Regex noob here struggling with this, which I know it will be easy for some of you regex gods out there!
Given the following:
title: Some title
date: 2022-08-15
tags: <value to extract>
identifier: 1234567
---------------------------
Some text
some more text
I would like a regex to match everything except the value of tags (ie the "<value to extract>" text).
For context, this is supposed to run on emacs (in case it matters).
EDIT: Just to clarify as per #phils question, all I care about extracting the tags value. However, this is via a package setting that asks for a regex string and I don't have much control over how it gets use. It seems to expect a regex to strip what I don't need from the string rather than matching what I do want, which is slightly annoying.. Also, the since it seems to match everything with \\(.\\), I'm guessing it's using the global flag?
Please let me know if any of this isn't clear.
Emacs regular expressions can't trivially express "not foo" for arbitrary values of foo. (The likes of PCRE have non-regular extensions for zero-width negative look-ahead/behind assertions, but in Emacs that sort of functionality is generally done with the support of lisp code1.)
You can still do it purely with regexp matching, but it's simply very cumbersome. An Emacs regexp which matches any line which does not begin with tags: is:
^\(?:$\|[^t]\|t[^a]\|ta[^g]\|tag[^s]\|tags[^:]\).*
or if you need to enter it in the elisp double-quoted read syntax for strings:
"^\\(?:$\\|[^t]\\|t[^a]\\|ta[^g]\\|tag[^s]\\|tags[^:]\\).*"
1 In lisp code you would instead simply check each line to see whether it does start with tags: and, if so, skip it (which is why Emacs generally gets away without the feature you're looking for, but of course that doesn't help you here).
After playing around with it for a bit and taken inspiration from #phils' answer, I've come up with the following:
"^\\(?:\\(#\\+\\)?\\(?:filetags:\s+\\|tags:\s+\\|title:.*\\|identifier:.*\\|date:.*\\)\\|.*\\)"
I've also added an extra \\(#\\+\\)? to account for org meta keys which would usually have the format #+key: value.
Related
I'm trying to see if its possible to extend an existing arbitrary regex by prepending or appending another regex to match within matches.
Take the following example:
The original regex is cat|car|bat so matching output is
cat
car
bat
I want to add to this regex and output only matches that start with 'ca',
cat
car
I specifically don't want to interpret a whole regex, which could be quite a long operation and then change its internal content to match produce the output as in:
^ca[tr]
or run the original regex and then the second one over the results. I'm taking the original regex as an argument in python but want to 'prefilter' the matches by adding the additional code.
This is probably a slight abuse of regex, but I'm still interested if it's possible. I have tried what I know of subgroups and the following examples but they're not giving me what I need.
Things I've tried:
^ca(cat|car|bat)
(?<=ca(cat|car|bat))
(?<=^ca(cat|car|bat))
It may not be possible but I'm interested in what any regex gurus think. I'm also interested if there is some way of doing this positionally if the length of the initial output is known.
A slightly more realistic example of the inital query might be [a-z]{4} but if I create (?<=^ca([a-z]{4})) it matches against 6 letter strings starting with ca, not 4 letter.
Thanks for any solutions and/or opinions on it.
EDIT: See solution including #Nick's contribution below. The tool I was testing this with (exrex) seems to have a slight bug that, following the examples given, would create matches 6 characters long.
You were not far off with what you tried, only you don't need a lookbehind, but rather a lookahead assertion, and a parenthesis was misplaced. The right thing is: Put the original pattern in parentheses, and prepend (?=ca):
(?=ca)(cat|car|bat)
(?=ca)([a-z]{4})
In the second example (without | alternative), the parentheses around the original pattern wouldn't be required.
Ok, thanks to #Armali I've come to the conclusion that (?=ca)(^[a-z]{4}$) works (see https://regexr.com/3f4vo). However, I'm trying this with the great exrex tool to attempt to produce matching strings, and it's producing matches that are 6 characters long rather than 4. This may be a limitation of exrex rather than the regex, which seems to work in other cases.
See #Nick's comment.
I've also raised an issue on the exrex GitHub for this.
So I read a lot about Negation in Regex but can't solve my problem in MS Word 2016.
How do I exclude a String, Word, Number(s) from being found?
Example:
<[A-Z]{2}[A-Z0-9]{9;11}> to search a String like XY123BBT22223
But how to exclude for example a specefic one like SEDWS12WW04?
Well it depends on what you need to achieve or is this a matter of curiosity... RegEx is not the same as the built-in Advanced Find with Wildcards; for that you need VBA.
Depending on your need, without using VBA, you could make use of space and return characters - something like this will work for the strings provided: [ ^13][A-Z]{2}[0-9]{1,}[A-Z]{1,}[0-9]{1,}[ ^13] (assuming you use normal carriage returns and spaces in your document)
Anyway, this is a good article on wildcard searches in MS Word: https://wordmvp.com/FAQs/General/UsingWildcards.htm
EDIT:
In light of your further comments you will probably want to look at section 8 of the linked article which explains grouping. For my proposed search you can use this to your advantage by creating 3 groups in your 'find' and only modifying the middle group, if indeed you do intend to modify. Using groups the search would look something like:
([ ^13])([A-Z]{2}[0-9]{1,}[A-Z]{1,}[0-9]{1,})([ ^13])
and the replace might look like this:
\1 SOMETHING \3
Note also: compared to a RegEx solution my suggestion is kinda lame, mainly because compared to RegEx, MS-Words find and replace (good as it is, and really it is) is kinda lame... it's hacky but it might work for you (although you might need to do a few searches).
BUT... if it really is REGEX that you want, well you can get access to this via VBA: How to Use/Enable (RegExp object) Regular Expression using VBA (MACRO) in word
And... then you will be able to use proper RegEx for find and replace, well almost - I'm under the impression that the VBA RegEx still has some quirks...
As already noted by others, this is not possible in Microsoft Word's flavor of regular expressions.
Instead, you should use standard regular expressions. It is actually possible to use standard regular expressions in MS Word if you use a special tool that integrates into Microsoft Word called Multiple Find & Replace (see http://www.translatortools.net/products/transtoolsplus/word-multiplefindreplace). This tool opens as a pane to the right of the document window and works just like the Advanced Find & Replace dialog. However, in addition to Word's existing search functionality, it can use the standard regular expressions syntax to search and replace any text within a Word document.
In your particular case, I would use this:
\b[A-Z]{2}[A-Z0-9]{9,11}\b(?<!\bSEDWS12WW04)
To explain, this searches for a word boundary + ID + word boundary, and then it looks back to make sure that the preceding string does not match [word boundary + excluded ID]. In a similar vein, you can do something like
(?<!\bSEDWS12WW04|\bSEDWS12WW05|\bSEDWS12WW05)
to exlude several IDs.
Multiple Find & Replace is quite powerful: you can add any number of expressions (either using regular expressions or using Word's standard search syntax) to a list and then search the document for all of them, replace everything, display all matches in a list and replace only specific matches, and a few more things.
I created this tool for translators and editors, but it is great for any advanced search/replace operations in Word, and I am sure you will find it very useful.
Best regards, Stanislav
According to the received wisdom MS Word (more or less) supports find/replace with use of regular expressions. I have a simple regular expression:
^(C[[:alpha:]]*)(\d*)(.*)$
That I'm running on the data:
indSIMDdecile
CSdeccrim12006
CSdeccrim12006
CSdeccrim12009
CSdeccrim12009
CSdeccrim12012
CSdeccrim12012
CSdeceduc12004
CSdeceduc12004
CSdeceduc12006
CSdeceduc12006
CSdeceduc12009
CSdeceduc12009
CSdeceduc12012
CSdeceduc12012
CSdecemp12004.x
I'm interested in returning the first word prior to the digit 1, which works as demonstrated on regex101 here.
Problem
I would like to the same but in MS Word (v. 15.18 on Mac). After getting error messages of trying to supply unsuitable syntax I learned that MS Word does not support to the full regex syntax. I simplified my expression to something on the lines:
but the search does not find any strings and nothing gets replaced. Hence my questions, is it possible to use MS Word on Mac with regex?
The linked help website hints that something like that should be possible, but so far now luck.
The simple answer is "no", if you mean "Does Mac Word have a UI feature that lets you use one of the modern dialects of regex?" Word's Find/Replace only supports its own Regular Expression syntax.
In this case, I think the following will give you what you need:
Find with wildcards:
(C)([!1]#)(1)
and a replace by
\1
(If you also had to find "C1", then that doesn't work, and unfortunately nor does
(C)([!1]{0,})(1)
because Word does not allow 0 in the {,} pattern)
But there is a problem with "#". If the text the "#" is looking for is long, the find/replace may fail. There is supposed to be a 255 limit, but it seems rather more arbitrary than that. (I have long suspected a buffer overrun type error in the Word code, but perhaps there is a simpler explanation).
If you mean, "is there any way to use modern regex with Word?", then the answer is "Yes, but you only get to operate on a copy of the text in the document. You will need to create your own code to do the 'replace' part of the find replace, and that means that you would have to deal with any of the issues such as preserving formatting that Word's built-in find/replace might get right for you.
On the Windows side, people who want a better regex than Word's often use VBScript's regexp object because it is easily used from VBA. VBA itself only really has the "like" operator, which also only has fairly crude pattern matching abilities. I think there are examples of VBScript rexexp use on StackOverflow. On the Mac side, you would either have to use VBA and "shell out" to one of the built-in Mac/Unix utilities to do your finding (and perhaps replacing), or perhaps use Applescript or Javascript application scripting to do it. As far as I can remember Applescript does not have a 'modern' regex built-in either.
[As a bit of history, Word's "regular expressions" were I think introduced in Word 6, around 1993, at a time when most dialects of regex were much more crude than they are today. I don't think Word's version has moved along much at all - it probably added some Unicode support at some point, but that's probably about it. I assume that people using modern regex don't regard it as regex at all, and I personally prefer not to call Word's Regular Expressions 'regex' precisely for that reason.]
Using regex I'm trying to remove empty bbcode tags. By empty I mean nothing in between them:
[tag][/tag]
If there is something between them then it should be kept.
I've searched a lot and played around with a regex tester but haven't come up with anything that works right.
Edit: I realize now why I was having a hard time with this. In addition to the example above, I also have one's like:
[url=http://www.somedomain.com/][/url]
I'm trying to cleanup bbcode when a form is submitted so it's not stored since it's unneeded.
In Javascript, you could do :
str.replace(/\[([^\[\]]*)\]\[\/\1\]/g, '');
The operative aspect of regex in this case is the use of internal backrefs; I'm not sure, off the top of my head, whether this is universally supported, but .NET, in any case, seems to use PCRE (is this true?).
The pattern, then, is [, a word, ][/, the same word, ]. If we assume the word has simply the quality of "does not contain ]", then an appropriate regex to match an empty tag is \[([^\]]+)\]\[/\1\], escaped as necessary in context.
For the second case, if assume the form [tag=arg][/tag], and that tag and arg each don't contain any ']' (not a reasonable assumption! but dealing with it is left as an exercise for the reader -- and I'm quite sure most bbcode implementations don't actually deal with that problem, either), one could use a regex \[([^\]=]+)(=[^\]]*)?\]\[/\1\].
I'm fairly new to figuring out how Regex works, but this one is just frustrating.
I have a massive XML document with a lot of <description>blahblahblah</description> tags. I want to basically remove any and all instances of <description></description>.
I'm using Eclipse and have tried a few examples of Regex I've found online, but nothing works.
<description>(.*?)</description>
Shouldn't that work?
EDIT:
Here is the actual code.
<description><![CDATA[<center><table><tr><th colspan='2' align='center'><em>Attributes</em></th></tr><tr bgcolor="#E3E3F3"><th>ID</th><td>308</td></tr></table></center>]]></description>
I'm not familiar with Eclipse, but I would expect its regex search facility to use Java's built-in regex flavor. You probably just need to check a box labeled "DOTALL" or "single-line" or something similar, or you can add the corresponding inline modifier to the regex:
(?s)<description>(.*?)</description>
That will allow the . to match newlines, which it doesn't by default.
EDIT: This is assuming there are newlines within the <description> element, which is the only reason I can think of why your regex wouldn't work. I'm also assuming you really are doing a regex search; is that automatic in Eclipse, or do you have to choose between regex and literal searching?