Bootstrapped by Boost

on YouTube ↗01h 23m

" summary

A real-world C++ search engine for unstructured medical text is built largely with Boost libraries: graph processing, parsing (Boost.Spirit), memory-mapped resources (Boost.Interprocess) for fast term lookups, logging, serialization, and a Python API via Boost.Python; a single-programmer codebase leverages ranges and containers heavily. Concrete issues: a link-time bug forces a Boost upgrade, Spirit grammars compile slowly but eliminate handwritten parsers, and mmap-backed terminology files deliver near-native access and actually make the system feasible.

AI-written summary · about

" chapters (8)

1. 0:00Introduction and context
2. 15:36Company and applications
3. 32:43Why we use Boost
4. 41:16Boost Spirit parser
5. 59:55Boost Python interop
6. 1:09:38Boost Interprocess demo
7. 1:14:31Missing Boost features
8. 1:20:16Q&A and closing

" related talks

• 05mLightning Talk: Searching with C++ - A Fulltext Index Search AlgorithmHana Dusikova · C++Now 2023
• 13mIntroducing Beast...Vinnie Falco · CppCon 2016
• 01h 22mGlean: C++ Code Indexing at MetaMichael Park · C++Now 2024
• 09mIntroducing brigandJoel Falcou, Edouard Alligand · CppCon 2015
• 01h 34mMetaprogramming in C++14Louis Dionne · ACCU 2017
• 01h 16mApplied C++20 CoroutinesJim Pascoe · ACCU 2023
• 59mSoftware Engineering Is About TradeoffsMateusz Pusz · NDC TechTown 2021
• 49mAn Engineering Approach to Optimising C++Dave Rowland · C++ on Sea 2023