Match all characters in a line between matching symbols or end of the row

Question

I want to match on all the words between the | and the end of the row (VEHICLE and all the Vechicle names). I don't want the extra lines matched

I've started with this regex but this wouldn't exclude the symbol |:

^\|.*$

So then I tried this syntax, but it leaves off the last word on the line:

(?<=\|)(.*)(?=\|)

Samples:

Text above I don't want matched

| VEHICLE | Truck | | Bike
| VEHICLE | Car          | | Scooter
| VEHICLE | Sedan | Mini Van    | 
| VEHICLE | Sedan | white, brown, black |     
| VEHICLE | Sedan | pack/cars   |

Text below that I don't want matched

I see that you have given some examples. Can you tell me which bits you want matched? — Sweeper, Nov 30 '18 at 08:13

score 1 · Answer 1 · answered Nov 30 '18 at 08:13

Repeat any character but a | in the middle, and in the lookahead, alternate between the | and the end of the line:

(?<=\|)([^\|]+)(?=\||$)

https://regex101.com/r/YmtEPE/1

Note that there probably isn't any need for the capturing group in the middle, it's equivalent to the entire match anyway.

score 1 · Answer 2 · answered Nov 30 '18 at 08:15

1

You may use

[^|\s](?:[^|]*[^|\s])?

See the regex demo

Details

[^|\s] - any char but | and whitespace
(?:[^|]*[^|\s])? - an optional sequence of any 0+ chars other than | followed with any char but | and whitespace.

answered Nov 30 '18 at 08:15

Wiktor Stribiżew

607,720
39
448
563

Your solution works great for this section of text but I have regular text in other rows in the same file. How would I modify the text you provided above to include ONLY lines that start with `|`?> – Tommy Wu Nov 30 '18 at 21:54
@TommyWu You probably might want `(?:\G(?!\A)\h*\|\h*|^\|\h*)\K(?:[^|\s](?:[^|]*[^|\s])?)?`, see [demo](https://regex101.com/r/e2Cuit/2). – Wiktor Stribiżew Dec 01 '18 at 15:22

vrintle · Answer 3 · 2018-12-01T10:00:54.087

1

You can go with:

(?<=\| *)[^|\s]+(?= *\||$)

Details:

[^|\s]+ - matches all characters except white-spaces and | 1+ times.
$ - matches end of a string (or a line when m flag is enabled).

Demo here

Update:

(?<=\| *)(?! +)[^|]+(?= *\||$)

edited Dec 01 '18 at 10:00

answered Nov 30 '18 at 08:24

vrintle

5,501
2
16
46

Your syntax seemed to work great but I noticed 2 issues: 1) It doesn't allow tab whitespace after any of the words 2) Looks like Sublime's Syntax highlighter (YAML) won't accept it. Any ideas? – Tommy Wu Nov 30 '18 at 22:29

The fourth bird · Accepted Answer · 2018-12-01T00:25:45.627

1

To match only the words, you could first match a pipe and a whitespace character. Then capture in a group one or more word characters followed by using a positive lookahead to check what is on the right side is either a whitespace character or the end of the string.

Your values are in the first capturing group.

\|\s(\w+)(?=\s\||$)

Explanation

\|\s Match | and a whitespace character
(\w+) Capture in a group 1+ whitespace characters
(?=\s\||$) Positive lookahead to assert what follows is either a whitespace character of the end of the string

Regex demo

Update:

To only match the words in Sublime including matching 1+ tabs you could use:

\|\h+\K\w+(?:[,\/]?\h*\w+)*(?=\h*(?:\||$))

Regex demo

Explanation

\|\h+ Match a pipe followed by 1+ horizontal whitespace characters
\K Forget what was currently matched
\w+ Match 1+ word characters
(?: Non capturing group
- [,\/]?\h*\w+ Match an optional comma or a forward slash followed by 0+ times a horizontal whitespace character and 1+ word characters
)* Close non capturing group and repeat 0+ times
(?= Positive lookahead to assert what follows is
- \h* Match 0+ horizontal whitespace characters
- (?:\||$) Match either a pipe or the end of the string
) Close positive lookahead

edited Dec 01 '18 at 00:25

answered Nov 30 '18 at 19:07

The fourth bird

154,723
16
55
70

The answer almost works but has 2 issues: 1. It doesn't capture words that are followed by tabs 2. It includes the `|`, which I want to exclude since I'm using this in Sublime syntax file – Tommy Wu Nov 30 '18 at 22:42
Apologies, realized another complication in my sample data that I should have called out. Sometimes the words to be captured are more than 1 word and/or are a comma separated list: a) Mini Van b) car, van, truck c) pack/cars How do I modify the \w+ line to capture these use cases – Tommy Wu Nov 30 '18 at 23:23
@TommyWu Do you mean like this? `\|\h+\K\w+(?:[,\/]?\h*\w+)*(?=\h*(?:\||$))` [See demo](https://regex101.com/r/Pbnhwb/1) – The fourth bird Nov 30 '18 at 23:27
that's exactly right, thank you! Now I just have to figure out why it's not working in my Sublime syntax highlighter. – Tommy Wu Nov 30 '18 at 23:38
@TommyWu When you use find and enable the regex, do you see your matches? – The fourth bird Nov 30 '18 at 23:46
while I see matches when I test out my data with regex101.com I do not see any matches in Sublime. Oddly, Sublime will save the changes I made to the syntax file, which usually is an indication it accepts the syntax. – Tommy Wu Nov 30 '18 at 23:49
@TommyWu That means that the regex works but you don't see the matches? I see the matches when I run this on ubuntu with Sublime version 3.1.1 and version 2.0.2 – The fourth bird Dec 01 '18 at 00:03
1

Sorry I misunderstood what you said. Yes, Sublime Search with regex will find the correct matches. The syntax file (YAML format) does not seem to find any matches – Tommy Wu Dec 01 '18 at 00:04
@TommyWu I see you asked a [new question](https://stackoverflow.com/questions/53567446/sublime-custom-syntax-highlighter-not-matching-the-same-as-search-with-regex) and when I read [this page](https://stackoverflow.com/questions/35343933/type-of-regex-used-by-sublime-text-3), I think the `\h` is not supported. You could use a character class which will match a space or a tab like `[ \t]` For example `\|[ \t]+\K\w+[^|[ \t]]+(?:[,\/]?[ \t]*\w+)*(?=[ \t]*(?:\||$))` [Regex demo](http://rubular.com/r/aN3QRRkWFj) – The fourth bird Dec 01 '18 at 09:58

Match all characters in a line between matching symbols or end of the row

4 Answers4