Match everything except the text and <br> tags

PeterJones

@dr-ramaanand said in Match everything except the text and <br> tags:

(?s)<div style="margin-bottom:-15px;width: 100%;background-color:#EBF4FB;">\s*<p[^<>]*+>\s*<span[^<>]*+>\s*<span[^<>]*+>\s*<span[^<>]*+>.+?</span>\s*</span>\s*</span>\s*</p>\s*</div>/s*</div>\s*<div class="container">\s*<div class="left">

In this case, your problem is easy. /s* doesn’t match what you think it matches. I think you meant \s*

Do you know how I found this? I started with a smaller part of your regex, saw that it matched, then slowly added more and more until it didn’t match; then I backed up and found the exact section that caused it to fail, and the solution was easy. It was simple debugging skills, which you need to learn if you are going to continue to manipulate data using regex.

dr ramaanand

@PeterJones Thank you very much. This Regular expression worked: (?s)<div style="margin-bottom:-15px;width: 100%;background-color:#EBF4FB;">\s*<p[^<>]*+>\s*<span[^<>]*+>\s*<span[^<>]*+>\s*<span[^<>]*+>.+?</span>\s*</span>\s*</span>\s*</p>\s*</div>\s*</div>\s*<div class="container">\s*<div class="left">

guy038

Hi, @dr-ramaanand, @peterjones and All,

Ah, I was a bit too slow and Peter just beats me ! Note that I used the same process than Peter to determine where the error occurs !

@dr-ramaanand, you just did a small typo error in the regex that you provided !

The correct regex, to match your text, is not that one, with a /s* syntax :

                                                                                                                                                                                      V
(?s)<div style="margin-bottom:-15px;width: 100%;background-color:#EBF4FB;">\s*<p[^<>]*+>\s*<span[^<>]*+>\s*<span[^<>]*+>\s*<span[^<>]*+>.+?</span>\s*</span>\s*</span>\s*</p>\s*</div>/s*</div>\s*<div class="container">\s*<div class="left">

but this one, with a correct \s* syntax :

                                                                                                                                                                                      V
(?s)<div style="margin-bottom:-15px;width: 100%;background-color:#EBF4FB;">\s*<p[^<>]*+>\s*<span[^<>]*+>\s*<span[^<>]*+>\s*<span[^<>]*+>.+?</span>\s*</span>\s*</span>\s*</p>\s*</div>\s*</div>\s*<div class="container">\s*<div class="left">

Remarks :

May be, it would be preferable to add a \s* syntax at the very end of your regex !
You could also simplify this regex, significantly, by using the version below :

SEARCH (?s)<div style="margin-bottom:-15px;width: 100%;background-color:#EBF4FB;">\s*.+?\s*<div class="left">\s*

BR

guy038

dr ramaanand

@guy038 I have more than one <div class="left">, so how do I make it stop searching after finding the first <div class="left"> ?

guy038

Hello, @dr-ramaanand and All,

To solve this case, I would use the following regex S/R :

SEARCH (?s)\A.+?\R\K\s*<div style="margin-bottom:-15px;width: 100%;background-color:#EBF4FB;">\s*.+?\s*<div class="left">

REPLACE Whatever you want to !

Note that I did not add, this time, the \s* part at the end of the search regex.

Also notice the two lazy syntaxes ( .+? ), right after \A and right before \s*<div class="left">, in order to select only the first section s*<div style=.....\s*<div class="left">, only !

BR

guy038

dr ramaanand

@guy038 I used this as a sample:-

<div style="margin-bottom:-15px;width: 100%;background-color:#EBF4FB;">
<div class="left">
<p class=MsoNormal><b><span style='font-size:13.5pt;line-height:115%;
    font-family:"Verdana","sans-serif";color:red'>SYNONYMS </span></b>
</p>
<div class="left">

Your Regular expression does not stop searching at the first occurrence of <div class="left">

dr ramaanand

@guy038 This RegEx helped stop searching as soon as it found a <p........>:-
(?s)\A.+?\R\K\s*<div style="margin-bottom:-15px;width: 100%;background-color:#EBF4FB;">\s*.+?\s*<div class="left">(?=\s*+<p[^<>]*+>)

guy038

Hi, @dr-ramaanand and All,

Ah, of course, if you add a <div class="left"> line, right after the first <div style="..... line, it will not work !

So, given this INPUT text, pasted in a new tab:

<div style="margin-bottom:-15px;width: 100%;background-color:#EBF4FB;">
<div class="left">
<p class=MsoNormal><b><span style='font-size:13.5pt;line-height:115%;
    font-family:"Verdana","sans-serif";color:red'>SYNONYMS </span></b>
</p>
<div class="left">

Simply, change the previous search regex by this new version :

(?s)\A.+?\R\s*\K<div style="margin-bottom:-15px;width: 100%;background-color:#EBF4FB;">\s*.*?\s*<div class="left">

Note the différence : between #EBF4FB;">\s* and \s*<div class="left">, I changed the part .+? by .*?

I also slightly change the position of the \K feature

Ax expected, this new regex will match the two consecutive lines :

<div style="margin-bottom:-15px;width: 100%;background-color:#EBF4FB;">
<div class="left">

BR

guy038

dr ramaanand

@guy038 This RegEx: (?s)\A.+?\R\K\s*<div style="margin-bottom:-15px;width: 100%;background-color:#EBF4FB;">.+(?=\s*+<div class="left">) would have stopped searching just before the second occurrence of <div class="left"> if the sample to be searched was like this:-

<div style="margin-bottom:-15px;width: 100%;background-color:#EBF4FB;">
<div class="left">
<div class="left">

guy038

@dr-ramaanand,

Yes, your regex does match the same amount of text as my version but my regex seems more simple and logic !

BR

guy038

dr ramaanand

@guy038 d’accord, merci beaucoup!

dr ramaanand

@guy038 your last RegEx finds the first occurrence of <div class="left"> even if there is some other text above it. Lovely!

guy038

Hi, @dr-ramaanand and All,

Again, I did not check all the possibilities before posting. Sorry for the NOISE !

So, the right regex to use should be :

(?s)\A.*?\s*\K<div style="margin-bottom:-15px;width: 100%;background-color:#EBF4FB;">\s*.*?\s*<div class="left">

This time, it will work if you pasted this text, in a new tab

<div style="margin-bottom:-15px;width: 100%;background-color:#EBF4FB;">
<div class="left">
<div class="left">

But it will also works, if you pasted the following text, in a new tab


First non-blank line
second line

Third line before the block to match

<div style="margin-bottom:-15px;width: 100%;background-color:#EBF4FB;">
<div class="left">
<div class="left">

Best Regards,

guy038

dr ramaanand

@guy038 I am not sure if I am allowed to do it (as the solution was provided by you), so I am requesting you to post the last Regular Expression you provided with the sample to be edited with a new heading, “How to find the first occurrence of a tag ?” so that people can search and find it online. Thank you!

guy038

Hello, @dr-ramaanand and All,

You said in your previous post :

… so I am requesting you to post the last Regular Expression you provided with the sample to be edited with a new heading, “How to find the first occurrence of a tag ?” so that people can search and find it online. Thank you!

But, actually, my regex finds the first occurrence of the <div class="left"> tag, AFTER a first occurrence of the <div style="margin-bottom:-15px;width: 100%;background-color:#EBF4FB;"> tag !

So, to my mind, the correct way to match the first occurrence of a specific tag, in current file, is to use the generic regex :

(?s-i)\A.*?\K<TAG Name(?: .*?)?>

Just replace the generic TAG Name value with a valid HTML tag

Note that, in case of the comment tag, replace the generic TAG Name, into the above regex, by the literal string !--.*?--

Similarly, the correct way to match the last occurrence of a specific tag, in current file, is to use the generic regex :

(?s-i)\A.*\K<TAG Name(?: .*?)?>

BR

guy038

dr ramaanand

@guy038 said in Match everything except the text and <br> tags:

(?s-i)\A.\K<TAG Name(?: .?)?>

I think that that should be (?s-i)\A.*\K<TAG Name(?:.*?)?> with no spaces anywhere in the middle

guy038

Hi, @dr-ramaanand and All,

In order to use a valid INPUT text to do some tests, just open the main page of our forum. Then hit the Ctrl + U shortcut to open the HTML source page of our forum and paste its contents in a new tab

My generic regex tries to match the syntax <TAG......, till the nearest > character and must be valid for any kind of tag.

Thus, I prefer to insert a space char to verify that the tag is a valid one . Indeed, this regex will match, either, tags like <head> or for example <span style="color:blue">blue</span>

If you replace the TAG Name in the generic regex (?s-i)\A.*?\K<TAG Name(?: .*?)?>, which matches the first tag, named TAG, in current file, you get, from the examples, the regexes :

(?s-i)\A.*?\K<head(?: .*?)?>
(?s-i)\A.*?\K<span(?: .*?)?>

Just test them against the HTML code source of our forum

Now, let’s suppose, for example, that you want to find out the first input ...> tag, AFTER the first img ......> tag, in the HTML code source of our forum :

Then, from my previous post, you would have to use the following regex :

(?s-i)\A.*?<img(?: .*?)?>.*?\K<input(?: .*?)?>

which matches, as expected, the following line :

<input autocomplete="off" type="text" class="form-control hidden" name="term" placeholder="Search"/>

BR

guy038

P.S. : You also replied in an old post, regarding this extra space char. However, I’ll not reply because this topic is old and not exactly related to the present discussion !

dr ramaanand

@guy038 Okay, thank you!