The list stays here, but the rule changes: read after retrieval exposes a gap, not before the work starts.
This is the spine. Labs and reviews are the reps around it.
Read narrowly after the matching lab exposes a gap. The sequence matters more than volume.
Read: ownership, borrowing, slices, enums, pattern matching, error handling, iterators, tests, I/O.
Do alongside the book. This turns passive Rust knowledge into muscle memory.
Good second pass if ownership/lifetimes still feel slippery.
Directly relevant: args, errors, stdin/files, stdout, testing CLI behavior.
Enough for parse-quote [-r] input.pcap.
Pick one:
Best broad book for Rust as a systems language.
Better after you already know Rust basics; useful for API design, traits, lifetimes, testing.
Also review slices, BufReader, BufWriter, Read, and Write in std docs.
Or use std from_be_bytes / from_le_bytes.
Only as a later reference. For the Tsuru sample, plain checked slicing is probably better.
Core PCAP file-format reference.
Useful explanation of variants, timestamps, link types.
Use this if packet headers, ports, byte order, and network layers are not second nature.
Learn priority queues thoroughly.
Reference only: heaps, sorting, asymptotic analysis. Do not try to read the whole book now.
Important mental model: what changes when input exceeds memory.
Must read after your first working implementation.
For microbenchmarks.
Pair with perf to profile where time actually goes.
System profiler reference for CPU and kernel-level behavior.
Good for malformed packet/property tests: parser never panics, short payload rejected, encoded/decode fields agree.
Start practical. The Tsuru packet spec is primary; real feed specs add context after that.
Treat the packet spec as your primary exchange spec.
Good real-world example of direct market-data feed design.
Useful for seeing UDP multicast, binary encoding, sequence/recovery concepts.
Another real-world feed-spec reference.
Best first market microstructure book. Read parts on limit orders, quotes, bid/ask spread, order precedence, liquidity.
More practitioner-oriented; useful for order types, execution, DMA, infrastructure.
Optional academic depth. Not necessary for the code sample.
Use these only after the parser is correct, tested, and measured.
Especially memory hierarchy, I/O, network programming, performance.
Memory, files, concurrency.
Heavy reference. Useful later, not needed immediately.