feat(tazjin/tgsa): add "web interface" for this r/3965

Change-Id: I1619f0ceec2f83f4728d7de006a08b2ce6148e59
Reviewed-on: https://cl.tvl.fyi/c/depot/+/5468
Tested-by: BuildkiteCI
Reviewed-by: tazjin <tazjin@tvl.su>

3 files changed, 442 insertions, 9 deletions

diff --git a/users/tazjin/tgsa/Cargo.lock b/users/tazjin/tgsa/Cargo.lock
index 1536f3a1c4..b1c154f0ab 100644
--- a/users/tazjin/tgsa/Cargo.lock
+++ b/users/tazjin/tgsa/Cargo.lock
@@ -3,24 +3,88 @@
 version = 3
 
 [[package]]
+name = "adler32"
+version = "1.2.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "aae1277d39aeec15cb388266ecc24b11c80469deae6067e17a1a7aa9e5c1f234"
+
+[[package]]
+name = "alloc-no-stdlib"
+version = "2.0.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "35ef4730490ad1c4eae5c4325b2a95f521d023e5c885853ff7aca0a6a1631db3"
+
+[[package]]
+name = "alloc-stdlib"
+version = "0.2.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "697ed7edc0f1711de49ce108c541623a0af97c6c60b2f6e2b65229847ac843c2"
+dependencies = [
+ "alloc-no-stdlib",
+]
+
+[[package]]
 name = "anyhow"
 version = "1.0.56"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "4361135be9122e0870de935d7c439aef945b9f9ddd4199a553b5270b49c82a27"
 
 [[package]]
+name = "ascii"
+version = "1.0.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "bbf56136a5198c7b01a49e3afcbef6cf84597273d298f54432926024107b0109"
+
+[[package]]
 name = "autocfg"
 version = "1.1.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "d468802bab17cbc0cc575e9b053f41e72aa36bfa6b7f55e3529ffa43161b97fa"
 
 [[package]]
+name = "base64"
+version = "0.13.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "904dfeac50f3cdaba28fc6f57fdcddb75f49ed61346676a78c4ffe55877802fd"
+
+[[package]]
 name = "bitflags"
 version = "1.3.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "bef38d45163c2f1dde094a7dfd33ccf595c92905c8f8f4fdc18d06fb1037718a"
 
 [[package]]
+name = "brotli"
+version = "3.3.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a1a0b1dbcc8ae29329621f8d4f0d835787c1c38bb1401979b49d13b0b305ff68"
+dependencies = [
+ "alloc-no-stdlib",
+ "alloc-stdlib",
+ "brotli-decompressor",
+]
+
+[[package]]
+name = "brotli-decompressor"
+version = "2.3.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "59ad2d4653bf5ca36ae797b1f4bb4dbddb60ce49ca4aed8a2ce4829f60425b80"
+dependencies = [
+ "alloc-no-stdlib",
+ "alloc-stdlib",
+]
+
+[[package]]
+name = "buf_redux"
+version = "0.8.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b953a6887648bb07a535631f2bc00fbdb2a2216f135552cb3f534ed136b9c07f"
+dependencies = [
+ "memchr",
+ "safemem",
+]
+
+[[package]]
 name = "byteorder"
 version = "1.4.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
@@ -39,12 +103,39 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "baf1de4339761588bc0619e3cbc0120ee582ebb74b53b4efbf79117bd2da40fd"
 
 [[package]]
+name = "chrono"
+version = "0.4.19"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "670ad68c9088c2a963aaa298cb369688cf3f9465ce5e2d4ca10e6e0098a1ce73"
+dependencies = [
+ "libc",
+ "num-integer",
+ "num-traits",
+ "winapi",
+]
+
+[[package]]
+name = "chunked_transfer"
+version = "1.4.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "fff857943da45f546682664a79488be82e69e43c1a7a2307679ab9afb3a66d2e"
+
+[[package]]
 name = "convert_case"
 version = "0.4.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "6245d59a3e82a7fc217c5828a6692dbc6dfb63a0c8c90495621f7b9d79704a0e"
 
 [[package]]
+name = "crc32fast"
+version = "1.3.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b540bd8bc810d3885c6ea91e2018302f68baba2129ab3e88f32389ee9370880d"
+dependencies = [
+ "cfg-if",
+]
+
+[[package]]
 name = "crimp"
 version = "0.2.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
@@ -113,6 +204,16 @@ dependencies = [
 ]
 
 [[package]]
+name = "deflate"
+version = "0.9.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5f95bf05dffba6e6cce8dfbb30def788154949ccd9aed761b472119c21e01c70"
+dependencies = [
+ "adler32",
+ "gzip-header",
+]
+
+[[package]]
 name = "derive_more"
 version = "0.99.17"
 source = "registry+https://github.com/rust-lang/crates.io-index"
@@ -147,6 +248,27 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "3a68a4904193147e0a8dec3314640e6db742afd5f6e634f428a6af230d9b3591"
 
 [[package]]
+name = "fastrand"
+version = "1.7.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "c3fcf0cee53519c866c09b5de1f6c56ff9d647101f81c1964fa632e148896cdf"
+dependencies = [
+ "instant",
+]
+
+[[package]]
+name = "filetime"
+version = "0.2.16"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "c0408e2626025178a6a7f7ffc05a25bc47103229f19c113755de7bf63816290c"
+dependencies = [
+ "cfg-if",
+ "libc",
+ "redox_syscall",
+ "winapi",
+]
+
+[[package]]
 name = "form_urlencoded"
 version = "1.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
@@ -207,6 +329,24 @@ dependencies = [
 ]
 
 [[package]]
+name = "gzip-header"
+version = "0.3.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "0131feb3d3bb2a5a238d8a4d09f6353b7ebfdc52e77bccbf4ea6eaa751dde639"
+dependencies = [
+ "crc32fast",
+]
+
+[[package]]
+name = "hermit-abi"
+version = "0.1.19"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "62b467343b94ba476dcb2500d242dadbb39557df889310ac77c5d99100aaac33"
+dependencies = [
+ "libc",
+]
+
+[[package]]
 name = "html5ever"
 version = "0.25.2"
 source = "registry+https://github.com/rust-lang/crates.io-index"
@@ -221,6 +361,12 @@ dependencies = [
 ]
 
 [[package]]
+name = "httparse"
+version = "1.7.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "6330e8a36bd8c859f3fa6d9382911fbb7147ec39807f63b923933a247240b9ba"
+
+[[package]]
 name = "idna"
 version = "0.2.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
@@ -232,6 +378,15 @@ dependencies = [
 ]
 
 [[package]]
+name = "instant"
+version = "0.1.12"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "7a5bbe824c507c5da5956355e86a746d82e0e1464f65d862cc5e71da70e94b2c"
+dependencies = [
+ "cfg-if",
+]
+
+[[package]]
 name = "itoa"
 version = "0.4.8"
 source = "registry+https://github.com/rust-lang/crates.io-index"
@@ -313,6 +468,46 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "a3e378b66a060d48947b590737b30a1be76706c8dd7b8ba0f2fe3989c68a853f"
 
 [[package]]
+name = "memchr"
+version = "2.4.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "308cc39be01b73d0d18f82a0e7b2a3df85245f84af96fdddc5d202d27e47b86a"
+
+[[package]]
+name = "mime"
+version = "0.3.16"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "2a60c7ce501c71e03a9c9c0d35b861413ae925bd979cc7a4e30d060069aaac8d"
+
+[[package]]
+name = "mime_guess"
+version = "2.0.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "4192263c238a5f0d0c6bfd21f336a313a4ce1c450542449ca191bb657b4642ef"
+dependencies = [
+ "mime",
+ "unicase",
+]
+
+[[package]]
+name = "multipart"
+version = "0.18.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "00dec633863867f29cb39df64a397cdf4a6354708ddd7759f70c7fb51c5f9182"
+dependencies = [
+ "buf_redux",
+ "httparse",
+ "log",
+ "mime",
+ "mime_guess",
+ "quick-error",
+ "rand 0.8.5",
+ "safemem",
+ "tempfile",
+ "twoway",
+]
+
+[[package]]
 name = "new_debug_unreachable"
 version = "1.0.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
@@ -325,6 +520,44 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "72ef4a56884ca558e5ddb05a1d1e7e1bfd9a68d9ed024c21704cc98872dae1bb"
 
 [[package]]
+name = "num-integer"
+version = "0.1.44"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d2cc698a63b549a70bc047073d2949cce27cd1c7b0a4a862d08a8031bc2801db"
+dependencies = [
+ "autocfg",
+ "num-traits",
+]
+
+[[package]]
+name = "num-traits"
+version = "0.2.14"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "9a64b1ec5cda2586e284722486d802acf1f7dbdc623e2bfc57e65ca1cd099290"
+dependencies = [
+ "autocfg",
+]
+
+[[package]]
+name = "num_cpus"
+version = "1.13.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "19e64526ebdee182341572e50e9ad03965aa510cd94427a4549448f285e957a1"
+dependencies = [
+ "hermit-abi",
+ "libc",
+]
+
+[[package]]
+name = "num_threads"
+version = "0.1.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "aba1801fb138d8e85e11d0fc70baf4fe1cdfffda7c6cd34a854905df588e5ed0"
+dependencies = [
+ "libc",
+]
+
+[[package]]
 name = "once_cell"
 version = "1.10.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
@@ -485,6 +718,12 @@ dependencies = [
 ]
 
 [[package]]
+name = "quick-error"
+version = "1.2.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "a1d01941d82fa2ab50be1e79e6714289dd7cde78eba4c074bc5a4374f650dfe0"
+
+[[package]]
 name = "quote"
 version = "1.0.18"
 source = "registry+https://github.com/rust-lang/crates.io-index"
@@ -584,6 +823,40 @@ dependencies = [
 ]
 
 [[package]]
+name = "remove_dir_all"
+version = "0.5.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "3acd125665422973a33ac9d3dd2df85edad0f4ae9b00dafb1a05e43a9f5ef8e7"
+dependencies = [
+ "winapi",
+]
+
+[[package]]
+name = "rouille"
+version = "3.5.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "18b2380c42510ef4a28b5f228a174c801e0dec590103e215e60812e2e2f34d05"
+dependencies = [
+ "base64",
+ "brotli",
+ "chrono",
+ "deflate",
+ "filetime",
+ "multipart",
+ "num_cpus",
+ "percent-encoding",
+ "rand 0.8.5",
+ "serde",
+ "serde_derive",
+ "serde_json",
+ "sha1",
+ "threadpool",
+ "time",
+ "tiny_http",
+ "url",
+]
+
+[[package]]
 name = "rustc_version"
 version = "0.4.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
@@ -599,6 +872,12 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "73b4b750c782965c211b42f022f59af1fbceabdd026623714f104152f1ec149f"
 
 [[package]]
+name = "safemem"
+version = "0.3.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ef703b7cb59335eae2eb93ceb664c0eb7ea6bf567079d843e09420219668e072"
+
+[[package]]
 name = "schannel"
 version = "0.1.19"
 source = "registry+https://github.com/rust-lang/crates.io-index"
@@ -663,6 +942,17 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "ce31e24b01e1e524df96f1c2fdd054405f8d7376249a5110886fb4b658484789"
 
 [[package]]
+name = "serde_derive"
+version = "1.0.136"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "08597e7152fcd306f41838ed3e37be9eaeed2b61c42e2117266a554fab4662f9"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "syn",
+]
+
+[[package]]
 name = "serde_json"
 version = "1.0.79"
 source = "registry+https://github.com/rust-lang/crates.io-index"
@@ -684,6 +974,21 @@ dependencies = [
 ]
 
 [[package]]
+name = "sha1"
+version = "0.6.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "c1da05c97445caa12d05e848c4a4fcbbea29e748ac28f7e80e9b010392063770"
+dependencies = [
+ "sha1_smol",
+]
+
+[[package]]
+name = "sha1_smol"
+version = "1.0.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ae1a47186c03a32177042e55dbc5fd5aee900b8e0069a8d70fba96a9375cd012"
+
+[[package]]
 name = "siphasher"
 version = "0.3.10"
 source = "registry+https://github.com/rust-lang/crates.io-index"
@@ -749,6 +1054,20 @@ dependencies = [
 ]
 
 [[package]]
+name = "tempfile"
+version = "3.3.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5cdb1ef4eaeeaddc8fbd371e5017057064af0911902ef36b39801f67cc6d79e4"
+dependencies = [
+ "cfg-if",
+ "fastrand",
+ "libc",
+ "redox_syscall",
+ "remove_dir_all",
+ "winapi",
+]
+
+[[package]]
 name = "tendril"
 version = "0.4.3"
 source = "registry+https://github.com/rust-lang/crates.io-index"
@@ -765,6 +1084,7 @@ version = "0.1.0"
 dependencies = [
  "anyhow",
  "crimp",
+ "rouille",
  "scraper",
  "url",
 ]
@@ -776,6 +1096,38 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "8eaa81235c7058867fa8c0e7314f33dcce9c215f535d1913822a2b3f5e289f3c"
 
 [[package]]
+name = "threadpool"
+version = "1.8.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "d050e60b33d41c19108b32cea32164033a9013fe3b46cbd4457559bfbf77afaa"
+dependencies = [
+ "num_cpus",
+]
+
+[[package]]
+name = "time"
+version = "0.3.9"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "c2702e08a7a860f005826c6815dcac101b19b5eb330c27fe4a5928fec1d20ddd"
+dependencies = [
+ "libc",
+ "num_threads",
+]
+
+[[package]]
+name = "tiny_http"
+version = "0.8.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "9ce51b50006056f590c9b7c3808c3bd70f0d1101666629713866c227d6e58d39"
+dependencies = [
+ "ascii",
+ "chrono",
+ "chunked_transfer",
+ "log",
+ "url",
+]
+
+[[package]]
 name = "tinyvec"
 version = "1.5.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
@@ -791,6 +1143,24 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "cda74da7e1a664f795bb1f8a87ec406fb89a02522cf6e50620d016add6dbbf5c"
 
 [[package]]
+name = "twoway"
+version = "0.1.8"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "59b11b2b5241ba34be09c3cc85a36e56e48f9888862e19cedf23336d35316ed1"
+dependencies = [
+ "memchr",
+]
+
+[[package]]
+name = "unicase"
+version = "2.6.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "50f37be617794602aabbeee0be4f259dc1778fabe05e2d67ee8f79326d5cb4f6"
+dependencies = [
+ "version_check",
+]
+
+[[package]]
 name = "unicode-bidi"
 version = "0.3.7"
 source = "registry+https://github.com/rust-lang/crates.io-index"
@@ -842,6 +1212,12 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "accd4ea62f7bb7a82fe23066fb0957d48ef677f6eeb8215f372f52e48bb32426"
 
 [[package]]
+name = "version_check"
+version = "0.9.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "49874b5167b65d7193b8aba1567f5c7d93d001cafc34600cee003eda787e483f"
+
+[[package]]
 name = "wasi"
 version = "0.9.0+wasi-snapshot-preview1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
diff --git a/users/tazjin/tgsa/Cargo.toml b/users/tazjin/tgsa/Cargo.toml
index 06e13f2b38..c4f40fd8a1 100644
--- a/users/tazjin/tgsa/Cargo.toml
+++ b/users/tazjin/tgsa/Cargo.toml
@@ -6,5 +6,6 @@ edition = "2021"
 [dependencies]
 anyhow = "1.0"
 crimp = "0.2"
+rouille = "3.5"
 scraper = "0.12"
 url = "2.2"
diff --git a/users/tazjin/tgsa/src/main.rs b/users/tazjin/tgsa/src/main.rs
index cfb1c7223f..d23430d7ee 100644
--- a/users/tazjin/tgsa/src/main.rs
+++ b/users/tazjin/tgsa/src/main.rs
@@ -10,6 +10,27 @@ impl TgLink {
     fn to_url(&self) -> String {
         format!("https://t.me/{}/{}?embed=1", self.username, self.message_id)
     }
+
+    fn parse(url: &str) -> Option<Self> {
+        let url = url.strip_prefix("/")?;
+        let parsed = url::Url::parse(url).ok()?;
+
+        if parsed.host()? != url::Host::Domain("t.me") {
+            // only t.me links are supported
+            return None;
+        }
+
+        let parts = parsed.path_segments()?.collect::<Vec<&str>>();
+        if parts.len() != 2 {
+            // only message links are supported
+            return None;
+        }
+
+        Some(TgLink {
+            username: parts[0].into(),
+            message_id: parts[1].parse().ok()?,
+        })
+    }
 }
 
 fn fetch_embed(link: &TgLink) -> Result<String> {
@@ -138,18 +159,53 @@ fn to_bbcode(link: &TgLink, msg: &TgMessage) -> Result<String> {
     Ok(out)
 }
 
+fn handle_tg_link(link: &TgLink) -> Result<String> {
+    let embed = fetch_embed(&link)?;
+    let msg = parse_tgmessage(&embed)?;
+    to_bbcode(&link, &msg).context("failed to make bbcode")
+}
+
 fn main() {
     crimp::init();
 
-    let link = TgLink {
-        // username: "RWApodcast".into(),
-        // message_id: 113,
-        username: "intelslava".into(),
-        message_id: 25483,
-    };
+    rouille::start_server("0.0.0.0:8472", move |request| {
+        match TgLink::parse(request.raw_url()) {
+            None => rouille::Response::text(
+                r#"tgsa
+----
 
-    let embed = fetch_embed(&link).unwrap();
-    let msg = parse_tgmessage(&embed).unwrap();
+this is a stupid program that lets you turn telegram message links
+into BBcode suitable for pasting on somethingawful dot com
 
-    println!("{}", to_bbcode(&link, &msg).unwrap());
+you can use it by putting a valid telegram message link in the url and
+waiting a few seconds (yes it's currently slow, yes it's SA's fault,
+yes I could work around but can't be bothered atm)
+
+for example:
+
+  https://tgsa.tazj.in/https://t.me/RWApodcast/113
+
+yes, that looks stupid, but it works
+
+if you see this message and think you did the above correctly, you
+didn't. try again. idiot.
+
+pm me on the forums if this makes you mad or something.
+"#,
+            ),
+            Some(link) => {
+                let result = handle_tg_link(&link);
+                match result {
+                    Ok(bbcode) => rouille::Response::text(bbcode),
+                    Err(err) => rouille::Response::text(format!(
+                        r#"something broke: {}
+
+nobody has been alerted about this and it has probably not been
+logged. pm me on the forums if you think it's important enough."#,
+                        err
+                    )),
+                }
+            }
+        }
+    });
 }