Regular expression to replace lines in a text which starts with a specific char

Question

I'm working on a regular expression but I just can't make it work.

With a text like this:

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Nunc ultricies congue feugiat. 

Tom wrote:
> blablabla
> this is very interesting 
> blabla blabla
> I ate a apple yesterday
> this is very interesting 
> blabla blabla

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Nunc ultricies congue feugiat.

I Would like to match the whole part of the text that starts with "Tom wrote:" and that ends with last line starting with a ">" in order to remove it and to get only the 2 lorem ipsum sentences.

I don't know if I'm clear enough...

Edit: I was thinking about a regexp that match everything that starts with "\w wrote: " and that stops when it matches a newline followed by a char that is not a ">"

Edit2: Found a solution :

\w+ wrote:(\n>[^\n]*)*

I tried: (\w+ wrote:)|(>.*?\n) But it doesn't take the last line if it doesn't end with a newline Plus it isn't correct as it would also remove lines with starts with a ">" but which are not preceded by "xxx wrote :" — philippe87, Oct 30 '12 at 15:26
Giving your question a good title means that other people with a similar issue can find your question and (hopefully) an answer. — Lee Taylor, Oct 30 '12 at 15:29
You may have wanted this, but your posted solution will match lines that look like "Bob wrote:" even if it has no lines starting with ">" after it. — Matt, Oct 31 '12 at 11:38

score 1 · Answer 1 · answered Oct 30 '12 at 15:27

1

Practice your regular expressions in a tool like this one. Makes it very easy to visualize what your regex is doing.

Give this one a go and adjust as necessary:

Tom wrote:(.|\s)*>.*

answered Oct 30 '12 at 15:27

ubiquibacon

10,451
28
109
179

Thank you but I already use Kodos which works perfectly fine. – philippe87 Oct 30 '12 at 15:28
What do you mean? Matches everything after what? – ubiquibacon Oct 30 '12 at 15:47
It means that it will match all lines after even if they do not start with a ">" – philippe87 Oct 31 '12 at 07:38
The regex I provided is correct, you just need a global/multi-line flag as Igor suggested. – ubiquibacon Oct 31 '12 at 16:33

score 1 · Accepted Answer · edited May 23 '17 at 12:24

1

Matching regular expressions across multiple lines requires specifying multiline matching.

See this answer for details: Regular expression matching a multiline block of text

edited May 23 '17 at 12:24

Community

1
1

answered Oct 30 '12 at 15:33

Igor Levicki

1,017
10
17

With mutliline, matching of ^ and $ is different. Try using ^ to match line which doesn't start with >. – Igor Levicki Oct 30 '12 at 15:42

Matt · Answer 3 · 2012-10-30T15:59:53.417

1

It looks like this is what you want (adjust newline characters as necessary for your system)

\w+ wrote:\n(>.*\n)*(>.*)

http://regexr.com?32l21

edited Oct 30 '12 at 15:59

answered Oct 30 '12 at 15:42

Matt

3,651
3
16
35

Thank you for your answer but I found one myself that works too :) – philippe87 Oct 31 '12 at 07:30

Regular expression to replace lines in a text which starts with a specific char

3 Answers3