Description

TL;DR

Thanks in advance for advices :)

• Whiskers regex - as-is, with boost::regex:

  /<([a-zA-Z0-9_$-]+)>|<#([a-zA-Z0-9_$-]+)>(.*?)</\2>|<\?([a-zA-Z0-9_$-]+)>(.*?)(<!\4>(.*?))?</\4>/g

• Whiskers regex - to-be, with std::regex:

  /<([a-zA-Z0-9_$-]+)>|<#([a-zA-Z0-9_$-]+)>((?:.|\r|\n)*?)</\2>|<\?([a-zA-Z0-9_$-]+)>((?:.|\n|\r)*?)(<!\4>((?:.|\n|\r)*?))?</\4>/g

Details

This PR replaces boost::regex with std::regex, as suggested in #6971 and #6997.
The biggest difference between boost::regex and std::regex is the behavior of dots:

• In boost::regex, .* matches over line breaks
• In std::regex, .* does not match newline characters(\r, \n).

To address the issue, a change is made to the Whiskers regular expression:

1. Replace . with non-capturing group (?:.|\r|\n)

And two compilation options are added to the MSVC case:

1. -DBOOST_REGEX_NO_LIB : It prevents automatic linking to libboost_regex.
2. -D_REGEX_MAX_STACK_COUNT=200000L: Microsoft implementation of std::regex has too tight recursion depth limit. This option increases it to 200,000 - which, IMHO, is sufficient for every sane use case.

Checklist

• Code compiles correctly
• All tests are passing
• New tests have been created which fail without the change (if possible)
• README / documentation was extended, if necessary
• Changelog entry (if change is visible to the user)
• Used meaningful commit messages

Part of #7259.