be what the OP wanted.

[guy038]

Hello, Hi, @johnnychangprogrammer, @scott-sumner, @claudia-frank and All,

Ah, Scott, very clever, indeed ! My fault was to test my regexes, against a 200 lines text which doesn’t allow a second match, anyway :-((

So, obviously, my regexes are not reliable, in @johnnychangprogrammer’s case, and the Scott regex S/R is the unique correct solution !

Best Regards,

guy038

[Scott Sumner]

@guy038

HAha…I tested with {2} in place of {145}…much less work to create test data that way…I just changed it to 145 for posting!

:-D

Of course, if we could ever get the \A to work… :-(