deplace a block to an other place

[Alan Kilborn]

@PeterJones

How did you make the little checkboxes in text? That’s pretty cool. I’ve read about some of the markdown used on this forum, but not all of it is clear to me. Not that I’m saying the checked-checkboxes are markdown…

I’m talking about this:
☑ Regular Expression

How would one do an empty checkbox as well as a checked one?

Thanks!

[PeterJones]

Copy/paste. :-) I have an always-open scratchpad in NPP that includes some often-used characters:

Checkboxes: `☐ ☑ ☒ ✓ ✗`

Windows’ charmap.exe > Character Set: Unicode > Group By: Unicode Subrange > Symbols & Dingbats has the checkboxes in the first row when I’m using Font: DejaVu Sans Mono.

I also like using http://www.fileformat.info/info/unicode/char/search.htm to find Unicode characters by name or idea (like “arrow” or “box” or “check”). (I’ve actually got it as one of my custom search engines in Firefox, with the keyword “unicode”, so I just have to type “unicode ballot box” in my address bar to get http://www.fileformat.info/info/unicode/char/search.htm?q=ballot+box&preview=entity.

[Pouemes44]

hello all
a lot of thanks for your answer
i am french and not perfectly fluent in english
thanks peterjones
i have tried
Find What = (<noscript.?>.?</noscript>)(.*)(^-=-=-= INSERT TEXT AFTER THIS LINE =-=-=-$)
Replace With = $1$2$3\r\n$1

my files are like this

1.html
<noscript class"text" lang=“example”>
line1
line2
line3
line4
</noscript>
some text…
etc…
etc
<img… and …style=“display:none;”></div>

2.html
<noscript class"text" lang=“otherexample”>
otherline1
otherline2
otherline3
otherline4
</noscript>
othersome text…
etc…
etc etc
<img… other and …style=“display:none;”></div>

i would to move (to move = déplacer in french) in each file all this <noscript… tag with the tag (generally inside there are 4 lines… can be 5)) i would to move it in each page just after style=“display:none;”></div>
and i would to do this for more than 400 files in knowing that all my noscript are differents in each page
warning; in each file i have 2 noscript tag one is just <noscript> the other is like <noscript class"text" lang=“example”>
its this second that i would to move

not easy :-)

[PeterJones]

What you said you tried for the search/replace didn’t quite match my examples, but that’s okay. I will customize it based on what you’ve given me:

• Find = (<noscript[^>]*?class=*["“][^"“”]*?["”][^>]*?lang=["“][^"“”]*?["”]>.*?</noscript>\R*)(.*?)(<img[^>]*?style=["“]display:none;["”]></div>)
• Replace = $2$3\r\n$1
• Assumptions/Changes
  • I used [>]*? to include various unknown attributes, assuming they don’t contain a >: if they do, this won’t work
  • I used expressions that were ambivalent about smart quotes “” and normal quotes "", because you switched back and forth. Personally, I doubt there are actual smart quotes for the attributes in your HTML, but this allows my expression to work on exactly the text that you put in your examples, or on text that I think is well-formed HTML
  • Similary, I used class=* to allow class="text" or class"text", because I believe class"text" is invalid HTML, but I wanted the expression to work on your example text, just in case you are processing invalid HTML.
  • I allowed anything (or, rather, any non-quote character) inside the class="text" and lang="example", so it’s not restricted to exactly text or example: I saw you had example and otherexample, so I wanted to make sure that text and othertext would both work, too.
  • Since there is a language barrier, I am assuming by “to move = déplacer” you mean “move it, completely deleting the block from the beginning, and instead putting it after the <img display:none;">...</div>”

So that should work manually on the two example files you gave me: when I copy/pasted from your post into two files, then ran the exact FIND and REPLACE listed above once per file, it successfully found the noscript and moved it after the img.

HOWEVER (and this is a big “however”), that won’t be fun for hundreds of files. I tried having NPP open with just the two example files, then I used the Replace All in All Opened Documents option, and it replaced it in both files. Then I copied the originals so I had 100 files, and I opened them ALL in Notepad++, and the “Replace All in ALl Opened Documents” still did what I wanted. So it might work if you just open all 400 hundred files, though there’s no guarantee: I might do it in smaller groups of 50-100 files, depending on how large your files are.

If you are going to be frequently making changes like this, but the rules are changing, I highly recommend learning a scripting language: I would have done this from the command line with Perl; many others here would have used Python or Lua. The benefit of the latter two is that there are plugins for automating Notepad++ using the Python Script Plugin or the LuaScript Plugin. @Claudia-Frank could proably write the Python Script to open each of the 400 files, apply my regex, and save and close them without much difficulty. Unfortunately, I am not a Python expert, so it would take me hours more than I have available to muddle through it. (It may even be that @go2to’s ecobyte-link provides such a scripting language, or a simple interface for applying a particular regex to a multitude of files.)

[Pouemes44]

a great thanks peterjones
perhaps the solution for me will be to joint each file x2 then remove the lines i dont want

in dos do you know how to join in one command each file of a directoty with itself
for one file its easy
copy name.htm +name.htm d:\temp\name.htm

time to sleep for me i shall search for this tomorrow
*do you know this french free texteditor http://www.gbesoft.fr/gbepad.php
its not bad, i use it in association wth notepad++

[Claudia Frank]

A script could look like this, so if you are interested in
a python script solution let me know and we can work it out.

# -*- coding: utf-8 -*-

import os
FIND_WHAT = u'FIND_WHAT_REGULAR_EXPRESSION'
REPLACE_WITH = u'REPLACE_WITH_EXPRESSION'

DIRECTORY_OF_FILES = u'THE_DIRECTORY'   # like c:\\temp\\directory - doublebackslash needed

os.chdir(DIRECTORY_OF_FILES)
list_of_files = [x for x in os.listdir('.') if x.endswith('.html')]
for file in list_of_files:
    notepad.open(file)
    editor.rereplace(FIND_WHAT, REPLACE_WITH)
    notepad.save()
    notepad.close()

This script looks for html files in a defined directory.
If your files are under certain different subdirectories we need to modify the code.

Cheers
Claudia

[guy038]

Hi, @kat75,

Here is my contribution, to your problem ;-)

Let’s suppose you have two files 1.html and 2.html, as below :

<!-- 1.html -->
Some text
before the
"noscript" tag
<noscript class"text" lang="example">
line1
line2
line3
line4
</noscript>
bla
bla
bla
blah
<img… and …style="display:none;"></div>
And some text,
located after
............
<noscript>
............

and

<!-- 2.html -->
bla
bla
<noscript>
bla

<noscript class"text" lang="otherexample">
Other line1
Other line2
Other line3
Other line4
Other line5
</noscript>
Other text…
etc…
etc…
etc…
<img… other and …style="display:none;"></div>
Other Text
..........
till the END of the file

• Move back, at the very beginning of current file ( Ctrl + Origin )

• Open the Replace dialog ( Ctrl + H )

• Select the Regular expression search mode

• In the Search what: box, type in the regex (?s)(<noscript class.+?</noscript>\R)(.+?style="display:none;"></div>\R)

• In the Replace with: box, type in the regex \2\1

• Click on the Replace All button

=> Your two html files should be modified, as below :

<!-- 1.html -->
Some text
before the
"noscript" tag
bla
bla
bla
blah
<img… and …style="display:none;"></div>
<noscript class"text" lang="example">
line1
line2
line3
line4
</noscript>
And some text,
located after
............
<noscript>
............

and

<!-- 2.html -->
bla
bla
<noscript>
bla

Other text…
etc…
etc…
etc…
<img… other and …style="display:none;"></div>
<noscript class"text" lang="otherexample">
Other line1
Other line2
Other line3
Other line4
Other line5
</noscript>
Other Text
..........
till the END of the file

I do hope that it’s the kind of Search/Replacement what you expect to !

---

Notes :

• I suppose that the ending tag </noscript> ends the current line, in your html files

• I suppose the the text <img………style="display:none;"></div> ends the current line, too

• As usual, the (?s) modifier means that a dot regex character represents any single character ( Standard or EOL chars )

• Then, the part <noscript class.+?</noscript>\R looks for any complete individual noscript block ( Note that the form .+? stands for the smallest non-null range of characters, between the two strings <noscript class and </noscript> )

• This noscript block, with the final EOL characters ( \R ), is stored as group 1, due to the enclosed parentheses

• Finally the part .+?style="display:none;"></div>\R searches for the smallest range of characters till the string style=“display:none;”></div>

• Again, this block of lines, with the final EOL characters ( \R ), is stored as group 2, due to the enclosed parentheses

• In replacement, we just rewrite these two groups, in a reverse order ! ( \2\1 )

---

Remarks :

• As previously said, for 400 files about, you should use the Replace in Files feature ( Ctrl + Shift + F )

• Group all your html files, which have to be modified, in a particular folder

• Do a BACKUP of this folder ( One never knows ! )

• Fill up the Filters and Directory boxes, too

• Click on the Replace in Files button, to perform a global replacement, on all your files

• IMPORTANT : Do NOT use the v7.4 or v7.4.1 version of N++. Indeed, with these last versions, there a bug, when multiple S/R are performed : a part of the files scanned are not changed -:((( Refer to :

https://notepad-plus-plus.org/community/topic/13865/replace-all-in-all-open-documents-problem-in-7-4-1/1

Cheers,

guy038

[Pouemes44]

hello
just awake
thanks peter, claudia and guy
guy its magic, seems to work fine, i didnt thank that it was so easily possible
i must know how it works and donwload a previous version of notepad++ i have v7.41
really fantastic
a great thanks

[glennfromiowa]

@kat75 Guy often performs magic with Regular Expressions. :) And I believe he speaks French also.

A good starting point for Regular Expressions is http://www.regular-expressions.info/

[Pouemes44]

thanks glen for the link
i shall try to learn regular expressions, there are really possibilities

[Pouemes44]

hello
i have a question
i am trying to understand how what has given to me guy works
(?s)(<noscript class.+?</noscript>\R)(.+?style=“display:none;”></div>\R)

if i try to get an other sequence between a a tag by example… i got nothing! what is wrong?

(?s)(<a href.+?</a>\R)

with an img tag it doesnt take the first “> why? :-) thanks
(?s)(<img class.+?”>\R)

[Claudia Frank]

@kat75

without having the data we cannot say for sure what went wrong.
So, in theory, the hyperlink regex would match everyhting within the tag if
closing tags is directly followed by an eol char (no space)
and the image regex would match starting with <img class followed by
anything until ”> appears, followed directly by an eol char.

Cheers
Claudia

[Pouemes44]

at the end it must be a eol character, ok i understand now

[Pouemes44]

i would to say thanks to all, i have well succeed to move my block and begin to understand a bit notepad++
is there a way to send private messages on notepad community, i dont find how

[Pouemes44]

Hello
do you think that its possible to move a block not to the bottom but to the top?
i dont succeed
i mean that in this example, how to move <img etc… and …style=“display:none;”></div> just before <noscript class"text" lang=“example”>

1.html
<noscript class"text" lang=“example”>
line1
line2
line3
line4
</noscript>
some text…
etc…
etc
<img… and …style=“display:none;”></div>

2.html
<noscript class"text" lang=“otherexample”>
otherline1
otherline2
otherline3
otherline4
</noscript>
othersome text…
etc…
etc etc
<img… other and …style=“display:none;”></div>

[guy038]

Hello, @pouemes44, and All,

Here is, below, the general method in order to switch two consecutive blocks of text, separated, or not, by some stuff text

So let’s start with the example text, below :

bla
bla
bla
Block 1
of
some
text
End block
bla bla
bla bla
bla bla
bla bla
Block 2
with some
other text
End block
bla bla bla
bla bla bla
bla bla bla

To correctly determine the limits of your block of text, whatever it is, you, necessarily, need to know :

• The location of the beginning of your block. In our example, I suppose that it’s the regex ^Block \d+

• The location of the end of your block. In our case, I suppose it’s the string End block

IMPORTANT : In our example, the second block has the same limits as Block 1, but it could, perfectly, have some other limit definitions !

So, from the above definitions, we can build the regex S/R, below :

SEARCH (?s)(^Block \d+.+?End block)(.*?)(^Block \d+.+?End block)

REPLACE \3\2\1

Notes :

• The leading (?s) modifier means that the special dot character represents, absolutely, any single character. Now :

• The first part, (^Block \d+.+?End block, represents a complete individual block of text, stored as group 1

• The third part, (^Block \d+.+?End block), stands, again, for an other complete block of text, stored as group 3

• The middle part, (.*?), is the shortest range, even empty, of any character, between these two consecutive blocks of text

• In replacement, we just switch the group 1 and 3, with group 2 standing as a pivot

And, after a click on the Replace All button, you should get the expected text , below :

bla
bla
bla
Block 2
with some
other text
End block
bla bla
bla bla
bla bla
bla bla
Block 1
of
some
text
End block
bla bla bla
bla bla bla
bla bla bla

Just note that, if the two blocks are strictly consecutive, without the bla bla lines, as below :

bla
bla
bla
Block 1
of
some
text
End block
Block 2
with some
other text
End block
bla bla bla
bla bla bla
bla bla bla

We still get the correct modified text :

bla
bla
bla
Block 2
with some
other text
End block
Block 1
of
some
text
End block
bla bla bla
bla bla bla
bla bla bla

---

Now, Pouemes44, let’s apply this general method to your particular problem :

The general template, of your first block, is :

<noscript......
......
......
......
......
</noscript>

The general template, of your second block, is :

<img.........
.......
.......
.......
style="display:none;"></div>

Therefore, the corresponding regex S/R, in your case, is, obviously :

SEARCH (?s)(^<noscript.+?</noscript>)(.*?)(^<img.+?style="display:none;"></div>)

REPLACE \3\2\1

So, considering your example, below :

1.html
<noscript class"text" lang="example">
line1
line2
line3
line4
</noscript>
some text…
etc…
etc
<img… and …style="display:none;"></div>

2.html
<noscript class"text" lang="otherexample">
otherline1
otherline2
otherline3
otherline4
</noscript>
othersome text…
etc…
etc etc
<img… other and …style="display:none;"></div>

After performing the above S/R, you’ll get the modified text, as expected to :

1.html
<img… and …style="display:none;"></div>
some text…
etc…
etc
<noscript class"text" lang="example">
line1
line2
line3
line4
</noscript>

2.html
<img… other and …style="display:none;"></div>
othersome text…
etc…
etc etc
<noscript class"text" lang="otherexample">
otherline1
otherline2
otherline3
otherline4
</noscript>

Et voilà !!

Best Regards,

guy038

P.S. : Beware that, on our NodeBB site, text, containing starting and ending usual simple and double quotes, are changed into their Unicode equivalents, below :

• Starting simple quote ‘, of Unicode code-point \x{2018}, instead of the single quote sign ' ( x{0027 )

• Ending simple quote ’, of Unicode code-point \x{2019}, instead of the single quote sign ' ( x{0027 )

• Starting double quote “, of Unicode code-point \x{201C}, instead of the double quote sign " ( x{0022 )

• Ending double quote ”, of Unicode code-point \x{201D}, instead of the double quote sign " ( x{0022 )

Notes :

• For ANSI encoded texts, to avoid the nasty message Find: Invalid regular expression, the **correct regex syntaxes are, respectively, \x91, \x92, \x93 and \x94

BTW :

• Remember that the \x{####} regex syntax can be used :

  • For search of a true ASCII character, between \x{0000} and \x{007F}, in ANSI encoded files

  • For search of any Unicode character, between \x{0000} and \x{FFFF}, in UNICODE encoded files

• Remember that the \x{##} regex syntax can be used :

  • For search of a true ASCII character, between \x{00} and \x{7F}, in ANSI encoded files

  • For search of any Unicode character, between \x{00} and \x{FF}, in UNICODE encoded files

• To end with, remember that the \x## regex syntax can be used :

  • For search of any Unicode character, between \x00 and \xFF, either, in UNICODE or ANSI encoded files

[Pouemes44]

a great thanks guy
that was this part (.*?) i missed

[Pouemes44]

i have an other question
if i search (?s)(<h1.+?</h1>\R) i got a result in my file

but if i search (?s)(<a.+?</a>\R) to get <a href=“mypage.htm” title=page">page</a> i got nothing why? what is wrong?

[guy038]

Hello, @pouemes44,

• In the regex (?s)(^Block \d+.+?End block)(.*?)(^Block \d+.+?End block), of my previous post, the part (.*?) represents the shortest range of characters, even empty, stored as group 2, between the two consecutive blocks of text, that are to be swapped !

Now, if we consider the HTML example text, below :

<td height="15">
    <font size="2" color="black" face="arial, verdana"><b>Lire un message / dossier : Reçus</b></font> 
</td>

<td height="15">
    <font size="2" color="black" face="arial, verdana"><b>Lire un message / dossier : Reçus</b></font> 
</td>

Beware about the two different behaviours :

• The regex (?s)<td.+</td> looks for the largest range of characters .+, between the strings <td and </td> => It matches all the text, at once

• The regex (?s)<td.+?</td> looks for the shortest range of characters .+?, between the strings <td and </td> => it matches, successively, each block <td…</td>

---

Now, regarding your HTML text :

<a href=“mypage.htm” title=page">page</a>

I suppose that you do NOT get a match, using the regex (?s)(<a.+?</a>\R), because, probably, it’s the last line of your current file, which is NOT followed by any End of line character !

Indeed, in that case, the ending part \R cannot match anything. So the overall match fails :-((

Two solutions :

• Use the regex (?s)(<a.+?</a>\R?). With that syntax, the \R part is optional

• Add a line break to this last line !

For a better visualization of the End of line characters, just click on the ¶ icone

Cheers,

guy038

[Pouemes44]

Hello Guy
i just see your answer
a great thanks, yes work with (?s)(<a.+?</a>\R?)
a real thanks for all your explanations, i already succeed to do wonderful things with them.