# compression

> written against zig 0.16.

## std.compress.zstd (decode)

zig std can decompress zstandard — no vendored libzstd needed for
plain-frame decode (this is what let zat consume jss archive blocks with
zero new dependencies):

```zig
var in: std.Io.Reader = .fixed(compressed);
var d: std.compress.zstd.Decompress = .init(&in, &.{}, .{});
_ = try d.reader.streamRemaining(&allocating_writer.writer);
```

**measure before shipping it on a hot path**: on real jss archive blocks
(atproto-bench jss lane, 58 frames / 61 MB decompressed) std's decoder took
~410 ms where libzstd took 22 ms — ~20× — and the gap survived every
buffering strategy (allocating writer, reused window + discard,
preallocated). zat ended up vendoring facebook/zstd v1.5.7 decode-only
(10 C files, ~150-line build.zig addition, cross-compiles) and got 2× ahead
of the go reference. std zstd is fine when the wire is the bottleneck.

sharp edges, verified in the 0.16 source:

- **decode only.** there is no compressor in std.
- **`verify_checksum: true` panics** — "not implemented yet" in
  `Decompress.zig`. frame content CRCs are silently skipped, so integrity
  must come from an outer layer (zat uses the jss segment xxh3 checksum;
  `std.hash.XxHash3` has streaming init/update/final).
- **no dictionary support** (`DictionaryIdFlagUnsupported`). fine for
  plain frames; not for dictionary-compressed streams like jetstream
  subscribe-v2 wire compression.
- with a non-empty buffer, `init` asserts capacity ≥
  `options.window_len + zstd.block_size_max`; the empty-buffer + writer
  form (above) is the simple path for buffered whole-frame decode.

## sources
- zat — vendored zstd v1.5.7 decode-only, jss archive block decode