From 80f8ede0bbc9799d5199707e1e1ad8e80e4ca7ac Mon Sep 17 00:00:00 2001 From: Griffin Smith Date: Sun, 7 Mar 2021 15:29:59 -0500 Subject: Initial commit --- .envrc | 1 + .gitignore | 1 + Cargo.lock | 789 +++++++++++++++++++++++++++++++++++++++++++++++ Cargo.toml | 18 ++ ach/.gitignore | 2 + ach/Makefile | 15 + ach/functions.ach | 3 + ach/simple.ach | 1 + shell.nix | 20 ++ src/ast/mod.rs | 166 ++++++++++ src/codegen/llvm.rs | 282 +++++++++++++++++ src/codegen/mod.rs | 26 ++ src/commands/compile.rs | 28 ++ src/commands/eval.rs | 30 ++ src/commands/mod.rs | 5 + src/common/env.rs | 40 +++ src/common/error.rs | 50 +++ src/common/mod.rs | 4 + src/compiler.rs | 84 +++++ src/interpreter/error.rs | 16 + src/interpreter/mod.rs | 125 ++++++++ src/interpreter/value.rs | 134 ++++++++ src/main.rs | 31 ++ src/parser/mod.rs | 445 ++++++++++++++++++++++++++ 24 files changed, 2316 insertions(+) create mode 100644 .envrc create mode 100644 .gitignore create mode 100644 Cargo.lock create mode 100644 Cargo.toml create mode 100644 ach/.gitignore create mode 100644 ach/Makefile create mode 100644 ach/functions.ach create mode 100644 ach/simple.ach create mode 100644 shell.nix create mode 100644 src/ast/mod.rs create mode 100644 src/codegen/llvm.rs create mode 100644 src/codegen/mod.rs create mode 100644 src/commands/compile.rs create mode 100644 src/commands/eval.rs create mode 100644 src/commands/mod.rs create mode 100644 src/common/env.rs create mode 100644 src/common/error.rs create mode 100644 src/common/mod.rs create mode 100644 src/compiler.rs create mode 100644 src/interpreter/error.rs create mode 100644 src/interpreter/mod.rs create mode 100644 src/interpreter/value.rs create mode 100644 src/main.rs create mode 100644 src/parser/mod.rs diff --git a/.envrc b/.envrc new file mode 100644 index 000000000000..051d09d292a8 --- /dev/null +++ b/.envrc @@ -0,0 +1 @@ +eval "$(lorri direnv)" diff --git a/.gitignore b/.gitignore new file mode 100644 index 000000000000..ea8c4bf7f35f --- /dev/null +++ b/.gitignore @@ -0,0 +1 @@ +/target diff --git a/Cargo.lock b/Cargo.lock new file mode 100644 index 000000000000..485e9f4cdfcb --- /dev/null +++ b/Cargo.lock @@ -0,0 +1,789 @@ +# This file is automatically @generated by Cargo. +# It is not intended for manual editing. +[[package]] +name = "achilles" +version = "0.1.0" +dependencies = [ + "anyhow", + "clap", + "derive_more", + "inkwell", + "llvm-sys", + "nom", + "nom-trace", + "pratt", + "proptest", + "test-strategy", + "thiserror", +] + +[[package]] +name = "aho-corasick" +version = "0.7.15" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "7404febffaa47dac81aa44dba71523c9d069b1bdc50a77db41195149e17f68e5" +dependencies = [ + "memchr", +] + +[[package]] +name = "anyhow" +version = "1.0.38" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "afddf7f520a80dbf76e6f50a35bca42a2331ef227a28b3b6dc5c2e2338d114b1" + +[[package]] +name = "arrayvec" +version = "0.5.2" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "23b62fc65de8e4e7f52534fb52b0f3ed04746ae267519eef2a83941e8085068b" + +[[package]] +name = "atty" +version = "0.2.14" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "d9b39be18770d11421cdb1b9947a45dd3f37e93092cbf377614828a319d5fee8" +dependencies = [ + "hermit-abi", + "libc", + "winapi", +] + +[[package]] +name = "autocfg" +version = "1.0.1" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "cdb031dd78e28731d87d56cc8ffef4a8f36ca26c38fe2de700543e627f8a464a" + +[[package]] +name = "bit-set" +version = "0.5.2" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "6e11e16035ea35e4e5997b393eacbf6f63983188f7a2ad25bfb13465f5ad59de" +dependencies = [ + "bit-vec", +] + +[[package]] +name = "bit-vec" +version = "0.6.3" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "349f9b6a179ed607305526ca489b34ad0a41aed5f7980fa90eb03160b69598fb" + +[[package]] +name = "bitflags" +version = "1.2.1" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "cf1de2fe8c75bc145a2f577add951f8134889b4795d47466a54a5c846d691693" + +[[package]] +name = "bitvec" +version = "0.19.5" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "8942c8d352ae1838c9dda0b0ca2ab657696ef2232a20147cf1b30ae1a9cb4321" +dependencies = [ + "funty", + "radium", + "tap", + "wyz", +] + +[[package]] +name = "byteorder" +version = "1.4.2" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "ae44d1a3d5a19df61dd0c8beb138458ac2a53a7ac09eba97d55592540004306b" + +[[package]] +name = "cc" +version = "1.0.67" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "e3c69b077ad434294d3ce9f1f6143a2a4b89a8a2d54ef813d85003a4fd1137fd" + +[[package]] +name = "cfg-if" +version = "1.0.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "baf1de4339761588bc0619e3cbc0120ee582ebb74b53b4efbf79117bd2da40fd" + +[[package]] +name = "clap" +version = "3.0.0-beta.2" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "4bd1061998a501ee7d4b6d449020df3266ca3124b941ec56cf2005c3779ca142" +dependencies = [ + "atty", + "bitflags", + "clap_derive", + "indexmap", + "lazy_static", + "os_str_bytes", + "strsim", + "termcolor", + "textwrap", + "unicode-width", + "vec_map", +] + +[[package]] +name = "clap_derive" +version = "3.0.0-beta.2" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "370f715b81112975b1b69db93e0b56ea4cd4e5002ac43b2da8474106a54096a1" +dependencies = [ + "heck", + "proc-macro-error", + "proc-macro2", + "quote", + "syn", +] + +[[package]] +name = "derive_more" +version = "0.99.11" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "41cb0e6161ad61ed084a36ba71fbba9e3ac5aee3606fb607fe08da6acbcf3d8c" +dependencies = [ + "proc-macro2", + "quote", + "syn", +] + +[[package]] +name = "either" +version = "1.6.1" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "e78d4f1cc4ae33bbfc157ed5d5a5ef3bc29227303d595861deb238fcec4e9457" + +[[package]] +name = "fnv" +version = "1.0.7" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "3f9eec918d3f24069decb9af1554cad7c880e2da24a9afd88aca000531ab82c1" + +[[package]] +name = "funty" +version = "1.1.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "fed34cd105917e91daa4da6b3728c47b068749d6a62c59811f06ed2ac71d9da7" + +[[package]] +name = "getrandom" +version = "0.2.2" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "c9495705279e7140bf035dde1f6e750c162df8b625267cd52cc44e0b156732c8" +dependencies = [ + "cfg-if", + "libc", + "wasi", +] + +[[package]] +name = "hashbrown" +version = "0.9.1" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "d7afe4a420e3fe79967a00898cc1f4db7c8a49a9333a29f8a4bd76a253d5cd04" + +[[package]] +name = "heck" +version = "0.3.2" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "87cbf45460356b7deeb5e3415b5563308c0a9b057c85e12b06ad551f98d0a6ac" +dependencies = [ + "unicode-segmentation", +] + +[[package]] +name = "hermit-abi" +version = "0.1.18" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "322f4de77956e22ed0e5032c359a0f1273f1f7f0d79bfa3b8ffbc730d7fbcc5c" +dependencies = [ + "libc", +] + +[[package]] +name = "indexmap" +version = "1.6.2" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "824845a0bf897a9042383849b02c1bc219c2383772efcd5c6f9766fa4b81aef3" +dependencies = [ + "autocfg", + "hashbrown", +] + +[[package]] +name = "inkwell" +version = "0.1.0" +source = "git+https://github.com/TheDan64/inkwell?branch=master#a2db15b0bd1c06d71763585ae10d9ea4e775da0c" +dependencies = [ + "either", + "inkwell_internals", + "libc", + "llvm-sys", + "once_cell", + "parking_lot", + "regex", +] + +[[package]] +name = "inkwell_internals" +version = "0.3.0" +source = "git+https://github.com/TheDan64/inkwell?branch=master#a2db15b0bd1c06d71763585ae10d9ea4e775da0c" +dependencies = [ + "proc-macro2", + "quote", + "syn", +] + +[[package]] +name = "instant" +version = "0.1.9" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "61124eeebbd69b8190558df225adf7e4caafce0d743919e5d6b19652314ec5ec" +dependencies = [ + "cfg-if", +] + +[[package]] +name = "lazy_static" +version = "1.4.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "e2abad23fbc42b3700f2f279844dc832adb2b2eb069b2df918f455c4e18cc646" + +[[package]] +name = "lexical-core" +version = "0.7.5" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "21f866863575d0e1d654fbeeabdc927292fdf862873dc3c96c6f753357e13374" +dependencies = [ + "arrayvec", + "bitflags", + "cfg-if", + "ryu", + "static_assertions", +] + +[[package]] +name = "libc" +version = "0.2.88" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "03b07a082330a35e43f63177cc01689da34fbffa0105e1246cf0311472cac73a" + +[[package]] +name = "llvm-sys" +version = "110.0.1" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "21ede189444b8c78907e5d36da5dabcf153170fcff9c1dba48afc4b33c7e19f0" +dependencies = [ + "cc", + "lazy_static", + "libc", + "regex", + "semver", +] + +[[package]] +name = "lock_api" +version = "0.4.2" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "dd96ffd135b2fd7b973ac026d28085defbe8983df057ced3eb4f2130b0831312" +dependencies = [ + "scopeguard", +] + +[[package]] +name = "memchr" +version = "2.3.4" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "0ee1c47aaa256ecabcaea351eae4a9b01ef39ed810004e298d2511ed284b1525" + +[[package]] +name = "nom" +version = "6.1.2" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "e7413f999671bd4745a7b624bd370a569fb6bc574b23c83a3c5ed2e453f3d5e2" +dependencies = [ + "bitvec", + "funty", + "lexical-core", + "memchr", + "version_check", +] + +[[package]] +name = "nom-trace" +version = "0.2.1" +source = "git+https://github.com/glittershark/nom-trace?branch=nom-6#6168d2e15cc51efd12d80260159b76a764dba138" +dependencies = [ + "nom", +] + +[[package]] +name = "num-traits" +version = "0.2.14" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "9a64b1ec5cda2586e284722486d802acf1f7dbdc623e2bfc57e65ca1cd099290" +dependencies = [ + "autocfg", +] + +[[package]] +name = "once_cell" +version = "1.7.2" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "af8b08b04175473088b46763e51ee54da5f9a164bc162f615b91bc179dbf15a3" + +[[package]] +name = "os_str_bytes" +version = "2.4.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "afb2e1c3ee07430c2cf76151675e583e0f19985fa6efae47d6848a3e2c824f85" + +[[package]] +name = "parking_lot" +version = "0.11.1" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "6d7744ac029df22dca6284efe4e898991d28e3085c706c972bcd7da4a27a15eb" +dependencies = [ + "instant", + "lock_api", + "parking_lot_core", +] + +[[package]] +name = "parking_lot_core" +version = "0.8.3" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "fa7a782938e745763fe6907fc6ba86946d72f49fe7e21de074e08128a99fb018" +dependencies = [ + "cfg-if", + "instant", + "libc", + "redox_syscall", + "smallvec", + "winapi", +] + +[[package]] +name = "pest" +version = "2.1.3" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "10f4872ae94d7b90ae48754df22fd42ad52ce740b8f370b03da4835417403e53" +dependencies = [ + "ucd-trie", +] + +[[package]] +name = "ppv-lite86" +version = "0.2.10" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "ac74c624d6b2d21f425f752262f42188365d7b8ff1aff74c82e45136510a4857" + +[[package]] +name = "pratt" +version = "0.3.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "e31bbc12f7936a7b195790dd6d9b982b66c54f45ff6766decf25c44cac302dce" + +[[package]] +name = "proc-macro-error" +version = "1.0.4" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "da25490ff9892aab3fcf7c36f08cfb902dd3e71ca0f9f9517bea02a73a5ce38c" +dependencies = [ + "proc-macro-error-attr", + "proc-macro2", + "quote", + "syn", + "version_check", +] + +[[package]] +name = "proc-macro-error-attr" +version = "1.0.4" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "a1be40180e52ecc98ad80b184934baf3d0d29f979574e439af5a55274b35f869" +dependencies = [ + "proc-macro2", + "quote", + "version_check", +] + +[[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 = "proptest" +version = "1.0.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "1e0d9cc07f18492d879586c92b485def06bc850da3118075cd45d50e9c95b0e5" +dependencies = [ + "bit-set", + "bitflags", + "byteorder", + "lazy_static", + "num-traits", + "quick-error 2.0.0", + "rand", + "rand_chacha", + "rand_xorshift", + "regex-syntax", + "rusty-fork", + "tempfile", +] + +[[package]] +name = "quick-error" +version = "1.2.3" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "a1d01941d82fa2ab50be1e79e6714289dd7cde78eba4c074bc5a4374f650dfe0" + +[[package]] +name = "quick-error" +version = "2.0.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "3ac73b1112776fc109b2e61909bc46c7e1bf0d7f690ffb1676553acce16d5cda" + +[[package]] +name = "quote" +version = "1.0.9" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "c3d0b9745dc2debf507c8422de05d7226cc1f0644216dfdfead988f9b1ab32a7" +dependencies = [ + "proc-macro2", +] + +[[package]] +name = "radium" +version = "0.5.3" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "941ba9d78d8e2f7ce474c015eea4d9c6d25b6a3327f9832ee29a4de27f91bbb8" + +[[package]] +name = "rand" +version = "0.8.3" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "0ef9e7e66b4468674bfcb0c81af8b7fa0bb154fa9f28eb840da5c447baeb8d7e" +dependencies = [ + "libc", + "rand_chacha", + "rand_core", + "rand_hc", +] + +[[package]] +name = "rand_chacha" +version = "0.3.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "e12735cf05c9e10bf21534da50a147b924d555dc7a547c42e6bb2d5b6017ae0d" +dependencies = [ + "ppv-lite86", + "rand_core", +] + +[[package]] +name = "rand_core" +version = "0.6.2" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "34cf66eb183df1c5876e2dcf6b13d57340741e8dc255b48e40a26de954d06ae7" +dependencies = [ + "getrandom", +] + +[[package]] +name = "rand_hc" +version = "0.3.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "3190ef7066a446f2e7f42e239d161e905420ccab01eb967c9eb27d21b2322a73" +dependencies = [ + "rand_core", +] + +[[package]] +name = "rand_xorshift" +version = "0.3.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "d25bf25ec5ae4a3f1b92f929810509a2f53d7dca2f50b794ff57e3face536c8f" +dependencies = [ + "rand_core", +] + +[[package]] +name = "redox_syscall" +version = "0.2.5" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "94341e4e44e24f6b591b59e47a8a027df12e008d73fd5672dbea9cc22f4507d9" +dependencies = [ + "bitflags", +] + +[[package]] +name = "regex" +version = "1.4.3" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "d9251239e129e16308e70d853559389de218ac275b515068abc96829d05b948a" +dependencies = [ + "aho-corasick", + "memchr", + "regex-syntax", + "thread_local", +] + +[[package]] +name = "regex-syntax" +version = "0.6.22" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "b5eb417147ba9860a96cfe72a0b93bf88fee1744b5636ec99ab20c1aa9376581" + +[[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 = "rusty-fork" +version = "0.3.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "cb3dcc6e454c328bb824492db107ab7c0ae8fcffe4ad210136ef014458c1bc4f" +dependencies = [ + "fnv", + "quick-error 1.2.3", + "tempfile", + "wait-timeout", +] + +[[package]] +name = "ryu" +version = "1.0.5" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "71d301d4193d031abdd79ff7e3dd721168a9572ef3fe51a1517aba235bd8f86e" + +[[package]] +name = "scopeguard" +version = "1.1.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "d29ab0c6d3fc0ee92fe66e2d99f700eab17a8d57d1c1d3b748380fb20baa78cd" + +[[package]] +name = "semver" +version = "0.11.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "f301af10236f6df4160f7c3f04eec6dbc70ace82d23326abad5edee88801c6b6" +dependencies = [ + "semver-parser", +] + +[[package]] +name = "semver-parser" +version = "0.10.2" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "00b0bef5b7f9e0df16536d3961cfb6e84331c065b4066afb39768d0e319411f7" +dependencies = [ + "pest", +] + +[[package]] +name = "smallvec" +version = "1.6.1" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "fe0f37c9e8f3c5a4a66ad655a93c74daac4ad00c441533bf5c6e7990bb42604e" + +[[package]] +name = "static_assertions" +version = "1.1.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "a2eb9349b6444b326872e140eb1cf5e7c522154d69e7a0ffb0fb81c06b37543f" + +[[package]] +name = "strsim" +version = "0.10.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "73473c0e59e6d5812c5dfe2a064a6444949f089e20eec9a2e5506596494e4623" + +[[package]] +name = "syn" +version = "1.0.61" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "ed22b90a0e734a23a7610f4283ac9e5acfb96cbb30dfefa540d66f866f1c09c5" +dependencies = [ + "proc-macro2", + "quote", + "unicode-xid", +] + +[[package]] +name = "tap" +version = "1.0.1" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "55937e1799185b12863d447f42597ed69d9928686b8d88a1df17376a097d8369" + +[[package]] +name = "tempfile" +version = "3.2.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "dac1c663cfc93810f88aed9b8941d48cabf856a1b111c29a40439018d870eb22" +dependencies = [ + "cfg-if", + "libc", + "rand", + "redox_syscall", + "remove_dir_all", + "winapi", +] + +[[package]] +name = "termcolor" +version = "1.1.2" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "2dfed899f0eb03f32ee8c6a0aabdb8a7949659e3466561fc0adf54e26d88c5f4" +dependencies = [ + "winapi-util", +] + +[[package]] +name = "test-strategy" +version = "0.1.1" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "2328963c69243416e811c88066d18f670792b2e36e17fa57f4b1a124f85d18a8" +dependencies = [ + "proc-macro2", + "quote", + "syn", +] + +[[package]] +name = "textwrap" +version = "0.12.1" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "203008d98caf094106cfaba70acfed15e18ed3ddb7d94e49baec153a2b462789" +dependencies = [ + "unicode-width", +] + +[[package]] +name = "thiserror" +version = "1.0.24" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "e0f4a65597094d4483ddaed134f409b2cb7c1beccf25201a9f73c719254fa98e" +dependencies = [ + "thiserror-impl", +] + +[[package]] +name = "thiserror-impl" +version = "1.0.24" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "7765189610d8241a44529806d6fd1f2e0a08734313a35d5b3a556f92b381f3c0" +dependencies = [ + "proc-macro2", + "quote", + "syn", +] + +[[package]] +name = "thread_local" +version = "1.1.3" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "8018d24e04c95ac8790716a5987d0fec4f8b27249ffa0f7d33f1369bdfb88cbd" +dependencies = [ + "once_cell", +] + +[[package]] +name = "ucd-trie" +version = "0.1.3" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "56dee185309b50d1f11bfedef0fe6d036842e3fb77413abef29f8f8d1c5d4c1c" + +[[package]] +name = "unicode-segmentation" +version = "1.7.1" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "bb0d2e7be6ae3a5fa87eed5fb451aff96f2573d2694942e40543ae0bbe19c796" + +[[package]] +name = "unicode-width" +version = "0.1.8" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "9337591893a19b88d8d87f2cec1e73fad5cdfd10e5a6f349f498ad6ea2ffb1e3" + +[[package]] +name = "unicode-xid" +version = "0.2.1" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "f7fe0bb3479651439c9112f72b6c505038574c9fbb575ed1bf3b797fa39dd564" + +[[package]] +name = "vec_map" +version = "0.8.2" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "f1bddf1187be692e79c5ffeab891132dfb0f236ed36a43c7ed39f1165ee20191" + +[[package]] +name = "version_check" +version = "0.9.2" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "b5a972e5669d67ba988ce3dc826706fb0a8b01471c088cb0b6110b805cc36aed" + +[[package]] +name = "wait-timeout" +version = "0.2.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "9f200f5b12eb75f8c1ed65abd4b2db8a6e1b138a20de009dacee265a2498f3f6" +dependencies = [ + "libc", +] + +[[package]] +name = "wasi" +version = "0.10.2+wasi-snapshot-preview1" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "fd6fbd9a79829dd1ad0cc20627bf1ed606756a7f77edff7b66b7064f9cb327c6" + +[[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-util" +version = "0.1.5" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "70ec6ce85bb158151cae5e5c87f95a8e97d2c0c4b001223f33a334e3ce5de178" +dependencies = [ + "winapi", +] + +[[package]] +name = "winapi-x86_64-pc-windows-gnu" +version = "0.4.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "712e227841d057c1ee1cd2fb22fa7e5a5461ae8e48fa2ca79ec42cfc1931183f" + +[[package]] +name = "wyz" +version = "0.2.0" +source = "registry+https://github.com/rust-lang/crates.io-index" +checksum = "85e60b0d1b5f99db2556934e21937020776a5d31520bf169e851ac44e6420214" diff --git a/Cargo.toml b/Cargo.toml new file mode 100644 index 000000000000..eda15e554661 --- /dev/null +++ b/Cargo.toml @@ -0,0 +1,18 @@ +[package] +name = "achilles" +version = "0.1.0" +authors = ["Griffin Smith "] +edition = "2018" + +[dependencies] +anyhow = "1.0.38" +clap = "3.0.0-beta.2" +derive_more = "0.99.11" +inkwell = { git = "https://github.com/TheDan64/inkwell", branch = "master", features = ["llvm11-0"] } +llvm-sys = "110.0.1" +nom = "6.1.2" +nom-trace = { git = "https://github.com/glittershark/nom-trace", branch = "nom-6" } +pratt = "0.3.0" +proptest = "1.0.0" +test-strategy = "0.1.1" +thiserror = "1.0.24" diff --git a/ach/.gitignore b/ach/.gitignore new file mode 100644 index 000000000000..e8423ae351b8 --- /dev/null +++ b/ach/.gitignore @@ -0,0 +1,2 @@ +*.ll +*.o diff --git a/ach/Makefile b/ach/Makefile new file mode 100644 index 000000000000..869a0d0f8a3e --- /dev/null +++ b/ach/Makefile @@ -0,0 +1,15 @@ +default: simple + +%.ll: %.ach + cargo run -- compile $< -o $@ -f llvm + +%.o: %.ll + llc $< -o $@ -filetype=obj + +%: %.o + clang $< -o $@ + +.PHONY: clean + +clean: + @rm -f *.ll *.o simple diff --git a/ach/functions.ach b/ach/functions.ach new file mode 100644 index 000000000000..8d91564c1559 --- /dev/null +++ b/ach/functions.ach @@ -0,0 +1,3 @@ +fn id x = x +fn plus x y = x + y +fn main = plus (id 2) 7 diff --git a/ach/simple.ach b/ach/simple.ach new file mode 100644 index 000000000000..20f1677235c0 --- /dev/null +++ b/ach/simple.ach @@ -0,0 +1 @@ +fn main = let x = 2; y = 3 in x + y diff --git a/shell.nix b/shell.nix new file mode 100644 index 000000000000..cdf74db415ca --- /dev/null +++ b/shell.nix @@ -0,0 +1,20 @@ +with import (builtins.fetchTarball { + url = "https://github.com/nixos/nixpkgs/archive/93a812bb9f9c398bd5b9636ab3674dcfe8cfb884.tar.gz"; + sha256 = "14zzsgnigd7vjbrpzm1s4qsknm73sci38ss00x96wamz6psaxyah"; +}) {}; + +mkShell { + buildInputs = [ + clang_11 + llvm_11.lib + llvmPackages_11.bintools + llvmPackages_11.clang + llvmPackages_11.libclang.lib + zlib + ncurses + libxml2 + libffi + pkg-config + ]; + +} diff --git a/src/ast/mod.rs b/src/ast/mod.rs new file mode 100644 index 000000000000..2dcf955fe67c --- /dev/null +++ b/src/ast/mod.rs @@ -0,0 +1,166 @@ +use std::borrow::Cow; +use std::convert::TryFrom; +use std::fmt::{self, Display, Formatter}; + +#[derive(Debug, PartialEq, Eq)] +pub struct InvalidIdentifier<'a>(Cow<'a, str>); + +#[derive(Debug, PartialEq, Eq, Hash)] +pub struct Ident<'a>(pub Cow<'a, str>); + +impl<'a> From<&'a Ident<'a>> for &'a str { + fn from(id: &'a Ident<'a>) -> Self { + id.0.as_ref() + } +} + +impl<'a> Display for Ident<'a> { + fn fmt(&self, f: &mut Formatter<'_>) -> fmt::Result { + write!(f, "{}", self.0) + } +} + +impl<'a> Ident<'a> { + pub fn to_owned(&self) -> Ident<'static> { + Ident(Cow::Owned(self.0.clone().into_owned())) + } + + pub fn from_str_unchecked(s: &'a str) -> Self { + debug_assert!(is_valid_identifier(s)); + Self(Cow::Borrowed(s)) + } + + pub fn from_string_unchecked(s: String) -> Self { + debug_assert!(is_valid_identifier(&s)); + Self(Cow::Owned(s)) + } +} + +pub fn is_valid_identifier(s: &S) -> bool +where + S: AsRef + ?Sized, +{ + s.as_ref() + .chars() + .any(|c| !c.is_alphanumeric() || !"_".contains(c)) +} + +impl<'a> TryFrom<&'a str> for Ident<'a> { + type Error = InvalidIdentifier<'a>; + + fn try_from(s: &'a str) -> Result { + if is_valid_identifier(s) { + Ok(Ident(Cow::Borrowed(s))) + } else { + Err(InvalidIdentifier(Cow::Borrowed(s))) + } + } +} + +impl<'a> TryFrom for Ident<'a> { + type Error = InvalidIdentifier<'static>; + + fn try_from(s: String) -> Result { + if is_valid_identifier(&s) { + Ok(Ident(Cow::Owned(s))) + } else { + Err(InvalidIdentifier(Cow::Owned(s))) + } + } +} + +#[derive(Debug, PartialEq, Eq)] +pub enum BinaryOperator { + /// `+` + Add, + + /// `-` + Sub, + + /// `*` + Mul, + + /// `/` + Div, + + /// `^` + Pow, + + /// `==` + Equ, + + /// `!=` + Neq, +} + +#[derive(Debug, PartialEq, Eq)] +pub enum UnaryOperator { + /// ! + Not, + + /// - + Neg, +} + +#[derive(Debug, PartialEq, Eq)] +pub enum Literal { + Int(u64), +} + +#[derive(Debug, PartialEq, Eq)] +pub enum Expr<'a> { + Ident(Ident<'a>), + + Literal(Literal), + + UnaryOp { + op: UnaryOperator, + rhs: Box>, + }, + + BinaryOp { + lhs: Box>, + op: BinaryOperator, + rhs: Box>, + }, + + Let { + bindings: Vec<(Ident<'a>, Expr<'a>)>, + body: Box>, + }, + + If { + condition: Box>, + then: Box>, + else_: Box>, + }, +} + +#[derive(Debug, PartialEq, Eq)] +pub struct Fun<'a> { + pub name: Ident<'a>, + pub args: Vec>, + pub body: Expr<'a>, +} + +#[derive(Debug, PartialEq, Eq)] +pub enum Decl<'a> { + Fun(Fun<'a>), +} + +#[derive(Debug, PartialEq, Eq, Clone, Copy)] +pub enum Type { + Int, + Float, + Bool, +} + +impl Display for Type { + fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result { + match self { + Self::Int => f.write_str("int"), + Self::Float => f.write_str("float"), + Self::Bool => f.write_str("bool"), + } + } +} diff --git a/src/codegen/llvm.rs b/src/codegen/llvm.rs new file mode 100644 index 000000000000..ff92c3373176 --- /dev/null +++ b/src/codegen/llvm.rs @@ -0,0 +1,282 @@ +use std::path::Path; +use std::result; + +use inkwell::basic_block::BasicBlock; +use inkwell::builder::Builder; +pub use inkwell::context::Context; +use inkwell::module::Module; +use inkwell::support::LLVMString; +use inkwell::types::FunctionType; +use inkwell::values::{BasicValueEnum, FunctionValue}; +use inkwell::IntPredicate; +use thiserror::Error; + +use crate::ast::{BinaryOperator, Decl, Expr, Fun, Ident, Literal, UnaryOperator}; +use crate::common::env::Env; + +#[derive(Debug, PartialEq, Eq, Error)] +pub enum Error { + #[error("Undefined variable {0}")] + UndefinedVariable(Ident<'static>), + + #[error("LLVM Error: {0}")] + LLVMError(String), +} + +impl From for Error { + fn from(s: LLVMString) -> Self { + Self::LLVMError(s.to_string()) + } +} + +pub type Result = result::Result; + +pub struct Codegen<'ctx, 'ast> { + context: &'ctx Context, + pub module: Module<'ctx>, + builder: Builder<'ctx>, + env: Env<'ast, BasicValueEnum<'ctx>>, + function: Option>, +} + +impl<'ctx, 'ast> Codegen<'ctx, 'ast> { + pub fn new(context: &'ctx Context, module_name: &str) -> Self { + let module = context.create_module(module_name); + let builder = context.create_builder(); + Self { + context, + module, + builder, + env: Default::default(), + function: None, + } + } + + pub fn new_function<'a>( + &'a mut self, + name: &str, + ty: FunctionType<'ctx>, + ) -> &'a FunctionValue<'ctx> { + self.function = Some(self.module.add_function(name, ty, None)); + let basic_block = self.append_basic_block("entry"); + self.builder.position_at_end(basic_block); + self.function.as_ref().unwrap() + } + + pub fn finish_function(&self, res: &BasicValueEnum<'ctx>) { + self.builder.build_return(Some(res)); + } + + pub fn append_basic_block(&self, name: &str) -> BasicBlock<'ctx> { + self.context + .append_basic_block(self.function.unwrap(), name) + } + + pub fn codegen_expr(&mut self, expr: &'ast Expr<'ast>) -> Result> { + match expr { + Expr::Ident(id) => self + .env + .resolve(id) + .cloned() + .ok_or_else(|| Error::UndefinedVariable(id.to_owned())), + Expr::Literal(Literal::Int(i)) => { + let ty = self.context.i64_type(); + Ok(BasicValueEnum::IntValue(ty.const_int(*i, false))) + } + Expr::UnaryOp { op, rhs } => { + let rhs = self.codegen_expr(rhs)?; + match op { + UnaryOperator::Not => unimplemented!(), + UnaryOperator::Neg => Ok(BasicValueEnum::IntValue( + self.builder.build_int_neg(rhs.into_int_value(), "neg"), + )), + } + } + Expr::BinaryOp { lhs, op, rhs } => { + let lhs = self.codegen_expr(lhs)?; + let rhs = self.codegen_expr(rhs)?; + match op { + BinaryOperator::Add => { + Ok(BasicValueEnum::IntValue(self.builder.build_int_add( + lhs.into_int_value(), + rhs.into_int_value(), + "add", + ))) + } + BinaryOperator::Sub => { + Ok(BasicValueEnum::IntValue(self.builder.build_int_sub( + lhs.into_int_value(), + rhs.into_int_value(), + "add", + ))) + } + BinaryOperator::Mul => { + Ok(BasicValueEnum::IntValue(self.builder.build_int_sub( + lhs.into_int_value(), + rhs.into_int_value(), + "add", + ))) + } + BinaryOperator::Div => { + Ok(BasicValueEnum::IntValue(self.builder.build_int_signed_div( + lhs.into_int_value(), + rhs.into_int_value(), + "add", + ))) + } + BinaryOperator::Pow => unimplemented!(), + BinaryOperator::Equ => { + Ok(BasicValueEnum::IntValue(self.builder.build_int_compare( + IntPredicate::EQ, + lhs.into_int_value(), + rhs.into_int_value(), + "eq", + ))) + } + BinaryOperator::Neq => todo!(), + } + } + Expr::Let { bindings, body } => { + self.env.push(); + for (id, val) in bindings { + let val = self.codegen_expr(val)?; + self.env.set(id, val); + } + let res = self.codegen_expr(body); + self.env.pop(); + res + } + Expr::If { + condition, + then, + else_, + } => { + let then_block = self.append_basic_block("then"); + let else_block = self.append_basic_block("else"); + let join_block = self.append_basic_block("join"); + let condition = self.codegen_expr(condition)?; + self.builder.build_conditional_branch( + condition.into_int_value(), + then_block, + else_block, + ); + self.builder.position_at_end(then_block); + let then_res = self.codegen_expr(then)?; + self.builder.build_unconditional_branch(join_block); + + self.builder.position_at_end(else_block); + let else_res = self.codegen_expr(else_)?; + self.builder.build_unconditional_branch(join_block); + + self.builder.position_at_end(join_block); + let phi = self.builder.build_phi(self.context.i64_type(), "join"); + phi.add_incoming(&[(&then_res, then_block), (&else_res, else_block)]); + Ok(phi.as_basic_value()) + } + } + } + + pub fn codegen_decl(&mut self, decl: &'ast Decl<'ast>) -> Result<()> { + match decl { + Decl::Fun(Fun { name, args, body }) => { + let i64_type = self.context.i64_type(); + self.new_function( + name.into(), + i64_type.fn_type( + args.iter() + .map(|_| i64_type.into()) + .collect::>() + .as_slice(), + false, + ), + ); + self.env.push(); + for (i, arg) in args.iter().enumerate() { + self.env + .set(arg, self.function.unwrap().get_nth_param(i as u32).unwrap()); + } + let res = self.codegen_expr(body)?; + self.env.pop(); + self.finish_function(&res); + Ok(()) + } + } + } + + pub fn codegen_main(&mut self, expr: &'ast Expr<'ast>) -> Result<()> { + self.new_function("main", self.context.i64_type().fn_type(&[], false)); + let res = self.codegen_expr(expr)?; + self.finish_function(&res); + Ok(()) + } + + pub fn print_to_file

(&self, path: P) -> Result<()> + where + P: AsRef, + { + Ok(self.module.print_to_file(path)?) + } + + pub fn binary_to_file

(&self, path: P) -> Result<()> + where + P: AsRef, + { + if self.module.write_bitcode_to_path(path.as_ref()) { + Ok(()) + } else { + Err(Error::LLVMError( + "Error writing bitcode to output path".to_owned(), + )) + } + } +} + +#[cfg(test)] +mod tests { + use inkwell::execution_engine::JitFunction; + use inkwell::OptimizationLevel; + + use super::*; + + fn jit_eval(expr: &str) -> anyhow::Result { + let expr = crate::parser::expr(expr).unwrap().1; + + let context = Context::create(); + let mut codegen = Codegen::new(&context, "test"); + let execution_engine = codegen + .module + .create_jit_execution_engine(OptimizationLevel::None) + .unwrap(); + + codegen.new_function("test", context.i64_type().fn_type(&[], false)); + let res = codegen.codegen_expr(&expr)?; + codegen.finish_function(&res); + + unsafe { + let fun: JitFunction T> = + execution_engine.get_function("test")?; + Ok(fun.call()) + } + } + + #[test] + fn add_literals() { + assert_eq!(jit_eval::("1 + 2").unwrap(), 3); + } + + #[test] + fn variable_shadowing() { + assert_eq!( + jit_eval::("let x = 1 in (let x = 2 in x) + x").unwrap(), + 3 + ); + } + + #[test] + fn eq() { + assert_eq!( + jit_eval::("let x = 1 in if x == 1 then 2 else 4").unwrap(), + 2 + ); + } +} diff --git a/src/codegen/mod.rs b/src/codegen/mod.rs new file mode 100644 index 000000000000..4620b6b48e84 --- /dev/null +++ b/src/codegen/mod.rs @@ -0,0 +1,26 @@ +pub mod llvm; + +use inkwell::execution_engine::JitFunction; +use inkwell::OptimizationLevel; +pub use llvm::*; + +use crate::ast::Expr; +use crate::common::Result; + +pub fn jit_eval(expr: &Expr) -> Result { + let context = Context::create(); + let mut codegen = Codegen::new(&context, "eval"); + let execution_engine = codegen + .module + .create_jit_execution_engine(OptimizationLevel::None) + .map_err(Error::from)?; + codegen.new_function("eval", context.i64_type().fn_type(&[], false)); + let res = codegen.codegen_expr(&expr)?; + codegen.finish_function(&res); + + unsafe { + let fun: JitFunction T> = + execution_engine.get_function("eval").unwrap(); + Ok(fun.call()) + } +} diff --git a/src/commands/compile.rs b/src/commands/compile.rs new file mode 100644 index 000000000000..e16b8c87a659 --- /dev/null +++ b/src/commands/compile.rs @@ -0,0 +1,28 @@ +use std::path::PathBuf; + +use clap::Clap; + +use crate::common::Result; +use crate::compiler::{self, CompilerOptions}; + +#[derive(Clap)] +pub struct Compile { + file: PathBuf, + + #[clap(short = 'o')] + out_file: PathBuf, + + #[clap(flatten)] + options: CompilerOptions, +} + +impl Compile { + pub fn run(self) -> Result<()> { + eprintln!( + ">>> {} -> {}", + &self.file.to_string_lossy(), + self.out_file.to_string_lossy() + ); + compiler::compile_file(&self.file, &self.out_file, &self.options) + } +} diff --git a/src/commands/eval.rs b/src/commands/eval.rs new file mode 100644 index 000000000000..112bee64625b --- /dev/null +++ b/src/commands/eval.rs @@ -0,0 +1,30 @@ +use clap::Clap; + +use crate::codegen; +use crate::interpreter; +use crate::parser; +use crate::Result; + +/// Evaluate an expression and print its result +#[derive(Clap)] +pub struct Eval { + /// JIT-compile with LLVM instead of interpreting + #[clap(long)] + jit: bool, + + /// Expression to evaluate + expr: String, +} + +impl Eval { + pub fn run(self) -> Result<()> { + let (_, parsed) = parser::expr(&self.expr)?; + let result = if self.jit { + codegen::jit_eval::(&parsed)?.into() + } else { + interpreter::eval(&parsed)? + }; + println!("{}", result); + Ok(()) + } +} diff --git a/src/commands/mod.rs b/src/commands/mod.rs new file mode 100644 index 000000000000..9c0038dabfb1 --- /dev/null +++ b/src/commands/mod.rs @@ -0,0 +1,5 @@ +pub mod compile; +pub mod eval; + +pub use compile::Compile; +pub use eval::Eval; diff --git a/src/common/env.rs b/src/common/env.rs new file mode 100644 index 000000000000..8b5cde49e9e4 --- /dev/null +++ b/src/common/env.rs @@ -0,0 +1,40 @@ +use std::collections::HashMap; + +use crate::ast::Ident; + +/// A lexical environment +#[derive(Debug, PartialEq, Eq)] +pub struct Env<'ast, V>(Vec, V>>); + +impl<'ast, V> Default for Env<'ast, V> { + fn default() -> Self { + Self::new() + } +} + +impl<'ast, V> Env<'ast, V> { + pub fn new() -> Self { + Self(vec![Default::default()]) + } + + pub fn push(&mut self) { + self.0.push(Default::default()); + } + + pub fn pop(&mut self) { + self.0.pop(); + } + + pub fn set(&mut self, k: &'ast Ident<'ast>, v: V) { + self.0.last_mut().unwrap().insert(k, v); + } + + pub fn resolve<'a>(&'a self, k: &'ast Ident<'ast>) -> Option<&'a V> { + for ctx in self.0.iter().rev() { + if let Some(res) = ctx.get(k) { + return Some(res); + } + } + None + } +} diff --git a/src/common/error.rs b/src/common/error.rs new file mode 100644 index 000000000000..f3f3023ceaf8 --- /dev/null +++ b/src/common/error.rs @@ -0,0 +1,50 @@ +use std::{io, result}; + +use thiserror::Error; + +use crate::{codegen, interpreter, parser}; + +#[derive(Error, Debug)] +pub enum Error { + #[error(transparent)] + IOError(#[from] io::Error), + + #[error("Error parsing input: {0}")] + ParseError(#[from] parser::Error), + + #[error("Error evaluating expression: {0}")] + EvalError(#[from] interpreter::Error), + + #[error("Compile error: {0}")] + CodegenError(#[from] codegen::Error), + + #[error("{0}")] + Message(String), +} + +impl From for Error { + fn from(s: String) -> Self { + Self::Message(s) + } +} + +impl<'a> From>> for Error { + fn from(e: nom::Err>) -> Self { + use nom::error::Error as NomError; + use nom::Err::*; + + Self::ParseError(match e { + Incomplete(i) => Incomplete(i), + Error(NomError { input, code }) => Error(NomError { + input: input.to_owned(), + code, + }), + Failure(NomError { input, code }) => Failure(NomError { + input: input.to_owned(), + code, + }), + }) + } +} + +pub type Result = result::Result; diff --git a/src/common/mod.rs b/src/common/mod.rs new file mode 100644 index 000000000000..af5974a116fb --- /dev/null +++ b/src/common/mod.rs @@ -0,0 +1,4 @@ +pub(crate) mod env; +pub(crate) mod error; + +pub use error::{Error, Result}; diff --git a/src/compiler.rs b/src/compiler.rs new file mode 100644 index 000000000000..5f8e1ef4fa03 --- /dev/null +++ b/src/compiler.rs @@ -0,0 +1,84 @@ +use std::fmt::{self, Display}; +use std::path::Path; +use std::str::FromStr; +use std::{fs, result}; + +use clap::Clap; +use test_strategy::Arbitrary; + +use crate::codegen::{self, Codegen}; +use crate::common::Result; +use crate::parser; + +#[derive(Debug, Clone, Copy, PartialEq, Eq, Hash, Arbitrary)] +pub enum OutputFormat { + LLVM, + Bitcode, +} + +impl Default for OutputFormat { + fn default() -> Self { + Self::Bitcode + } +} + +impl FromStr for OutputFormat { + type Err = String; + + fn from_str(s: &str) -> result::Result { + match s { + "llvm" => Ok(Self::LLVM), + "binary" => Ok(Self::Bitcode), + _ => Err(format!( + "Invalid output format {}, expected one of {{llvm, binary}}", + s + )), + } + } +} + +impl Display for OutputFormat { + fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result { + match self { + OutputFormat::LLVM => f.write_str("llvm"), + OutputFormat::Bitcode => f.write_str("binary"), + } + } +} + +#[derive(Clap, Debug, PartialEq, Eq, Default)] +pub struct CompilerOptions { + #[clap(long, short = 'f', default_value)] + format: OutputFormat, +} + +pub fn compile_file(input: &Path, output: &Path, options: &CompilerOptions) -> Result<()> { + let src = fs::read_to_string(input)?; + let (_, decls) = parser::toplevel(&src)?; // TODO: statements + let context = codegen::Context::create(); + let mut codegen = Codegen::new( + &context, + &input + .file_stem() + .map_or("UNKNOWN".to_owned(), |s| s.to_string_lossy().into_owned()), + ); + for decl in &decls { + codegen.codegen_decl(decl)?; + } + match options.format { + OutputFormat::LLVM => codegen.print_to_file(output)?, + OutputFormat::Bitcode => codegen.binary_to_file(output)?, + } + Ok(()) +} + +#[cfg(test)] +mod tests { + use super::*; + use test_strategy::proptest; + + #[proptest] + fn output_format_display_from_str_round_trip(of: OutputFormat) { + assert_eq!(OutputFormat::from_str(&of.to_string()), Ok(of)); + } +} diff --git a/src/interpreter/error.rs b/src/interpreter/error.rs new file mode 100644 index 000000000000..e0299d180553 --- /dev/null +++ b/src/interpreter/error.rs @@ -0,0 +1,16 @@ +use std::result; + +use thiserror::Error; + +use crate::ast::{Ident, Type}; + +#[derive(Debug, PartialEq, Eq, Error)] +pub enum Error { + #[error("Undefined variable {0}")] + UndefinedVariable(Ident<'static>), + + #[error("Unexpected type {actual}, expected type {expected}")] + InvalidType { actual: Type, expected: Type }, +} + +pub type Result = result::Result; diff --git a/src/interpreter/mod.rs b/src/interpreter/mod.rs new file mode 100644 index 000000000000..adff3568c2c3 --- /dev/null +++ b/src/interpreter/mod.rs @@ -0,0 +1,125 @@ +mod error; +mod value; + +pub use self::error::{Error, Result}; +pub use self::value::Value; +use crate::ast::{BinaryOperator, Expr, Ident, Literal, UnaryOperator}; +use crate::common::env::Env; + +#[derive(Debug, Default)] +pub struct Interpreter<'a> { + env: Env<'a, Value>, +} + +impl<'a> Interpreter<'a> { + pub fn new() -> Self { + Self::default() + } + + fn resolve(&self, var: &'a Ident<'a>) -> Result { + self.env + .resolve(var) + .cloned() + .ok_or_else(|| Error::UndefinedVariable(var.to_owned())) + } + + pub fn eval(&mut self, expr: &'a Expr<'a>) -> Result { + match expr { + Expr::Ident(id) => self.resolve(id), + Expr::Literal(Literal::Int(i)) => Ok((*i).into()), + Expr::UnaryOp { op, rhs } => { + let rhs = self.eval(rhs)?; + match op { + UnaryOperator::Neg => -rhs, + _ => unimplemented!(), + } + } + Expr::BinaryOp { lhs, op, rhs } => { + let lhs = self.eval(lhs)?; + let rhs = self.eval(rhs)?; + match op { + BinaryOperator::Add => lhs + rhs, + BinaryOperator::Sub => lhs - rhs, + BinaryOperator::Mul => lhs * rhs, + BinaryOperator::Div => lhs / rhs, + BinaryOperator::Pow => todo!(), + BinaryOperator::Equ => Ok(lhs.eq(&rhs).into()), + BinaryOperator::Neq => todo!(), + } + } + Expr::Let { bindings, body } => { + self.env.push(); + for (id, val) in bindings { + let val = self.eval(val)?; + self.env.set(id, val); + } + let res = self.eval(body)?; + self.env.pop(); + Ok(res) + } + Expr::If { + condition, + then, + else_, + } => { + let condition = self.eval(condition)?; + if *(condition.into_type::()?) { + self.eval(then) + } else { + self.eval(else_) + } + } + } + } +} + +pub fn eval<'a>(expr: &'a Expr<'a>) -> Result { + let mut interpreter = Interpreter::new(); + interpreter.eval(expr) +} + +#[cfg(test)] +mod tests { + use std::convert::TryFrom; + + use super::value::{TypeOf, Val}; + use super::*; + use BinaryOperator::*; + + fn int_lit(i: u64) -> Box> { + Box::new(Expr::Literal(Literal::Int(i))) + } + + fn parse_eval(src: &str) -> T + where + for<'a> &'a T: TryFrom<&'a Val>, + T: Clone + TypeOf, + { + let expr = crate::parser::expr(src).unwrap().1; + let res = eval(&expr).unwrap(); + res.into_type::().unwrap().clone() + } + + #[test] + fn simple_addition() { + let expr = Expr::BinaryOp { + lhs: int_lit(1), + op: Mul, + rhs: int_lit(2), + }; + let res = eval(&expr).unwrap(); + assert_eq!(*res.into_type::().unwrap(), 2); + } + + #[test] + fn variable_shadowing() { + let res = parse_eval::("let x = 1 in (let x = 2 in x) + x"); + assert_eq!(res, 3); + } + + #[test] + fn conditional_with_equals() { + let res = parse_eval::("let x = 1 in if x == 1 then 2 else 4"); + assert_eq!(res, 2); + } +} diff --git a/src/interpreter/value.rs b/src/interpreter/value.rs new file mode 100644 index 000000000000..69e4d4ffeb96 --- /dev/null +++ b/src/interpreter/value.rs @@ -0,0 +1,134 @@ +use std::convert::TryFrom; +use std::fmt::{self, Display}; +use std::ops::{Add, Div, Mul, Neg, Sub}; +use std::rc::Rc; + +use derive_more::{Deref, From, TryInto}; + +use super::{Error, Result}; +use crate::ast::Type; + +#[derive(Debug, PartialEq, From, TryInto)] +#[try_into(owned, ref)] +pub enum Val { + Int(i64), + Float(f64), + Bool(bool), +} + +impl From for Val { + fn from(i: u64) -> Self { + Self::from(i as i64) + } +} + +impl Display for Val { + fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result { + match self { + Val::Int(x) => x.fmt(f), + Val::Float(x) => x.fmt(f), + Val::Bool(x) => x.fmt(f), + } + } +} + +impl Val { + pub fn type_(&self) -> Type { + match self { + Val::Int(_) => Type::Int, + Val::Float(_) => Type::Float, + Val::Bool(_) => Type::Bool, + } + } + + pub fn into_type<'a, T>(&'a self) -> Result<&'a T> + where + T: TypeOf + 'a + Clone, + &'a T: TryFrom<&'a Self>, + { + <&T>::try_from(self).map_err(|_| Error::InvalidType { + actual: self.type_(), + expected: ::type_of(), + }) + } +} + +#[derive(Debug, PartialEq, Clone, Deref)] +pub struct Value(Rc); + +impl Display for Value { + fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result { + self.0.fmt(f) + } +} + +impl From for Value +where + Val: From, +{ + fn from(x: T) -> Self { + Self(Rc::new(x.into())) + } +} + +impl Neg for Value { + type Output = Result; + + fn neg(self) -> Self::Output { + Ok((-self.into_type::()?).into()) + } +} + +impl Add for Value { + type Output = Result; + + fn add(self, rhs: Self) -> Self::Output { + Ok((self.into_type::()? + rhs.into_type::()?).into()) + } +} + +impl Sub for Value { + type Output = Result; + + fn sub(self, rhs: Self) -> Self::Output { + Ok((self.into_type::()? - rhs.into_type::()?).into()) + } +} + +impl Mul for Value { + type Output = Result; + + fn mul(self, rhs: Self) -> Self::Output { + Ok((self.into_type::()? * rhs.into_type::()?).into()) + } +} + +impl Div for Value { + type Output = Result; + + fn div(self, rhs: Self) -> Self::Output { + Ok((self.into_type::()? / rhs.into_type::()?).into()) + } +} + +pub trait TypeOf { + fn type_of() -> Type; +} + +impl TypeOf for i64 { + fn type_of() -> Type { + Type::Int + } +} + +impl TypeOf for bool { + fn type_of() -> Type { + Type::Bool + } +} + +impl TypeOf for f64 { + fn type_of() -> Type { + Type::Float + } +} diff --git a/src/main.rs b/src/main.rs new file mode 100644 index 000000000000..c31fda32dbc4 --- /dev/null +++ b/src/main.rs @@ -0,0 +1,31 @@ +use clap::Clap; + +pub mod ast; +pub mod codegen; +pub(crate) mod commands; +pub(crate) mod common; +pub mod compiler; +pub mod interpreter; +pub mod parser; + +pub use common::{Error, Result}; + +#[derive(Clap)] +struct Opts { + #[clap(subcommand)] + subcommand: Command, +} + +#[derive(Clap)] +enum Command { + Eval(commands::Eval), + Compile(commands::Compile), +} + +fn main() -> anyhow::Result<()> { + let opts = Opts::parse(); + match opts.subcommand { + Command::Eval(eval) => Ok(eval.run()?), + Command::Compile(compile) => Ok(compile.run()?), + } +} diff --git a/src/parser/mod.rs b/src/parser/mod.rs new file mode 100644 index 000000000000..811450da0d61 --- /dev/null +++ b/src/parser/mod.rs @@ -0,0 +1,445 @@ +use nom::character::complete::{digit1, multispace0, multispace1}; +use nom::error::{ErrorKind, ParseError}; +use nom::{ + alt, char, complete, delimited, do_parse, flat_map, many0, map, named, parse_to, + separated_list0, separated_list1, tag, tuple, +}; +use pratt::{Affix, Associativity, PrattParser, Precedence}; + +use crate::ast::{BinaryOperator, Decl, Expr, Fun, Ident, Literal, UnaryOperator}; + +pub type Error = nom::Err>; + +#[derive(Debug)] +enum TokenTree<'a> { + Prefix(UnaryOperator), + // Postfix(char), + Infix(BinaryOperator), + Primary(Expr<'a>), + Group(Vec>), +} + +named!(prefix(&str) -> TokenTree, map!(alt!( + complete!(char!('-')) => { |_| UnaryOperator::Neg } | + complete!(char!('!')) => { |_| UnaryOperator::Not } +), TokenTree::Prefix)); + +named!(infix(&str) -> TokenTree, map!(alt!( + complete!(tag!("==")) => { |_| BinaryOperator::Equ } | + complete!(tag!("!=")) => { |_| BinaryOperator::Neq } | + complete!(char!('+')) => { |_| BinaryOperator::Add } | + complete!(char!('-')) => { |_| BinaryOperator::Sub } | + complete!(char!('*')) => { |_| BinaryOperator::Mul } | + complete!(char!('/')) => { |_| BinaryOperator::Div } | + complete!(char!('^')) => { |_| BinaryOperator::Pow } +), TokenTree::Infix)); + +named!(primary(&str) -> TokenTree, alt!( + do_parse!( + multispace0 >> + char!('(') >> + multispace0 >> + group: group >> + multispace0 >> + char!(')') >> + multispace0 >> + (TokenTree::Group(group)) + ) | + delimited!(multispace0, simple_expr, multispace0) => { |s| TokenTree::Primary(s) } +)); + +named!( + rest(&str) -> Vec<(TokenTree, Vec, TokenTree)>, + many0!(tuple!( + infix, + delimited!(multispace0, many0!(prefix), multispace0), + primary + // many0!(postfix) + )) +); + +named!(group(&str) -> Vec, do_parse!( + prefix: many0!(prefix) + >> primary: primary + // >> postfix: many0!(postfix) + >> rest: rest + >> ({ + let mut res = prefix; + res.push(primary); + // res.append(&mut postfix); + for (infix, mut prefix, primary/*, mut postfix*/) in rest { + res.push(infix); + res.append(&mut prefix); + res.push(primary); + // res.append(&mut postfix); + } + res + }) +)); + +fn token_tree(i: &str) -> nom::IResult<&str, Vec> { + group(i) +} + +struct ExprParser; + +impl<'a, I> PrattParser for ExprParser +where + I: Iterator>, +{ + type Error = pratt::NoError; + type Input = TokenTree<'a>; + type Output = Expr<'a>; + + fn query(&mut self, input: &Self::Input) -> Result { + use BinaryOperator::*; + use UnaryOperator::*; + + Ok(match input { + TokenTree::Infix(Add) => Affix::Infix(Precedence(6), Associativity::Left), + TokenTree::Infix(Sub) => Affix::Infix(Precedence(6), Associativity::Left), + TokenTree::Infix(Mul) => Affix::Infix(Precedence(7), Associativity::Left), + TokenTree::Infix(Div) => Affix::Infix(Precedence(7), Associativity::Left), + TokenTree::Infix(Pow) => Affix::Infix(Precedence(8), Associativity::Right), + TokenTree::Infix(Equ) => Affix::Infix(Precedence(4), Associativity::Right), + TokenTree::Infix(Neq) => Affix::Infix(Precedence(4), Associativity::Right), + TokenTree::Prefix(Neg) => Affix::Prefix(Precedence(6)), + TokenTree::Prefix(Not) => Affix::Prefix(Precedence(6)), + TokenTree::Primary(_) => Affix::Nilfix, + TokenTree::Group(_) => Affix::Nilfix, + }) + } + + fn primary(&mut self, input: Self::Input) -> Result { + Ok(match input { + TokenTree::Primary(expr) => expr, + TokenTree::Group(group) => self.parse(&mut group.into_iter()).unwrap(), + _ => unreachable!(), + }) + } + + fn infix( + &mut self, + lhs: Self::Output, + op: Self::Input, + rhs: Self::Output, + ) -> Result { + let op = match op { + TokenTree::Infix(op) => op, + _ => unreachable!(), + }; + Ok(Expr::BinaryOp { + lhs: Box::new(lhs), + op, + rhs: Box::new(rhs), + }) + } + + fn prefix(&mut self, op: Self::Input, rhs: Self::Output) -> Result { + let op = match op { + TokenTree::Prefix(op) => op, + _ => unreachable!(), + }; + + Ok(Expr::UnaryOp { + op, + rhs: Box::new(rhs), + }) + } + + fn postfix( + &mut self, + _lhs: Self::Output, + _op: Self::Input, + ) -> Result { + unreachable!() + } +} + +fn ident<'a, E>(i: &'a str) -> nom::IResult<&'a str, Ident, E> +where + E: ParseError<&'a str>, +{ + let mut chars = i.chars(); + if let Some(f) = chars.next() { + if f.is_alphabetic() || f == '_' { + let mut idx = 1; + for c in chars { + if !(c.is_alphanumeric() || c == '_') { + break; + } + idx += 1; + } + Ok((&i[idx..], Ident::from_str_unchecked(&i[..idx]))) + } else { + Err(nom::Err::Error(E::from_error_kind(i, ErrorKind::Satisfy))) + } + } else { + Err(nom::Err::Error(E::from_error_kind(i, ErrorKind::Eof))) + } +} + +named!(int(&str) -> Literal, map!(flat_map!(digit1, parse_to!(u64)), Literal::Int)); + +named!(literal(&str) -> Expr, map!(alt!(int), Expr::Literal)); + +named!(binding(&str) -> (Ident, Expr), do_parse!( + multispace0 + >> ident: ident + >> multispace0 + >> char!('=') + >> multispace0 + >> expr: expr + >> (ident, expr) +)); + +named!(let_(&str) -> Expr, do_parse!( + tag!("let") + >> multispace0 + >> bindings: separated_list1!(alt!(char!(';') | char!('\n')), binding) + >> multispace0 + >> tag!("in") + >> multispace0 + >> body: expr + >> (Expr::Let { + bindings, + body: Box::new(body) + }) +)); + +named!(if_(&str) -> Expr, do_parse! ( + tag!("if") + >> multispace0 + >> condition: expr + >> multispace0 + >> tag!("then") + >> multispace0 + >> then: expr + >> multispace0 + >> tag!("else") + >> multispace0 + >> else_: expr + >> (Expr::If { + condition: Box::new(condition), + then: Box::new(then), + else_: Box::new(else_) + }) +)); + +named!(ident_expr(&str) -> Expr, map!(ident, Expr::Ident)); + +named!(simple_expr(&str) -> Expr, alt!( + let_ | + if_ | + literal | + ident_expr +)); + +named!(pub expr(&str) -> Expr, alt!( + map!(token_tree, |tt| { + ExprParser.parse(&mut tt.into_iter()).unwrap() + }) | + simple_expr)); + +////// + +named!(fun(&str) -> Fun, do_parse!( + tag!("fn") + >> multispace0 + >> name: ident + >> multispace1 + >> args: separated_list0!(multispace1, ident) + >> multispace0 + >> char!('=') + >> multispace0 + >> body: expr + >> (Fun { + name, + args, + body + }) +)); + +named!(pub decl(&str) -> Decl, alt!( + fun => { |f| Decl::Fun(f) } +)); + +named!(pub toplevel(&str) -> Vec, separated_list0!(multispace1, decl)); + +#[cfg(test)] +mod tests { + use std::convert::{TryFrom, TryInto}; + + use super::*; + use BinaryOperator::*; + use Expr::{BinaryOp, If, Let, UnaryOp}; + use UnaryOperator::*; + + fn ident_expr(s: &str) -> Box { + Box::new(Expr::Ident(Ident::try_from(s).unwrap())) + } + + macro_rules! test_parse { + ($parser: ident, $src: expr) => {{ + let (rem, res) = $parser($src).unwrap(); + assert!( + rem.is_empty(), + "non-empty remainder: \"{}\", parsed: {:?}", + rem, + res + ); + res + }}; + } + + mod operators { + use super::*; + + #[test] + fn mul_plus() { + let (rem, res) = expr("x*y+z").unwrap(); + assert!(rem.is_empty()); + assert_eq!( + res, + BinaryOp { + lhs: Box::new(BinaryOp { + lhs: ident_expr("x"), + op: Mul, + rhs: ident_expr("y") + }), + op: Add, + rhs: ident_expr("z") + } + ) + } + + #[test] + fn mul_plus_ws() { + let (rem, res) = expr("x * y + z").unwrap(); + assert!(rem.is_empty(), "non-empty remainder: \"{}\"", rem); + assert_eq!( + res, + BinaryOp { + lhs: Box::new(BinaryOp { + lhs: ident_expr("x"), + op: Mul, + rhs: ident_expr("y") + }), + op: Add, + rhs: ident_expr("z") + } + ) + } + + #[test] + fn unary() { + let (rem, res) = expr("x * -z").unwrap(); + assert!(rem.is_empty(), "non-empty remainder: \"{}\"", rem); + assert_eq!( + res, + BinaryOp { + lhs: ident_expr("x"), + op: Mul, + rhs: Box::new(UnaryOp { + op: Neg, + rhs: ident_expr("z"), + }) + } + ) + } + + #[test] + fn mul_literal() { + let (rem, res) = expr("x * 3").unwrap(); + assert!(rem.is_empty()); + assert_eq!( + res, + BinaryOp { + lhs: ident_expr("x"), + op: Mul, + rhs: Box::new(Expr::Literal(Literal::Int(3))), + } + ) + } + + #[test] + fn equ() { + let res = test_parse!(expr, "x * 7 == 7"); + assert_eq!( + res, + BinaryOp { + lhs: Box::new(BinaryOp { + lhs: ident_expr("x"), + op: Mul, + rhs: Box::new(Expr::Literal(Literal::Int(7))) + }), + op: Equ, + rhs: Box::new(Expr::Literal(Literal::Int(7))) + } + ) + } + } + + #[test] + fn let_complex() { + let res = test_parse!(expr, "let x = 1; y = x * 7 in (x + y) * 4"); + assert_eq!( + res, + Let { + bindings: vec![ + ( + Ident::try_from("x").unwrap(), + Expr::Literal(Literal::Int(1)) + ), + ( + Ident::try_from("y").unwrap(), + Expr::BinaryOp { + lhs: ident_expr("x"), + op: Mul, + rhs: Box::new(Expr::Literal(Literal::Int(7))) + } + ) + ], + body: Box::new(Expr::BinaryOp { + lhs: Box::new(Expr::BinaryOp { + lhs: ident_expr("x"), + op: Add, + rhs: ident_expr("y"), + }), + op: Mul, + rhs: Box::new(Expr::Literal(Literal::Int(4))), + }) + } + ) + } + + #[test] + fn if_simple() { + let res = test_parse!(expr, "if x == 8 then 9 else 20"); + assert_eq!( + res, + If { + condition: Box::new(BinaryOp { + lhs: ident_expr("x"), + op: Equ, + rhs: Box::new(Expr::Literal(Literal::Int(8))), + }), + then: Box::new(Expr::Literal(Literal::Int(9))), + else_: Box::new(Expr::Literal(Literal::Int(20))) + } + ) + } + + #[test] + fn fn_decl() { + let res = test_parse!(decl, "fn id x = x"); + assert_eq!( + res, + Decl::Fun(Fun { + name: "id".try_into().unwrap(), + args: vec!["x".try_into().unwrap()], + body: *ident_expr("x"), + }) + ) + } +} -- cgit 1.4.1