refactor(file): add ripgrep search service (#22295)

This commit is contained in:
Kit Langton
2026-04-13 10:04:32 -04:00
committed by GitHub
parent 321bf1f8e1
commit 6fdb8ab90d
12 changed files with 665 additions and 348 deletions

View File

@@ -1,4 +1,5 @@
import { EOL } from "os"
import { AppRuntime } from "../../../effect/app-runtime"
import { Ripgrep } from "../../../file/ripgrep"
import { Instance } from "../../../project/instance"
import { bootstrap } from "../../bootstrap"
@@ -76,12 +77,18 @@ const SearchCommand = cmd({
description: "Limit number of results",
}),
async handler(args) {
const results = await Ripgrep.search({
cwd: process.cwd(),
pattern: args.pattern,
glob: args.glob as string[] | undefined,
limit: args.limit,
await bootstrap(process.cwd(), async () => {
const results = await AppRuntime.runPromise(
Ripgrep.Service.use((svc) =>
svc.search({
cwd: Instance.directory,
pattern: args.pattern,
glob: args.glob as string[] | undefined,
limit: args.limit,
}),
),
)
process.stdout.write(JSON.stringify(results.items, null, 2) + EOL)
})
process.stdout.write(JSON.stringify(results, null, 2) + EOL)
},
})

View File

@@ -96,6 +96,7 @@ export namespace Ripgrep {
export type Result = z.infer<typeof Result>
export type Match = z.infer<typeof Match>
export type Item = Match["data"]
export type Begin = z.infer<typeof Begin>
export type End = z.infer<typeof End>
export type Summary = z.infer<typeof Summary>
@@ -289,6 +290,13 @@ export namespace Ripgrep {
follow?: boolean
maxDepth?: number
}) => Stream.Stream<string, PlatformError>
readonly search: (input: {
cwd: string
pattern: string
glob?: string[]
limit?: number
follow?: boolean
}) => Effect.Effect<{ items: Item[]; partial: boolean }, PlatformError | Error>
}
export class Service extends Context.Service<Service, Interface>()("@opencode/Ripgrep") {}
@@ -298,6 +306,32 @@ export namespace Ripgrep {
Effect.gen(function* () {
const spawner = yield* ChildProcessSpawner
const afs = yield* AppFileSystem.Service
const bin = Effect.fn("Ripgrep.path")(function* () {
return yield* Effect.promise(() => filepath())
})
const args = Effect.fn("Ripgrep.args")(function* (input: {
mode: "files" | "search"
glob?: string[]
hidden?: boolean
follow?: boolean
maxDepth?: number
limit?: number
pattern?: string
}) {
const out = [yield* bin(), input.mode === "search" ? "--json" : "--files", "--glob=!.git/*"]
if (input.follow) out.push("--follow")
if (input.hidden !== false) out.push("--hidden")
if (input.maxDepth !== undefined) out.push(`--max-depth=${input.maxDepth}`)
if (input.glob) {
for (const g of input.glob) {
out.push(`--glob=${g}`)
}
}
if (input.limit) out.push(`--max-count=${input.limit}`)
if (input.mode === "search") out.push("--no-messages")
if (input.pattern) out.push("--", input.pattern)
return out
})
const files = Effect.fn("Ripgrep.files")(function* (input: {
cwd: string
@@ -306,7 +340,7 @@ export namespace Ripgrep {
follow?: boolean
maxDepth?: number
}) {
const rgPath = yield* Effect.promise(() => filepath())
const rgPath = yield* bin()
const isDir = yield* afs.isDir(input.cwd)
if (!isDir) {
return yield* Effect.die(
@@ -318,23 +352,76 @@ export namespace Ripgrep {
)
}
const args = [rgPath, "--files", "--glob=!.git/*"]
if (input.follow) args.push("--follow")
if (input.hidden !== false) args.push("--hidden")
if (input.maxDepth !== undefined) args.push(`--max-depth=${input.maxDepth}`)
if (input.glob) {
for (const g of input.glob) {
args.push(`--glob=${g}`)
}
}
const cmd = yield* args({
mode: "files",
glob: input.glob,
hidden: input.hidden,
follow: input.follow,
maxDepth: input.maxDepth,
})
return spawner
.streamLines(ChildProcess.make(args[0], args.slice(1), { cwd: input.cwd }))
.streamLines(ChildProcess.make(cmd[0], cmd.slice(1), { cwd: input.cwd }))
.pipe(Stream.filter((line: string) => line.length > 0))
})
const search = Effect.fn("Ripgrep.search")(function* (input: {
cwd: string
pattern: string
glob?: string[]
limit?: number
follow?: boolean
}) {
return yield* Effect.scoped(
Effect.gen(function* () {
const cmd = yield* args({
mode: "search",
glob: input.glob,
follow: input.follow,
limit: input.limit,
pattern: input.pattern,
})
const handle = yield* spawner.spawn(
ChildProcess.make(cmd[0], cmd.slice(1), {
cwd: input.cwd,
stdin: "ignore",
}),
)
const [stdout, stderr, code] = yield* Effect.all(
[
Stream.mkString(Stream.decodeText(handle.stdout)),
Stream.mkString(Stream.decodeText(handle.stderr)),
handle.exitCode,
],
{ concurrency: "unbounded" },
)
if (code !== 0 && code !== 1 && code !== 2) {
return yield* Effect.fail(new Error(`ripgrep failed: ${stderr}`))
}
const items = stdout
.trim()
.split(/\r?\n/)
.filter(Boolean)
.map((line) => JSON.parse(line))
.map((parsed) => Result.parse(parsed))
.filter((row): row is Match => row.type === "match")
.map((row) => row.data)
return {
items,
partial: code === 2,
}
}),
)
})
return Service.of({
files: (input) => Stream.unwrap(files(input)),
search,
})
}),
)
@@ -401,46 +488,4 @@ export namespace Ripgrep {
return lines.join("\n")
}
export async function search(input: {
cwd: string
pattern: string
glob?: string[]
limit?: number
follow?: boolean
}) {
const args = [`${await filepath()}`, "--json", "--hidden", "--glob=!.git/*"]
if (input.follow) args.push("--follow")
if (input.glob) {
for (const g of input.glob) {
args.push(`--glob=${g}`)
}
}
if (input.limit) {
args.push(`--max-count=${input.limit}`)
}
args.push("--")
args.push(input.pattern)
const result = await Process.text(args, {
cwd: input.cwd,
nothrow: true,
})
if (result.code !== 0) {
return []
}
// Handle both Unix (\n) and Windows (\r\n) line endings
const lines = result.text.trim().split(/\r?\n/).filter(Boolean)
// Parse JSON lines from ripgrep output
return lines
.map((line) => JSON.parse(line))
.map((parsed) => Result.parse(parsed))
.filter((r) => r.type === "match")
.map((r) => r.data)
}
}

View File

@@ -1,6 +1,7 @@
import { Hono } from "hono"
import { describeRoute, validator, resolver } from "hono-openapi"
import z from "zod"
import { AppRuntime } from "../../effect/app-runtime"
import { File } from "../../file"
import { Ripgrep } from "../../file/ripgrep"
import { LSP } from "../../lsp"
@@ -34,12 +35,10 @@ export const FileRoutes = lazy(() =>
),
async (c) => {
const pattern = c.req.valid("query").pattern
const result = await Ripgrep.search({
cwd: Instance.directory,
pattern,
limit: 10,
})
return c.json(result)
const result = await AppRuntime.runPromise(
Ripgrep.Service.use((svc) => svc.search({ cwd: Instance.directory, pattern, limit: 10 })),
)
return c.json(result.items)
},
)
.get(

View File

@@ -1,11 +1,8 @@
import z from "zod"
import { Effect } from "effect"
import * as Stream from "effect/Stream"
import { Effect, Option } from "effect"
import { Tool } from "./tool"
import { Filesystem } from "../util/filesystem"
import { Ripgrep } from "../file/ripgrep"
import { ChildProcess } from "effect/unstable/process"
import { ChildProcessSpawner } from "effect/unstable/process/ChildProcessSpawner"
import { AppFileSystem } from "../filesystem"
import DESCRIPTION from "./grep.txt"
import { Instance } from "../project/instance"
@@ -17,7 +14,8 @@ const MAX_LINE_LENGTH = 2000
export const GrepTool = Tool.define(
"grep",
Effect.gen(function* () {
const spawner = yield* ChildProcessSpawner
const fs = yield* AppFileSystem.Service
const rg = yield* Ripgrep.Service
return {
description: DESCRIPTION,
@@ -28,6 +26,11 @@ export const GrepTool = Tool.define(
}),
execute: (params: { pattern: string; path?: string; include?: string }, ctx: Tool.Context) =>
Effect.gen(function* () {
const empty = {
title: params.pattern,
metadata: { matches: 0, truncated: false },
output: "No files found",
}
if (!params.pattern) {
throw new Error("pattern is required")
}
@@ -43,92 +46,58 @@ export const GrepTool = Tool.define(
},
})
let searchPath = params.path ?? Instance.directory
searchPath = path.isAbsolute(searchPath) ? searchPath : path.resolve(Instance.directory, searchPath)
const searchPath = AppFileSystem.resolve(
path.isAbsolute(params.path ?? Instance.directory)
? (params.path ?? Instance.directory)
: path.join(Instance.directory, params.path ?? "."),
)
yield* assertExternalDirectoryEffect(ctx, searchPath, { kind: "directory" })
const rgPath = yield* Effect.promise(() => Ripgrep.filepath())
const args = ["-nH", "--hidden", "--no-messages", "--field-match-separator=|", "--regexp", params.pattern]
if (params.include) {
args.push("--glob", params.include)
}
args.push(searchPath)
const result = yield* rg.search({
cwd: searchPath,
pattern: params.pattern,
glob: params.include ? [params.include] : undefined,
})
const result = yield* Effect.scoped(
Effect.gen(function* () {
const handle = yield* spawner.spawn(
ChildProcess.make(rgPath, args, {
stdin: "ignore",
}),
)
if (result.items.length === 0) return empty
const [output, errorOutput] = yield* Effect.all(
[Stream.mkString(Stream.decodeText(handle.stdout)), Stream.mkString(Stream.decodeText(handle.stderr))],
{ concurrency: 2 },
)
const exitCode = yield* handle.exitCode
return { output, errorOutput, exitCode }
}),
const rows = result.items.map((item) => ({
path: AppFileSystem.resolve(
path.isAbsolute(item.path.text) ? item.path.text : path.join(searchPath, item.path.text),
),
line: item.line_number,
text: item.lines.text,
}))
const times = new Map(
(yield* Effect.forEach(
[...new Set(rows.map((row) => row.path))],
Effect.fnUntraced(function* (file) {
const info = yield* fs.stat(file).pipe(Effect.catch(() => Effect.succeed(undefined)))
if (!info || info.type === "Directory") return undefined
return [
file,
info.mtime.pipe(
Option.map((time) => time.getTime()),
Option.getOrElse(() => 0),
) ?? 0,
] as const
}),
{ concurrency: 16 },
)).filter((entry): entry is readonly [string, number] => Boolean(entry)),
)
const matches = rows.flatMap((row) => {
const mtime = times.get(row.path)
if (mtime === undefined) return []
return [{ ...row, mtime }]
})
const { output, errorOutput, exitCode } = result
// Exit codes: 0 = matches found, 1 = no matches, 2 = errors (but may still have matches)
// With --no-messages, we suppress error output but still get exit code 2 for broken symlinks etc.
// Only fail if exit code is 2 AND no output was produced
if (exitCode === 1 || (exitCode === 2 && !output.trim())) {
return {
title: params.pattern,
metadata: { matches: 0, truncated: false },
output: "No files found",
}
}
if (exitCode !== 0 && exitCode !== 2) {
throw new Error(`ripgrep failed: ${errorOutput}`)
}
const hasErrors = exitCode === 2
// Handle both Unix (\n) and Windows (\r\n) line endings
const lines = output.trim().split(/\r?\n/)
const matches = []
for (const line of lines) {
if (!line) continue
const [filePath, lineNumStr, ...lineTextParts] = line.split("|")
if (!filePath || !lineNumStr || lineTextParts.length === 0) continue
const lineNum = parseInt(lineNumStr, 10)
const lineText = lineTextParts.join("|")
const stats = Filesystem.stat(filePath)
if (!stats) continue
matches.push({
path: filePath,
modTime: stats.mtime.getTime(),
lineNum,
lineText,
})
}
matches.sort((a, b) => b.modTime - a.modTime)
matches.sort((a, b) => b.mtime - a.mtime)
const limit = 100
const truncated = matches.length > limit
const finalMatches = truncated ? matches.slice(0, limit) : matches
if (finalMatches.length === 0) {
return {
title: params.pattern,
metadata: { matches: 0, truncated: false },
output: "No files found",
}
}
if (finalMatches.length === 0) return empty
const totalMatches = matches.length
const outputLines = [`Found ${totalMatches} matches${truncated ? ` (showing first ${limit})` : ""}`]
@@ -143,10 +112,8 @@ export const GrepTool = Tool.define(
outputLines.push(`${match.path}:`)
}
const truncatedLineText =
match.lineText.length > MAX_LINE_LENGTH
? match.lineText.substring(0, MAX_LINE_LENGTH) + "..."
: match.lineText
outputLines.push(` Line ${match.lineNum}: ${truncatedLineText}`)
match.text.length > MAX_LINE_LENGTH ? match.text.substring(0, MAX_LINE_LENGTH) + "..." : match.text
outputLines.push(` Line ${match.line}: ${truncatedLineText}`)
}
if (truncated) {
@@ -156,7 +123,7 @@ export const GrepTool = Tool.define(
)
}
if (hasErrors) {
if (result.partial) {
outputLines.push("")
outputLines.push("(Some paths were inaccessible and skipped)")
}