Bazel: remove non-working fake `tree-sitter-extractor` workaround

The `.cargo/config.toml` override based workaround wasn't really
working, as while `cargo build|check` was reading that, `cargo metadata`
wasn't, ending up in a completely broken IDE experience.

For the moment, we just use a unified workspace `Cargo.toml` for all
extractors using the shared tree-sitter code, which has the downside of
making bazel pull in dependencies for all of them, and not being able to
do sparse checkouts for them. We should investigate and rivist this in
the future.

.gitignore

@@ -68,3 +68,6 @@ node_modules/
 
 # Jetbrains IDE files
 .idea
+
+# cargo build directory
+/target

Cargo.lock

@@ -92,9 +92,9 @@ dependencies = [
 
 [[package]]
 name = "anyhow"
-version = "1.0.86"
+version = "1.0.87"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "b3d1d046238990b9cf5bcde22a3fb3584ee5cf65fb2765f454ed428c7a0063da"
+checksum = "10f00e1f6e58a40e807377c75c6a7f97bf9044fab57816f2414e6f5f4499d7b8"
 
 [[package]]
 name = "arrayvec"
@@ -153,9 +153,9 @@ checksum = "79296716171880943b8470b5f8d03aa55eb2e645a4874bdbb28adb49162e012c"
 
 [[package]]
 name = "bytemuck"
-version = "1.17.1"
+version = "1.18.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "773d90827bc3feecfb67fab12e24de0749aad83c74b9504ecde46237b5cd24e2"
+checksum = "94bbb0ad554ad961ddc5da507a12a29b14e4ae5bda06b19f575a3e6079d2e2ae"
 
 [[package]]
 name = "byteorder"
@@ -197,9 +197,9 @@ dependencies = [
 
 [[package]]
 name = "cc"
-version = "1.1.16"
+version = "1.1.18"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "e9d013ecb737093c0e86b151a7b837993cf9ec6c502946cfb44bedc392421e0b"
+checksum = "b62ac837cdb5cb22e10a256099b4fc502b1dfe560cb282963a974d7abd80e476"
 dependencies = [
  "shlex",
 ]
@@ -338,6 +338,23 @@ dependencies = [
  "tree-sitter-ql",
 ]
 
+[[package]]
+name = "codeql-extractor-ruby"
+version = "0.1.0"
+dependencies = [
+ "clap",
+ "codeql-extractor",
+ "encoding",
+ "lazy_static",
+ "rayon",
+ "regex",
+ "tracing",
+ "tracing-subscriber",
+ "tree-sitter",
+ "tree-sitter-embedded-template",
+ "tree-sitter-ruby",
+]
+
 [[package]]
 name = "codeql-rust"
 version = "0.1.0"
@@ -654,9 +671,9 @@ dependencies = [
 
 [[package]]
 name = "globset"
-version = "0.4.14"
+version = "0.4.15"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "57da3b9b5b85bd66f31093f8c408b90a74431672542466497dcbdfdc02034be1"
+checksum = "15f1ce686646e7f1e19bf7d5533fe443a45dbfb990e00629110797578b42fb19"
 dependencies = [
  "aho-corasick",
  "bstr",
@@ -1911,18 +1928,18 @@ dependencies = [
 
 [[package]]
 name = "serde"
-version = "1.0.209"
+version = "1.0.210"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "99fce0ffe7310761ca6bf9faf5115afbc19688edd00171d81b1bb1b116c63e09"
+checksum = "c8e3592472072e6e22e0a54d5904d9febf8508f65fb8552499a1abc7d1078c3a"
 dependencies = [
  "serde_derive",
 ]
 
 [[package]]
 name = "serde_derive"
-version = "1.0.209"
+version = "1.0.210"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "a5831b979fd7b5439637af1752d535ff49f4860c0f341d1baeb6faf0f4242170"
+checksum = "243902eda00fad750862fc144cea25caca5e20d615af0a81bee94ca738f1df1f"
 dependencies = [
  "proc-macro2",
  "quote",
@@ -2203,6 +2220,15 @@ dependencies = [
  "tree-sitter-language",
 ]
 
+[[package]]
+name = "tree-sitter-embedded-template"
+version = "0.23.0"
+source = "git+https://github.com/tree-sitter/tree-sitter-embedded-template.git?rev=62b0a6e45900a7dff7c37da95fec20a09968ba52#62b0a6e45900a7dff7c37da95fec20a09968ba52"
+dependencies = [
+ "cc",
+ "tree-sitter-language",
+]
+
 [[package]]
 name = "tree-sitter-json"
 version = "0.23.0"
@@ -2227,6 +2253,15 @@ dependencies = [
  "tree-sitter-language",
 ]
 
+[[package]]
+name = "tree-sitter-ruby"
+version = "0.23.0"
+source = "git+https://github.com/tree-sitter/tree-sitter-ruby.git?rev=a66579f70d6f50ffd81a16fc3d3358e2ac173c88#a66579f70d6f50ffd81a16fc3d3358e2ac173c88"
+dependencies = [
+ "cc",
+ "tree-sitter-language",
+]
+
 [[package]]
 name = "triomphe"
 version = "0.1.13"
@@ -2254,9 +2289,9 @@ dependencies = [
 
 [[package]]
 name = "unicode-ident"
-version = "1.0.12"
+version = "1.0.13"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "3354b9ac3fae1ff6755cb6db53683adb661634f67557942dea4facebec0fee4b"
+checksum = "e91b56cd4cadaeb79bbf1a5645f6b4f8dc5bde8834ad5894a8db35fda9efa1fe"
 
 [[package]]
 name = "unicode-properties"

Cargo.toml

@@ -0,0 +1,14 @@
+# This is the shared workspace file for extractor using shared/tree-sitter/extractor
+[workspace]
+
+resolver = "2"
+members = [
+    "shared/tree-sitter-extractor",
+    "ruby/extractor",
+    "rust/extractor",
+]
+
+[patch.crates-io]
+# patch for build script bug preventing bazel build
+# see https://github.com/rust-lang/rustc_apfloat/pull/17
+rustc_apfloat = { git = "https://github.com/redsun82/rustc_apfloat.git", rev = "096d585100636bc2e9f09d7eefec38c5b334d47b" }

MODULE.bazel

@@ -47,35 +47,23 @@ cp.from_cargo(
 )
 use_repo(cp, "py_deps")
 
-# crate_ruby, but shortened due to windows file paths
+# deps for ruby+rust, but shortened due to windows file paths
 r = use_extension(
     "@rules_rust//crate_universe:extension.bzl",
     "crate",
     isolate = True,
 )
 r.from_cargo(
-    name = "rd",
-    cargo_lockfile = "//ruby/extractor:Cargo.lock",
+    name = "r",
+    cargo_lockfile = "//:Cargo.lock",
     manifests = [
+        "//:Cargo.toml",
         "//ruby/extractor:Cargo.toml",
-        "//ruby/extractor/codeql-extractor-fake-crate:Cargo.toml",
-    ],
-)
-use_repo(r, ruby_deps = "rd")
-
-rsp = use_extension(
-    "@rules_rust//crate_universe:extension.bzl",
-    "crate",
-    isolate = True,
-)
-rsp.from_cargo(
-    name = "rs_deps",
-    cargo_lockfile = "//rust/extractor:Cargo.lock",
-    manifests = [
         "//rust/extractor:Cargo.toml",
+        "//shared/tree-sitter-extractor:Cargo.toml",
     ],
 )
-use_repo(rsp, rust_deps = "rs_deps")
+use_repo(r, tree_sitter_extractors_deps = "r")
 
 dotnet = use_extension("@rules_dotnet//dotnet:extensions.bzl", "dotnet")
 dotnet.toolchain(dotnet_version = "8.0.101")

config/identical-files.json

@@ -355,10 +355,5 @@
   "Python model summaries test extension": [
     "python/ql/test/library-tests/dataflow/model-summaries/InlineTaintTest.ext.yml",
     "python/ql/test/library-tests/dataflow/model-summaries/NormalDataflowTest.ext.yml"
-  ],
-  "shared tree-sitter extractor cargo.toml": [
-    "shared/tree-sitter-extractor/Cargo.toml",
-    "ruby/extractor/codeql-extractor-fake-crate/Cargo.toml",
-    "rust/extractor/codeql-extractor-fake-crate/Cargo.toml"
   ]
 }

ruby/extractor/.cargo/config.toml

@@ -1 +0,0 @@
-paths = ["../../shared/tree-sitter-extractor"]

ruby/extractor/BUILD.bazel

@@ -1,4 +1,4 @@
-load("@ruby_deps//:defs.bzl", "aliases", "all_crate_deps")
+load("@tree_sitter_extractors_deps//:defs.bzl", "aliases", "all_crate_deps")
 load("//misc/bazel:rust.bzl", "codeql_rust_binary")
 
 codeql_rust_binary(

ruby/extractor/Cargo.toml

@@ -1,4 +1,3 @@
-[workspace]
 [package]
 name = "codeql-extractor-ruby"
 description = "CodeQL Ruby extractor"
@@ -16,8 +15,6 @@ tracing-subscriber = { version = "0.3.3", features = ["env-filter"] }
 rayon = "1.5.0"
 regex = "1.7.1"
 encoding = "0.2"
-lazy_static = "1.4.0" 
-# Ideally, we'd like to pull this in via a relative path.
-# However, our bazel/rust tooling chokes on this, c.f. https://github.com/bazelbuild/rules_rust/issues/1525
-# Therefore, we have a pretty bad hack in place instead, see README.md in the codeql-extractor-fake-crate directory.
-codeql-extractor = { path = "codeql-extractor-fake-crate" }
+lazy_static = "1.4.0"
+
+codeql-extractor = { path = "../../shared/tree-sitter-extractor" }

ruby/extractor/codeql-extractor-fake-crate/Cargo.toml

@@ -1,26 +0,0 @@
-[package]
-name = "codeql-extractor"
-version = "0.2.0"
-edition = "2021"
-authors = ["GitHub"]
-
-[dependencies]
-flate2 = "1.0"
-globset = "0.4"
-tree-sitter = ">= 0.23.0"
-tracing = "0.1"
-tracing-subscriber = { version = "0.3.3", features = ["env-filter"] }
-rayon = "1.5.0"
-regex = "1.7.1"
-encoding = "0.2"
-lazy_static = "1.4.0"
-serde = { version = "1.0", features = ["derive"] }
-serde_json = "1.0"
-chrono = { version = "0.4.19", features = ["serde"] }
-num_cpus = "1.14.0"
-
-[dev-dependencies]
-tree-sitter-ql = { git = "https://github.com/tree-sitter/tree-sitter-ql" }
-tree-sitter-json = {git = "https://github.com/tree-sitter/tree-sitter-json" }
-rand = "0.8.5"
-

ruby/extractor/codeql-extractor-fake-crate/README.md

@@ -1,7 +0,0 @@
-We're presenting a fake crate in this workspace that ensures that the correct crate dependencies from the shared tree sitter
-extractor can be parsed by Bazel (which doesn't resolve path dependencies outside of the cargo workspace unfortunately).
-
-The sync-identical-files script keeps this up-to-date.
-For local development and IDEs, we override the path to `codeql-extractor` using the `.cargo/config.toml` mechanism.
-Bazel doesn't actually do anything with path dependencies except to pull in their dependency tree, so we manually
-specify the dependency from the ruby extractor to the shared extractor in `BUILD.bazel`.

ruby/extractor/codeql-extractor-fake-crate/src/lib.rs

@@ -1 +0,0 @@
-

rust/extractor/.cargo/config.toml

@@ -1 +0,0 @@
-paths = ["../../shared/tree-sitter-extractor"]

rust/extractor/BUILD.bazel

@@ -1,4 +1,4 @@
-load("@rust_deps//:defs.bzl", "aliases", "all_crate_deps")
+load("@tree_sitter_extractors_deps//:defs.bzl", "aliases", "all_crate_deps")
 load("//misc/bazel:rust.bzl", "codeql_rust_binary")
 
 codeql_rust_binary(

rust/extractor/Cargo.toml

@@ -1,5 +1,3 @@
-[workspace]
-
 [package]
 name = "codeql-rust"
 version = "0.1.0"
@@ -27,9 +25,4 @@ triomphe = "0.1.13"
 # Ideally, we'd like to pull this in via a relative path.
 # However, our bazel/rust tooling chokes on this, c.f. https://github.com/bazelbuild/rules_rust/issues/1525
 # Therefore, we have a pretty bad hack in place instead, see README.md in the codeql-extractor-fake-crate directory.
-codeql-extractor = { path = "codeql-extractor-fake-crate" }
-
-[patch.crates-io]
-# patch for build script bug preventing bazel build
-# see https://github.com/rust-lang/rustc_apfloat/pull/17
-rustc_apfloat = { git = "https://github.com/redsun82/rustc_apfloat.git", rev = "096d585100636bc2e9f09d7eefec38c5b334d47b" }
+codeql-extractor = { path = "../../shared/tree-sitter-extractor" }

rust/extractor/codeql-extractor-fake-crate/Cargo.toml

@@ -1,26 +0,0 @@
-[package]
-name = "codeql-extractor"
-version = "0.2.0"
-edition = "2021"
-authors = ["GitHub"]
-
-[dependencies]
-flate2 = "1.0"
-globset = "0.4"
-tree-sitter = ">= 0.23.0"
-tracing = "0.1"
-tracing-subscriber = { version = "0.3.3", features = ["env-filter"] }
-rayon = "1.5.0"
-regex = "1.7.1"
-encoding = "0.2"
-lazy_static = "1.4.0"
-serde = { version = "1.0", features = ["derive"] }
-serde_json = "1.0"
-chrono = { version = "0.4.19", features = ["serde"] }
-num_cpus = "1.14.0"
-
-[dev-dependencies]
-tree-sitter-ql = { git = "https://github.com/tree-sitter/tree-sitter-ql" }
-tree-sitter-json = {git = "https://github.com/tree-sitter/tree-sitter-json" }
-rand = "0.8.5"
-

rust/extractor/codeql-extractor-fake-crate/README.md

@@ -1,7 +0,0 @@
-We're presenting a fake crate in this workspace that ensures that the correct crate dependencies from the shared tree sitter
-extractor can be parsed by Bazel (which doesn't resolve path dependencies outside of the cargo workspace unfortunately).
-
-The sync-identical-files script keeps this up-to-date.
-For local development and IDEs, we override the path to `codeql-extractor` using the `.cargo/config.toml` mechanism.
-Bazel doesn't actually do anything with path dependencies except to pull in their dependency tree, so we manually
-specify the dependency from the ruby extractor to the shared extractor in `BUILD.bazel`.

rust/extractor/codeql-extractor-fake-crate/src/lib.rs

@@ -1 +0,0 @@
-

shared/tree-sitter-extractor/BUILD.bazel

@@ -1,5 +1,5 @@
-load("@ruby_deps//:defs.bzl", "aliases", "all_crate_deps")
 load("@rules_rust//rust:defs.bzl", "rust_library")
+load("@tree_sitter_extractors_deps//:defs.bzl", "aliases", "all_crate_deps")
 
 package(default_visibility = ["//visibility:public"])
 
@@ -12,5 +12,5 @@ rust_library(
     compile_data = [
         "src/generator/prefix.dbscheme",
     ],
-    deps = all_crate_deps(package_name = "ruby/extractor/codeql-extractor-fake-crate"),
+    deps = all_crate_deps(),
 )