fix(886): parse cargo diagnostics in run_check/run_build/run_lint
Before: tool_run_check (and run_build/run_lint via run_script_tool)
returned the entire cargo log verbatim in `output`. For runs with many
errors the response routinely exceeded the MCP token cap, was dumped
to a tool-results file, and the agent had to scrape it with python3
just to see the error list — burning many turns on file archaeology
for what should be a one-look operation. Real example: 864's coder
hit `result (143,708 characters) exceeds maximum allowed tokens` and
spent ~8 turns extracting 3 errors.
Now:
- New `service::shell::parse_diagnostics` parses `error[CODE]:` /
`warning[CODE]:` headers + their `--> file:line` markers into
structured `Diagnostic { kind, code, message, file, line }`.
- `tool_run_check` (and the run_build/run_lint shared body) returns
`{ passed, exit_code, errors: [...], warnings: [...], summary }`.
Raw `output` is dropped from the default response.
- New `verbose: bool` argument (default false) restores the raw
output for callers who actually need it.
- Updated the existing tool_run_check test to assert the new
contract (150 errors → 150 structured entries, response < 50KB).
Skipped run_tests in this pass — its parser would need to recognise
test-runner output (different format from cargo); will land separately.
Closes 886.
This commit is contained in:
@@ -10,6 +10,10 @@
|
||||
pub mod io;
|
||||
/// Pure command-safety checks, blocked-binary lists, and output truncation.
|
||||
pub mod path_guard;
|
||||
/// Cargo / rustc diagnostic parser — extracts structured errors and warnings
|
||||
/// from raw cargo output. Used by run_check / run_build / run_lint MCP tools
|
||||
/// (bug 886).
|
||||
pub mod parse_diagnostics;
|
||||
|
||||
#[allow(unused_imports)]
|
||||
pub use path_guard::{
|
||||
|
||||
@@ -0,0 +1,203 @@
|
||||
//! Parser for cargo compiler output (errors + warnings).
|
||||
//!
|
||||
//! Cargo emits diagnostics in a recognisable two-line shape:
|
||||
//!
|
||||
//! ```text
|
||||
//! error[E0061]: this function takes 4 arguments but 3 arguments were supplied
|
||||
//! --> server/src/agents/pool/auto_assign/merge.rs:86:29
|
||||
//! ```
|
||||
//!
|
||||
//! or, without a code:
|
||||
//!
|
||||
//! ```text
|
||||
//! warning: unused import: `Foo`
|
||||
//! --> server/src/lib.rs:3:5
|
||||
//! ```
|
||||
//!
|
||||
//! This module extracts those headers into structured [`Diagnostic`] entries so
|
||||
//! that MCP shell tools (`run_check`, `run_build`, `run_lint`) can return a
|
||||
//! compact error list instead of the full multi-hundred-KB cargo log.
|
||||
//! See bug 886.
|
||||
use regex::Regex;
|
||||
use serde::Serialize;
|
||||
use std::sync::OnceLock;
|
||||
|
||||
/// One compiler diagnostic extracted from cargo output.
|
||||
#[derive(Debug, Clone, PartialEq, Eq, Serialize)]
|
||||
pub struct Diagnostic {
|
||||
/// Either `"error"` or `"warning"`.
|
||||
pub kind: String,
|
||||
/// Rust diagnostic code if present, e.g. `"E0061"`. `None` for plain
|
||||
/// `error: ...` / `warning: ...` headers.
|
||||
pub code: Option<String>,
|
||||
/// The message text on the same line as the header.
|
||||
pub message: String,
|
||||
/// Source file path from the `-->` line, when present.
|
||||
pub file: Option<String>,
|
||||
/// 1-based line number from the `-->` line, when present.
|
||||
pub line: Option<u32>,
|
||||
}
|
||||
|
||||
/// Compact summary of a diagnostics scan.
|
||||
#[derive(Debug, Clone, PartialEq, Eq, Serialize)]
|
||||
pub struct DiagnosticSummary {
|
||||
pub error_count: usize,
|
||||
pub warning_count: usize,
|
||||
}
|
||||
|
||||
/// Parse cargo / rustc style output into structured diagnostics.
|
||||
///
|
||||
/// Recognises lines starting with `error[CODE]:`, `error:`, `warning[CODE]:`,
|
||||
/// `warning:`. If the next non-blank line starts with ` --> file:line(:col)?`
|
||||
/// the location is attached to the diagnostic. Lines that do not match either
|
||||
/// pattern are ignored, so the parser tolerates arbitrary surrounding output
|
||||
/// (Compiling/Checking progress, source snippets, etc.).
|
||||
pub fn parse_diagnostics(output: &str) -> Vec<Diagnostic> {
|
||||
static HEADER_RE: OnceLock<Regex> = OnceLock::new();
|
||||
static LOC_RE: OnceLock<Regex> = OnceLock::new();
|
||||
let header_re = HEADER_RE.get_or_init(|| {
|
||||
// Anchored to start-of-line: "error" or "warning", optional [CODE], colon, message.
|
||||
Regex::new(r"^(error|warning)(?:\[([A-Za-z0-9]+)\])?: (.+)$").unwrap()
|
||||
});
|
||||
let loc_re = LOC_RE
|
||||
.get_or_init(|| Regex::new(r"^\s*-->\s+([^:\s][^:]*):(\d+)(?::\d+)?\s*$").unwrap());
|
||||
|
||||
let lines: Vec<&str> = output.lines().collect();
|
||||
let mut diagnostics = Vec::new();
|
||||
|
||||
for (i, line) in lines.iter().enumerate() {
|
||||
let Some(caps) = header_re.captures(line) else {
|
||||
continue;
|
||||
};
|
||||
let kind = caps.get(1).unwrap().as_str().to_string();
|
||||
let code = caps.get(2).map(|m| m.as_str().to_string());
|
||||
let message = caps.get(3).unwrap().as_str().to_string();
|
||||
|
||||
// Look ahead a few lines for the `-->` location marker. Cargo
|
||||
// typically puts it 1 line after the header, but allow up to 3 in
|
||||
// case of formatting quirks.
|
||||
let mut file = None;
|
||||
let mut line_num = None;
|
||||
for look in 1..=3 {
|
||||
if let Some(next) = lines.get(i + look)
|
||||
&& let Some(loc) = loc_re.captures(next)
|
||||
{
|
||||
file = Some(loc.get(1).unwrap().as_str().to_string());
|
||||
line_num = loc.get(2).unwrap().as_str().parse::<u32>().ok();
|
||||
break;
|
||||
}
|
||||
}
|
||||
|
||||
diagnostics.push(Diagnostic {
|
||||
kind,
|
||||
code,
|
||||
message,
|
||||
file,
|
||||
line: line_num,
|
||||
});
|
||||
}
|
||||
|
||||
diagnostics
|
||||
}
|
||||
|
||||
/// Count errors vs warnings in a parsed diagnostics list.
|
||||
pub fn summarise(diagnostics: &[Diagnostic]) -> DiagnosticSummary {
|
||||
let error_count = diagnostics.iter().filter(|d| d.kind == "error").count();
|
||||
let warning_count = diagnostics.iter().filter(|d| d.kind == "warning").count();
|
||||
DiagnosticSummary {
|
||||
error_count,
|
||||
warning_count,
|
||||
}
|
||||
}
|
||||
|
||||
#[cfg(test)]
|
||||
mod tests {
|
||||
use super::*;
|
||||
|
||||
#[test]
|
||||
fn parses_error_with_code_and_location() {
|
||||
let input = "error[E0061]: this function takes 4 arguments but 3 arguments were supplied\n --> server/src/agents/pool/auto_assign/merge.rs:86:29\n |\n86 | foo();\n";
|
||||
let diags = parse_diagnostics(input);
|
||||
assert_eq!(diags.len(), 1);
|
||||
assert_eq!(diags[0].kind, "error");
|
||||
assert_eq!(diags[0].code.as_deref(), Some("E0061"));
|
||||
assert!(diags[0].message.contains("4 arguments"));
|
||||
assert_eq!(
|
||||
diags[0].file.as_deref(),
|
||||
Some("server/src/agents/pool/auto_assign/merge.rs")
|
||||
);
|
||||
assert_eq!(diags[0].line, Some(86));
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn parses_warning_without_code() {
|
||||
let input = "warning: unused import: `Foo`\n --> server/src/lib.rs:3:5\n";
|
||||
let diags = parse_diagnostics(input);
|
||||
assert_eq!(diags.len(), 1);
|
||||
assert_eq!(diags[0].kind, "warning");
|
||||
assert!(diags[0].code.is_none());
|
||||
assert_eq!(diags[0].message, "unused import: `Foo`");
|
||||
assert_eq!(diags[0].file.as_deref(), Some("server/src/lib.rs"));
|
||||
assert_eq!(diags[0].line, Some(3));
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn parses_header_without_location() {
|
||||
// Plain `error:` / `warning:` headers without a `-->` marker still
|
||||
// produce a diagnostic — just with file/line as None.
|
||||
let input = "error: could not compile `huskies` (bin \"huskies\") due to 3 previous errors";
|
||||
let diags = parse_diagnostics(input);
|
||||
assert_eq!(diags.len(), 1);
|
||||
assert_eq!(diags[0].kind, "error");
|
||||
assert!(diags[0].file.is_none());
|
||||
assert!(diags[0].line.is_none());
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn ignores_non_diagnostic_lines() {
|
||||
let input = " Checking huskies v0.10.4\n Compiling foo v0.1.0\n Finished dev profile\n";
|
||||
assert!(parse_diagnostics(input).is_empty());
|
||||
}
|
||||
|
||||
#[test]
|
||||
fn parses_three_compile_errors_from_real_cargo_output() {
|
||||
// Regression for bug 886: real-world output from 864's failing run.
|
||||
let input = " Checking huskies v0.10.4 (/workspace/server)\n\
|
||||
warning: unused imports: `ItemMetadata` and `OwnedItemMetadata`\n \
|
||||
--> server/src/db/mod.rs:25:5\n\
|
||||
|\n\
|
||||
25 | ItemMetadata, OwnedItemMetadata, delete_item, move_item_stage, next_item_number,\n \
|
||||
| ^^^^^^^^^^^^ ^^^^^^^^^^^^^^^^^\n\n\
|
||||
error[E0061]: this function takes 4 arguments but 3 arguments were supplied\n \
|
||||
--> server/src/agents/pool/auto_assign/merge.rs:86:29\n\
|
||||
|\n\
|
||||
86 | crate::db::write_item_with_content(story_id, \"4_merge\", &updated);\n\n\
|
||||
error[E0061]: this function takes 4 arguments but 3 arguments were supplied\n \
|
||||
--> server/src/agents/pool/pipeline/advance/helpers.rs:91:9\n\
|
||||
|\n\
|
||||
91 | crate::db::write_item_with_content(story_id, &stage, &updated);\n\n\
|
||||
error[E0061]: this function takes 4 arguments but 3 arguments were supplied\n \
|
||||
--> server/src/chat/commands/unblock.rs:103:13\n\
|
||||
|\n\
|
||||
103| crate::db::write_item_with_content(story_id, &stage, &updated);\n\n\
|
||||
For more information about this error, try `rustc --explain E0061`.\n\
|
||||
warning: `huskies` (bin \"huskies\") generated 1 warning\n\
|
||||
error: could not compile `huskies` (bin \"huskies\") due to 3 previous errors; 1 warning emitted\n";
|
||||
|
||||
let diags = parse_diagnostics(input);
|
||||
let summary = summarise(&diags);
|
||||
|
||||
// 3 compile errors + 1 unused-import warning + 1 generated-warning summary line + 1 final "could not compile" error
|
||||
assert_eq!(summary.error_count, 4);
|
||||
assert_eq!(summary.warning_count, 2);
|
||||
|
||||
// The 3 main compile errors must be present with file + line.
|
||||
let e0061: Vec<&Diagnostic> =
|
||||
diags.iter().filter(|d| d.code.as_deref() == Some("E0061")).collect();
|
||||
assert_eq!(e0061.len(), 3);
|
||||
let files: Vec<Option<&str>> = e0061.iter().map(|d| d.file.as_deref()).collect();
|
||||
assert!(files.contains(&Some("server/src/agents/pool/auto_assign/merge.rs")));
|
||||
assert!(files.contains(&Some("server/src/agents/pool/pipeline/advance/helpers.rs")));
|
||||
assert!(files.contains(&Some("server/src/chat/commands/unblock.rs")));
|
||||
}
|
||||
}
|
||||
Reference in New Issue
Block a user