diff --git a/Cargo.lock b/Cargo.lock index 3c6b01c69..11c843b60 100644 --- a/Cargo.lock +++ b/Cargo.lock @@ -346,6 +346,7 @@ dependencies = [ "iron 0.5.1 (registry+https://github.com/rust-lang/crates.io-index)", "kuchiki 0.8.0 (registry+https://github.com/rust-lang/crates.io-index)", "log 0.4.8 (registry+https://github.com/rust-lang/crates.io-index)", + "lol_html 0.2.0 (registry+https://github.com/rust-lang/crates.io-index)", "mime_guess 2.0.3 (registry+https://github.com/rust-lang/crates.io-index)", "notify 4.0.15 (registry+https://github.com/rust-lang/crates.io-index)", "once_cell 1.4.0 (registry+https://github.com/rust-lang/crates.io-index)", @@ -478,6 +479,24 @@ dependencies = [ "subtle 2.2.3 (registry+https://github.com/rust-lang/crates.io-index)", ] +[[package]] +name = "cssparser" +version = "0.25.9" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "autocfg 0.1.7 (registry+https://github.com/rust-lang/crates.io-index)", + "cssparser-macros 0.3.6 (registry+https://github.com/rust-lang/crates.io-index)", + "dtoa-short 0.3.2 (registry+https://github.com/rust-lang/crates.io-index)", + "itoa 0.4.5 (registry+https://github.com/rust-lang/crates.io-index)", + "matches 0.1.8 (registry+https://github.com/rust-lang/crates.io-index)", + "phf 0.7.24 (registry+https://github.com/rust-lang/crates.io-index)", + "proc-macro2 1.0.13 (registry+https://github.com/rust-lang/crates.io-index)", + "procedural-masquerade 0.1.7 (registry+https://github.com/rust-lang/crates.io-index)", + "quote 1.0.6 (registry+https://github.com/rust-lang/crates.io-index)", + "smallvec 0.6.13 (registry+https://github.com/rust-lang/crates.io-index)", + "syn 1.0.22 (registry+https://github.com/rust-lang/crates.io-index)", +] + [[package]] name = "cssparser" version = "0.27.2" @@ -494,6 +513,18 @@ dependencies = [ "syn 1.0.22 (registry+https://github.com/rust-lang/crates.io-index)", ] +[[package]] +name = "cssparser-macros" +version = "0.3.6" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "phf_codegen 0.7.24 (registry+https://github.com/rust-lang/crates.io-index)", + "proc-macro2 1.0.13 (registry+https://github.com/rust-lang/crates.io-index)", + "procedural-masquerade 0.1.7 (registry+https://github.com/rust-lang/crates.io-index)", + "quote 1.0.6 (registry+https://github.com/rust-lang/crates.io-index)", + "syn 1.0.22 (registry+https://github.com/rust-lang/crates.io-index)", +] + [[package]] name = "cssparser-macros" version = "0.6.0" @@ -1320,6 +1351,23 @@ dependencies = [ "cfg-if 0.1.10 (registry+https://github.com/rust-lang/crates.io-index)", ] +[[package]] +name = "lol_html" +version = "0.2.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "bitflags 1.2.1 (registry+https://github.com/rust-lang/crates.io-index)", + "cfg-if 0.1.10 (registry+https://github.com/rust-lang/crates.io-index)", + "cssparser 0.25.9 (registry+https://github.com/rust-lang/crates.io-index)", + "encoding_rs 0.8.23 (registry+https://github.com/rust-lang/crates.io-index)", + "lazy_static 1.4.0 (registry+https://github.com/rust-lang/crates.io-index)", + "lazycell 1.2.1 (registry+https://github.com/rust-lang/crates.io-index)", + "memchr 2.3.3 (registry+https://github.com/rust-lang/crates.io-index)", + "safemem 0.3.3 (registry+https://github.com/rust-lang/crates.io-index)", + "selectors 0.21.0 (registry+https://github.com/rust-lang/crates.io-index)", + "thiserror 1.0.20 (registry+https://github.com/rust-lang/crates.io-index)", +] + [[package]] name = "mac" version = "0.1.1" @@ -1966,6 +2014,11 @@ dependencies = [ "unicode-xid 0.2.0 (registry+https://github.com/rust-lang/crates.io-index)", ] +[[package]] +name = "procedural-masquerade" +version = "0.1.7" +source = "registry+https://github.com/rust-lang/crates.io-index" + [[package]] name = "procfs" version = "0.7.9" @@ -2439,6 +2492,24 @@ dependencies = [ "libc 0.2.74 (registry+https://github.com/rust-lang/crates.io-index)", ] +[[package]] +name = "selectors" +version = "0.21.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "bitflags 1.2.1 (registry+https://github.com/rust-lang/crates.io-index)", + "cssparser 0.25.9 (registry+https://github.com/rust-lang/crates.io-index)", + "fxhash 0.2.1 (registry+https://github.com/rust-lang/crates.io-index)", + "log 0.4.8 (registry+https://github.com/rust-lang/crates.io-index)", + "matches 0.1.8 (registry+https://github.com/rust-lang/crates.io-index)", + "phf 0.7.24 (registry+https://github.com/rust-lang/crates.io-index)", + "phf_codegen 0.7.24 (registry+https://github.com/rust-lang/crates.io-index)", + "precomputed-hash 0.1.1 (registry+https://github.com/rust-lang/crates.io-index)", + "servo_arc 0.1.1 (registry+https://github.com/rust-lang/crates.io-index)", + "smallvec 0.6.13 (registry+https://github.com/rust-lang/crates.io-index)", + "thin-slice 0.1.1 (registry+https://github.com/rust-lang/crates.io-index)", +] + [[package]] name = "selectors" version = "0.22.0" @@ -2594,6 +2665,14 @@ dependencies = [ "unidecode 0.2.0 (registry+https://github.com/rust-lang/crates.io-index)", ] +[[package]] +name = "smallvec" +version = "0.6.13" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "maybe-uninit 2.0.0 (registry+https://github.com/rust-lang/crates.io-index)", +] + [[package]] name = "smallvec" version = "1.4.0" @@ -2886,6 +2965,24 @@ name = "thin-slice" version = "0.1.1" source = "registry+https://github.com/rust-lang/crates.io-index" +[[package]] +name = "thiserror" +version = "1.0.20" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "thiserror-impl 1.0.20 (registry+https://github.com/rust-lang/crates.io-index)", +] + +[[package]] +name = "thiserror-impl" +version = "1.0.20" +source = "registry+https://github.com/rust-lang/crates.io-index" +dependencies = [ + "proc-macro2 1.0.13 (registry+https://github.com/rust-lang/crates.io-index)", + "quote 1.0.6 (registry+https://github.com/rust-lang/crates.io-index)", + "syn 1.0.22 (registry+https://github.com/rust-lang/crates.io-index)", +] + [[package]] name = "thread_local" version = "1.0.1" @@ -3499,7 +3596,9 @@ dependencies = [ "checksum crossbeam-queue 0.2.1 (registry+https://github.com/rust-lang/crates.io-index)" = "c695eeca1e7173472a32221542ae469b3e9aac3a4fc81f7696bcad82029493db" "checksum crossbeam-utils 0.7.2 (registry+https://github.com/rust-lang/crates.io-index)" = "c3c7c73a2d1e9fc0886a08b93e98eb643461230d5f1925e4036204d5f2e261a8" "checksum crypto-mac 0.8.0 (registry+https://github.com/rust-lang/crates.io-index)" = "b584a330336237c1eecd3e94266efb216c56ed91225d634cb2991c5f3fd1aeab" +"checksum cssparser 0.25.9 (registry+https://github.com/rust-lang/crates.io-index)" = "fbe18ca4efb9ba3716c6da66cc3d7e673bf59fa576353011f48c4cfddbdd740e" "checksum cssparser 0.27.2 (registry+https://github.com/rust-lang/crates.io-index)" = "754b69d351cdc2d8ee09ae203db831e005560fc6030da058f86ad60c92a9cb0a" +"checksum cssparser-macros 0.3.6 (registry+https://github.com/rust-lang/crates.io-index)" = "5bb1c84e87c717666564ec056105052331431803d606bd45529b28547b611eef" "checksum cssparser-macros 0.6.0 (registry+https://github.com/rust-lang/crates.io-index)" = "dfae75de57f2b2e85e8768c3ea840fd159c8f33e2b6522c7835b7abac81be16e" "checksum csv 1.1.3 (registry+https://github.com/rust-lang/crates.io-index)" = "00affe7f6ab566df61b4be3ce8cf16bc2576bca0963ceb0955e45d514bf9a279" "checksum csv-core 0.1.10 (registry+https://github.com/rust-lang/crates.io-index)" = "2b2466559f260f48ad25fe6317b3c8dac77b5bdb5763ac7d9d6103530663bc90" @@ -3594,6 +3693,7 @@ dependencies = [ "checksum lock_api 0.4.1 (registry+https://github.com/rust-lang/crates.io-index)" = "28247cc5a5be2f05fbcd76dd0cf2c7d3b5400cb978a28042abcd4fa0b3f8261c" "checksum log 0.3.9 (registry+https://github.com/rust-lang/crates.io-index)" = "e19e8d5c34a3e0e2223db8e060f9e8264aeeb5c5fc64a4ee9965c062211c024b" "checksum log 0.4.8 (registry+https://github.com/rust-lang/crates.io-index)" = "14b6052be84e6b71ab17edffc2eeabf5c2c3ae1fdb464aae35ac50c67a44e1f7" +"checksum lol_html 0.2.0 (registry+https://github.com/rust-lang/crates.io-index)" = "169299b3b58aa5cd8ad25fd8fe984e93748046d24c80f05aaadd9022f95423ec" "checksum mac 0.1.1 (registry+https://github.com/rust-lang/crates.io-index)" = "c41e0c4fef86961ac6d6f8a82609f55f31b05e4fce149ac5710e439df7619ba4" "checksum maplit 1.0.2 (registry+https://github.com/rust-lang/crates.io-index)" = "3e2e65a1a2e43cfcb47a895c4c8b10d1f4a61097f9f254f183aee60cad9c651d" "checksum markup5ever 0.10.0 (registry+https://github.com/rust-lang/crates.io-index)" = "aae38d669396ca9b707bfc3db254bc382ddb94f57cc5c235f34623a669a01dab" @@ -3668,6 +3768,7 @@ dependencies = [ "checksum proc-macro-hack 0.5.15 (registry+https://github.com/rust-lang/crates.io-index)" = "0d659fe7c6d27f25e9d80a1a094c223f5246f6a6596453e09d7229bf42750b63" "checksum proc-macro-nested 0.1.4 (registry+https://github.com/rust-lang/crates.io-index)" = "8e946095f9d3ed29ec38de908c22f95d9ac008e424c7bcae54c75a79c527c694" "checksum proc-macro2 1.0.13 (registry+https://github.com/rust-lang/crates.io-index)" = "53f5ffe53a6b28e37c9c1ce74893477864d64f74778a93a4beb43c8fa167f639" +"checksum procedural-masquerade 0.1.7 (registry+https://github.com/rust-lang/crates.io-index)" = "8f1383dff4092fe903ac180e391a8d4121cc48f08ccf850614b0290c6673b69d" "checksum procfs 0.7.9 (registry+https://github.com/rust-lang/crates.io-index)" = "c434e93ef69c216e68e4f417c927b4f31502c3560b72cfdb6827e2321c5c6b3e" "checksum prometheus 0.7.0 (registry+https://github.com/rust-lang/crates.io-index)" = "5567486d5778e2c6455b1b90ff1c558f29e751fc018130fa182e15828e728af1" "checksum quick-error 1.2.3 (registry+https://github.com/rust-lang/crates.io-index)" = "a1d01941d82fa2ab50be1e79e6714289dd7cde78eba4c074bc5a4374f650dfe0" @@ -3713,6 +3814,7 @@ dependencies = [ "checksum scopeguard 1.1.0 (registry+https://github.com/rust-lang/crates.io-index)" = "d29ab0c6d3fc0ee92fe66e2d99f700eab17a8d57d1c1d3b748380fb20baa78cd" "checksum security-framework 0.4.4 (registry+https://github.com/rust-lang/crates.io-index)" = "64808902d7d99f78eaddd2b4e2509713babc3dc3c85ad6f4c447680f3c01e535" "checksum security-framework-sys 0.4.3 (registry+https://github.com/rust-lang/crates.io-index)" = "17bf11d99252f512695eb468de5516e5cf75455521e69dfe343f3b74e4748405" +"checksum selectors 0.21.0 (registry+https://github.com/rust-lang/crates.io-index)" = "1b86b100bede4f651059740afc3b6cb83458d7401cb7c1ad96d8a11e91742c86" "checksum selectors 0.22.0 (registry+https://github.com/rust-lang/crates.io-index)" = "df320f1889ac4ba6bc0cdc9c9af7af4bd64bb927bccdf32d81140dc1f9be12fe" "checksum semver 0.9.0 (registry+https://github.com/rust-lang/crates.io-index)" = "1d7eb9ef2c18661902cc47e535f9bc51b78acd254da71d375c2f6720d9a40403" "checksum semver-parser 0.7.0 (registry+https://github.com/rust-lang/crates.io-index)" = "388a1df253eca08550bef6c72392cfe7c30914bf41df5269b68cbd6ff8f570a3" @@ -3731,6 +3833,7 @@ dependencies = [ "checksum siphasher 0.3.3 (registry+https://github.com/rust-lang/crates.io-index)" = "fa8f3741c7372e75519bd9346068370c9cdaabcc1f9599cbcf2a2719352286b7" "checksum slab 0.4.2 (registry+https://github.com/rust-lang/crates.io-index)" = "c111b5bd5695e56cffe5129854aa230b39c93a305372fdbb2668ca2394eea9f8" "checksum slug 0.1.1 (registry+https://github.com/rust-lang/crates.io-index)" = "39af1ce888a1253c8b9fcfa36626557650fb487c013620a743262d2769a3e9f3" +"checksum smallvec 0.6.13 (registry+https://github.com/rust-lang/crates.io-index)" = "f7b0758c52e15a8b5e3691eae6cc559f08eee9406e548a4477ba4e67770a82b6" "checksum smallvec 1.4.0 (registry+https://github.com/rust-lang/crates.io-index)" = "c7cb5678e1615754284ec264d9bb5b4c27d2018577fd90ac0ceb578591ed5ee4" "checksum socket2 0.3.12 (registry+https://github.com/rust-lang/crates.io-index)" = "03088793f677dce356f3ccc2edb1b314ad191ab702a5de3faf49304f7e104918" "checksum spin 0.5.2 (registry+https://github.com/rust-lang/crates.io-index)" = "6e63cff320ae2c57904679ba7cb63280a3dc4613885beafb148ee7bf9aa9042d" @@ -3761,6 +3864,8 @@ dependencies = [ "checksum termcolor 1.1.0 (registry+https://github.com/rust-lang/crates.io-index)" = "bb6bfa289a4d7c5766392812c0a1f4c1ba45afa1ad47803c11e1f407d846d75f" "checksum textwrap 0.11.0 (registry+https://github.com/rust-lang/crates.io-index)" = "d326610f408c7a4eb6f51c37c330e496b08506c9457c9d34287ecc38809fb060" "checksum thin-slice 0.1.1 (registry+https://github.com/rust-lang/crates.io-index)" = "8eaa81235c7058867fa8c0e7314f33dcce9c215f535d1913822a2b3f5e289f3c" +"checksum thiserror 1.0.20 (registry+https://github.com/rust-lang/crates.io-index)" = "7dfdd070ccd8ccb78f4ad66bf1982dc37f620ef696c6b5028fe2ed83dd3d0d08" +"checksum thiserror-impl 1.0.20 (registry+https://github.com/rust-lang/crates.io-index)" = "bd80fc12f73063ac132ac92aceea36734f04a1d93c1240c6944e23a3b8841793" "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 time 0.2.16 (registry+https://github.com/rust-lang/crates.io-index)" = "3a51cadc5b1eec673a685ff7c33192ff7b7603d0b75446fb354939ee615acb15" diff --git a/Cargo.toml b/Cargo.toml index ae52fd984..e9b155df5 100644 --- a/Cargo.toml +++ b/Cargo.toml @@ -27,7 +27,6 @@ backtrace = "0.3" failure = { version = "0.1.3", features = ["backtrace"] } comrak = { version = "0.8", default-features = false } toml = "0.5" -kuchiki = "0.8" schemamama = "0.3" schemamama_postgres = "0.3" systemstat = "0.1.4" @@ -41,6 +40,7 @@ path-slash = "0.1.3" once_cell = { version = "1.4.0", features = ["parking_lot"] } base64 = "0.12.1" strum = { version = "0.18.0", features = ["derive"] } +lol_html = "0.2" # Async tokio = { version = "0.2.22", features = ["rt-threaded"] } @@ -81,6 +81,7 @@ procfs = "0.7" [dev-dependencies] criterion = "0.3" +kuchiki = "0.8" rand = "0.7.3" [[bench]] diff --git a/src/utils/html.rs b/src/utils/html.rs index 9a24b4561..cce275a74 100644 --- a/src/utils/html.rs +++ b/src/utils/html.rs @@ -1,40 +1,64 @@ -use crate::error::Result; -use failure::err_msg; -use kuchiki::traits::TendrilSink; -use kuchiki::NodeRef; - -/// Extracts the contents of the `
` and `` tags from an HTML document, as well as the -/// classes on the `` tag, if any. -pub fn extract_head_and_body(html: &str) -> Result<(String, String, String)> { - let dom = kuchiki::parse_html().one(html); - - let head = dom - .select_first("head") - .map_err(|_| err_msg("couldn't find tag in rustdoc output"))?; - let body = dom - .select_first("body") - .map_err(|_| err_msg("couldn't find tag in rustdoc output"))?; - - let class = body - .attributes - .borrow() - .get("class") - .map(|v| v.to_owned()) - .unwrap_or_default(); - - Ok((serialize(head.as_node()), serialize(body.as_node()), class)) -} +use crate::web::page::TemplateData; +use lol_html::errors::RewritingError; +use tera::Context; -fn serialize(v: &NodeRef) -> String { - let mut contents = Vec::new(); - for child in v.children() { - child - .serialize(&mut contents) - .expect("serialization failed"); - } - String::from_utf8(contents).expect("non utf-8 html") +pub(crate) fn rewrite_lol( + html: &str, + ctx: Context, + templates: &TemplateData, +) -> Result