Front page | perl.perl5.porters |
Postings from November 2022
The regex engine is a huge amount of code divided into several conceptual components along with tons of ancillary code, all crammed basically into two files. regcomp.c has 26k lines, and regexec.c has 11k lines. The nature of some of the code somewhat biases towards large functions, but nevertheless we have several large functions in both of the files, arguably code that could be separated into more reasonably sized compilation units. So for instance the regex compiler is divided into the toker/parser and a peephole optimizer, debug code, and ancillary logic, for instance all of the trie code could be in its own file. The regex execution engine is divided into a specialized "find the start" function and a more generalized "regex execution engine". These are related, but really need not be in the same files. Some of this code could be reused. How many people know we have a complete levenstein distance computation code hidden in the regex engine? (Besides Karl. :-) We could expose this to more areas of the code (mistyped scalars comes to mind), and maybe also via a builtin (since it is in the core why not let people use it? (for instance Merijn pointed out that cpan could use it if it was available). The counter argument for splitting the code would be that we would lose some sense of the history of the code via standard blame invocation. Blame can be told to do the work to see past these splits, but it is not always perfect at it. Personally I think having more structure would make it easier to maintain and understand, but I am not sure I fully understand all the factors that need considering when doing a reorg like this. I have noticed adding a new .c file to the code base is a pain, but it seems manageable if a bit tedius. But are there any other concerns I should know about or take into account? Anybody with experience adding a new .c file have any guidance I can follow. Does anybody have any thoughts on this? Relatively few people work on the regex code, does anybody have any strong feelings about this? Any guidance to provide? cheers, Yves -- perl -Mre=debug -e "/just|another|perl|hacker/"Thread Next