Delete duplicated expressions with keeping the first one

Ziad Aborami · Mar 2, 2019, 2:36 PM

@Alan-Kilborn @PeterJones
Thanks so much, the python method works well!

@PeterJones
I was running the python from cmd with python.exe
I have created a new script and save it in Notepad++ scripts
When I have run it from scripts in Notepad++, it has worked well in 2 seconds without lagging.
Finally the problem was solved!

Ekopalypse · Mar 2, 2019, 3:32 PM

It is only needed to import from Npp if the current file isn’t the main file executed.
Like if you create a module or package.

guy038 · Mar 3, 2019, 12:03 PM

Hi, @ziad-aborami, and All,

While waiting for your post, I noticed some errors in your file ! Not a lot : only 7 :-))

Here are the corrections, to be made to your new.txt file that you attach to your e-mail. These modifications are listed, in reverse order, in order to not change the numbering during the manual modifications !

First this simple change to respect the <tr class="content">........</tr> multi-lines block, for each entry
- Add a line <tr class="content">, before the first line of your file
- Save your file, which is, now, 350,579 lines long
Then, here are the 7 modifications to do :
- Add the line <tr class="content">, BEFORE the line 346,171 ( </> )
- Change the entire line 346,149 ( <tr class>“content”> ) into the line <tr class="content">
- Change the entire line 343,625 ( </> ) into the <tr class="content">
- Add the line </tr>, AFTER the line 250,404 ( <td class=“content” width=“50%” valign=“top”>Herabsetzung</td> )
- Add the two lines <tr class="content"> and </>, BEFORE the line 250,399 ( Rabaisssement )
- Add the line <tr class="content">, BEFORE the line 48,800 ( </> )
- Add the line </tr>, AFTER the line 48,799 ( </td> )

Your file should be, now, 350,585 lines long

If you count the number of its <tr class="content">........</tr> multi-lines blocks, with the regex, below :

SEARCXH (?-s)^<tr class="content">\R</>\R(?s).+?</tr>

You should get 38784 blocks or entries, in your German-German dictionary !

Although, I haven’t got any clue about the exact text that you would like to keep, let’s me give it a blind try !

I’m starting with the file being modified, as above !

Now :

Open the Replace dialog ( Ctrl + H )
SEARCH (?i-s)^(<tr class="content">\R</>\R(.+)\R(?:.+\R){4,70}?)</tr>\R<tr class="content">\R</>\R\2\R
REPLACE \1</>\r\n
Check the Wrap around option and/or move to the very beginning of your file
Select the Regular expression search mode
Click on the Replace All button, repeatedly, till the message Replace All: 0 occurrences were replaced occurs, at bottom of the dialog

Hence, the results :

After a 1ST click on the "Replace All" button  =>  10,583 replacements done, in 12s
After a 2ND click on the "Replace All" button  =>   4,119 replacements done, in 12s
After a 3RD click on the "Replace All" button  =>     912 replacements done, in 11s

But the 4th replacement was incorrect ! It said 2 replacements but, actually, only the first was OK, as the second wrongly selects all file contents :-((. So I went back, one step, with Ctrl + Z

Then I marked all remaining entries of dictionary and place them in a new tab. After detecting the duplicates entries and deleting the others, it remained 20 entries, in 2 or 3 copies, listed below :

Abischnitt			Abischnitt
Erneuerung			Erneuerung
großartig			großartig
hervorragend		hervorragend
Muster				Muster
Verbindung			Verbindung
vornehm				vornehm
Vorrang				Vorrang
Wiederherstellung	Wiederherstellung
Überbleibsel		Überbleibsel
Übereinstimmung		Übereinstimmung			Übereinstimmung
Überlegenheit		Überlegenheit
Überlegung			Überlegung
übereinstimmen		übereinstimmen			übereinstimmen
übereinstimmend		übereinstimmend			übereinstimmend
überlegen			überlegen				überlegen
übersinnlich		übersinnlich	
überspannt			überspannt
übertrieben			übertrieben
üppig				üppig

Finally, for each of these 20 entries, I simply changed the lines :

</td>
</tr>
<tr class="content">
</>
üppig
<div style="margin-left:1em"><span style="color:darkblue"><b>üppig</b></span> <i class="p" style="color:green"><n></i>

as below :

</td>
</>
<div style="margin-left:1em"><span style="color:darkblue"><b>üppig</b></span> <i class="p" style="color:green"><n></i>

Best Regards,

guy038

IMPORTANT :

My first attempt of the search regex, against the corrected file, was :

SEARCH (?-s)^(<tr class="content">\R</>\R(.+)\R(?s).+?)</tr>\R<tr class="content">\R</>\R\2\R

But it did not work at all ! It just modified 8 entries and wrongly selected all the file contents, at the end !

Then, I decided to omit the smallest range of characters (?s).+?) between the entry (.+)\R and the closing tag </tr> and to replace it with the syntax (?:.+\R)+?, which grabs the smallest range of non-empty lines, in a non-capturing group, giving the regex :

SEARCH (?-s)^(<tr class="content">\R</>\R(.+)\R(?:.+\R)+?)</tr>\R<tr class="content">\R</>\R\2\R

It was even worse ! This time, I just got 1 result : the whole file contents, again :-((

Then, I verified how many lines was located after an entry, till the nearest </tr> closing tag. And it happens that this initial range is between 4 and 11 lines. However, as the goal is to gather duplicates entries, after some replacements, it would increase the size of all the multi-lines areas <tr class="content">...........</tr> !

So I decided of a compromise, using the range {4,70}, in the search regex below, which detects and replaces, correctly, 99,9% of all the occurrences, with a manual modification of 20 entries, only :-))

SEARCH (?i-s)^(<tr class="content">\R</>\R(.+)\R(?:.+\R){4,70}?)</tr>\R<tr class="content">\R</>\R\2\R

The initial file, of 350585 lines long, which contained 38,784 entries, whose some of them were duplicates, now contains 23,146 uniques entries, for 303,669 lines, only !

Ziad Aborami · Mar 6, 2019, 6:47 PM

@guy038
Thanks so much!
I have fixed the errors.
The command of replacment worked fine!
Thanks a lot for your time
You are generous!