May 1st @ 1:30 PM
Duration: 50 minutes
Talk type: Presentation
Level: Intermediate, Advanced
The talk gives an up-close view of how a regular expression libraries work behind the scenes, dispelling a few common myths in the process. The talk consists of two parts with equatorial line being between the generic and the D specific.
The first part is intentionally language agnostic and focuses on the core principles behind the regular expressions. The key point of this part is the two fundamental ways of matching regular expression, their respective strengths and weaknesses.
The second part dives into how all of this fits into std.regex, a modern regular expression library written in D. Important practical considerations are discussed, and some unique D features that aid such an undertaking are highlighted. Among other things this includes an in-depth overview of meta-programming tricks used to generate optimized machine code for a regular expression during the compile-time (phew!).
Benefits: Acquire the knowledge of how regular expressions work from the inside out. It will greatly aid in furthering the understanding of how to use them and when not to use them. The last but not least it's an opportunity to learn how to take advantage of the meta-programming in D.
Speaker Bio: I'm Dmitry Olshansky, a young all-around researcher. My interests are wildly changing over the time and include text processing, compilers, robotics, embedded systems, parallel and concurrent programming, scalable network systems and the current pursuit being AI. Among my works the most publicly recognized today is the standard regular expression library for D (a.k.a. std.regex).