cassiopeia: adding new format lexer built on logos

3 years ago · afd8a74e43
parent 585078cd7a
commit afd8a74e43
7 changed files with 308 additions and 2 deletions
--- a/apps/cassiopeia/.envrc
+++ b/apps/cassiopeia/.envrc
@ -0,0 +1 @@
+eval "$(lorri direnv)"
--- a/apps/cassiopeia/Cargo.lock
+++ b/apps/cassiopeia/Cargo.lock
@ -0,0 +1,175 @@
+# This file is automatically @generated by Cargo.
+# It is not intended for manual editing.
+[[package]]
+name = "autocfg"
+version = "1.0.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "cdb031dd78e28731d87d56cc8ffef4a8f36ca26c38fe2de700543e627f8a464a"
+
+[[package]]
+name = "beef"
+version = "0.4.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "474a626a67200bd107d44179bb3d4fc61891172d11696609264589be6a0e6a43"
+
+[[package]]
+name = "cassiopeia"
+version = "0.1.0"
+dependencies = [
+ "chrono",
+ "logos",
+]
+
+[[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",
+ "time",
+ "winapi",
+]
+
+[[package]]
+name = "fnv"
+version = "1.0.7"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "3f9eec918d3f24069decb9af1554cad7c880e2da24a9afd88aca000531ab82c1"
+
+[[package]]
+name = "libc"
+version = "0.2.81"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "1482821306169ec4d07f6aca392a4681f66c75c9918aa49641a2595db64053cb"
+
+[[package]]
+name = "logos"
+version = "0.11.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b91c49573597a5d6c094f9031617bb1fed15c0db68c81e6546d313414ce107e4"
+dependencies = [
+ "logos-derive",
+]
+
+[[package]]
+name = "logos-derive"
+version = "0.11.5"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "797b1f8a0571b331c1b47e7db245af3dc634838da7a92b3bef4e30376ae1c347"
+dependencies = [
+ "beef",
+ "fnv",
+ "proc-macro2",
+ "quote",
+ "regex-syntax",
+ "syn",
+ "utf8-ranges",
+]
+
+[[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 = "proc-macro2"
+version = "1.0.24"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "1e0704ee1a7e00d7bb417d0770ea303c1bccbabf0ef1667dae92b5967f5f8a71"
+dependencies = [
+ "unicode-xid",
+]
+
+[[package]]
+name = "quote"
+version = "1.0.7"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "aa563d17ecb180e500da1cfd2b028310ac758de548efdd203e18f283af693f37"
+dependencies = [
+ "proc-macro2",
+]
+
+[[package]]
+name = "regex-syntax"
+version = "0.6.21"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "3b181ba2dcf07aaccad5448e8ead58db5b742cf85dfe035e2227f137a539a189"
+
+[[package]]
+name = "syn"
+version = "1.0.54"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "9a2af957a63d6bd42255c359c93d9bfdb97076bd3b820897ce55ffbfbf107f44"
+dependencies = [
+ "proc-macro2",
+ "quote",
+ "unicode-xid",
+]
+
+[[package]]
+name = "time"
+version = "0.1.44"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "6db9e6914ab8b1ae1c260a4ae7a49b6c5611b40328a735b21862567685e73255"
+dependencies = [
+ "libc",
+ "wasi",
+ "winapi",
+]
+
+[[package]]
+name = "unicode-xid"
+version = "0.2.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "f7fe0bb3479651439c9112f72b6c505038574c9fbb575ed1bf3b797fa39dd564"
+
+[[package]]
+name = "utf8-ranges"
+version = "1.0.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "b4ae116fef2b7fea257ed6440d3cfcff7f190865f170cdad00bb6465bf18ecba"
+
+[[package]]
+name = "wasi"
+version = "0.10.0+wasi-snapshot-preview1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "1a143597ca7c7793eff794def352d41792a93c481eb1042423ff7ff72ba2c31f"
+
+[[package]]
+name = "winapi"
+version = "0.3.9"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "5c839a674fcd7a98952e593242ea400abe93992746761e38641405d28b00f419"
+dependencies = [
+ "winapi-i686-pc-windows-gnu",
+ "winapi-x86_64-pc-windows-gnu",
+]
+
+[[package]]
+name = "winapi-i686-pc-windows-gnu"
+version = "0.4.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "ac3b87c63620426dd9b991e5ce0329eff545bccbbb34f3be09ff6fb6ab51b7b6"
+
+[[package]]
+name = "winapi-x86_64-pc-windows-gnu"
+version = "0.4.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "712e227841d057c1ee1cd2fb22fa7e5a5461ae8e48fa2ca79ec42cfc1931183f"
--- a/apps/cassiopeia/Cargo.toml
+++ b/apps/cassiopeia/Cargo.toml
@ -5,4 +5,5 @@ authors = ["Mx Kookie <kookie@spacekookie.de>"]
 edition = "2018"

 [dependencies]
-chrono = "*"
+chrono = "*"
+logos = "0.11"
--- a/apps/cassiopeia/shell.nix
+++ b/apps/cassiopeia/shell.nix
@ -0,0 +1,8 @@
+with import <nixpkgs> {};
+
+stdenv.mkDerivation {
+  name = "cassiopeia";
+  buildInputs = with pkgs; [
+    rustracer rustup clangStdenv
+  ];
+}
--- a/apps/cassiopeia/src/format/lexer.rs
+++ b/apps/cassiopeia/src/format/lexer.rs
@ -0,0 +1,116 @@
+//! Cassiopeia file lexer
+
+use logos::Logos;
+
+/// A basic line lexer type
+///
+/// This lexer distinguishes between comments, and keyword lines.  It
+/// does not attempt to parse the line specifics.  This is what the
+/// content lexer is for.
+#[derive(Logos, Debug, PartialEq)]
+enum Line {
+
+    #[token("HEADER")]
+    Header,
+
+    #[token("START")]
+    Start,
+
+    #[token("STOP")]
+    Stop,
+
+    #[token("INVOICE")]
+    Invoice,
+    
+    #[regex(r"\w+=[^,$]+[,$]")]
+    HeaderData,
+
+    // FIXME: this will have a leading whitespace that we could remove
+    // with ^\w, but logos does not support this at the moment
+    #[regex(r"[0-9-:+ ]+")]
+    Date,
+
+    #[token(" ", logos::skip)]
+    Space,
+    
+    #[error]
+    Error,
+}
+
+
+// pub fn test_this() {
+//     // let mut lex = Line::lexer("HEADER version=0.0.0,location=Berlin,");
+//     let mut lex = Line::lexer("START 2020-11-11 13:00:00+01:00");
+
+//     while let Some(t) = lex.next() {
+//         println!("{:?}: {}", t, lex.slice());
+//     }
+// }
+
+
+#[test]
+fn basic_header() {
+    let mut lex = Line::lexer("HEADER version=0.0.0,location=Berlin Lichtenberg,");
+
+    assert_eq!(lex.next(), Some(Line::Header));
+    assert_eq!(lex.span(), 0..6);
+    assert_eq!(lex.slice(), "HEADER");
+
+    assert_eq!(lex.next(), Some(Line::HeaderData));
+    assert_eq!(lex.span(), 7..21);
+    assert_eq!(lex.slice(), "version=0.0.0,");
+
+    assert_eq!(lex.next(), Some(Line::HeaderData));
+    assert_eq!(lex.span(), 21..49);
+    assert_eq!(lex.slice(), "location=Berlin Lichtenberg,");
+
+    assert_eq!(lex.next(), None);
+}
+
+
+#[test]
+fn basic_start() {
+    let mut lex = Line::lexer("START 2020-11-11 13:00:00+01:00");
+
+    assert_eq!(lex.next(), Some(Line::Start));
+    assert_eq!(lex.span(), 0..5);
+    assert_eq!(lex.slice(), "START");
+
+    assert_eq!(lex.next(), Some(Line::Date));
+    assert_eq!(lex.span(), 5..31);
+    assert_eq!(lex.slice(), " 2020-11-11 13:00:00+01:00");
+
+    assert_eq!(lex.next(), None);
+}
+
+
+#[test]
+fn basic_stop() {
+    let mut lex = Line::lexer("STOP 2020-11-11 13:00:00+01:00");
+
+    assert_eq!(lex.next(), Some(Line::Stop));
+    assert_eq!(lex.span(), 0..4);
+    assert_eq!(lex.slice(), "STOP");
+
+    assert_eq!(lex.next(), Some(Line::Date));
+    assert_eq!(lex.span(), 4..30);
+    assert_eq!(lex.slice(), " 2020-11-11 13:00:00+01:00");
+
+    assert_eq!(lex.next(), None);
+}
+
+
+#[test]
+fn basic_invoice() {
+    let mut lex = Line::lexer("INVOICE 2020-11-11 13:00:00+01:00");
+
+    assert_eq!(lex.next(), Some(Line::Invoice));
+    assert_eq!(lex.span(), 0..7);
+    assert_eq!(lex.slice(), "INVOICE");
+
+    assert_eq!(lex.next(), Some(Line::Date));
+    assert_eq!(lex.span(), 7..33);
+    assert_eq!(lex.slice(), " 2020-11-11 13:00:00+01:00");
+
+    assert_eq!(lex.next(), None);
+}
--- a/apps/cassiopeia/src/format/mod.rs
+++ b/apps/cassiopeia/src/format/mod.rs
@ -0,0 +1,3 @@
+//! cassiopeia file format handling
+
+pub(crate) mod lexer;
--- a/apps/cassiopeia/src/main.rs
+++ b/apps/cassiopeia/src/main.rs
@ -1,3 +1,5 @@
+mod format;
+
 fn main() {
-    println!("Hello, world!");
+
 }