regex: Parsing html tags in other tags / links and titles

Robin Cruise · Mar 9, 2021, 2:43 PM

also, the same basic formula, for another part:

FIND: (title="(.*?)">.*?)(mr-1"></i> ).*?(<span>)

REPLACE BY: \1\3\2 \4

Check this 2 regex formulas I wrote in the last 2 posts, and see what has changed

guy038 · Mar 9, 2021, 7:46 PM

Hi, @Robin-cruise and All,

When I said, in my previous post :

More generally, you would like that any line, as below, in a <ul .............</ul> section :

¤¤¤¤<li><a href="https://my-website.com/••••.html" title="••••">••••</a></li>

is changed into :

<div class="categories-name">
¤¤¤¤<a href="https://my-website.com/••••.html" title="••••">
¤¤¤¤<p class="font-16 color-grey text-capitalize"><i class="fa fa-angle-right font-14 color-blue mr-1"></i> •••• <span>••••</span> </p>
¤¤¤¤</a>
</div>

This can be achieved with the following regex S/R, without any group :

SEARCH ^\h*<li><a href="https://my\-website.com/.+?\.html" title=".+?">.+?</a></li>

REPLACE <div class="categories-name">\r\n <a href="https://my-website.com/••••.html" title="••••">\r\n <p class="font-16 color-grey text-capitalize"><i class="fa fa-angle-right font-14 color-blue mr-1"></i>••••<span>••••</span> </p>\r\n </a>\r\n</div>

So, from any initial line, inside a <ul .............</ul> section, as below :

            <li><a href="https://my-website.com/••••.html" title="••••">••••</a></li>

You get these output lines :

<div class="categories-name">
    <a href="https://my-website.com/••••.html" title="••••">
    <p class="font-16 color-grey text-capitalize"><i class="fa fa-angle-right font-14 color-blue mr-1"></i>••••<span>••••</span> </p>
    </a>
</div>

Just try it out !

Now, the problem is HOW to perform this S/R, ONLY on lines inside these <ul> sections, below :

<ul id="myNavigation">
....
....
....
</ul>

To do that, we’ll use, again, the generic regex, already discussed, with :

BSR = <ul id="myNavigation">
ESR = (?!</ul>)
SR = ^\h*<li><a href="https://my\-website.com/.+?\.html" title=".+?">.+?</a></li>
RR = <div class="categories-name">\r\n <a href="https://my-website.com/••••.html" title="••••">\r\n <p class="font-16 color-grey text-capitalize"><i class="fa fa-angle-right font-14 color-blue mr-1"></i> •••• <span>••••</span> </p>\r\n </a>\r\n</div>

Leading to this general S/R :

SEARCH (?s-i)(<ul id="myNavigation">|(?!\A)\G)((?!</ul>).)*?\K^\h*<li><a href="https://my\-website.com/.+?\.html" title=".+?">.+?</a></li>

REPLACE <div class="categories-name">\r\n <a href="https://my-website.com/••••.html" title="••••">\r\n <p class="font-16 color-grey text-capitalize"><i class="fa fa-angle-right font-14 color-blue mr-1"></i>••••<span>••••</span> </p>\r\n </a>\r\n</div>

Notes :

You must use N++ v7.9.1 or a later release, which correctly handles the \A assertion !
In replacement, everything is literal, but the parts \r\n, standing for Windows EOL
Of course, AFTER the replacement, you’ll have to change any •••• zone with its exact contents !

Let’s try this final regex S/R :

Place this sample text, below, in a N++ new tab :

...
...
<ul id="myNavigation">
    <li><a href="https://my-website.com/••••.html" title="••••">••••</a></li>
    <li><a href="https://my-website.com/••••.html" title="••••">••••</a></li>
    <li><a href="https://my-website.com/••••.html" title="••••">••••</a></li>
....
....
....
    <li><a href="https://my-website.com/••••.html" title="••••">••••</a></li>
</ul>

    <li><a href="https://my-website.com/••••.html" title="••••">••••</a></li>
    <li><a href="https://my-website.com/••••.html" title="••••">••••</a></li>
    <li><a href="https://my-website.com/••••.html" title="••••">••••</a></li>

<ul id="myNavigation">
    <li><a href="https://my-website.com/••••.html" title="••••">••••</a></li>
....
....
....
    <li><a href="https://my-website.com/••••.html" title="••••">••••</a></li>
    <li><a href="https://my-website.com/••••.html" title="••••">••••</a></li>
    <li><a href="https://my-website.com/••••.html" title="••••">••••</a></li>
</ul>


    <li><a href="https://my-website.com/••••.html" title="••••">••••</a></li>
    <li><a href="https://my-website.com/••••.html" title="••••">••••</a></li>
    <li><a href="https://my-website.com/••••.html" title="••••">••••</a></li>

Perform the last regex S/R, above
You’re left with the expected text :

...
...
<ul id="myNavigation">
<div class="categories-name">
    <a href="https://my-website.com/••••.html" title="••••">
    <p class="font-16 color-grey text-capitalize"><i class="fa fa-angle-right font-14 color-blue mr-1"></i>••••<span>••••</span> </p>
    </a>
</div>
<div class="categories-name">
    <a href="https://my-website.com/••••.html" title="••••">
    <p class="font-16 color-grey text-capitalize"><i class="fa fa-angle-right font-14 color-blue mr-1"></i>••••<span>••••</span> </p>
    </a>
</div>
<div class="categories-name">
    <a href="https://my-website.com/••••.html" title="••••">
    <p class="font-16 color-grey text-capitalize"><i class="fa fa-angle-right font-14 color-blue mr-1"></i>••••<span>••••</span> </p>
    </a>
</div>
....
....
....
<div class="categories-name">
    <a href="https://my-website.com/••••.html" title="••••">
    <p class="font-16 color-grey text-capitalize"><i class="fa fa-angle-right font-14 color-blue mr-1"></i>••••<span>••••</span> </p>
    </a>
</div>
</ul>

    <li><a href="https://my-website.com/••••.html" title="••••">••••</a></li>
    <li><a href="https://my-website.com/••••.html" title="••••">••••</a></li>
    <li><a href="https://my-website.com/••••.html" title="••••">••••</a></li>

<ul id="myNavigation">
<div class="categories-name">
    <a href="https://my-website.com/••••.html" title="••••">
    <p class="font-16 color-grey text-capitalize"><i class="fa fa-angle-right font-14 color-blue mr-1"></i>••••<span>••••</span> </p>
    </a>
</div>
....
....
....
<div class="categories-name">
    <a href="https://my-website.com/••••.html" title="••••">
    <p class="font-16 color-grey text-capitalize"><i class="fa fa-angle-right font-14 color-blue mr-1"></i>••••<span>••••</span> </p>
    </a>
</div>
<div class="categories-name">
    <a href="https://my-website.com/••••.html" title="••••">
    <p class="font-16 color-grey text-capitalize"><i class="fa fa-angle-right font-14 color-blue mr-1"></i>••••<span>••••</span> </p>
    </a>
</div>
<div class="categories-name">
    <a href="https://my-website.com/••••.html" title="••••">
    <p class="font-16 color-grey text-capitalize"><i class="fa fa-angle-right font-14 color-blue mr-1"></i>••••<span>••••</span> </p>
    </a>
</div>
</ul>


    <li><a href="https://my-website.com/••••.html" title="••••">••••</a></li>
    <li><a href="https://my-website.com/••••.html" title="••••">••••</a></li>
    <li><a href="https://my-website.com/••••.html" title="••••">••••</a></li>

You can verify that the remaining lines <li><a href="https://my-website.com/••••.html" title="••••">••••</a></li> are only lines, which are outside a <ul id="myNavigation">............</ul> multi-lines section, so not replaced at all !

Best Regards,

guy038

Oups !! I forgot to delete, either, the lines <ul id="myNavigation"> and </ul>. Quite easy with :

SEARCH (?-i)<(ul id="myNavigation"|/ul)>\R

REPLACE Leave EMPTY

Robin Cruise · Mar 9, 2021, 9:29 PM

This post is deleted!

Robin Cruise · Mar 9, 2021, 9:39 PM

@guy038 yes, sir. Thank you. You succeded to transform the part 1 with part 2.

My problem, or I didn’t understand to good, are those ••••

How can I change those many •••• with different words and numbers ? If I have 200 lines, I need 2 months to change those ••••

I return to my regex,

please .matches newlines and run this 2 regex

FIND: (<li><a href="https://my-website.com/(.*?)" title=".*?)(\h+<a href="https://my-website.com/).*?(" title=")

REPLACE BY: \1\3\2\4

and

FIND: (title="(.*?)">.*?)(mr-1"></i> ).*?(<span>)

REPLACE BY: \1\3\2 \4

With this 2 regex, I manage to make a parsing. I copy a information from part 1 to part 2. Of course, is an example of how the formula should work

guy038 · Mar 9, 2021, 9:52 PM

Hello @Robin-cruise,

If you want me to test your two regex S/R ( Let’s say respectively, S/R A and S/R B )

I need an exact raw text which must be processed with the SR A
I need any additional action, if any, which is to be done, after processing S/R A and before processing S/R B
I need an exact raw text which must be processed with the S/R B ( Of course, it may be the result of the S/R A on the initial text )

BR

guy038

Robin Cruise · Mar 11, 2021, 9:25 AM

@guy038 @Alan-Kilborn This is the solution. Yes, it can be done with regex, but must integrated in powershell. Works super !

$file1 = 'c:\Folder1\file1.html'
$file2 = 'c:\Folder1\file2.html'
$result = 'c:\Folder1\result.html'
$link=@()
$title=@()
$number=@()
Get-Content -Path $file1 -Delimiter '</li>'|ForEach-Object{
    $_|ForEach-Object{
        if($_ -match '(?<=href=").+?(?=")'){$link += $Matches.Values}
        if($_ -match '(?<=title=").+?(?=")'){$title += $Matches.Values}
        if($_ -match '(?<=\()\d+(?=\))'){$number += $Matches.Values}
    }
}
$content = Get-Content -Path $file2 -Delimiter '</div>'
for($i=0;$i -lt $content.Count;$i++){
    $content[$i] | ForEach-Object{
        if($_ -match '(?<=href=").+?(?=")'){$link2 = $Matches.Values}
        if($_ -match '(?<=title=").+?(?=")'){$title2 = $Matches.Values}
        if($_ -match '(?<=<span>)\d+(?=</span>)'){$number2 = $Matches.Values}
    }
    $content[$i] -replace $link2, $link[$i] -replace $title2, $title[$i] -replace $number2, $number[$i] | Out-File -FilePath $result -Append
}

Source: https://docs.microsoft.com/en-us/answers/questions/307621/powershell-copy-strings-from-some-html-tags-to-ano.html

Alan Kilborn · Mar 11, 2021, 12:31 PM

@Robin-Cruise said in regex: Parsing html tags in other tags / links and titles:

This is the solution. Yes, it can be done with regex, but must integrated in powershell.

Ok… but this is really off-topic for a Notepad++ forum.
Whatever you’re doing could I’m sure be done in many different ways.
This isn’t a general “data manipulation” forum.

guy038 · Mar 11, 2021, 12:43 PM

Hi, @Robin-cruise,

Nice ! So you, finally, got a working solution.

Out of curiosity, could you provide me a short sample of your initial files file1.html with several sections <ul......</ul> and file2.html and the corresponding result.html file, created by your Powershell script

This could help me to understand which you really wanted to achieve and why I could not imagine about suitable and useful regexes !

Thanks for your cooperation !

Best Regards,

guy038

Robin Cruise · Mar 11, 2021, 2:56 PM

File1.html

<ul id="myNavigation">
    <li><a href="https://my-website.com/page-1.html" title="Page 1">Page 1 (34)</a></li>
    <li><a href="https://my-website.com/page-2.html" title="Page 2">Page 2 (29)</a></li>
    <li><a href="https://my-website.com/page-3.html" title="Page-3">Page 3 (11)</a></li>
</ul>

File2.html

<div class="categories-name">
   <a href="https://my-website.com/page-66.html" title="Page 66">
   <p class="font-16 color-grey text-capitalize"><i class="fa fa-angle-right font-14 color-blue mr-1"></i> Page 66 <span>27</span> </p>
  </a>
</div>
<div class="categories-name">
   <a href="https://my-website.com/page-67.html" title="Page 67">
   <p class="font-16 color-grey text-capitalize"><i class="fa fa-angle-right font-14 color-blue mr-1"></i> Page 67 <span>24</span> </p>
   </a>
</div>
<div class="categories-name">
   <a href="https://my-website.com/page-68.html" title="Page 68">
   <p class="font-16 color-grey text-capitalize"><i class="fa fa-angle-right font-14 color-blue mr-1"></i> Page 68 <span>07</span> </p>
   </a>
</div>

@guy038 @Alan-Kilborn
Thank you. Please understand that finding this solution, was not easy.

First, I knew that @guy038 is one of the best REGEX developer on earth. He already gave a solution, in another post, to a PARSING html. so I thought he would succeed.

Second, in another post, someone else gave a solution with PowerShell, of another problem. I remember also that post.

So, I knew that this problem can be resolve, but I had to try both options.

Now, maybe will be a better way to handle regex and files, by integrate notepad++ them into another program such as PowerShell.

guy038 · Mar 11, 2021, 3:01 PM

Hi, @robin-cruise,

But, could you provide, also, the result.html file ?

Thanks

BR

guy038

Robin Cruise · Mar 11, 2021, 3:01 PM

@Alan-Kilborn said in regex: Parsing html tags in other tags / links and titles:

but this is really off-topic for a Notepad++ forum

No, sir. Is an opportunity to understand that a problem has many solution, even if people told you that isn’t possible.

By the way, the same problem can be done in LINUX, much easier…with bash script, but that’s another story.

Robin Cruise · Mar 11, 2021, 3:06 PM

@guy038 result.html is just a blank file, put in the same folder with File1.html and File2.html

After you run the code in PowerShell, the output result will appear into result.html

please test the PowerShell code. You will see in result.html the structure html from the File2.html but with the values extracted from File1.html

guy038 · Mar 11, 2021, 3:09 PM

Hi, @robin-cruise,

OK but could you provide the result.html file AFTER running the Powershell script !

BR

guy038

Sorry, with my old XP SP3 machine is a bit out of date !

Robin Cruise · Mar 11, 2021, 3:13 PM

@guy038 this is the result.html after run the PowerShell code

Robin Cruise · Mar 11, 2021, 3:24 PM

the single problem in PowerShell is that not every regex formulas that works in notepad++ works there. Seems that PowerShell don’t like \K or \G or replacement such as (?2\2)

guy038 · Mar 11, 2021, 7:46 PM

Hello, @robin-cruise, @alan-kilborn and All,

OMG, now I understand the all story ! In fact, you’re did not express your needs correctly, in your first post ! The File-2.tml is useless and we just can go from the result_1.hmtl file to the result.html file !

Indeed, in your first post, you said :

<ul id="myNavigation">
    <li><a href="https://my-website.com/page-1.html" title="Page 1">Page 1 (34)</a></li>
    <li><a href="https://my-website.com/page-2.html" title="Page 2">Page 2 (29)</a></li>
    <li><a href="https://my-website.com/page-3.html" title="Page-3">Page 3 (11)</a></li>
</ul>

MUST BECOME:

<div class="categories-name">
   <a href="https://my-website.com/page-66.html" title="Page 66">
   <p class="font-16 color-grey text-capitalize"><i class="fa fa-angle-right font-14 color-blue mr-1"></i> Page 66 <span>27</span> </p>
  </a>
</div>
<div class="categories-name">
   <a href="https://my-website.com/page-67.html" title="Page 67">
   <p class="font-16 color-grey text-capitalize"><i class="fa fa-angle-right font-14 color-blue mr-1"></i> Page 67 <span>24</span> </p>
   </a>
</div>
<div class="categories-name">
   <a href="https://my-website.com/page-68.html" title="Page 68">
   <p class="font-16 color-grey text-capitalize"><i class="fa fa-angle-right font-14 color-blue mr-1"></i> Page 68 <span>07</span> </p>
   </a>
</div>

can something like this be done with regex?

But… you SHOULD have written :

<ul id="myNavigation">
    <li><a href="https://my-website.com/page-1.html" title="Page 1">Page 1 (34)</a></li>
    <li><a href="https://my-website.com/page-2.html" title="Page 2">Page 2 (29)</a></li>
    <li><a href="https://my-website.com/page-3.html" title="Page-3">Page 3 (11)</a></li>
</ul>

MUST BECOME

<div class="categories-name">
   <a href="https://my-website.com/page-1.html" title="Page 1">
   <p class="font-16 color-grey text-capitalize"><i class="fa fa-angle-right font-14 color-blue mr-1"></i> Page 1 <span>34</span> </p>
  </a>
</div>
<div class="categories-name">
   <a href="https://my-website.com/page-2.html" title="Page 2">
   <p class="font-16 color-grey text-capitalize"><i class="fa fa-angle-right font-14 color-blue mr-1"></i> Page 2 <span>29</span> </p>
   </a>
</div>
<div class="categories-name">
   <a href="https://my-website.com/page-3.html" title="Page-3">
   <p class="font-16 color-grey text-capitalize"><i class="fa fa-angle-right font-14 color-blue mr-1"></i> Page-3 <span>11</span> </p>
   </a>
</div>

can something like this be done with regex?

And yes, this transformation can, indeed, be solved with ONE regex S/R only !!!

What misled me was the fact that your output file contained other values than the ones indicated in the initial file ! ( Page 66 <span>27</span>, Page 67 <span>24</span>, Page 68 <span>07</span>... which are completely out of scope :-(( I vainly looked for a link between, for example, the strings "Page 1">Page 1 (34) and Page 66 <span>27</span> !

So, Robin, now, it’s fairly easy to get the right regex S/R !

From this initial text

===========================================================================================================================================
<ul id="myNavigation">
    <li><a href="https://my-website.com/page-1.html" title="Page 1">Page 1 (34)</a></li>
    <li><a href="https://my-website.com/page-2.html" title="Page 2">Page 2 (29)</a></li>
    <li><a href="https://my-website.com/page-3.html" title="Page-3">Page 3 (11)</a></li>
</ul>
===========================================================================================================================================

with the following regex S/R :

SEARCH (?-i)^\h*<li><a href="https://my\-website.com/(.+?)" title="(.+?)">.+?\((.+?)\)</a></li>|<(ul id="myNavigation"|/ul)>\R

REPLACE ?4:<div class="categories-name">\r\n\x20\x20\x20<a href="https://my-website.com/\1" title="\2">\r\n\x20\x20\x20<p class="font-16 color-grey text-capitalize"><i class="fa fa-angle-right font-14 color-blue mr-1"></i>\x20\2\x20<span>\3</span> </p>\r\n\x20\x20\x20</a>\r\n</div>

we do get the expected text, totally identical to your result.html contents ! Moreover, you do not need the Powershell script any more !

===========================================================================================================================================
<div class="categories-name">
   <a href="https://my-website.com/page-1.html" title="Page 1">
   <p class="font-16 color-grey text-capitalize"><i class="fa fa-angle-right font-14 color-blue mr-1"></i> Page 1 <span>34</span> </p>
   </a>
</div>
<div class="categories-name">
   <a href="https://my-website.com/page-2.html" title="Page 2">
   <p class="font-16 color-grey text-capitalize"><i class="fa fa-angle-right font-14 color-blue mr-1"></i> Page 2 <span>29</span> </p>
   </a>
</div>
<div class="categories-name">
   <a href="https://my-website.com/page-3.html" title="Page-3">
   <p class="font-16 color-grey text-capitalize"><i class="fa fa-angle-right font-14 color-blue mr-1"></i> Page-3 <span>11</span> </p>
   </a>
</div>
===========================================================================================================================================

Et voilà !

Notes :

The groups used are :
- Group 1 : the part after website.com/ till the double quote, not included
- Group 2 : the title part
- Group 3 : the string between parentheses, before </a>
- Group 4 : either the literal string <ul id="myNavigation"> or the literal string </ul>, which must be deleted, in replacement
Ths regex string \x20\x20\x20 occurs three times, in the replacement regex, and correspond to the needed leading spaces in each <div ......</div> section. Changed it as desired !

Best Regards,

guy038

P.S. :

And, if you want to change the lines <li>.........</li>, ONLY inside the <ul...............</ul> sections, use this second regex S/R, derived of the generic regex S/R, discussed in other topics :

SEARCH (?s-i)(?:<ul id="myNavigation">|(?!\A)\G)(?:(?!</ul>).)*?\K^\h*<li><a href="https://my\-website.com/(.+?)" title="(.+?)">.+?\((.+?)\)</a></li>

REPLACE <div class="categories-name">\r\n\x20\x20\x20<a href="https://my-website.com/\1" title="\2">\r\n\x20\x20\x20<p class="font-16 color-grey text-capitalize"><i class="fa fa-angle-right font-14 color-blue mr-1"></i>\x20\2\x20<span>\3</span> </p>\r\n\x20\x20\x20</a>\r\n</div>

Now, end the job, getting rid of the literal string <ul id="myNavigation"> and the literal string </ul>, with, simply :

SEARCH (?-i)<(ul id="myNavigation"|/ul)>\R

REPLACE Leave EMPTY

We need consecutive regexes S/R because the values <ul id="myNavigation"> and </ul> are used as anchors, in the first generic S/R and must not be deleted during the first regex process !

Alan Kilborn · Mar 11, 2021, 6:57 PM

@guy038

Hi Guy… it is too funny.
Now, before solving a problem, you have to solve another problem: What the problem actually is!
It’s good that you have the talent and perseverance to do that; well, it’s good for the OP I suppose.

Robin Cruise · Mar 11, 2021, 7:36 PM

@guy038 thank you, I knew that you can do it. I believe, you are the best regex developer in the entire word.

I understood one thing. Never say “it can’t be done” and even if it can’t be done in one way, you can always find another way. Be open to another programs.

Much more important: if you give someone a new perspective, that someone will be able to see clearly the problem and to find the solution much easy.

So, my friends. Thanks a lot, from now on I know that many people understood also the power of PowerShell, and it can be used very often to modify files by integrate regex.

guy038 · Mar 12, 2021, 2:47 AM

Hi, @robin-cruise and All,

You said :

I believe, you are the best regex developer in the entire word.

I would say definitively not the best !

For instance, just think of this man here and here .

Jan Goyvaerts :

Created the fantastic regular-expression.info site and still maintains it !
Developed all these valuable regex tools : PowerGREP, RegexBuddy, RegexMagic and EditPad Pro

This is just an example of the many developers, involved in regular expressions !

BR

guy038

Robin Cruise · Mar 12, 2021, 8:20 AM

of course there are others, but only @guy038 helped me and all notepad++ community with so hard to find solutions.

and I am using notepad++ and grepWin (and as for the Batch Process I using a great tool named TextCrawler)