refactor: add Workspace scanner (#4682)

Cargo.toml

@@ -188,7 +188,7 @@ indexmap             = { version = "2.6.0" }
 insta                = "1.41.1"
 natord               = "1.0.9"
 oxc_resolver         = "1.12.0"
-papaya               = "0.1.5"
+papaya               = "0.1.6"
 proc-macro2          = "1.0.86"
 quickcheck           = "1.0.3"
 quickcheck_macros    = "1.0.0"

biome.json

@@ -1,6 +1,6 @@
 {
   "$schema": "./packages/@biomejs/biome/configuration_schema.json",
-  "assists": {
+  "assist": {
     "enabled": true,
     "ignore": [
       "./packages/@biomejs/biome/configuration_schema.json"

crates/biome_cli/src/commands/mod.rs

@@ -1,6 +1,6 @@
 use crate::changed::{get_changed_files, get_staged_files};
 use crate::cli_options::{cli_options, CliOptions, CliReporter, ColorsArg};
-use crate::execute::Stdin;
+use crate::execute::{ReportMode, Stdin};
 use crate::logging::LoggingKind;
 use crate::{
     execute_mode, setup_cli_subscriber, CliDiagnostic, CliSession, Execution, LoggingLevel, VERSION,
@@ -822,6 +822,17 @@ pub(crate) trait CommandRunner: Sized {
         })?;
 
         let execution = self.get_execution(cli_options, console, workspace)?;
+
+        if execution.traversal_mode().should_scan_project() {
+            let result = workspace.scan_current_project_folder(())?;
+            if cli_options.verbose && matches!(execution.report_mode(), ReportMode::Terminal { .. })
+            {
+                console.log(markup! {
+                    <Info>"Scanned project folder in "<Emphasis>{result.duration}</Emphasis>"."</Info>
+                });
+            }
+        }
+
         Ok((execution, paths))
     }
 

crates/biome_cli/src/execute/migrate.rs

@@ -14,7 +14,9 @@ use biome_json_parser::{parse_json_with_cache, JsonParserOptions};
 use biome_json_syntax::{JsonFileSource, JsonRoot};
 use biome_migrate::{migrate_configuration, ControlFlow};
 use biome_rowan::{AstNode, NodeCache};
-use biome_service::workspace::{ChangeFileParams, FixAction, FormatFileParams, OpenFileParams};
+use biome_service::workspace::{
+    ChangeFileParams, FileContent, FixAction, FormatFileParams, OpenFileParams,
+};
 use std::borrow::Cow;
 use std::ffi::OsStr;
 use std::path::PathBuf;
@@ -66,7 +68,7 @@ pub(crate) fn run(migrate_payload: MigratePayload) -> Result<(), CliDiagnostic>
     let biome_path = BiomePath::new(configuration_file_path.as_path());
     workspace.open_file(OpenFileParams {
         path: biome_path.clone(),
-        content: biome_config_content.to_string(),
+        content: FileContent::FromClient(biome_config_content.to_string()),
         version: 0,
         document_file_source: Some(JsonFileSource::json().into()),
     })?;

crates/biome_cli/src/execute/mod.rs

@@ -23,7 +23,8 @@ use biome_diagnostics::{category, Category};
 use biome_fs::BiomePath;
 use biome_grit_patterns::GritTargetLanguage;
 use biome_service::workspace::{
-    FeatureName, FeaturesBuilder, FixFileMode, FormatFileParams, OpenFileParams, PatternId,
+    FeatureName, FeaturesBuilder, FileContent, FixFileMode, FormatFileParams, OpenFileParams,
+    PatternId,
 };
 use std::ffi::OsString;
 use std::fmt::{Display, Formatter};
@@ -228,6 +229,16 @@ impl Display for TraversalMode {
     }
 }
 
+impl TraversalMode {
+    pub fn should_scan_project(&self) -> bool {
+        matches!(self, Self::CI { .. })
+            || matches!(
+                self,
+                Self::Check { stdin,.. } | Self::Lint { stdin, .. } if stdin.is_none()
+            )
+    }
+}
+
 /// Tells to the execution of the traversal how the information should be reported
 #[derive(Copy, Clone, Debug)]
 pub enum ReportMode {
@@ -529,7 +540,7 @@ pub fn execute_mode(
                     })?;
                     let report_file = BiomePath::new("_report_output.json");
                     session.app.workspace.open_file(OpenFileParams {
-                        content,
+                        content: FileContent::FromClient(content),
                         path: report_file.clone(),
                         version: 0,
                         document_file_source: None,

crates/biome_cli/src/execute/process_file/workspace_file.rs

@@ -2,7 +2,7 @@ use crate::execute::diagnostics::{ResultExt, ResultIoExt};
 use crate::execute::process_file::SharedTraversalOptions;
 use biome_diagnostics::{category, Error};
 use biome_fs::{BiomePath, File, OpenOptions};
-use biome_service::workspace::{FileGuard, OpenFileParams};
+use biome_service::workspace::{FileContent, FileGuard, OpenFileParams};
 use biome_service::{Workspace, WorkspaceError};
 use std::ffi::OsStr;
 use std::path::{Path, PathBuf};
@@ -40,7 +40,7 @@ impl<'ctx, 'app> WorkspaceFile<'ctx, 'app> {
                 document_file_source: None,
                 path: biome_path,
                 version: 0,
-                content: input.clone(),
+                content: FileContent::FromClient(input.clone()),
             },
         )
         .with_file_path_and_code(path.display().to_string(), category!("internalError/fs"))?;

crates/biome_cli/src/execute/std_in.rs

@@ -9,8 +9,8 @@ use biome_diagnostics::PrintDiagnostic;
 use biome_fs::BiomePath;
 use biome_service::file_handlers::{AstroFileHandler, SvelteFileHandler, VueFileHandler};
 use biome_service::workspace::{
-    ChangeFileParams, DropPatternParams, FeaturesBuilder, FixFileParams, FormatFileParams,
-    OpenFileParams, OrganizeImportsParams, SupportsFeatureParams,
+    ChangeFileParams, DropPatternParams, FeaturesBuilder, FileContent, FixFileParams,
+    FormatFileParams, OpenFileParams, OrganizeImportsParams, SupportsFeatureParams,
 };
 use biome_service::WorkspaceError;
 use std::borrow::Cow;
@@ -48,7 +48,7 @@ pub(crate) fn run<'a>(
             workspace.open_file(OpenFileParams {
                 path: biome_path.clone(),
                 version: 0,
-                content: content.into(),
+                content: FileContent::FromClient(content.into()),
                 document_file_source: None,
             })?;
             let printed = workspace.format_file(FormatFileParams {
@@ -80,7 +80,7 @@ pub(crate) fn run<'a>(
         workspace.open_file(OpenFileParams {
             path: biome_path.clone(),
             version: 0,
-            content: content.into(),
+            content: FileContent::FromClient(content.into()),
             document_file_source: None,
         })?;
         // apply fix file of the linter

crates/biome_cli/src/execute/traverse.rs

@@ -25,10 +25,7 @@ use std::{
     ffi::OsString,
     panic::catch_unwind,
     path::PathBuf,
-    sync::{
-        atomic::{AtomicUsize, Ordering},
-        Once,
-    },
+    sync::atomic::{AtomicUsize, Ordering},
     thread,
     time::{Duration, Instant},
 };
@@ -45,8 +42,6 @@ pub(crate) fn traverse(
     cli_options: &CliOptions,
     mut inputs: Vec<OsString>,
 ) -> Result<TraverseResult, CliDiagnostic> {
-    init_thread_pool();
-
     if inputs.is_empty() {
         match &execution.traversal_mode {
             TraversalMode::Check { .. }
@@ -155,19 +150,6 @@ pub(crate) fn traverse(
     })
 }
 
-/// This function will setup the global Rayon thread pool the first time it's called
-///
-/// This is currently only used to assign friendly debug names to the threads of the pool
-fn init_thread_pool() {
-    static INIT_ONCE: Once = Once::new();
-    INIT_ONCE.call_once(|| {
-        rayon::ThreadPoolBuilder::new()
-            .thread_name(|index| format!("biome::worker_{index}"))
-            .build_global()
-            .expect("failed to initialize the global thread pool");
-    });
-}
-
 /// Initiate the filesystem traversal tasks with the provided input paths and
 /// run it to completion, returning the duration of the process and the evaluated paths
 fn traverse_inputs(

crates/biome_cli/tests/snap_test.rs

@@ -150,11 +150,10 @@ fn redact_snapshot(input: &str) -> Option<Cow<'_, str>> {
     let mut output = Cow::Borrowed(input);
 
     // There are some logs that print the timing, and we can't snapshot that message
-    // otherwise at each run we invalid the previous snapshot.
+    // otherwise at each run we invalidate the previous snapshot.
     //
     // This is a workaround, and it might not work for all cases.
-    let the_match = TIME_REGEX.find(output.as_ref()).map(|f| f.start()..f.end());
-    if let Some(found) = the_match {
+    while let Some(found) = TIME_REGEX.find(&output).map(|f| f.start()..f.end()) {
         output.to_mut().replace_range(found, " <TIME>.");
     }
 

crates/biome_cli/tests/snapshots/main_cases_diagnostics/max_diagnostics_verbose.snap

@@ -1,7 +1,6 @@
 ---
 source: crates/biome_cli/tests/snap_test.rs
 expression: content
-snapshot_kind: text
 ---
 ## `src/file.js`
 
@@ -22,6 +21,10 @@ ci ━━━━━━━━━━━━━━━━━━━━━━━━━
 
 # Emitted Messages
 
+```block
+Scanned project folder in <TIME>.
+```
+
 ```block
 src/folder_1/package-lock.json project  VERBOSE  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 

crates/biome_cli/tests/snapshots/main_commands_check/print_verbose.snap

@@ -22,6 +22,10 @@ check ━━━━━━━━━━━━━━━━━━━━━━━━
 
 # Emitted Messages
 
+```block
+Scanned project folder in <TIME>.
+```
+
 ```block
 check.js:1:1 lint/style/useWhile  FIXABLE  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 

crates/biome_cli/tests/snapshots/main_commands_check/print_verbose_write.snap

@@ -11,6 +11,10 @@ while (true);
 
 # Emitted Messages
 
+```block
+Scanned project folder in <TIME>.
+```
+
 ```block
 Checked 1 file in <TIME>. Fixed 1 file.
 ```

crates/biome_cli/tests/snapshots/main_commands_check/unsupported_file_verbose.snap

@@ -22,6 +22,10 @@ internalError/io ━━━━━━━━━━━━━━━━━━━━━
 
 # Emitted Messages
 
+```block
+Scanned project folder in <TIME>.
+```
+
 ```block
 check.txt files/missingHandler  VERBOSE  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 

crates/biome_cli/tests/snapshots/main_commands_ci/print_verbose.snap

@@ -22,6 +22,10 @@ ci ━━━━━━━━━━━━━━━━━━━━━━━━━
 
 # Emitted Messages
 
+```block
+Scanned project folder in <TIME>.
+```
+
 ```block
 ci.js:1:1 lint/style/useWhile  FIXABLE  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 

crates/biome_cli/tests/snapshots/main_commands_lint/print_verbose.snap

@@ -22,6 +22,10 @@ lint ━━━━━━━━━━━━━━━━━━━━━━━━━
 
 # Emitted Messages
 
+```block
+Scanned project folder in <TIME>.
+```
+
 ```block
 check.js:1:1 lint/style/useWhile  FIXABLE  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 

crates/biome_cli/tests/snapshots/main_commands_lint/unsupported_file_verbose.snap

@@ -22,6 +22,10 @@ internalError/io ━━━━━━━━━━━━━━━━━━━━━
 
 # Emitted Messages
 
+```block
+Scanned project folder in <TIME>.
+```
+
 ```block
 check.txt files/missingHandler  VERBOSE  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
 

crates/biome_fs/src/interner.rs

@@ -22,7 +22,8 @@ impl PathInterner {
         (interner, recv)
     }
 
-    /// Insert the path.
+    /// Inserts the path.
+    ///
     /// Returns `true` if the path was not previously inserted.
     pub fn intern_path(&self, path: PathBuf) -> bool {
         let result = self.storage.pin().insert(path.clone());

crates/biome_lsp/src/handlers/text_document.rs

@@ -2,7 +2,8 @@ use crate::utils::apply_document_changes;
 use crate::{documents::Document, session::Session};
 use anyhow::Result;
 use biome_service::workspace::{
-    ChangeFileParams, CloseFileParams, DocumentFileSource, GetFileContentParams, OpenFileParams,
+    ChangeFileParams, CloseFileParams, DocumentFileSource, FileContent, GetFileContentParams,
+    OpenFileParams,
 };
 use tower_lsp::lsp_types;
 use tracing::{error, field};
@@ -31,7 +32,7 @@ pub(crate) async fn did_open(
     session.workspace.open_file(OpenFileParams {
         path: biome_path,
         version,
-        content,
+        content: FileContent::FromClient(content),
         document_file_source: Some(language_hint),
     })?;
 

crates/biome_lsp/src/server.rs

@@ -609,6 +609,7 @@ impl ServerFactory {
         workspace_method!(builder, is_path_ignored);
         workspace_method!(builder, update_settings);
         workspace_method!(builder, register_project_folder);
+        workspace_method!(builder, scan_current_project_folder);
         workspace_method!(builder, unregister_project_folder);
         workspace_method!(builder, open_file);
         workspace_method!(builder, set_manifest_for_project);

crates/biome_service/Cargo.toml

@@ -54,12 +54,14 @@ biome_rowan              = { workspace = true, features = ["serde"] }
 biome_string_case        = { workspace = true }
 biome_text_edit          = { workspace = true }
 bpaf                     = { workspace = true }
+crossbeam                = { workspace = true }
 enumflags2               = { workspace = true, features = ["serde"] }
 getrandom                = { workspace = true, features = ["js"] }
 ignore                   = { workspace = true }
 indexmap                 = { workspace = true, features = ["serde"] }
 oxc_resolver             = { workspace = true }
 papaya                   = { workspace = true }
+rayon                    = { workspace = true }
 regex                    = { workspace = true }
 rustc-hash               = { workspace = true }
 schemars                 = { workspace = true, optional = true }