Debian Code Search doesn’t crawl, but it indexes all software in Debian, which is typically helpful enough :)

I guess code search is just too niche with no marketability to ever improve

There's also at least one version of the OpenCart admin search which was designed so poorly that it didn't actually search based on whether the term was anywhere in the page title/content, but whether the title started with the search term (I think they'd screwed up the MySQL syntax). That was interesting.

Either way, search on a system not designed as a search engine is usually okay at best, absolutely terrible at worst.

                /* evil! how this this happen */                                        \

I'd hate to be the poor soul who encounters this error message and finds no information about it except the source code.

Specifically, the branch is only reachable if the PID is not what was expected, or the location that the macro was supposed to longjmp to is not set, which would be fatal defects.

You can still use it as an outsider to search Chromium:

https://cs.chromium.org/

https://cs.chromium.org/search/?q=%5Cb%5BA-Z0-9._%25%2B-%5D%...

Considering that one can't 'index by regex', I wonder how the backend works - does it actually run the regex against every source code line in the repository in a few hundred milliseconds? Sounds expensive! They must have thousands of machines! Maybe thats why they don't have a public codesearch tool for the whole internet...

A lot of ordinary British things are seen in the U.S. through a filter of... let's call it "quaintness."

Swear words. Thatched roofs. Music on AM radio.

So, yeah. Whatever offensive word you heard probably didn't have the same impact on the other side of the Atlantic. There's probably some reciprocity, too.

Brits have their own quaintness filter for some of the words that NorthAms see as out of bounds.

https://cs.chromium.org/