Auto merge of #9525 - willcrichton:example-analyzer, r=alexcrichton

Scrape code examples from examples/ directory for Rustdoc

Adds support for the functionality described in rust-lang/rfcs#3123

Matching changes to rustdoc are here: rust-lang/rust#85833

src/bin/cargo/commands/doc.rs

@@ -1,6 +1,7 @@
 use crate::command_prelude::*;
 
-use cargo::ops::{self, DocOptions};
+use anyhow::anyhow;
+use cargo::ops::{self, CompileFilter, DocOptions, FilterRule, LibRule};
 
 pub fn cli() -> App {
     subcommand("doc")
@@ -19,6 +20,13 @@ pub fn cli() -> App {
         )
         .arg(opt("no-deps", "Don't build documentation for dependencies"))
         .arg(opt("document-private-items", "Document private items"))
+        .arg(
+            opt(
+                "scrape-examples",
+                "Scrape examples to include as function documentation",
+            )
+            .value_name("FLAGS"),
+        )
         .arg_jobs()
         .arg_targets_lib_bin_example(
             "Document only this package's library",
@@ -48,6 +56,33 @@ pub fn exec(config: &mut Config, args: &ArgMatches<'_>) -> CliResult {
         args.compile_options(config, mode, Some(&ws), ProfileChecking::Custom)?;
     compile_opts.rustdoc_document_private_items = args.is_present("document-private-items");
 
+    // TODO(wcrichto): move scrape example configuration into Cargo.toml before stabilization
+    // See: https://github.com/rust-lang/cargo/pull/9525#discussion_r728470927
+    compile_opts.rustdoc_scrape_examples = match args.value_of("scrape-examples") {
+        Some(s) => Some(match s {
+            "all" => CompileFilter::new_all_targets(),
+            "examples" => CompileFilter::new(
+                LibRule::False,
+                FilterRule::none(),
+                FilterRule::none(),
+                FilterRule::All,
+                FilterRule::none(),
+            ),
+            _ => {
+                return Err(CliError::from(anyhow!(
+                    r#"--scrape-examples must take "all" or "examples" as an argument"#
+                )));
+            }
+        }),
+        None => None,
+    };
+
+    if compile_opts.rustdoc_scrape_examples.is_some() {
+        config
+            .cli_unstable()
+            .fail_if_stable_opt("--scrape-examples", 9910)?;
+    }
+
     let doc_opts = DocOptions {
         open_result: args.is_present("open"),
         compile_opts,

src/cargo/core/compiler/build_config.rs

@@ -149,6 +149,8 @@ pub enum CompileMode {
     Doc { deps: bool },
     /// A target that will be tested with `rustdoc`.
     Doctest,
+    /// An example or library that will be scraped for function calls by `rustdoc`.
+    Docscrape,
     /// A marker for Units that represent the execution of a `build.rs` script.
     RunCustomBuild,
 }
@@ -166,6 +168,7 @@ impl ser::Serialize for CompileMode {
             Bench => "bench".serialize(s),
             Doc { .. } => "doc".serialize(s),
             Doctest => "doctest".serialize(s),
+            Docscrape => "docscrape".serialize(s),
             RunCustomBuild => "run-custom-build".serialize(s),
         }
     }
@@ -187,6 +190,11 @@ impl CompileMode {
         self == CompileMode::Doctest
     }
 
+    /// Returns `true` if this is scraping examples for documentation.
+    pub fn is_doc_scrape(self) -> bool {
+        self == CompileMode::Docscrape
+    }
+
     /// Returns `true` if this is any type of test (test, benchmark, doc test, or
     /// check test).
     pub fn is_any_test(self) -> bool {

src/cargo/core/compiler/build_context/mod.rs

@@ -47,6 +47,9 @@ pub struct BuildContext<'a, 'cfg> {
     /// The dependency graph of units to compile.
     pub unit_graph: UnitGraph,
 
+    /// Reverse-dependencies of documented units, used by the rustdoc --scrape-examples flag.
+    pub scrape_units: Vec<Unit>,
+
     /// The list of all kinds that are involved in this build
     pub all_kinds: HashSet<CompileKind>,
 }
@@ -61,6 +64,7 @@ impl<'a, 'cfg> BuildContext<'a, 'cfg> {
         target_data: RustcTargetData<'cfg>,
         roots: Vec<Unit>,
         unit_graph: UnitGraph,
+        scrape_units: Vec<Unit>,
     ) -> CargoResult<BuildContext<'a, 'cfg>> {
         let all_kinds = unit_graph
             .keys()
@@ -79,6 +83,7 @@ impl<'a, 'cfg> BuildContext<'a, 'cfg> {
             target_data,
             roots,
             unit_graph,
+            scrape_units,
             all_kinds,
         })
     }

src/cargo/core/compiler/build_context/target_info.rs

@@ -452,7 +452,10 @@ impl TargetInfo {
                 }
             }
             CompileMode::Check { .. } => Ok((vec![FileType::new_rmeta()], Vec::new())),
-            CompileMode::Doc { .. } | CompileMode::Doctest | CompileMode::RunCustomBuild => {
+            CompileMode::Doc { .. }
+            | CompileMode::Doctest
+            | CompileMode::Docscrape
+            | CompileMode::RunCustomBuild => {
                 panic!("asked for rustc output for non-rustc mode")
             }
         }

src/cargo/core/compiler/context/compilation_files.rs

@@ -191,7 +191,9 @@ impl<'a, 'cfg: 'a> CompilationFiles<'a, 'cfg> {
     /// Returns the directory where the artifacts for the given unit are
     /// initially created.
     pub fn out_dir(&self, unit: &Unit) -> PathBuf {
-        if unit.mode.is_doc() {
+        // Docscrape units need to have doc/ set as the out_dir so sources for reverse-dependencies
+        // will be put into doc/ and not into deps/ where the *.examples files are stored.
+        if unit.mode.is_doc() || unit.mode.is_doc_scrape() {
             self.layout(unit.kind).doc().to_path_buf()
         } else if unit.mode.is_doc_test() {
             panic!("doc tests do not have an out dir");
@@ -417,6 +419,17 @@ impl<'a, 'cfg: 'a> CompilationFiles<'a, 'cfg> {
                 // but Cargo does not know about that.
                 vec![]
             }
+            CompileMode::Docscrape => {
+                let path = self
+                    .deps_dir(unit)
+                    .join(format!("{}.examples", unit.buildkey()));
+                vec![OutputFile {
+                    path,
+                    hardlink: None,
+                    export_path: None,
+                    flavor: FileFlavor::Normal,
+                }]
+            }
             CompileMode::Test
             | CompileMode::Build
             | CompileMode::Bench

src/cargo/core/compiler/context/mod.rs

@@ -80,6 +80,10 @@ pub struct Context<'a, 'cfg> {
     /// compilation is happening (only object, only bitcode, both, etc), and is
     /// precalculated early on.
     pub lto: HashMap<Unit, Lto>,
+
+    /// Map of Doc/Docscrape units to metadata for their -Cmetadata flag.
+    /// See Context::find_metadata_units for more details.
+    pub metadata_for_doc_units: HashMap<Unit, Metadata>,
 }
 
 impl<'a, 'cfg> Context<'a, 'cfg> {
@@ -120,6 +124,7 @@ impl<'a, 'cfg> Context<'a, 'cfg> {
             rustc_clients: HashMap::new(),
             pipelining,
             lto: HashMap::new(),
+            metadata_for_doc_units: HashMap::new(),
         })
     }
 
@@ -134,6 +139,7 @@ impl<'a, 'cfg> Context<'a, 'cfg> {
         self.prepare()?;
         custom_build::build_map(&mut self)?;
         self.check_collisions()?;
+        self.compute_metadata_for_doc_units();
 
         // We need to make sure that if there were any previous docs
         // already compiled, they were compiled with the same Rustc version that we're currently
@@ -620,4 +626,40 @@ impl<'a, 'cfg> Context<'a, 'cfg> {
 
         Ok(client)
     }
+
+    /// Finds metadata for Doc/Docscrape units.
+    ///
+    /// rustdoc needs a -Cmetadata flag in order to recognize StableCrateIds that refer to
+    /// items in the crate being documented. The -Cmetadata flag used by reverse-dependencies
+    /// will be the metadata of the Cargo unit that generated the current library's rmeta file,
+    /// which should be a Check unit.
+    ///
+    /// If the current crate has reverse-dependencies, such a Check unit should exist, and so
+    /// we use that crate's metadata. If not, we use the crate's Doc unit so at least examples
+    /// scraped from the current crate can be used when documenting the current crate.
+    pub fn compute_metadata_for_doc_units(&mut self) {
+        for unit in self.bcx.unit_graph.keys() {
+            if !unit.mode.is_doc() && !unit.mode.is_doc_scrape() {
+                continue;
+            }
+
+            let matching_units = self
+                .bcx
+                .unit_graph
+                .keys()
+                .filter(|other| {
+                    unit.pkg == other.pkg
+                        && unit.target == other.target
+                        && !other.mode.is_doc_scrape()
+                })
+                .collect::<Vec<_>>();
+            let metadata_unit = matching_units
+                .iter()
+                .find(|other| other.mode.is_check())
+                .or_else(|| matching_units.iter().find(|other| other.mode.is_doc()))
+                .unwrap_or(&unit);
+            self.metadata_for_doc_units
+                .insert(unit.clone(), self.files().metadata(metadata_unit));
+        }
+    }
 }

src/cargo/core/compiler/mod.rs

@@ -165,7 +165,7 @@ fn compile<'cfg>(
         let force = exec.force_rebuild(unit) || force_rebuild;
         let mut job = fingerprint::prepare_target(cx, unit, force)?;
         job.before(if job.freshness() == Freshness::Dirty {
-            let work = if unit.mode.is_doc() {
+            let work = if unit.mode.is_doc() || unit.mode.is_doc_scrape() {
                 rustdoc(cx, unit)?
             } else {
                 rustc(cx, unit, exec)?
@@ -647,6 +647,42 @@ fn rustdoc(cx: &mut Context<'_, '_>, unit: &Unit) -> CargoResult<Work> {
         rustdoc.args(args);
     }
 
+    let metadata = cx.metadata_for_doc_units[&unit];
+    rustdoc.arg("-C").arg(format!("metadata={}", metadata));
+
+    let scrape_output_path = |unit: &Unit| -> CargoResult<PathBuf> {
+        let output_dir = cx.files().deps_dir(unit);
+        Ok(output_dir.join(format!("{}.examples", unit.buildkey())))
+    };
+
+    if unit.mode.is_doc_scrape() {
+        debug_assert!(cx.bcx.scrape_units.contains(unit));
+
+        rustdoc.arg("-Zunstable-options");
+
+        rustdoc
+            .arg("--scrape-examples-output-path")
+            .arg(scrape_output_path(unit)?);
+
+        // Only scrape example for items from crates in the workspace, to reduce generated file size
+        for pkg in cx.bcx.ws.members() {
+            rustdoc
+                .arg("--scrape-examples-target-crate")
+                .arg(pkg.name());
+        }
+    } else if cx.bcx.scrape_units.len() > 0 && cx.bcx.ws.is_member(&unit.pkg) {
+        // We only pass scraped examples to packages in the workspace
+        // since examples are only coming from reverse-dependencies of workspace packages
+
+        rustdoc.arg("-Zunstable-options");
+
+        for scrape_unit in &cx.bcx.scrape_units {
+            rustdoc
+                .arg("--with-examples")
+                .arg(scrape_output_path(scrape_unit)?);
+        }
+    }
+
     build_deps_args(&mut rustdoc, cx, unit)?;
     rustdoc::add_root_urls(cx, unit, &mut rustdoc)?;
 

src/cargo/core/compiler/timings.rs

@@ -176,6 +176,7 @@ impl<'cfg> Timings<'cfg> {
             CompileMode::Bench => target.push_str(" (bench)"),
             CompileMode::Doc { .. } => target.push_str(" (doc)"),
             CompileMode::Doctest => target.push_str(" (doc test)"),
+            CompileMode::Docscrape => target.push_str(" (doc scrape)"),
             CompileMode::RunCustomBuild => target.push_str(" (run)"),
         }
         let unit_time = UnitTime {

src/cargo/core/compiler/unit_dependencies.rs

@@ -47,6 +47,7 @@ struct State<'a, 'cfg> {
     target_data: &'a RustcTargetData<'cfg>,
     profiles: &'a Profiles,
     interner: &'a UnitInterner,
+    scrape_units: &'a [Unit],
 
     /// A set of edges in `unit_dependencies` where (a, b) means that the
     /// dependency from a to b was added purely because it was a dev-dependency.
@@ -61,6 +62,7 @@ pub fn build_unit_dependencies<'a, 'cfg>(
     features: &'a ResolvedFeatures,
     std_resolve: Option<&'a (Resolve, ResolvedFeatures)>,
     roots: &[Unit],
+    scrape_units: &[Unit],
     std_roots: &HashMap<CompileKind, Vec<Unit>>,
     global_mode: CompileMode,
     target_data: &'a RustcTargetData<'cfg>,
@@ -91,6 +93,7 @@ pub fn build_unit_dependencies<'a, 'cfg>(
         target_data,
         profiles,
         interner,
+        scrape_units,
         dev_dependency_edges: HashSet::new(),
     };
 
@@ -253,6 +256,7 @@ fn compute_deps(
         if !dep.is_transitive()
             && !unit.target.is_test()
             && !unit.target.is_example()
+            && !unit.mode.is_doc_scrape()
             && !unit.mode.is_any_test()
         {
             return false;
@@ -467,6 +471,25 @@ fn compute_deps_doc(
     if unit.target.is_bin() || unit.target.is_example() {
         ret.extend(maybe_lib(unit, state, unit_for)?);
     }
+
+    // Add all units being scraped for examples as a dependency of Doc units.
+    if state.ws.is_member(&unit.pkg) {
+        for scrape_unit in state.scrape_units.iter() {
+            // This needs to match the FeaturesFor used in cargo_compile::generate_targets.
+            let unit_for = UnitFor::new_host(scrape_unit.target.proc_macro());
+            deps_of(scrape_unit, state, unit_for)?;
+            ret.push(new_unit_dep(
+                state,
+                scrape_unit,
+                &scrape_unit.pkg,
+                &scrape_unit.target,
+                unit_for,
+                scrape_unit.kind,
+                scrape_unit.mode,
+            )?);
+        }
+    }
+
     Ok(ret)
 }
 
@@ -558,7 +581,7 @@ fn dep_build_script(
 /// Choose the correct mode for dependencies.
 fn check_or_build_mode(mode: CompileMode, target: &Target) -> CompileMode {
     match mode {
-        CompileMode::Check { .. } | CompileMode::Doc { .. } => {
+        CompileMode::Check { .. } | CompileMode::Doc { .. } | CompileMode::Docscrape => {
             if target.for_host() {
                 // Plugin and proc macro targets should be compiled like
                 // normal.
@@ -695,6 +718,14 @@ fn connect_run_custom_build_deps(state: &mut State<'_, '_>) {
                         && other.unit.target.is_linkable()
                         && other.unit.pkg.manifest().links().is_some()
                 })
+                // Avoid cycles when using the doc --scrape-examples feature:
+                // Say a workspace has crates A and B where A has a build-dependency on B.
+                // The Doc units for A and B will have a dependency on the Docscrape for both A and B.
+                // So this would add a dependency from B-build to A-build, causing a cycle:
+                //   B (build) -> A (build) -> B(build)
+                // See the test scrape_examples_avoid_build_script_cycle for a concrete example.
+                // To avoid this cycle, we filter out the B -> A (docscrape) dependency.
+                .filter(|(_parent, other)| !other.unit.mode.is_doc_scrape())
                 // Skip dependencies induced via dev-dependencies since
                 // connections between `links` and build scripts only happens
                 // via normal dependencies. Otherwise since dev-dependencies can

src/cargo/core/profiles.rs

@@ -323,7 +323,9 @@ impl Profiles {
                         (InternedString::new("dev"), None)
                     }
                 }
-                CompileMode::Doc { .. } => (InternedString::new("doc"), None),
+                CompileMode::Doc { .. } | CompileMode::Docscrape => {
+                    (InternedString::new("doc"), None)
+                }
             }
         } else {
             (self.requested_profile, None)