Regex examples

[theducks]

Originally Posted by DiapDealer

Be be or not be twice?

A William S find

[DiapDealer]

In my desire to be concise as possible, I would probably represent Shakespeare's famous line in regex as:

b?

[dwlamb]

Hello,

I am trying to come up with a method to search for blocks of HTML in which dialogue was started but a new paragraph was started before the closing quotation mark.

I have a string of

\>.*".*[^(")]

Basically I want to find

• text starting after the opening <p> tag and before an opening quotation mark
• the opening quotation mark
• the block of text up to the end of the paragraph, provided there is no closing quotation mark before the \n

Is this feasible?

[rubeus]

deleted. strange, my posting was declared off topic. Dumb postings above not. Now you can block me

[dwlamb]

rubeus:
The only inference of replace all has been in your reply, rubeus. My intent is clear: find

The criteria I want would only look for a dialogue that is started but not closed by the time a </p>\n happens. The search matched, I would then visually inspect why it matches which would deal with possible situations you specify.

[rubeus]

If you are only looking for one aspect, my posting was off topic: But to be honest: it wasnt really clear that you are interested only in findings.

Well, you just skiupped al other problems i mentioned, so i guess you're no longer interested in my opinion. I'm fine with that.

Last famous words: using the off topcsmiley was rude,in my opinion. Bye.

[eschwartz]

Find:

<p( [^<>]+)?>["“]((?!</p>).)*[^"”]</p>

Assumes the paragraph starts with a quote mark, accounts for smart or dumb quotes, matches optional attributes in the para tag, and matches tags in the paragraph assuming they aren't the para tag (see HERE for how).

This one finds text before the quotes too:

<p( [^<>]+)?>((?!</p>).)*["“]((?!</p>).)*[^"”]</p>

[dwlamb]

Hi,

Thank you for responding. I tried your syntax and it is catching any paragraph that has quoted text properly formed or not. Is it possible you can further break down what your search strings do so that I can piece-meal learn and experiment? I read the item from stackoverflow but it was fuzzy to me.

To give more of an example of the text I am searching for, I have these examples:

<p class="some_class">"Phasellus in ante ac lectus," said Ceasar. Vestibulum neque nisi, dapibus quis, "Sed et magna eget orci.</p>

<p class="some_class">Proin non ex ex? Aliquam. Nam tristique scelerisque orci, et."</p>

or

<p class="some_class">"Sed eget nulla vel augue. Proin id metus sed mi.</p>

<p class="some_class">Sed sed eleifend ex! Phasellus justo lectus, tempor id mollis."</p>

Thanks for the help.

Originally Posted by eschwartz

Find:

<p( [^<>]+)?>["“]((?!</p>).)*[^"”]</p>

Assumes the paragraph starts with a quote mark, accounts for smart or dumb quotes, matches optional attributes in the para tag, and matches tags in the paragraph assuming they aren't the para tag (see HERE for how).

This one finds text before the quotes too:

<p( [^<>]+)?>((?!</p>).)*["“]((?!</p>).)*[^"”]</p>

[signum]

In my opinion, I don't think it's possible to write a single regular expression to do what you want. REs are not a general programming language and are quite limited in some respects. I wrote a short program to step through an htm file and find the number of ldquos, rdquos and quots in each paragraph. If the number of ldquos didn't equal the number of rdquos or the number of quots was odd, it adds a class="unbal" to the paragraph. Regular expressions are usually incapable of counting, comparing two registers or taking the modulus of a number. I would be pleased to be shown wrong, using the RE syntax in Sigil.

[eschwartz]

Originally Posted by signum

In my opinion, I don't think it's possible to write a single regular expression to do what you want. REs are not a general programming language and are quite limited in some respects. I wrote a short program to step through an htm file and find the number of ldquos, rdquos and quots in each paragraph. If the number of ldquos didn't equal the number of rdquos or the number of quots was odd, it adds a class="unbal" to the paragraph. Regular expressions are usually incapable of counting, comparing two registers or taking the modulus of a number. I would be pleased to be shown wrong, using the RE syntax in Sigil.

That is also a different case.

Note that my, WORKING, solution is not a true regular expression either. But regular expressions are not nearly as useful as regex+lookaround, so it was decided that regular expressions need not in fact be regular.

[Buchstabensalat]

Hi,

As a newbie to the secrets of regex I am searching a regex formula finding (and then correcting) words that contain one or more upper case letters in them, excluding the first letter from the search. Example:

1. neWbie or nEwBie -> newbie
2. NeWbie or NEwBie -> Newbie

Please help. Thank you.

Regards

Buchstabensalat

[darrenmcg]

@ dwlamb

These are two I've been using to find something similar, maybe they can help...

“(?>[^\r\n“”]*)(?!\r?\n“)(?!”)
‘(?>[^\r\n‘’]*)(?!\r?\n’)(?!’)

[Doitsu]

Originally Posted by Buchstabensalat

1. neWbie or nEwBie -> newbie

Find:([[:alpha:]]+)
Replace:\L\1\E

Originally Posted by Buchstabensalat

2. NeWbie or NEwBie -> Newbie

Find:([[:upper:]])([[:alpha:]]+)
Replace:\1\L\2\E

Note that these expressions will also find some non-mixed case words and won't work with accented characters and umlauts.

You can find more examples here.

[Buchstabensalat]

Originally Posted by Doitsu

Find:([[:alpha:]]+)
Replace:\L\1\E



Find:([[:upper:]])([[:alpha:]]+)
Replace:\1\L\2\E

Note that these expressions will also find some non-mixed case words and won't work with accented characters and umlauts.

You can find more examples here.

Hi,

Thank you.

Using Sigil`s search (regex mode) your first string finds every single word in the text, the second string matches every uppercase word in my text and not only those that meet my search criteria ( uppercase letter(s) within a word). Is there a possiblity to match only those words with upercase letters in the word ?

Thanks again

Regards,

Buchstabensalat

[signum]

Originally Posted by eschwartz

That is also a different case.

Note that my, WORKING, solution is not a true regular expression either. But regular expressions are not nearly as useful as regex+lookaround, so it was decided that regular expressions need not in fact be regular.

Yeah, it works, sorta. But it matches all paragraphs with quotation marks, whether or not they are "valid", i.e. matched or balanced. This was first reported by the OP in msg #428, above.