Revert "Revert "Rust: allow to specify more cargo configuration options""

MODULE.bazel

@@ -68,7 +68,7 @@ use_repo(py_deps, "vendor__anyhow-1.0.44", "vendor__cc-1.0.70", "vendor__clap-2.
 # deps for ruby+rust
 # keep in sync by running `misc/bazel/3rdparty/update_cargo_deps.sh`
 tree_sitter_extractors_deps = use_extension("//misc/bazel/3rdparty:tree_sitter_extractors_extension.bzl", "r")
-use_repo(tree_sitter_extractors_deps, "vendor__anyhow-1.0.93", "vendor__argfile-0.2.1", "vendor__chrono-0.4.38", "vendor__clap-4.5.20", "vendor__encoding-0.2.33", "vendor__figment-0.10.19", "vendor__flate2-1.0.34", "vendor__glob-0.3.1", "vendor__globset-0.4.15", "vendor__itertools-0.10.5", "vendor__itertools-0.13.0", "vendor__lazy_static-1.5.0", "vendor__log-0.4.22", "vendor__num-traits-0.2.19", "vendor__num_cpus-1.16.0", "vendor__proc-macro2-1.0.89", "vendor__quote-1.0.37", "vendor__ra_ap_base_db-0.0.232", "vendor__ra_ap_hir-0.0.232", "vendor__ra_ap_hir_def-0.0.232", "vendor__ra_ap_hir_expand-0.0.232", "vendor__ra_ap_ide_db-0.0.232", "vendor__ra_ap_load-cargo-0.0.232", "vendor__ra_ap_parser-0.0.232", "vendor__ra_ap_paths-0.0.232", "vendor__ra_ap_project_model-0.0.232", "vendor__ra_ap_span-0.0.232", "vendor__ra_ap_syntax-0.0.232", "vendor__ra_ap_vfs-0.0.232", "vendor__rand-0.8.5", "vendor__rayon-1.10.0", "vendor__regex-1.11.1", "vendor__serde-1.0.214", "vendor__serde_json-1.0.132", "vendor__serde_with-3.11.0", "vendor__stderrlog-0.6.0", "vendor__syn-2.0.87", "vendor__tracing-0.1.40", "vendor__tracing-subscriber-0.3.18", "vendor__tree-sitter-0.24.4", "vendor__tree-sitter-embedded-template-0.23.2", "vendor__tree-sitter-json-0.24.8", "vendor__tree-sitter-ql-0.23.1", "vendor__tree-sitter-ruby-0.23.1", "vendor__triomphe-0.1.14", "vendor__ungrammar-1.16.1")
+use_repo(tree_sitter_extractors_deps, "vendor__anyhow-1.0.93", "vendor__argfile-0.2.1", "vendor__chrono-0.4.38", "vendor__clap-4.5.20", "vendor__encoding-0.2.33", "vendor__figment-0.10.19", "vendor__flate2-1.0.34", "vendor__glob-0.3.1", "vendor__globset-0.4.15", "vendor__itertools-0.10.5", "vendor__itertools-0.13.0", "vendor__lazy_static-1.5.0", "vendor__log-0.4.22", "vendor__num-traits-0.2.19", "vendor__num_cpus-1.16.0", "vendor__proc-macro2-1.0.89", "vendor__quote-1.0.37", "vendor__ra_ap_base_db-0.0.232", "vendor__ra_ap_cfg-0.0.232", "vendor__ra_ap_hir-0.0.232", "vendor__ra_ap_hir_def-0.0.232", "vendor__ra_ap_hir_expand-0.0.232", "vendor__ra_ap_ide_db-0.0.232", "vendor__ra_ap_intern-0.0.232", "vendor__ra_ap_load-cargo-0.0.232", "vendor__ra_ap_parser-0.0.232", "vendor__ra_ap_paths-0.0.232", "vendor__ra_ap_project_model-0.0.232", "vendor__ra_ap_span-0.0.232", "vendor__ra_ap_syntax-0.0.232", "vendor__ra_ap_vfs-0.0.232", "vendor__rand-0.8.5", "vendor__rayon-1.10.0", "vendor__regex-1.11.1", "vendor__serde-1.0.214", "vendor__serde_json-1.0.132", "vendor__serde_with-3.11.0", "vendor__stderrlog-0.6.0", "vendor__syn-2.0.87", "vendor__tracing-0.1.40", "vendor__tracing-subscriber-0.3.18", "vendor__tree-sitter-0.24.4", "vendor__tree-sitter-embedded-template-0.23.2", "vendor__tree-sitter-json-0.24.8", "vendor__tree-sitter-ql-0.23.1", "vendor__tree-sitter-ruby-0.23.1", "vendor__triomphe-0.1.14", "vendor__ungrammar-1.16.1")
 
 dotnet = use_extension("@rules_dotnet//dotnet:extensions.bzl", "dotnet")
 dotnet.toolchain(dotnet_version = "8.0.101")

rust/ast-generator/Cargo.toml

@@ -3,7 +3,7 @@ name = "ast-generator"
 version = "0.1.0"
 edition = "2021"
 
-# When updating these dependencies, run `misc/bazel/3rdparty/update_cargo_deps.sh`
+# When updating these dependencies, run `rust/update_cargo_deps.sh`
 [dependencies]
 ungrammar = "1.16.1"
 proc-macro2 = "1.0.33"

rust/codeql-extractor.yml

@@ -35,3 +35,22 @@ options:
       reduce execution time of consecutive extractor runs. By default, a new scratch
       directory is used for each run.
     type: string
+  cargo_target:
+    title: Target architecture
+    description: >
+      Target architecture to use for analysis, analogous to `cargo --target`. By
+      default the host architecture is used.
+    type: string
+  cargo_features:
+    title: Cargo features to turn on
+    description: >
+      Comma-separated list of features to turn on. If any value is `*` all features
+      are turned on. By default only default cargo features are enabled. Can be
+      repeated.
+    type: array
+  cargo_cfg_overrides:
+    title: Cargo cfg overrides
+    description: >
+      Comma-separated list of cfg settings to enable, or disable if prefixed with `-`.
+      Can be repeated.
+    type: array

rust/extractor/Cargo.toml

@@ -3,7 +3,7 @@ name = "codeql-rust"
 version = "0.1.0"
 edition = "2021"
 
-# When updating these dependencies, run `misc/bazel/3rdparty/update_cargo_deps.sh`
+# When updating these dependencies, run `rust/update_cargo_deps.sh`
 [dependencies]
 anyhow = "1.0.86"
 clap = { version = "4.5.16", features = ["derive"] }
@@ -22,6 +22,8 @@ ra_ap_syntax = "0.0.232"
 ra_ap_vfs = "0.0.232"
 ra_ap_parser = "0.0.232"
 ra_ap_span = "0.0.232"
+ra_ap_cfg = "0.0.232"
+ra_ap_intern = "0.0.232"
 serde = "1.0.209"
 serde_with = "3.9.0"
 stderrlog = "0.6.0"

rust/extractor/macros/Cargo.toml

@@ -6,7 +6,7 @@ edition = "2021"
 [lib]
 proc-macro = true
 
-# When updating these dependencies, run `misc/bazel/3rdparty/update_cargo_deps.sh`
+# When updating these dependencies, run `rust/update_cargo_deps.sh`
 [dependencies]
 quote = "1.0.37"
 syn = { version = "2.0.77", features = ["full"] }

rust/extractor/macros/src/lib.rs

@@ -1,47 +1,74 @@
 use proc_macro::TokenStream;
 use quote::{format_ident, quote};
+use syn::{Ident, Type};
+
+fn get_type_tip(t: &Type) -> Option<&Ident> {
+    let syn::Type::Path(path) = t else {
+        return None;
+    };
+    let segment = path.path.segments.last()?;
+    Some(&segment.ident)
+}
 
 /// Allow all fields in the extractor config to be also overrideable by extractor CLI flags
 #[proc_macro_attribute]
 pub fn extractor_cli_config(_attr: TokenStream, item: TokenStream) -> TokenStream {
     let ast = syn::parse_macro_input!(item as syn::ItemStruct);
     let name = &ast.ident;
+    let fields = ast
+        .fields
+        .iter()
+        .map(|f| {
+            if f.ident.as_ref().is_some_and(|i| i != "inputs")
+                && get_type_tip(&f.ty).is_some_and(|i| i == "Vec")
+            {
+                quote! {
+                    #[serde(deserialize_with="deserialize_newline_or_comma_separated")]
+                    #f
+                }
+            } else {
+                quote! { #f }
+            }
+        })
+        .collect::<Vec<_>>();
     let cli_name = format_ident!("Cli{}", name);
     let cli_fields = ast
         .fields
         .iter()
         .map(|f| {
             let id = f.ident.as_ref().unwrap();
             let ty = &f.ty;
-            if let syn::Type::Path(p) = ty {
-                if p.path.is_ident(&format_ident!("bool")) {
-                    return quote! {
-                        #[arg(long)]
-                        #[serde(skip_serializing_if="<&bool>::not")]
-                        #id: bool,
-                    };
+            let type_tip = get_type_tip(ty);
+            if type_tip.is_some_and(|i| i == "bool") {
+                quote! {
+                    #[arg(long)]
+                    #[serde(skip_serializing_if="<&bool>::not")]
+                    #id: bool
                 }
-                if p.path.segments.len() == 1 && p.path.segments[0].ident == "Option" {
-                    return quote! {
-                        #[arg(long)]
-                        #id: #ty,
-                    };
+            } else if type_tip.is_some_and(|i| i == "Option") {
+                quote! {
+                    #[arg(long)]
+                    #f
                 }
-            }
-            if id == &format_ident!("verbose") {
+            } else if id == &format_ident!("verbose") {
                 quote! {
                     #[arg(long, short, action=clap::ArgAction::Count)]
                     #[serde(skip_serializing_if="u8::is_zero")]
-                    #id: u8,
+                    #id: u8
                 }
             } else if id == &format_ident!("inputs") {
                 quote! {
-                    #id: #ty,
+                    #f
+                }
+            } else if type_tip.is_some_and(|i| i == "Vec") {
+                quote! {
+                    #[arg(long)]
+                    #id: Option<String>
                 }
             } else {
                 quote! {
                     #[arg(long)]
-                    #id: Option<#ty>,
+                    #id: Option<#ty>
                 }
             }
         })
@@ -66,7 +93,9 @@ pub fn extractor_cli_config(_attr: TokenStream, item: TokenStream) -> TokenStrea
     let gen = quote! {
         #[serde_with::apply(_ => #[serde(default)])]
         #[derive(Deserialize, Default)]
-        #ast
+        pub struct #name {
+            #(#fields),*
+        }
 
         impl Debug for #name {
             fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
@@ -80,7 +109,7 @@ pub fn extractor_cli_config(_attr: TokenStream, item: TokenStream) -> TokenStrea
         #[derive(clap::Parser, Serialize)]
         #[command(about, long_about = None)]
         struct #cli_name {
-            #(#cli_fields)*
+            #(#cli_fields),*
         }
     };
     gen.into()

rust/extractor/src/config.rs

@@ -1,13 +1,21 @@
+mod deserialize_vec;
+
 use anyhow::Context;
 use clap::Parser;
 use codeql_extractor::trap;
+use deserialize_vec::deserialize_newline_or_comma_separated;
 use figment::{
     providers::{Env, Format, Serialized, Yaml},
     value::Value,
     Figment,
 };
 use itertools::Itertools;
+use log::warn;
 use num_traits::Zero;
+use ra_ap_cfg::{CfgAtom, CfgDiff};
+use ra_ap_intern::Symbol;
+use ra_ap_paths::Utf8PathBuf;
+use ra_ap_project_model::{CargoConfig, CargoFeatures, CfgOverrides, RustLibSource};
 use rust_extractor_macros::extractor_cli_config;
 use serde::{Deserialize, Serialize};
 use std::fmt::Debug;
@@ -38,6 +46,9 @@ pub struct Config {
     pub trap_dir: PathBuf,
     pub source_archive_dir: PathBuf,
     pub cargo_target_dir: Option<PathBuf>,
+    pub cargo_target: Option<String>,
+    pub cargo_features: Vec<String>,
+    pub cargo_cfg_overrides: Vec<String>,
     pub verbose: u8,
     pub compression: Compression,
     pub inputs: Vec<PathBuf>,
@@ -52,7 +63,7 @@ impl Config {
             .context("expanding parameter files")?;
         let cli_args = CliConfig::parse_from(args);
         let mut figment = Figment::new()
-            .merge(Env::prefixed("CODEQL_"))
+            .merge(Env::raw().only(["CODEQL_VERBOSE"].as_slice()))
             .merge(Env::prefixed("CODEQL_EXTRACTOR_RUST_"))
             .merge(Env::prefixed("CODEQL_EXTRACTOR_RUST_OPTION_"))
             .merge(Serialized::defaults(cli_args));
@@ -78,4 +89,81 @@ impl Config {
         }
         figment.extract().context("loading configuration")
     }
+
+    pub fn to_cargo_config(&self) -> CargoConfig {
+        let sysroot = Some(RustLibSource::Discover);
+
+        let target_dir = self
+            .cargo_target_dir
+            .clone()
+            .unwrap_or_else(|| self.scratch_dir.join("target"));
+        let target_dir = Utf8PathBuf::from_path_buf(target_dir).ok();
+
+        let features = if self.cargo_features.is_empty() {
+            Default::default()
+        } else if self.cargo_features.contains(&"*".to_string()) {
+            CargoFeatures::All
+        } else {
+            CargoFeatures::Selected {
+                features: self.cargo_features.clone(),
+                no_default_features: false,
+            }
+        };
+
+        let target = self.cargo_target.clone();
+
+        let cfg_overrides = to_cfg_overrides(&self.cargo_cfg_overrides);
+
+        CargoConfig {
+            sysroot,
+            target_dir,
+            features,
+            target,
+            cfg_overrides,
+            ..Default::default()
+        }
+    }
+}
+
+fn to_cfg_override(spec: &str) -> CfgAtom {
+    if let Some((key, value)) = spec.split_once("=") {
+        CfgAtom::KeyValue {
+            key: Symbol::intern(key),
+            value: Symbol::intern(value),
+        }
+    } else {
+        CfgAtom::Flag(Symbol::intern(spec))
+    }
+}
+
+fn to_cfg_overrides(specs: &Vec<String>) -> CfgOverrides {
+    let mut enabled_cfgs = Vec::new();
+    let mut disabled_cfgs = Vec::new();
+    let mut has_test_explicitly_enabled = false;
+    for spec in specs {
+        if spec.starts_with("-") {
+            disabled_cfgs.push(to_cfg_override(&spec[1..]));
+        } else {
+            enabled_cfgs.push(to_cfg_override(spec));
+            if spec == "test" {
+                has_test_explicitly_enabled = true;
+            }
+        }
+    }
+    if !has_test_explicitly_enabled {
+        disabled_cfgs.push(to_cfg_override("test"));
+    }
+    if let Some(global) = CfgDiff::new(enabled_cfgs, disabled_cfgs) {
+        CfgOverrides {
+            global,
+            ..Default::default()
+        }
+    } else {
+        warn!("non-disjoint cfg overrides, ignoring: {}", specs.join(", "));
+        CfgOverrides {
+            global: CfgDiff::new(Vec::new(), vec![to_cfg_override("test")])
+                .expect("disabling one cfg should always succeed"),
+            ..Default::default()
+        }
+    }
 }