Increase capnproto size limit

Bump limit for Mash

Make Clippy happy

Fix comment. Decrease limit

Author: audy

README.md

@@ -12,7 +12,17 @@ This repository provides a library and command-line interface that reimplements
 You may build Finch from source, which requires Rust >= `1.49`. Rust's Cargo package manager (see [rustup](https://www.rustup.rs) for Cargo installation instructions) can automatically build and install Finch with `cargo install finch_cli`.
 If you require python bindings, you must take extra steps (see [python support](#python-support)). Alternatively, [download a prebuilt binary](https://github.com/onecodex/finch-rs/releases) or install from [PyPi](https://pypi.org/project/finch-sketch/) `pip install finch-sketch`.
 
+### Building ###
+
+To compile from source:
+
+```sh
+# from finch-rs
+cargo +nightly-2023-06-28 build --release
+```
+
 ### Example Usage ###
+
 To get started, we first compute sketches for several FASTA or FASTQ files. These sketches are compact, sampled representations of the underlying genomic data, and what allow `finch` to rapidly estimate distances between datasets. Sketching files uses the `finch sketch` command:
 
 ```

cli/Cargo.toml

@@ -16,7 +16,7 @@ path = "src/main.rs"
 
 [dependencies]
 clap = "2.33.0"
-finch = "0.6"
+finch = { path = "../lib" }
 serde_json = "1"
 anyhow = "1"
 

lib/src/serialization/json.rs

@@ -9,7 +9,6 @@ use serde::{Deserialize, Serialize};
 use crate::bail;
 use crate::errors::FinchResult;
 use crate::filtering::FilterParams;
-pub use crate::serialization::mash::{read_mash_file, write_mash_file};
 use crate::serialization::Sketch;
 use crate::sketch_schemes::{KmerCount, SketchParams};
 

lib/src/serialization/mash.rs

@@ -58,7 +58,11 @@ pub fn write_mash_file(file: &mut dyn Write, sketches: &[Sketch]) -> FinchResult
 }
 
 pub fn read_mash_file(file: &mut dyn BufRead) -> FinchResult<Vec<Sketch>> {
-    let options = *message::ReaderOptions::new().traversal_limit_in_words(Some(1024 * 1024 * 1024));
+    let options = *message::ReaderOptions::new().traversal_limit_in_words(
+        // measured in words
+        // 1 word = 8 bytes
+        Some(2 * 1024 * 1024 * 1024),
+    );
     let reader = capnp_serialize::read_message(file, options)?;
     let mash_data: min_hash::Reader = reader.get_root::<min_hash::Reader>()?;
 

lib/src/serialization/mod.rs

@@ -171,7 +171,11 @@ pub fn write_finch_file(file: &mut dyn Write, sketches: &[Sketch]) -> FinchResul
 }
 
 pub fn read_finch_file(file: &mut dyn BufRead) -> FinchResult<Vec<Sketch>> {
-    let options = *message::ReaderOptions::new().traversal_limit_in_words(Some(1024 * 1024 * 1024));
+    let options = *message::ReaderOptions::new().traversal_limit_in_words(
+        // measured in words
+        // 1 word = 8 bytes
+        Some(2 * 1024 * 1024 * 1024),
+    );
     let reader = capnp_serialize::read_message(file, options)?;
     let cap_data: multisketch::Reader = reader.get_root::<multisketch::Reader>()?;
     let cap_sketches = cap_data.get_sketches()?;