handle archives

CHANGELOG.md

@@ -9,8 +9,19 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ### Added
 
+- Added a method `Cache::cached_path_with_options` and a corresponding `Options` struct.
+- Added ability to automatically extract archives through the `Cache::cached_path_with_options` method.
 - Added integration tests.
 
+### Changed
+
+- `Meta` struct is no longer public.
+- `Cache::cached_path_in_subdir` is now deprecated.
+
+### Removed
+
+- Removed the `only_keep_latest` setting for the `Cache`.
+
 ## [v0.4.2](https://github.com/epwalsh/rust-cached-path/releases/tag/v0.4.3) - 2020-09-11
 
 ### Changed

Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "cached-path"
-version = "0.4.3"
+version = "0.4.4"
 authors = ["epwalsh <epwalsh10@gmail.com>"]
 edition = "2018"
 keywords = ["http", "caching"]
@@ -32,6 +32,10 @@ serde_json = "1.0"
 rand = "0.7"
 glob = "0.3"
 thiserror = "1.0"
+flate2 = "1.0"
+tar = "0.4"
+zip = "0.5"
+zip-extensions = "0.5"
 env_logger = { version = "0.7", optional = true }
 structopt = { version = "0.3", optional = true }
 anyhow = { version = "1.0", optional = true }

README.md

@@ -32,8 +32,9 @@ cached version:
 ```rust
 use cached_path::cached_path;
 
-let path =
-cached_path("https://github.com/epwalsh/rust-cached-path/blob/master/README.md").unwrap();
+let path = cached_path(
+    "https://github.com/epwalsh/rust-cached-path/blob/master/README.md"
+).unwrap();
 assert!(path.is_file());
 ```
 
@@ -54,13 +55,32 @@ assert_eq!(path.to_str().unwrap(), "README.md");
 
 ```bash
 # From the command line:
-$ cached-path https://github.com/epwalsh/rust-cached-path/blob/master/README.md
+$ cached-path README.md
+README.md
+```
+
+For resources that are archives, like `*.tar.gz` files, `cached-path` can also
+automatically extract the files:
+
+```rust
+use cached_path::{cached_path_with_options, Options};
+
+let path = cached_path_with_options(
+    "https://raw.githubusercontent.com/epwalsh/rust-cached-path/master/test_fixtures/utf-8_sample/archives/utf-8.tar.gz",
+    &Options::default().extract(),
+).unwrap();
+assert!(path.is_dir());
+```
+
+```bash
+# From the command line:
+$ cached-path --extract https://raw.githubusercontent.com/epwalsh/rust-cached-path/master/test_fixtures/utf-8_sample/archives/utf-8.tar.gz
 README.md
 ```
 
-It's easy to customize the cache location, the HTTP client, and other options
+It's also easy to customize the cache location, the HTTP client, and other options
 using a [`CacheBuilder`](https://docs.rs/cached-path/*/cached_path/struct.CacheBuilder.html) to construct a custom
-[`Cache`](https://docs.rs/cached-path/*/cached_path/struct.Cache.html) object. This is also the recommended thing
+[`Cache`](https://docs.rs/cached-path/*/cached_path/struct.Cache.html) object. This is the recommended thing
 to do if your application makes multiple calls to `cached_path`, since it avoids the overhead
 of creating a new HTTP client on each call:
 

src/archives.rs

@@ -0,0 +1,56 @@
+use crate::error::Error;
+use flate2::read::GzDecoder;
+use std::fs::{self, File};
+use std::path::Path;
+use tempfile::tempdir_in;
+use zip_extensions::read::zip_extract;
+
+/// Supported archive types.
+pub(crate) enum ArchiveFormat {
+    TarGz,
+    Zip,
+}
+
+impl ArchiveFormat {
+    /// Parse archive type from resource extension.
+    pub(crate) fn parse_from_extension(resource: &str) -> Result<Self, Error> {
+        if resource.ends_with(".tar.gz") {
+            Ok(Self::TarGz)
+        } else if resource.ends_with(".zip") {
+            Ok(Self::Zip)
+        } else {
+            Err(Error::ExtractionError("unsupported archive format".into()))
+        }
+    }
+}
+
+pub(crate) fn extract_archive<P: AsRef<Path>>(
+    path: P,
+    target: P,
+    format: &ArchiveFormat,
+) -> Result<(), Error> {
+    // We'll first extract to a temp directory in the same parent as the target directory.
+    let target_parent_dir = target.as_ref().parent().unwrap();
+    let temp_target = tempdir_in(target_parent_dir)?;
+
+    match format {
+        ArchiveFormat::TarGz => {
+            let tar_gz = File::open(path)?;
+            let tar = GzDecoder::new(tar_gz);
+            let mut archive = tar::Archive::new(tar);
+            archive.unpack(&temp_target)?;
+        }
+        ArchiveFormat::Zip => {
+            zip_extract(
+                &path.as_ref().to_path_buf(),
+                &temp_target.path().to_path_buf(),
+            )
+            .map_err(|e| Error::ExtractionError(format!("{:?}", e)))?;
+        }
+    };
+
+    // Now rename the temp directory to the final target directory.
+    fs::rename(temp_target, target)?;
+
+    Ok(())
+}