server: add ability to slurp contents from site

Cargo.lock

@@ -72,7 +72,7 @@ version = "3.3.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "64e6d1c7838db705c9b756557ee27c384ce695a1c51a6fe528784cb1c6840170"
 dependencies = [
- "html5ever",
+ "html5ever 0.26.0",
  "maplit",
  "once_cell",
  "tendril",
@@ -272,6 +272,12 @@ dependencies = [
  "bytemuck",
 ]
 
+[[package]]
+name = "atomic-waker"
+version = "1.1.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "1505bd5d3d116872e7271a6d4e16d81d0c8570876c8de68093a09ac269d8aac0"
+
 [[package]]
 name = "atomic_hooks"
 version = "0.1.6"
@@ -328,6 +334,12 @@ version = "0.21.7"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "9d297deb1925b89f2ccc13d7635fa0714f12c87adce1c75356b39ca9b7178567"
 
+[[package]]
+name = "base64"
+version = "0.22.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "72b3254f16251a8381aa12e40e3c4d2f0199f8c6508fbecb9d91f575e0fbb8c6"
+
 [[package]]
 name = "base64ct"
 version = "1.6.0"
@@ -586,11 +598,11 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "0250ac93bbccb4f0a892507a4580178edddef5e8267650e294b4fe00597b0da8"
 dependencies = [
  "cssparser 0.31.2",
- "html5ever",
+ "html5ever 0.26.0",
  "indexmap 2.2.6",
  "pico-args",
  "rayon",
- "reqwest",
+ "reqwest 0.11.27",
  "rustc-hash",
  "selectors 0.25.0",
  "smallvec",
@@ -838,6 +850,12 @@ dependencies = [
  "paste",
 ]
 
+[[package]]
+name = "ego-tree"
+version = "0.6.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "12a0bb14ac04a9fcf170d0bbbef949b44cc492f4452bd20c095636956f653642"
+
 [[package]]
 name = "either"
 version = "1.13.0"
@@ -941,6 +959,21 @@ version = "1.0.7"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "3f9eec918d3f24069decb9af1554cad7c880e2da24a9afd88aca000531ab82c1"
 
+[[package]]
+name = "foreign-types"
+version = "0.3.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f6f339eb8adc052cd2ca78910fda869aefa38d22d5cb648e6485e4d3fc06f3b1"
+dependencies = [
+ "foreign-types-shared",
+]
+
+[[package]]
+name = "foreign-types-shared"
+version = "0.1.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "00b0228411908ca8685dba7fc2cdd70ec9990a6e753e89b6ac91a84c40fbaf4b"
+
 [[package]]
 name = "form_urlencoded"
 version = "1.2.1"
@@ -1102,6 +1135,15 @@ dependencies = [
  "winapi",
 ]
 
+[[package]]
+name = "getopts"
+version = "0.2.21"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "14dbbfd5c71d70241ecf9e6f13737f7b5ce823821063188d7e46c41d371eebd5"
+dependencies = [
+ "unicode-width",
+]
+
 [[package]]
 name = "getrandom"
 version = "0.1.16"
@@ -1316,6 +1358,25 @@ dependencies = [
  "tracing",
 ]
 
+[[package]]
+name = "h2"
+version = "0.4.6"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "524e8ac6999421f49a846c2d4411f337e53497d8ec55d67753beffa43c5d9205"
+dependencies = [
+ "atomic-waker",
+ "bytes",
+ "fnv",
+ "futures-core",
+ "futures-sink",
+ "http 1.1.0",
+ "indexmap 2.2.6",
+ "slab",
+ "tokio",
+ "tokio-util",
+ "tracing",
+]
+
 [[package]]
 name = "handlebars"
 version = "4.5.0"
@@ -1444,12 +1505,26 @@ checksum = "bea68cab48b8459f17cf1c944c67ddc572d272d9f2b274140f223ecb1da4a3b7"
 dependencies = [
  "log",
  "mac",
- "markup5ever",
+ "markup5ever 0.11.0",
  "proc-macro2",
  "quote",
  "syn 1.0.109",
 ]
 
+[[package]]
+name = "html5ever"
+version = "0.27.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "c13771afe0e6e846f1e67d038d4cb29998a6779f93c809212e4e9c32efd244d4"
+dependencies = [
+ "log",
+ "mac",
+ "markup5ever 0.12.1",
+ "proc-macro2",
+ "quote",
+ "syn 2.0.69",
+]
+
 [[package]]
 name = "http"
 version = "0.2.12"
@@ -1483,6 +1558,29 @@ dependencies = [
  "pin-project-lite",
 ]
 
+[[package]]
+name = "http-body"
+version = "1.0.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "1efedce1fb8e6913f23e0c92de8e62cd5b772a67e7b3946df930a62566c93184"
+dependencies = [
+ "bytes",
+ "http 1.1.0",
+]
+
+[[package]]
+name = "http-body-util"
+version = "0.1.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "793429d76616a256bcb62c2a2ec2bed781c8307e797e2598c50010f2bee2544f"
+dependencies = [
+ "bytes",
+ "futures-util",
+ "http 1.1.0",
+ "http-body 1.0.1",
+ "pin-project-lite",
+]
+
 [[package]]
 name = "httparse"
 version = "1.9.4"
@@ -1511,9 +1609,9 @@ dependencies = [
  "futures-channel",
  "futures-core",
  "futures-util",
- "h2",
+ "h2 0.3.26",
  "http 0.2.12",
- "http-body",
+ "http-body 0.4.6",
  "httparse",
  "httpdate",
  "itoa 1.0.11",
@@ -1525,6 +1623,26 @@ dependencies = [
  "want",
 ]
 
+[[package]]
+name = "hyper"
+version = "1.4.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "50dfd22e0e76d0f662d429a5f80fcaf3855009297eab6a0a9f8543834744ba05"
+dependencies = [
+ "bytes",
+ "futures-channel",
+ "futures-util",
+ "h2 0.4.6",
+ "http 1.1.0",
+ "http-body 1.0.1",
+ "httparse",
+ "itoa 1.0.11",
+ "pin-project-lite",
+ "smallvec",
+ "tokio",
+ "want",
+]
+
 [[package]]
 name = "hyper-rustls"
 version = "0.24.2"
@@ -1533,10 +1651,63 @@ checksum = "ec3efd23720e2049821a693cbc7e65ea87c72f1c58ff2f9522ff332b1491e590"
 dependencies = [
  "futures-util",
  "http 0.2.12",
- "hyper",
+ "hyper 0.14.29",
- "rustls",
+ "rustls 0.21.12",
  "tokio",
- "tokio-rustls",
+ "tokio-rustls 0.24.1",
+]
+
+[[package]]
+name = "hyper-rustls"
+version = "0.27.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5ee4be2c948921a1a5320b629c4193916ed787a7f7f293fd3f7f5a6c9de74155"
+dependencies = [
+ "futures-util",
+ "http 1.1.0",
+ "hyper 1.4.1",
+ "hyper-util",
+ "rustls 0.23.12",
+ "rustls-pki-types",
+ "tokio",
+ "tokio-rustls 0.26.0",
+ "tower-service",
+]
+
+[[package]]
+name = "hyper-tls"
+version = "0.6.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "70206fc6890eaca9fde8a0bf71caa2ddfc9fe045ac9e5c70df101a7dbde866e0"
+dependencies = [
+ "bytes",
+ "http-body-util",
+ "hyper 1.4.1",
+ "hyper-util",
+ "native-tls",
+ "tokio",
+ "tokio-native-tls",
+ "tower-service",
+]
+
+[[package]]
+name = "hyper-util"
+version = "0.1.7"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "cde7055719c54e36e95e8719f95883f22072a48ede39db7fc17a4e1d5281e9b9"
+dependencies = [
+ "bytes",
+ "futures-channel",
+ "futures-util",
+ "http 1.1.0",
+ "http-body 1.0.1",
+ "hyper 1.4.1",
+ "pin-project-lite",
+ "socket2",
+ "tokio",
+ "tower",
+ "tower-service",
+ "tracing",
 ]
 
 [[package]]
@@ -1855,6 +2026,20 @@ dependencies = [
  "tendril",
 ]
 
+[[package]]
+name = "markup5ever"
+version = "0.12.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "16ce3abbeba692c8b8441d036ef91aea6df8da2c6b6e21c7e14d3c18e526be45"
+dependencies = [
+ "log",
+ "phf 0.11.2",
+ "phf_codegen 0.11.2",
+ "string_cache",
+ "string_cache_codegen",
+ "tendril",
+]
+
 [[package]]
 name = "matchers"
 version = "0.1.0"
@@ -1974,6 +2159,23 @@ dependencies = [
  "version_check",
 ]
 
+[[package]]
+name = "native-tls"
+version = "0.2.12"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a8614eb2c83d59d1c8cc974dd3f920198647674a0a035e1af1fa58707e317466"
+dependencies = [
+ "libc",
+ "log",
+ "openssl",
+ "openssl-probe",
+ "openssl-sys",
+ "schannel",
+ "security-framework",
+ "security-framework-sys",
+ "tempfile",
+]
+
 [[package]]
 name = "new_debug_unreachable"
 version = "1.0.6"
@@ -2100,6 +2302,50 @@ version = "1.19.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "3fdb12b2476b595f9358c5161aa467c2438859caa136dec86c26fdd2efe17b92"
 
+[[package]]
+name = "openssl"
+version = "0.10.66"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "9529f4786b70a3e8c61e11179af17ab6188ad8d0ded78c5529441ed39d4bd9c1"
+dependencies = [
+ "bitflags 2.6.0",
+ "cfg-if",
+ "foreign-types",
+ "libc",
+ "once_cell",
+ "openssl-macros",
+ "openssl-sys",
+]
+
+[[package]]
+name = "openssl-macros"
+version = "0.1.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a948666b637a0f465e8564c73e89d4dde00d72d4d473cc972f390fc3dcee7d9c"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "syn 2.0.69",
+]
+
+[[package]]
+name = "openssl-probe"
+version = "0.1.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ff011a302c396a5197692431fc1948019154afc178baf7d8e37367442a4601cf"
+
+[[package]]
+name = "openssl-sys"
+version = "0.9.103"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "7f9e8deee91df40a943c71b917e5874b951d32a802526c85721ce3b776c929d6"
+dependencies = [
+ "cc",
+ "libc",
+ "pkg-config",
+ "vcpkg",
+]
+
 [[package]]
 name = "owning_ref"
 version = "0.4.1"
@@ -2296,6 +2542,16 @@ dependencies = [
  "phf_shared 0.10.0",
 ]
 
+[[package]]
+name = "phf_codegen"
+version = "0.11.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "e8d39688d359e6b34654d328e262234662d16cc0f60ec8dcbe5e718709342a5a"
+dependencies = [
+ "phf_generator 0.11.2",
+ "phf_shared 0.11.2",
+]
+
 [[package]]
 name = "phf_generator"
 version = "0.8.0"
@@ -2760,11 +3016,11 @@ dependencies = [
  "encoding_rs",
  "futures-core",
  "futures-util",
- "h2",
+ "h2 0.3.26",
  "http 0.2.12",
- "http-body",
+ "http-body 0.4.6",
- "hyper",
+ "hyper 0.14.29",
- "hyper-rustls",
+ "hyper-rustls 0.24.2",
  "ipnet",
  "js-sys",
  "log",
@@ -2772,15 +3028,15 @@ dependencies = [
  "once_cell",
  "percent-encoding",
  "pin-project-lite",
- "rustls",
+ "rustls 0.21.12",
- "rustls-pemfile",
+ "rustls-pemfile 1.0.4",
  "serde",
  "serde_json",
  "serde_urlencoded",
- "sync_wrapper",
+ "sync_wrapper 0.1.2",
- "system-configuration",
+ "system-configuration 0.5.1",
  "tokio",
- "tokio-rustls",
+ "tokio-rustls 0.24.1",
  "tower-service",
  "url",
  "wasm-bindgen",
@@ -2790,6 +3046,50 @@ dependencies = [
  "winreg",
 ]
 
+[[package]]
+name = "reqwest"
+version = "0.12.7"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f8f4955649ef5c38cc7f9e8aa41761d48fb9677197daea9984dc54f56aad5e63"
+dependencies = [
+ "base64 0.22.1",
+ "bytes",
+ "encoding_rs",
+ "futures-channel",
+ "futures-core",
+ "futures-util",
+ "h2 0.4.6",
+ "http 1.1.0",
+ "http-body 1.0.1",
+ "http-body-util",
+ "hyper 1.4.1",
+ "hyper-rustls 0.27.2",
+ "hyper-tls",
+ "hyper-util",
+ "ipnet",
+ "js-sys",
+ "log",
+ "mime",
+ "native-tls",
+ "once_cell",
+ "percent-encoding",
+ "pin-project-lite",
+ "rustls-pemfile 2.1.3",
+ "serde",
+ "serde_json",
+ "serde_urlencoded",
+ "sync_wrapper 1.0.1",
+ "system-configuration 0.6.1",
+ "tokio",
+ "tokio-native-tls",
+ "tower-service",
+ "url",
+ "wasm-bindgen",
+ "wasm-bindgen-futures",
+ "web-sys",
+ "windows-registry",
+]
+
 [[package]]
 name = "ring"
 version = "0.17.8"
@@ -2887,7 +3187,7 @@ dependencies = [
  "either",
  "futures",
  "http 0.2.12",
- "hyper",
+ "hyper 0.14.29",
  "indexmap 2.2.6",
  "log",
  "memchr",
@@ -2966,10 +3266,23 @@ checksum = "3f56a14d1f48b391359b22f731fd4bd7e43c97f3c50eee276f3aa09c94784d3e"
 dependencies = [
  "log",
  "ring",
- "rustls-webpki",
+ "rustls-webpki 0.101.7",
  "sct",
 ]
 
+[[package]]
+name = "rustls"
+version = "0.23.12"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "c58f8c84392efc0a126acce10fa59ff7b3d2ac06ab451a33f2741989b806b044"
+dependencies = [
+ "once_cell",
+ "rustls-pki-types",
+ "rustls-webpki 0.102.6",
+ "subtle",
+ "zeroize",
+]
+
 [[package]]
 name = "rustls-pemfile"
 version = "1.0.4"
@@ -2979,6 +3292,22 @@ dependencies = [
  "base64 0.21.7",
 ]
 
+[[package]]
+name = "rustls-pemfile"
+version = "2.1.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "196fe16b00e106300d3e45ecfcb764fa292a535d7326a29a5875c579c7417425"
+dependencies = [
+ "base64 0.22.1",
+ "rustls-pki-types",
+]
+
+[[package]]
+name = "rustls-pki-types"
+version = "1.8.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "fc0a2ce646f8655401bb81e7927b812614bd5d91dbc968696be50603510fcaf0"
+
 [[package]]
 name = "rustls-webpki"
 version = "0.101.7"
@@ -2989,6 +3318,17 @@ dependencies = [
  "untrusted",
 ]
 
+[[package]]
+name = "rustls-webpki"
+version = "0.102.6"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "8e6b52d4fda176fd835fdc55a835d4a89b8499cad995885a21149d5ad62f852e"
+dependencies = [
+ "ring",
+ "rustls-pki-types",
+ "untrusted",
+]
+
 [[package]]
 name = "rustversion"
 version = "1.0.17"
@@ -3001,6 +3341,15 @@ version = "1.0.18"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "f3cb5ba0dc43242ce17de99c180e96db90b235b8a9fdc9543c96d2209116bd9f"
 
+[[package]]
+name = "schannel"
+version = "0.1.23"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "fbc91545643bcf3a0bbb6569265615222618bdf33ce4ffbbd13c4bbd4c093534"
+dependencies = [
+ "windows-sys 0.52.0",
+]
+
 [[package]]
 name = "scoped-tls"
 version = "1.0.1"
@@ -3013,6 +3362,22 @@ version = "1.2.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "94143f37725109f92c262ed2cf5e59bce7498c01bcc1502d7b9afe439a4e9f49"
 
+[[package]]
+name = "scraper"
+version = "0.20.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b90460b31bfe1fc07be8262e42c665ad97118d4585869de9345a84d501a9eaf0"
+dependencies = [
+ "ahash 0.8.11",
+ "cssparser 0.31.2",
+ "ego-tree",
+ "getopts",
+ "html5ever 0.27.0",
+ "once_cell",
+ "selectors 0.25.0",
+ "tendril",
+]
+
 [[package]]
 name = "sct"
 version = "0.7.1"
@@ -3023,6 +3388,29 @@ dependencies = [
  "untrusted",
 ]
 
+[[package]]
+name = "security-framework"
+version = "2.11.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "897b2245f0b511c87893af39b033e5ca9cce68824c4d7e7630b5a1d339658d02"
+dependencies = [
+ "bitflags 2.6.0",
+ "core-foundation",
+ "core-foundation-sys",
+ "libc",
+ "security-framework-sys",
+]
+
+[[package]]
+name = "security-framework-sys"
+version = "2.11.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "75da29fe9b9b08fe9d6b22b5b4bcbc75d8db3aa31e639aa56bb62e9d46bfceaf"
+dependencies = [
+ "core-foundation-sys",
+ "libc",
+]
+
 [[package]]
 name = "seed"
 version = "0.10.0"
@@ -3175,6 +3563,7 @@ dependencies = [
  "anyhow",
  "async-graphql",
  "async-graphql-rocket",
+ "async-trait",
  "css-inline",
  "glog",
  "html-escape",
@@ -3185,8 +3574,10 @@ dependencies = [
  "maplit",
  "memmap",
  "notmuch",
+ "reqwest 0.12.7",
  "rocket",
  "rocket_cors",
+ "scraper",
  "serde",
  "serde_json",
  "shared",
@@ -3676,6 +4067,15 @@ version = "0.1.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "2047c6ded9c721764247e62cd3b03c09ffc529b2ba5b10ec482ae507a4a70160"
 
+[[package]]
+name = "sync_wrapper"
+version = "1.0.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a7065abeca94b6a8a577f9bd45aa0867a2238b74e8eb67cf10d492bc39351394"
+dependencies = [
+ "futures-core",
+]
+
 [[package]]
 name = "system-configuration"
 version = "0.5.1"
@@ -3684,7 +4084,18 @@ checksum = "ba3a3adc5c275d719af8cb4272ea1c4a6d668a777f37e115f6d11ddbc1c8e0e7"
 dependencies = [
  "bitflags 1.2.1",
  "core-foundation",
- "system-configuration-sys",
+ "system-configuration-sys 0.5.0",
+]
+
+[[package]]
+name = "system-configuration"
+version = "0.6.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "3c879d448e9d986b661742763247d3693ed13609438cf3d006f51f5368a5ba6b"
+dependencies = [
+ "bitflags 2.6.0",
+ "core-foundation",
+ "system-configuration-sys 0.6.0",
 ]
 
 [[package]]
@@ -3697,6 +4108,16 @@ dependencies = [
  "libc",
 ]
 
+[[package]]
+name = "system-configuration-sys"
+version = "0.6.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "8e1d1b10ced5ca923a1fcb8d03e96b8d3268065d724548c0211415ff6ac6bac4"
+dependencies = [
+ "core-foundation-sys",
+ "libc",
+]
+
 [[package]]
 name = "tempfile"
 version = "3.10.1"
@@ -3839,13 +4260,34 @@ dependencies = [
  "syn 2.0.69",
 ]
 
+[[package]]
+name = "tokio-native-tls"
+version = "0.3.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "bbae76ab933c85776efabc971569dd6119c580d8f5d448769dec1764bf796ef2"
+dependencies = [
+ "native-tls",
+ "tokio",
+]
+
 [[package]]
 name = "tokio-rustls"
 version = "0.24.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "c28327cf380ac148141087fbfb9de9d7bd4e84ab5d2c28fbc911d753de8a7081"
 dependencies = [
- "rustls",
+ "rustls 0.21.12",
+ "tokio",
+]
+
+[[package]]
+name = "tokio-rustls"
+version = "0.26.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "0c7bc40d0e5a97695bb96e27995cd3a08538541b0a846f65bba7a359f36700d4"
+dependencies = [
+ "rustls 0.23.12",
+ "rustls-pki-types",
  "tokio",
 ]
 
@@ -3942,6 +4384,27 @@ dependencies = [
  "syn 1.0.109",
 ]
 
+[[package]]
+name = "tower"
+version = "0.4.13"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b8fa9be0de6cf49e536ce1851f987bd21a43b771b09473c3549a6c853db37c1c"
+dependencies = [
+ "futures-core",
+ "futures-util",
+ "pin-project",
+ "pin-project-lite",
+ "tokio",
+ "tower-layer",
+ "tower-service",
+]
+
+[[package]]
+name = "tower-layer"
+version = "0.3.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "121c2a6cda46980bb0fcd1647ffaf6cd3fc79a013de288782836f6df9c48780e"
+
 [[package]]
 name = "tower-service"
 version = "0.3.2"
@@ -4099,6 +4562,12 @@ version = "1.11.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "d4c87d22b6e3f4a18d4d40ef354e97c90fcb14dd91d7dc0aa9d8a1172ebf7202"
 
+[[package]]
+name = "unicode-width"
+version = "0.1.13"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "0336d538f7abc86d282a4189614dfaa90810dfc2c6f6427eaf88e16311dd225d"
+
 [[package]]
 name = "unicode-xid"
 version = "0.2.4"
@@ -4393,6 +4862,36 @@ dependencies = [
  "windows-targets 0.52.6",
 ]
 
+[[package]]
+name = "windows-registry"
+version = "0.2.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "e400001bb720a623c1c69032f8e3e4cf09984deec740f007dd2b03ec864804b0"
+dependencies = [
+ "windows-result",
+ "windows-strings",
+ "windows-targets 0.52.6",
+]
+
+[[package]]
+name = "windows-result"
+version = "0.2.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "1d1043d8214f791817bab27572aaa8af63732e11bf84aa21a45a78d6c317ae0e"
+dependencies = [
+ "windows-targets 0.52.6",
+]
+
+[[package]]
+name = "windows-strings"
+version = "0.1.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "4cd9b125c486025df0eabcb585e62173c6c9eddcec5d117d3b6e8c30e2ee4d10"
+dependencies = [
+ "windows-result",
+ "windows-targets 0.52.6",
+]
+
 [[package]]
 name = "windows-sys"
 version = "0.48.0"

server/Cargo.toml

@@ -11,6 +11,7 @@ ammonia = "3.3.0"
 anyhow = "1.0.79"
 async-graphql = { version = "6.0.11", features = ["log"] }
 async-graphql-rocket = "6.0.11"
+async-trait = "0.1.81"
 css-inline = "0.13.0"
 glog = "0.1.0"
 html-escape = "0.2.13"
@@ -21,8 +22,10 @@ mailparse = "0.15.0"
 maplit = "1.0.2"
 memmap = "0.7.0"
 notmuch = { path = "../notmuch" }
+reqwest = { version = "0.12.7", features = ["blocking"] }
 rocket = { version = "0.5.0-rc.2", features = [ "json" ] }
 rocket_cors = "0.6.0"
+scraper = "0.20.0"
 serde = { version = "1.0.147", features = ["derive"] }
 serde_json = "1.0.87"
 shared = { path = "../shared" }

server/src/lib.rs

@@ -3,13 +3,15 @@ pub mod graphql;
 pub mod newsreader;
 pub mod nm;
 
-use std::{convert::Infallible, str::FromStr};
+use std::{collections::HashMap, convert::Infallible, str::FromStr};
 
+use async_trait::async_trait;
 use css_inline::{CSSInliner, InlineError, InlineOptions};
 use linkify::{LinkFinder, LinkKind};
-use log::{error, info};
+use log::{error, info, warn};
 use lol_html::{element, errors::RewritingError, rewrite_str, text, RewriteStrSettings};
 use maplit::{hashmap, hashset};
+use scraper::{error::SelectorErrorKind, Html, Selector};
 use thiserror::Error;
 use url::Url;
 
@@ -19,23 +21,28 @@ use crate::newsreader::{
 const NON_EXISTENT_SITE_NAME: &'static str = "NO-SUCH-SITE";
 
 // TODO: figure out how to use Cow
-trait Transformer {
+#[async_trait]
-    fn should_run(&self, _html: &str) -> bool {
+trait Transformer: Send + Sync {
+    fn should_run(&self, addr: &Option<Url>, _html: &str) -> bool {
         true
     }
     // TODO: should html be something like `html_escape` uses:
     // <S: ?Sized + AsRef<str>>(text: &S) -> Cow<str>
-    fn transform(&self, html: &str) -> Result<String, TransformError>;
+    async fn transform(&self, addr: &Option<Url>, html: &str) -> Result<String, TransformError>;
 }
 
 // TODO: how would we make this more generic to allow good implementations of Transformer outside
 // of this module?
 #[derive(Error, Debug)]
 pub enum TransformError {
-    #[error("lol-html rewrite error")]
+    #[error("lol-html rewrite error: {0}")]
     RewritingError(#[from] RewritingError),
-    #[error("css inline error")]
+    #[error("css inline error: {0}")]
     InlineError(#[from] InlineError),
+    #[error("failed to fetch url error: {0}")]
+    ReqwestError(#[from] reqwest::Error),
+    #[error("failed to parse HTML: {0}")]
+    HtmlParsingError(String),
 }
 
 struct SanitizeHtml<'a> {
@@ -43,31 +50,34 @@ struct SanitizeHtml<'a> {
     base_url: &'a Option<Url>,
 }
 
+#[async_trait]
 impl<'a> Transformer for SanitizeHtml<'a> {
-    fn transform(&self, html: &str) -> Result<String, TransformError> {
+    async fn transform(&self, _: &Option<Url>, html: &str) -> Result<String, TransformError> {
         Ok(sanitize_html(html, self.cid_prefix, self.base_url)?)
     }
 }
 
 struct EscapeHtml;
 
+#[async_trait]
 impl Transformer for EscapeHtml {
-    fn should_run(&self, html: &str) -> bool {
+    fn should_run(&self, _: &Option<Url>, html: &str) -> bool {
         html.contains("&")
     }
-    fn transform(&self, html: &str) -> Result<String, TransformError> {
+    async fn transform(&self, _: &Option<Url>, html: &str) -> Result<String, TransformError> {
         Ok(html_escape::decode_html_entities(html).to_string())
     }
 }
 
 struct StripHtml;
 
+#[async_trait]
 impl Transformer for StripHtml {
-    fn should_run(&self, html: &str) -> bool {
+    fn should_run(&self, _: &Option<Url>, html: &str) -> bool {
         // Lame test
         html.contains("<")
     }
-    fn transform(&self, html: &str) -> Result<String, TransformError> {
+    async fn transform(&self, _: &Option<Url>, html: &str) -> Result<String, TransformError> {
         let mut text = String::new();
         let element_content_handlers = vec![text!("*", |t| {
             text += t.as_str();
@@ -87,8 +97,9 @@ impl Transformer for StripHtml {
 
 struct InlineStyle;
 
+#[async_trait]
 impl Transformer for InlineStyle {
-    fn transform(&self, html: &str) -> Result<String, TransformError> {
+    async fn transform(&self, _: &Option<Url>, html: &str) -> Result<String, TransformError> {
         let css = concat!(
             "/* chrome-default.css */\n",
             include_str!("chrome-default.css"),
@@ -118,29 +129,78 @@ impl Transformer for InlineStyle {
     }
 }
 
-struct AddOutlink(Option<url::Url>);
+struct AddOutlink;
 
+#[async_trait]
 impl Transformer for AddOutlink {
-    fn should_run(&self, html: &str) -> bool {
+    fn should_run(&self, link: &Option<Url>, html: &str) -> bool {
-        if let Some(link) = &self.0 {
+        if let Some(link) = link {
-            return link.scheme().starts_with("http") && !html.contains(link.as_str());
+            link.scheme().starts_with("http") && !html.contains(link.as_str())
+        } else {
+            false
         }
-        false
     }
-    fn transform(&self, html: &str) -> Result<String, TransformError> {
+    async fn transform(&self, link: &Option<Url>, html: &str) -> Result<String, TransformError> {
-        if let Some(url) = &self.0 {
+        if let Some(link) = link {
             Ok(format!(
                 r#"
                 {html}
                 <div><a href="{}">View on site</a></div>
                 "#,
-                url
+                link
             ))
         } else {
             Ok(html.to_string())
         }
     }
 }
+
+struct SlurpContents {
+    site_selectors: HashMap<String, Vec<Selector>>,
+}
+
+impl SlurpContents {
+    fn get_selectors(&self, link: &Url) -> Option<&[Selector]> {
+        for (host, selector) in self.site_selectors.iter() {
+            if link.host_str().map(|h| h.contains(host)).unwrap_or(false) {
+                return Some(&selector);
+            }
+        }
+        None
+    }
+}
+
+#[async_trait]
+impl Transformer for SlurpContents {
+    fn should_run(&self, link: &Option<Url>, html: &str) -> bool {
+        if let Some(link) = link {
+            return self.get_selectors(link).is_some();
+        }
+        false
+    }
+    async fn transform(&self, link: &Option<Url>, html: &str) -> Result<String, TransformError> {
+        let Some(link) = link else {
+            return Ok(html.to_string());
+        };
+        let Some(selectors) = self.get_selectors(&link) else {
+            return Ok(html.to_string());
+        };
+        let body = reqwest::get(link.as_str()).await?.text().await?;
+        let doc = Html::parse_document(&body);
+
+        let mut results = Vec::new();
+        for selector in selectors {
+            if let Some(frag) = doc.select(&selector).next() {
+                results.push(frag.html())
+            } else {
+                warn!("couldn't find '{:?}' in {}", selector, link);
+                return Ok(html.to_string());
+            }
+        }
+        Ok(results.join("<br><br>"))
+    }
+}
+
 pub fn linkify_html(text: &str) -> String {
     let mut finder = LinkFinder::new();
     let finder = finder.url_must_have_scheme(false).kinds(&[LinkKind::Url]);

server/src/newsreader.rs

@@ -1,6 +1,8 @@
 use std::hash::{DefaultHasher, Hash, Hasher};
 
 use log::info;
+use maplit::hashmap;
+use scraper::Selector;
 use sqlx::postgres::PgPool;
 use url::Url;
 
@@ -13,7 +15,7 @@ use crate::{
     compute_offset_limit,
     error::ServerError,
     graphql::{Body, Email, Html, Message, Tag, Thread, ThreadSummary},
-    AddOutlink, EscapeHtml, InlineStyle, SanitizeHtml, StripHtml, Transformer,
+    AddOutlink, EscapeHtml, InlineStyle, SanitizeHtml, SlurpContents, StripHtml, Transformer,
 };
 
 pub fn is_newsreader_search(query: &str) -> bool {
@@ -89,36 +91,34 @@ pub async fn search(
     .fetch_all(pool)
     .await?;
 
-    Ok(rows
+    let mut res = Vec::new();
-        .into_iter()
+    for (i, r) in rows.into_iter().enumerate() {
-        .enumerate()
+        let site = r.site.unwrap_or("UNKOWN TAG".to_string());
-        .map(|(i, r)| {
+        let mut tags = vec![format!("{TAG_PREFIX}{site}")];
-            let site = r.site.unwrap_or("UNKOWN TAG".to_string());
+        if !r.is_read.unwrap_or(true) {
-            let mut tags = vec![format!("{TAG_PREFIX}{site}")];
+            tags.push("unread".to_string());
-            if !r.is_read.unwrap_or(true) {
+        };
-                tags.push("unread".to_string());
+        let mut title = r.title.unwrap_or("NO TITLE".to_string());
-            };
+        title = clean_title(&title).await.expect("failed to clean title");
-            let mut title = r.title.unwrap_or("NO TITLE".to_string());
+        res.push((
-            title = clean_title(&title).expect("failed to clean title");
+            i as i32 + offset,
-            (
+            ThreadSummary {
-                i as i32 + offset,
+                thread: format!("{THREAD_PREFIX}{}", r.uid),
-                ThreadSummary {
+                timestamp: r
-                    thread: format!("{THREAD_PREFIX}{}", r.uid),
+                    .date
-                    timestamp: r
+                    .expect("post missing date")
-                        .date
+                    .assume_utc()
-                        .expect("post missing date")
+                    .unix_timestamp() as isize,
-                        .assume_utc()
+                date_relative: "TODO date_relative".to_string(),
-                        .unix_timestamp() as isize,
+                matched: 0,
-                    date_relative: "TODO date_relative".to_string(),
+                total: 1,
-                    matched: 0,
+                authors: r.name.unwrap_or_else(|| site.clone()),
-                    total: 1,
+                subject: title,
-                    authors: r.name.unwrap_or_else(|| site.clone()),
+                tags,
-                    subject: title,
+            },
-                    tags,
+        ));
-                },
+    }
-            )
+    Ok(res)
-        })
-        .collect())
 }
 
 pub async fn tags(pool: &PgPool, _needs_unread: bool) -> Result<Vec<Tag>, ServerError> {
@@ -197,8 +197,25 @@ pub async fn thread(pool: &PgPool, thread_id: String) -> Result<Thread, ServerEr
     // TODO: add site specific cleanups. For example:
     // * Grafana does <div class="image-wrapp"><img class="lazyload>"<img src="/media/...>"</img></div>
     // * Some sites appear to be HTML encoded, unencode them, i.e. imperialviolent
-    let mut body_tranformers: Vec<Box<dyn Transformer>> = vec![
+    let body_tranformers: Vec<Box<dyn Transformer>> = vec![
-        Box::new(AddOutlink(link.clone())),
+        // TODO: add a map of urls and selectors
+        Box::new(SlurpContents {
+            site_selectors: hashmap![
+                "hackaday.com".to_string() => vec![
+                    Selector::parse("div.entry-featured-image").unwrap(),
+                    Selector::parse("div.entry-content").unwrap()
+                ],
+                "mitchellh.com".to_string() => vec![Selector::parse("div.w-full").unwrap()],
+                "natwelch.com".to_string() => vec![
+                    Selector::parse("article div.prose").unwrap(),
+                ],
+                "slashdot.org".to_string() => vec![
+                    Selector::parse("span.story-byline").unwrap(),
+                    Selector::parse("div.p").unwrap(),
+                ],
+            ],
+        }),
+        Box::new(AddOutlink),
         Box::new(EscapeHtml),
         Box::new(InlineStyle),
         Box::new(SanitizeHtml {
@@ -207,15 +224,15 @@ pub async fn thread(pool: &PgPool, thread_id: String) -> Result<Thread, ServerEr
         }),
     ];
     for t in body_tranformers.iter() {
-        if t.should_run(&body) {
+        if t.should_run(&link, &body) {
-            body = t.transform(&body)?;
+            body = t.transform(&link, &body).await?;
         }
     }
     let body = Body::Html(Html {
         html: body,
         content_tree: "".to_string(),
     });
-    let title = clean_title(&r.title.unwrap_or("NO TITLE".to_string()))?;
+    let title = clean_title(&r.title.unwrap_or("NO TITLE".to_string())).await?;
     let from = Some(Email {
         name: r.name,
         addr: addr.map(|a| a.to_string()),
@@ -254,7 +271,7 @@ pub async fn set_read_status<'ctx>(
         .await?;
     Ok(true)
 }
-fn clean_title(title: &str) -> Result<String, ServerError> {
+async fn clean_title(title: &str) -> Result<String, ServerError> {
     // Make title HTML so html parsers work
     let mut title = format!("<html>{title}</html>");
     let title_tranformers: Vec<Box<dyn Transformer>> =
@@ -262,8 +279,8 @@ fn clean_title(title: &str) -> Result<String, ServerError> {
     // Make title HTML so html parsers work
     title = format!("<html>{title}</html>");
     for t in title_tranformers.iter() {
-        if t.should_run(&title) {
+        if t.should_run(&None, &title) {
-            title = t.transform(&title)?;
+            title = t.transform(&None, &title).await?;
         }
     }
     Ok(title)