diff --git a/Cargo.lock b/Cargo.lock
index b81f71a..2931082 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -500,6 +500,12 @@ dependencies = [
"tokio-core",
]
+[[package]]
+name = "glob"
+version = "0.3.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "9b919933a397b79c37e33b77bb2aa3dc8eb6e165ad809e58ff75bc7db2e34574"
+
[[package]]
name = "gumdrop"
version = "0.8.0"
@@ -826,6 +832,7 @@ dependencies = [
"anyhow",
"futures 0.3.5",
"github-rs",
+ "glob",
"gumdrop",
"http 0.1.21",
"linkify",
diff --git a/Cargo.toml b/Cargo.toml
index 8123650..be17c5d 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -12,6 +12,7 @@ version = "0.2.0"
anyhow = "1.0.32"
futures = "0.3"
github-rs = "0.7.0"
+glob = "*"
gumdrop = "0.8.0"
http = "0.1.21"
linkify = "0.4.0"
diff --git a/README.md b/README.md
index 7e3df8e..b0e9053 100644
--- a/README.md
+++ b/README.md
@@ -26,15 +26,14 @@ lychee can...
- disguise as a different user agent (like curl)
- optionally ignore SSL certificate errors
- run with a low memory/CPU footprint
-- check multiple files at once
+- check multiple files at once (supports globbing)
- support checking links from any website URL
- limit scheme (e.g. only check HTTPS links with "https")
SOON:
- automatically retry and backoff
-- check relative and absolute paths
-- support input files using wildcards
+- check relative and absolute URLs
- set timeout for HTTP requests in seconds. Disabled by default.
- accept custom headers (see https://github.com/rust-lang/crates.io/issues/788)
- use `HEAD` requests instead of `GET` to avoid network I/O
diff --git a/fixtures/TEST.html b/fixtures/TEST.html
new file mode 100644
index 0000000..4415bbc
--- /dev/null
+++ b/fixtures/TEST.html
@@ -0,0 +1,15 @@
+
+
+ Some more test links
+
+
+
+
+
diff --git a/src/checker.rs b/src/checker.rs
index 66729ee..73a978a 100644
--- a/src/checker.rs
+++ b/src/checker.rs
@@ -151,8 +151,10 @@ impl Checker {
return true;
}
}
- if Some(url.scheme().to_string()) != self.scheme {
- return true;
+ if let Some(scheme) = &self.scheme {
+ if url.scheme() != scheme {
+ return true;
+ }
}
false
}
diff --git a/src/main.rs b/src/main.rs
index 5acad49..d4f1cb5 100644
--- a/src/main.rs
+++ b/src/main.rs
@@ -2,6 +2,7 @@
extern crate log;
use anyhow::Result;
+use glob::glob;
use regex::RegexSet;
use std::env;
use std::{collections::HashSet, fs};
@@ -96,14 +97,25 @@ async fn collect_links(inputs: Vec) -> Result> {
let mut links = HashSet::new();
for input in inputs {
- let content = match Url::parse(&input) {
+ match Url::parse(&input) {
Ok(url) => {
let res = reqwest::get(url).await?;
- res.text().await?
+ let content = res.text().await?;
+ links.extend(extract_links(&content));
+ }
+ Err(_) => {
+ // Assume we got a single file or a glob on our hands
+ for entry in glob(&input)? {
+ match entry {
+ Ok(path) => {
+ let content = fs::read_to_string(path)?;
+ links.extend(extract_links(&content));
+ }
+ Err(e) => println!("{:?}", e),
+ }
+ }
}
- Err(_) => fs::read_to_string(input)?,
};
- links.extend(extract_links(&content));
}
Ok(links)
}