fix srcset parsing

3 years ago · 614a518475
parent 0533b287b7
commit 614a518475
4 changed files with 81 additions and 5 deletions
--- a/Cargo.lock
+++ b/Cargo.lock
@ -5,6 +5,14 @@ name = "adler"
 version = "0.2.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"

+[[package]]
+name = "aho-corasick"
+version = "0.7.15"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+dependencies = [
+ "memchr 2.3.3 (registry+https://github.com/rust-lang/crates.io-index)",
+]
+
 [[package]]
 name = "ansi_term"
 version = "0.11.0"
@ -634,6 +642,7 @@ dependencies = [
 "clap 2.33.3 (registry+https://github.com/rust-lang/crates.io-index)",
 "cssparser 0.27.2 (registry+https://github.com/rust-lang/crates.io-index)",
 "html5ever 0.24.1 (registry+https://github.com/rust-lang/crates.io-index)",
+ "regex 1.4.2 (registry+https://github.com/rust-lang/crates.io-index)",
 "reqwest 0.10.9 (registry+https://github.com/rust-lang/crates.io-index)",
 "sha2 0.9.2 (registry+https://github.com/rust-lang/crates.io-index)",
 "tempfile 3.1.0 (registry+https://github.com/rust-lang/crates.io-index)",
@ -1064,6 +1073,22 @@ name = "redox_syscall"
 version = "0.1.57"
 source = "registry+https://github.com/rust-lang/crates.io-index"

+[[package]]
+name = "regex"
+version = "1.4.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+dependencies = [
+ "aho-corasick 0.7.15 (registry+https://github.com/rust-lang/crates.io-index)",
+ "memchr 2.3.3 (registry+https://github.com/rust-lang/crates.io-index)",
+ "regex-syntax 0.6.21 (registry+https://github.com/rust-lang/crates.io-index)",
+ "thread_local 1.0.1 (registry+https://github.com/rust-lang/crates.io-index)",
+]
+
+[[package]]
+name = "regex-syntax"
+version = "0.6.21"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+
 [[package]]
 name = "remove_dir_all"
 version = "0.5.3"
@ -1304,6 +1329,14 @@ dependencies = [
 "unicode-width 0.1.8 (registry+https://github.com/rust-lang/crates.io-index)",
 ]

+[[package]]
+name = "thread_local"
+version = "1.0.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+dependencies = [
+ "lazy_static 1.4.0 (registry+https://github.com/rust-lang/crates.io-index)",
+]
+
 [[package]]
 name = "time"
 version = "0.1.43"
@ -1623,6 +1656,7 @@ dependencies = [

 [metadata]
 "checksum adler 0.2.3 (registry+https://github.com/rust-lang/crates.io-index)" = "ee2a4ec343196209d6594e19543ae87a39f96d5534d7174822a3ad825dd6ed7e"
+"checksum aho-corasick 0.7.15 (registry+https://github.com/rust-lang/crates.io-index)" = "7404febffaa47dac81aa44dba71523c9d069b1bdc50a77db41195149e17f68e5"
 "checksum ansi_term 0.11.0 (registry+https://github.com/rust-lang/crates.io-index)" = "ee49baf6cb617b853aa8d93bf420db2383fab46d314482ca2803b40d5fde979b"
 "checksum assert_cmd 1.0.1 (registry+https://github.com/rust-lang/crates.io-index)" = "c88b9ca26f9c16ec830350d309397e74ee9abdfd8eb1f71cb6ecc71a3fc818da"
 "checksum async-compression 0.3.5 (registry+https://github.com/rust-lang/crates.io-index)" = "9021768bcce77296b64648cc7a7460e3df99979b97ed5c925c38d1cc83778d98"
@ -1749,6 +1783,8 @@ dependencies = [
 "checksum rand_xorshift 0.1.1 (registry+https://github.com/rust-lang/crates.io-index)" = "cbf7e9e623549b0e21f6e97cf8ecf247c1a8fd2e8a992ae265314300b2455d5c"
 "checksum rdrand 0.4.0 (registry+https://github.com/rust-lang/crates.io-index)" = "678054eb77286b51581ba43620cc911abf02758c91f93f479767aed0f90458b2"
 "checksum redox_syscall 0.1.57 (registry+https://github.com/rust-lang/crates.io-index)" = "41cc0f7e4d5d4544e8861606a285bb08d3e70712ccc7d2b84d7c0ccfaf4b05ce"
+"checksum regex 1.4.2 (registry+https://github.com/rust-lang/crates.io-index)" = "38cf2c13ed4745de91a5eb834e11c00bcc3709e773173b2ce4c56c9fbde04b9c"
+"checksum regex-syntax 0.6.21 (registry+https://github.com/rust-lang/crates.io-index)" = "3b181ba2dcf07aaccad5448e8ead58db5b742cf85dfe035e2227f137a539a189"
 "checksum remove_dir_all 0.5.3 (registry+https://github.com/rust-lang/crates.io-index)" = "3acd125665422973a33ac9d3dd2df85edad0f4ae9b00dafb1a05e43a9f5ef8e7"
 "checksum reqwest 0.10.9 (registry+https://github.com/rust-lang/crates.io-index)" = "fb15d6255c792356a0f578d8a645c677904dc02e862bebe2ecc18e0c01b9a0ce"
 "checksum ryu 1.0.5 (registry+https://github.com/rust-lang/crates.io-index)" = "71d301d4193d031abdd79ff7e3dd721168a9572ef3fe51a1517aba235bd8f86e"
@ -1774,6 +1810,7 @@ dependencies = [
 "checksum tempfile 3.1.0 (registry+https://github.com/rust-lang/crates.io-index)" = "7a6e24d9338a0a5be79593e2fa15a648add6138caa803e2d5bc782c371732ca9"
 "checksum tendril 0.4.1 (registry+https://github.com/rust-lang/crates.io-index)" = "707feda9f2582d5d680d733e38755547a3e8fb471e7ba11452ecfd9ce93a5d3b"
 "checksum textwrap 0.11.0 (registry+https://github.com/rust-lang/crates.io-index)" = "d326610f408c7a4eb6f51c37c330e496b08506c9457c9d34287ecc38809fb060"
+"checksum thread_local 1.0.1 (registry+https://github.com/rust-lang/crates.io-index)" = "d40c6d1b69745a6ec6fb1ca717914848da4b44ae29d9b3080cbee91d72a69b14"
 "checksum time 0.1.43 (registry+https://github.com/rust-lang/crates.io-index)" = "ca8a50ef2360fbd1eeb0ecd46795a87a19024eb4b53c5dc916ca1fd95fe62438"
 "checksum tinyvec 0.3.3 (registry+https://github.com/rust-lang/crates.io-index)" = "53953d2d3a5ad81d9f844a32f14ebb121f50b650cd59d0ee2a07cf13c617efed"
 "checksum tokio 0.2.22 (registry+https://github.com/rust-lang/crates.io-index)" = "5d34ca54d84bf2b5b4d7d31e901a8464f7b60ac145a284fba25ceb801f2ddccd"
--- a/Cargo.toml
+++ b/Cargo.toml
@ -17,6 +17,7 @@ chrono = "0.4.19" # Used for formatting creation timestamp
 clap = "2.33.3"
 cssparser = "0.27.2"
 html5ever = "0.24.1"
+regex = "1.4.2" # Used for parsing srcset
 sha2 = "0.9.2" # Used for calculating checksums during integrity checks
 url = "2.2.0"

--- a/src/html.rs
+++ b/src/html.rs
@ -7,6 +7,7 @@ use html5ever::serialize::{serialize, SerializeOpts};
 use html5ever::tendril::{format_tendril, TendrilSink};
 use html5ever::tree_builder::{Attribute, TreeSink};
 use html5ever::{local_name, namespace_url, ns, LocalName};
+use regex::Regex;
 use reqwest::blocking::Client;
 use reqwest::Url;
 use sha2::{Digest, Sha256, Sha384, Sha512};
@ -156,8 +157,8 @@ pub fn embed_srcset(
    depth: u32,
 ) -> String {
    let mut array: Vec<SrcSetItem> = vec![];
-    let srcset_items: Vec<&str> = srcset.split(',').collect();
-    for srcset_item in srcset_items {
+    let re = Regex::new(r",\s+").unwrap();
+    for srcset_item in re.split(srcset) {
        let parts: Vec<&str> = srcset_item.trim().split_whitespace().collect();
        if parts.len() > 0 {
            let path = parts[0].trim();
--- a/src/tests/html/embed_srcset.rs
+++ b/src/tests/html/embed_srcset.rs
@ -24,8 +24,45 @@ mod passing {
        let embedded_css = html::embed_srcset(cache, &client, "", &srcset_value, &options, 0);

        assert_eq!(
+            embedded_css,
            format!("{} 1x, {} 2x", empty_image!(), empty_image!()),
-            embedded_css
+        );
+    }
+
+    #[test]
+    fn commas_within_file_names() {
+        let cache = &mut HashMap::new();
+        let client = Client::new();
+        let srcset_value = "small,s.png 1x, large,l.png 2x";
+        let mut options = Options::default();
+        options.no_images = true;
+        options.silent = true;
+        let embedded_css = html::embed_srcset(cache, &client, "", &srcset_value, &options, 0);
+
+        assert_eq!(
+            embedded_css,
+            format!("{} 1x, {} 2x", empty_image!(), empty_image!()),
+        );
+    }
+
+    #[test]
+    fn tabs_and_newlines_after_commas() {
+        let cache = &mut HashMap::new();
+        let client = Client::new();
+        let srcset_value = "small,s.png 1x,\nmedium,m.png 2x,\nlarge,l.png 3x";
+        let mut options = Options::default();
+        options.no_images = true;
+        options.silent = true;
+        let embedded_css = html::embed_srcset(cache, &client, "", &srcset_value, &options, 0);
+
+        assert_eq!(
+            embedded_css,
+            format!(
+                "{} 1x, {} 2x, {} 3x",
+                empty_image!(),
+                empty_image!(),
+                empty_image!()
+            ),
        );
    }
 }
@ -56,8 +93,8 @@ mod failing {
        let embedded_css = html::embed_srcset(cache, &client, "", &srcset_value, &options, 0);

        assert_eq!(
-            format!("{} 1x, {} 2x", empty_image!(), empty_image!()),
-            embedded_css
+            embedded_css,
+            format!("{} 1x, {} 2x,", empty_image!(), empty_image!()),
        );
    }
 }