Rollup merge of rust-lang#134531 - GuillaumeGomez:extract-doctests, r…

…=notriddle,aDotInTheVoid

[rustdoc] Add `--extract-doctests` command-line flag

Part of rust-lang#134529.

It was discussed with the Rust-for-Linux project recently that they needed a way to extract doctests so they can modify them and then run them more easily (look for "a way to extract doctests" [here](Rust-for-Linux/linux#2)).

For now, I output most of `ScrapedDoctest` fields in JSON format with `serde_json`. So it outputs the following information:

 * filename
 * line
 * langstr
 * text

cc `@ojeda`
r? `@notriddle`

src/doc/rustdoc/src/unstable-features.md

@@ -524,6 +524,8 @@ use `-o -`.
 
 ## `-w`/`--output-format`: output format
 
+### json
+
 `--output-format json` emits documentation in the experimental
 [JSON format](https://doc.rust-lang.org/nightly/nightly-rustc/rustdoc_json_types/). `--output-format html` has no effect,
 and is also accepted on stable toolchains.
@@ -542,6 +544,68 @@ It can also be used with `--show-coverage`. Take a look at its
 [documentation](#--show-coverage-calculate-the-percentage-of-items-with-documentation) for more
 information.
 
+### doctest
+
+`--output-format doctest` emits JSON on stdout which gives you information about doctests in the
+provided crate.
+
+Tracking issue: [#134529](https://github.com/rust-lang/rust/issues/134529)
+
+You can use this option like this:
+
+```bash
+rustdoc -Zunstable-options --output-format=doctest src/lib.rs
+```
+
+For this rust code:
+
+```rust
+/// ```
+/// let x = 12;
+/// ```
+pub trait Trait {}
+```
+
+The generated output (formatted) will look like this:
+
+```json
+{
+  "format_version": 1,
+  "doctests": [
+    {
+      "file": "foo.rs",
+      "line": 1,
+      "doctest_attributes": {
+        "original": "",
+        "should_panic": false,
+        "no_run": false,
+        "ignore": "None",
+        "rust": true,
+        "test_harness": false,
+        "compile_fail": false,
+        "standalone_crate": false,
+        "error_codes": [],
+        "edition": null,
+        "added_css_classes": [],
+        "unknown": []
+      },
+      "original_code": "let x = 12;",
+      "doctest_code": "#![allow(unused)]\nfn main() {\nlet x = 12;\n}",
+      "name": "foo.rs - Trait (line 1)"
+    }
+  ]
+}
+```
+
+ * `format_version` gives you the current version of the generated JSON. If we change the output in any way, the number will increase.
+ * `doctests` contains the list of doctests present in the crate.
+   * `file` is the file path where the doctest is located.
+   * `line` is the line where the doctest starts (so where the \`\`\` is located in the current code).
+   * `doctest_attributes` contains computed information about the attributes used on the doctests. For more information about doctest attributes, take a look [here](write-documentation/documentation-tests.html#attributes).
+   * `original_code` is the code as written in the source code before rustdoc modifies it.
+   * `doctest_code` is the code modified by rustdoc that will be run. If there is a fatal syntax error, this field will not be present.
+   * `name` is the name generated by rustdoc which represents this doctest.
+
 ## `--enable-per-target-ignores`: allow `ignore-foo` style filters for doctests
 
  * Tracking issue: [#64245](https://github.com/rust-lang/rust/issues/64245)

src/librustdoc/config.rs

@@ -33,6 +33,7 @@ pub(crate) enum OutputFormat {
     Json,
     #[default]
     Html,
+    Doctest,
 }
 
 impl OutputFormat {
@@ -48,6 +49,7 @@ impl TryFrom<&str> for OutputFormat {
         match value {
             "json" => Ok(OutputFormat::Json),
             "html" => Ok(OutputFormat::Html),
+            "doctest" => Ok(OutputFormat::Doctest),
             _ => Err(format!("unknown output format `{value}`")),
         }
     }
@@ -445,14 +447,42 @@ impl Options {
             }
         }
 
+        let show_coverage = matches.opt_present("show-coverage");
+        let output_format_s = matches.opt_str("output-format");
+        let output_format = match output_format_s {
+            Some(ref s) => match OutputFormat::try_from(s.as_str()) {
+                Ok(out_fmt) => out_fmt,
+                Err(e) => dcx.fatal(e),
+            },
+            None => OutputFormat::default(),
+        };
+
         // check for `--output-format=json`
-        if !matches!(matches.opt_str("output-format").as_deref(), None | Some("html"))
-            && !matches.opt_present("show-coverage")
-            && !nightly_options::is_unstable_enabled(matches)
-        {
-            dcx.fatal(
-                "the -Z unstable-options flag must be passed to enable --output-format for documentation generation (see https://github.com/rust-lang/rust/issues/76578)",
-            );
+        match (
+            output_format_s.as_ref().map(|_| output_format),
+            show_coverage,
+            nightly_options::is_unstable_enabled(matches),
+        ) {
+            (None | Some(OutputFormat::Json), true, _) => {}
+            (_, true, _) => {
+                dcx.fatal(format!(
+                    "`--output-format={}` is not supported for the `--show-coverage` option",
+                    output_format_s.unwrap_or_default(),
+                ));
+            }
+            // If `-Zunstable-options` is used, nothing to check after this point.
+            (_, false, true) => {}
+            (None | Some(OutputFormat::Html), false, _) => {}
+            (Some(OutputFormat::Json), false, false) => {
+                dcx.fatal(
+                    "the -Z unstable-options flag must be passed to enable --output-format for documentation generation (see https://github.com/rust-lang/rust/issues/76578)",
+                );
+            }
+            (Some(OutputFormat::Doctest), false, false) => {
+                dcx.fatal(
+                    "the -Z unstable-options flag must be passed to enable --output-format for documentation generation (see https://github.com/rust-lang/rust/issues/134529)",
+                );
+            }
         }
 
         let to_check = matches.opt_strs("check-theme");
@@ -704,29 +734,13 @@ impl Options {
             })
             .collect();
 
-        let show_coverage = matches.opt_present("show-coverage");
-
         let crate_types = match parse_crate_types_from_list(matches.opt_strs("crate-type")) {
             Ok(types) => types,
             Err(e) => {
                 dcx.fatal(format!("unknown crate type: {e}"));
             }
         };
 
-        let output_format = match matches.opt_str("output-format") {
-            Some(s) => match OutputFormat::try_from(s.as_str()) {
-                Ok(out_fmt) => {
-                    if !out_fmt.is_json() && show_coverage {
-                        dcx.fatal(
-                            "html output format isn't supported for the --show-coverage option",
-                        );
-                    }
-                    out_fmt
-                }
-                Err(e) => dcx.fatal(e),
-            },
-            None => OutputFormat::default(),
-        };
         let crate_name = matches.opt_str("crate-name");
         let bin_crate = crate_types.contains(&CrateType::Executable);
         let proc_macro_crate = crate_types.contains(&CrateType::ProcMacro);

src/librustdoc/doctest.rs

@@ -1,3 +1,4 @@
+mod extracted;
 mod make;
 mod markdown;
 mod runner;
@@ -30,7 +31,7 @@ use tempfile::{Builder as TempFileBuilder, TempDir};
 use tracing::debug;
 
 use self::rust::HirCollector;
-use crate::config::Options as RustdocOptions;
+use crate::config::{Options as RustdocOptions, OutputFormat};
 use crate::html::markdown::{ErrorCodes, Ignore, LangString, MdRelLine};
 use crate::lint::init_lints;
 
@@ -209,15 +210,8 @@ pub(crate) fn run(dcx: DiagCtxtHandle<'_>, input: Input, options: RustdocOptions
     let args_path = temp_dir.path().join("rustdoc-cfgs");
     crate::wrap_return(dcx, generate_args_file(&args_path, &options));
 
-    let CreateRunnableDocTests {
-        standalone_tests,
-        mergeable_tests,
-        rustdoc_options,
-        opts,
-        unused_extern_reports,
-        compiling_test_count,
-        ..
-    } = interface::run_compiler(config, |compiler| {
+    let extract_doctests = options.output_format == OutputFormat::Doctest;
+    let result = interface::run_compiler(config, |compiler| {
         let krate = rustc_interface::passes::parse(&compiler.sess);
 
         let collector = rustc_interface::create_and_enter_global_ctxt(compiler, krate, |tcx| {
@@ -226,22 +220,53 @@ pub(crate) fn run(dcx: DiagCtxtHandle<'_>, input: Input, options: RustdocOptions
             let opts = scrape_test_config(crate_name, crate_attrs, args_path);
             let enable_per_target_ignores = options.enable_per_target_ignores;
 
-            let mut collector = CreateRunnableDocTests::new(options, opts);
             let hir_collector = HirCollector::new(
                 ErrorCodes::from(compiler.sess.opts.unstable_features.is_nightly_build()),
                 enable_per_target_ignores,
                 tcx,
             );
             let tests = hir_collector.collect_crate();
-            tests.into_iter().for_each(|t| collector.add_test(t));
+            if extract_doctests {
+                let mut collector = extracted::ExtractedDocTests::new();
+                tests.into_iter().for_each(|t| collector.add_test(t, &opts, &options));
+
+                let stdout = std::io::stdout();
+                let mut stdout = stdout.lock();
+                if let Err(error) = serde_json::ser::to_writer(&mut stdout, &collector) {
+                    eprintln!();
+                    Err(format!("Failed to generate JSON output for doctests: {error:?}"))
+                } else {
+                    Ok(None)
+                }
+            } else {
+                let mut collector = CreateRunnableDocTests::new(options, opts);
+                tests.into_iter().for_each(|t| collector.add_test(t));
 
-            collector
+                Ok(Some(collector))
+            }
         });
         compiler.sess.dcx().abort_if_errors();
 
         collector
     });
 
+    let CreateRunnableDocTests {
+        standalone_tests,
+        mergeable_tests,
+        rustdoc_options,
+        opts,
+        unused_extern_reports,
+        compiling_test_count,
+        ..
+    } = match result {
+        Ok(Some(collector)) => collector,
+        Ok(None) => return,
+        Err(error) => {
+            eprintln!("{error}");
+            std::process::exit(1);
+        }
+    };
+
     run_tests(opts, &rustdoc_options, &unused_extern_reports, standalone_tests, mergeable_tests);
 
     let compiling_test_count = compiling_test_count.load(Ordering::SeqCst);