I need to scan text files and pull their text, but there are certain sections I want to ignore.
The sections to ignore start with:
--- start private ---
and end with:
--- end private ---
I don’t mind if the start and end phrases are included or not.
Example
test1
test2
--- start private ---
this text should not be returned by the regex
--- end private ---
test3
test4
--- start private ---
this text should also not be returned by the regex
--- end private ---
test5
Only the lines starting with “test” should be returned.
I have the .net 6 flavour of Regex to do this with, but I don’t have access to the .net code. I can only use regex to get what I need
12
The canonical approach in .NET would be to use Regex.Replace to replace everything that starts with --- start private --- and ends with --- end private --- with an empty string, leaving just the lines you want.
However, you mention in your question that you cannot use Regex.Replace, you can just parameterize an existing call to Regex.Match. So we need to get creative.
This should do it:
(?s)(^|--- end private ---).*?(--- start private ---|$)
It matches everything that
- starts at the start of the document (
^) or with--- end private ---and - ends with
--- start private ---or the end of the document ($).
(?s) is the single line flag to ensure that newlines are also matched by ..
The question mark at the end of .*? ensures that the match is non-greedy. Otherwise the regex would just match everything from the start to the end of the document.
If you want to exclude the separator lines, you can use look-ahead/look-behind assertions for matching the separator texts:
(?s)(^|(?<=--- end private ---)).*?((?=--- start private ---)|$)
3