Skip to content

Commit d987dbb

Browse files
Column encoding options and parallel writes (#56)
* Add options for controlling parquet encoding * Parallel writes at maximum speeeeed * Don't allow invalid encodings fs q * Improve docs for options to illuminate behavior * mega-test for safety * Handle potential deadlock
1 parent cd6e5d2 commit d987dbb

10 files changed

Lines changed: 4590 additions & 346 deletions

File tree

Cargo.lock

Lines changed: 236 additions & 36 deletions
Some generated files are not rendered by default. Learn more about customizing how changed files appear on GitHub.

Cargo.toml

Lines changed: 6 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -16,6 +16,7 @@ path = "src/main.rs"
1616

1717
[dependencies]
1818
anyhow = "^1.0.100"
19+
rayon = "^1.10"
1920
arrow = { version = "^57.0.0", features = ["default", "ipc_compression"] }
2021
clap = { version = "^4.5.53", features = ["default", "derive"] }
2122
datafusion = { git = "https://github.com/apache/datafusion.git", rev = "1cc9bcd52e5901f9085864c032c5f51a1932a075" }
@@ -48,10 +49,14 @@ arrow-schema = "56"
4849
default = []
4950

5051
[dev-dependencies]
51-
5252
assert_cmd = "^2.1.1"
53+
criterion = { version = "^0.8", features = ["html_reports"] }
5354
predicates = "^3.1.3"
5455

56+
[[bench]]
57+
name = "parquet_writer"
58+
harness = false
59+
5560
[profile.release]
5661
lto = true
5762
codegen-units = 1

0 commit comments

Comments
 (0)