Shorter readme page.

Byron · Feb 9, 2019 · 0ee3b3c · 0ee3b3c
1 parent b3ac65c
commit 0ee3b3c
Show file tree

Hide file tree

Showing 2 changed files with 25 additions and 34 deletions.
diff --git a/README.md b/README.md
@@ -1,9 +1,6 @@
 jwalk
 =======
 
-**Note** This is my first public crate. Looking testing and feedback (Code, API,
-Documentation) before posting 1.0.
-
 Fast recursive directory walk.
 
 - Walk is performed in parallel using rayon
@@ -15,19 +12,17 @@ Fast recursive directory walk.
 [![Documentation](https://docs.rs/jwalk/badge.svg)](https://docs.rs/jwalk)
 [![License](https://img.shields.io/crates/l/jwalk.svg)](https://github.com/rust-lang-nursery/jwalk.rs#license)
 
-### Documentation
-
-[docs.rs/jwalk](https://docs.rs/jwalk)
-
 ### Usage
 
-Add `jwalk` to your project's `Cargo.toml`:
+Add this to your `Cargo.toml`:
 
 ```toml
 [dependencies]
 jwalk = "0.1.0"
 ```
 
+See documentation next: [docs.rs/jwalk](https://docs.rs/jwalk).
+
 ### Example
 
 Recursively iterate over the "foo" directory sorting by name:
@@ -48,16 +43,10 @@ parallelism of `ignore` with `walkdir`s streaming iterator API.
 
 ### Why use this crate?
 
-Speed and flexibility.
-
-This crate is particularly fast when you want streamed sorted results. In
-this case it's much faster then `walkdir` and has much better latency then
-`ignore`.
-
-This crate's `process_entries` callback allows you to arbitrarily
-sort/filter/skip each directories entries before they are yielded. This
-processing happens in the thread pool and effects the directory traversal.
-It can be much faster then post processing the yielded entries.
+This crate is particularly fast when you want streamed sorted results. About 4x
+`walkdir` speed for sorted results with metadata in my tests. Also this crate's
+`process_entries` callback allows you to arbitrarily sort/filter/skip entries
+before they are yielded.
 
 ### Why not use this crate?
 
@@ -67,19 +56,5 @@ implementation.
 
 ### Benchmarks
 
-Time to walk linux's source tree:
-
-| Crate   | Options                        | Time      |
-|---------|--------------------------------|-----------|
-| jwalk   | unsorted, parallel             | 60.811 ms |
-| jwalk   | sorted, parallel               | 61.445 ms |
-| jwalk   | sorted, parallel, metadata     | 100.95 ms |
-| jwalk   | unsorted, parallel (2 threads) | 99.998 ms |
-| jwalk   | unsorted, serial               | 168.68 ms |
-| jwalk   | sorted, parallel, first 100    | 9.9794 ms |
-| ignore  | unsorted, parallel             | 74.251 ms |
-| ignore  | sorted, parallel               | 99.336 ms |
-| ignore  | sorted, parallel, metadata     | 134.26 ms |
-| walkdir | unsorted                       | 162.09 ms |
-| walkdir | sorted                         | 200.09 ms |
-| walkdir | sorted, metadata               | 422.74 ms |
+[Benchmarks](https://github.com/jessegrosjean/jwalk/blob/master/benches/benchmarks.md)
+comparing this crate with `jwalk` and `ignore`.
diff --git a/benches/benchmarks.md b/benches/benchmarks.md
@@ -0,0 +1,16 @@
+Time to walk linux's source tree:
+
+| Crate   | Options                        | Time      |
+|---------|--------------------------------|-----------|
+| jwalk   | unsorted, parallel             | 60.811 ms |
+| jwalk   | sorted, parallel               | 61.445 ms |
+| jwalk   | sorted, parallel, metadata     | 100.95 ms |
+| jwalk   | unsorted, parallel (2 threads) | 99.998 ms |
+| jwalk   | unsorted, serial               | 168.68 ms |
+| jwalk   | sorted, parallel, first 100    | 9.9794 ms |
+| ignore  | unsorted, parallel             | 74.251 ms |
+| ignore  | sorted, parallel               | 99.336 ms |
+| ignore  | sorted, parallel, metadata     | 134.26 ms |
+| walkdir | unsorted                       | 162.09 ms |
+| walkdir | sorted                         | 200.09 ms |
+| walkdir | sorted, metadata               | 422.74 ms |