File size: 11,550 Bytes
bc575bc
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
import { readFile as fsReadFile, stat as fsStat } from "node:fs/promises";
import { createInterface } from "node:readline";
import type { AgentTool } from "@earendil-works/pi-agent-core";
import { spawn } from "child_process";
import path from "path";
import { type Static, Type } from "typebox";
import { ensureTool } from "../../utils/tools-manager.ts";
import type { ExtensionContext, ToolDefinition } from "../extensions/types.ts";
import { resolveToCwd } from "./path-utils.ts";
import { grepRenderers } from "./renderers/grep.ts";
import { wrapToolDefinition } from "./tool-definition-wrapper.ts";
import {
	DEFAULT_MAX_BYTES,
	formatSize,
	GREP_MAX_LINE_LENGTH,
	type TruncationResult,
	truncateHead,
	truncateLine,
} from "./truncate.ts";

const grepSchema = Type.Object({
	pattern: Type.String({ description: "Search pattern (regex or literal string)" }),
	path: Type.Optional(Type.String({ description: "Directory or file to search (default: current directory)" })),
	glob: Type.Optional(Type.String({ description: "Filter files by glob pattern, e.g. '*.ts' or '**/*.spec.ts'" })),
	ignoreCase: Type.Optional(Type.Boolean({ description: "Case-insensitive search (default: false)" })),
	literal: Type.Optional(
		Type.Boolean({ description: "Treat pattern as literal string instead of regex (default: false)" }),
	),
	context: Type.Optional(
		Type.Number({ description: "Number of lines to show before and after each match (default: 0)" }),
	),
	limit: Type.Optional(Type.Number({ description: "Maximum number of matches to return (default: 100)" })),
});

export const grepToolSystemPromptContribution = {
	snippet: "Search file contents for patterns (respects .gitignore)",
	guidelines: [],
} as const;

export type GrepToolInput = Static<typeof grepSchema>;
const DEFAULT_LIMIT = 100;

export interface GrepToolDetails {
	truncation?: TruncationResult;
	matchLimitReached?: number;
	linesTruncated?: boolean;
}

/**
 * Pluggable operations for the grep tool.
 * Override these to delegate search to remote systems (for example SSH).
 */
export interface GrepOperations {
	/** Check if path is a directory. Throws if path does not exist. */
	isDirectory: (absolutePath: string) => Promise<boolean> | boolean;
	/** Read file contents for context lines */
	readFile: (absolutePath: string) => Promise<string> | string;
}

const defaultGrepOperations: GrepOperations = {
	isDirectory: async (p) => (await fsStat(p)).isDirectory(),
	readFile: (p) => fsReadFile(p, "utf-8"),
};

export interface GrepToolOptions {
	/** Custom operations for grep. Default: local filesystem plus ripgrep */
	operations?: GrepOperations;
}

export function createGrepToolDefinition(
	cwd: string,
	options?: GrepToolOptions,
): ToolDefinition<typeof grepSchema, GrepToolDetails | undefined> {
	const customOps = options?.operations;
	return {
		name: "grep",
		label: "grep",
		description: `Search file contents for a pattern. Returns matching lines with file paths and line numbers. Respects .gitignore. Output is truncated to ${DEFAULT_LIMIT} matches or ${DEFAULT_MAX_BYTES / 1024}KB (whichever is hit first). Long lines are truncated to ${GREP_MAX_LINE_LENGTH} chars.`,
		promptSnippet: grepToolSystemPromptContribution.snippet,
		parameters: grepSchema,
		async execute(
			_toolCallId,
			{
				pattern,
				path: searchDir,
				glob,
				ignoreCase,
				literal,
				context,
				limit,
			}: {
				pattern: string;
				path?: string;
				glob?: string;
				ignoreCase?: boolean;
				literal?: boolean;
				context?: number;
				limit?: number;
			},
			signal?: AbortSignal,
			_onUpdate?,
			ctx?: ExtensionContext,
		) {
			return new Promise((resolve, reject) => {
				if (signal?.aborted) {
					reject(new Error("Operation aborted"));
					return;
				}
				let settled = false;
				const settle = (fn: () => void) => {
					if (!settled) {
						settled = true;
						fn();
					}
				};

				(async () => {
					try {
						const rgPath = await ensureTool("rg");
						if (!rgPath) {
							settle(() => reject(new Error("ripgrep (rg) is not available and could not be downloaded")));
							return;
						}

						const searchPath = resolveToCwd(searchDir || ".", ctx?.cwd || cwd);
						const ops = customOps ?? defaultGrepOperations;
						let isDirectory: boolean;
						try {
							isDirectory = await ops.isDirectory(searchPath);
						} catch {
							settle(() => reject(new Error(`Path not found: ${searchPath}`)));
							return;
						}

						const contextValue = context && context > 0 ? context : 0;
						const effectiveLimit = Math.max(1, limit ?? DEFAULT_LIMIT);
						const formatPath = (filePath: string): string => {
							if (isDirectory) {
								const relative = path.relative(searchPath, filePath);
								if (relative && !relative.startsWith("..")) {
									return relative.replace(/\\/g, "/");
								}
							}
							return path.basename(filePath);
						};

						const fileCache = new Map<string, string[]>();
						const getFileLines = async (filePath: string): Promise<string[]> => {
							let lines = fileCache.get(filePath);
							if (!lines) {
								try {
									const content = await ops.readFile(filePath);
									lines = content.replace(/\r\n/g, "\n").replace(/\r/g, "\n").split("\n");
								} catch {
									lines = [];
								}
								fileCache.set(filePath, lines);
							}
							return lines;
						};

						const args: string[] = ["--json", "--line-number", "--color=never", "--hidden"];
						if (ignoreCase) args.push("--ignore-case");
						if (literal) args.push("--fixed-strings");
						if (glob) args.push("--glob", glob);
						args.push("--", pattern, searchPath);

						const child = spawn(rgPath, args, { stdio: ["ignore", "pipe", "pipe"] });
						const rl = createInterface({ input: child.stdout });
						let stderr = "";
						let matchCount = 0;
						let matchLimitReached = false;
						let linesTruncated = false;
						let aborted = false;
						let killedDueToLimit = false;
						const outputLines: string[] = [];

						const cleanup = () => {
							rl.close();
							signal?.removeEventListener("abort", onAbort);
						};
						const stopChild = (dueToLimit = false) => {
							if (!child.killed) {
								killedDueToLimit = dueToLimit;
								child.kill();
							}
						};
						const onAbort = () => {
							aborted = true;
							stopChild();
						};
						signal?.addEventListener("abort", onAbort, { once: true });
						child.stderr?.on("data", (chunk) => {
							stderr += chunk.toString();
						});

						const formatBlock = async (filePath: string, lineNumber: number): Promise<string[]> => {
							const relativePath = formatPath(filePath);
							const lines = await getFileLines(filePath);
							if (!lines.length) return [`${relativePath}:${lineNumber}: (unable to read file)`];
							const block: string[] = [];
							const start = contextValue > 0 ? Math.max(1, lineNumber - contextValue) : lineNumber;
							const end = contextValue > 0 ? Math.min(lines.length, lineNumber + contextValue) : lineNumber;
							for (let current = start; current <= end; current++) {
								const lineText = lines[current - 1] ?? "";
								const sanitized = lineText.replace(/\r/g, "");
								const isMatchLine = current === lineNumber;
								// Truncate long lines so grep output stays compact.
								const { text: truncatedText, wasTruncated } = truncateLine(sanitized);
								if (wasTruncated) linesTruncated = true;
								if (isMatchLine) block.push(`${relativePath}:${current}: ${truncatedText}`);
								else block.push(`${relativePath}-${current}- ${truncatedText}`);
							}
							return block;
						};

						// Collect matches during streaming, then format them after rg exits.
						const matches: Array<{ filePath: string; lineNumber: number; lineText?: string }> = [];
						rl.on("line", (line) => {
							if (!line.trim() || matchCount >= effectiveLimit) return;
							let event: any;
							try {
								event = JSON.parse(line);
							} catch {
								return;
							}
							if (event.type === "match") {
								matchCount++;
								const filePath = event.data?.path?.text;
								const lineNumber = event.data?.line_number;
								const lineText = event.data?.lines?.text;
								if (filePath && typeof lineNumber === "number")
									matches.push({ filePath, lineNumber, lineText });
								if (matchCount >= effectiveLimit) {
									matchLimitReached = true;
									stopChild(true);
								}
							}
						});

						child.on("error", (error) => {
							cleanup();
							settle(() => reject(new Error(`Failed to run ripgrep: ${error.message}`)));
						});
						child.on("close", async (code) => {
							cleanup();
							if (aborted) {
								settle(() => reject(new Error("Operation aborted")));
								return;
							}
							if (!killedDueToLimit && code !== 0 && code !== 1) {
								const errorMsg = stderr.trim() || `ripgrep exited with code ${code}`;
								settle(() => reject(new Error(errorMsg)));
								return;
							}
							if (matchCount === 0) {
								settle(() =>
									resolve({ content: [{ type: "text", text: "No matches found" }], details: undefined }),
								);
								return;
							}

							// Format matches after streaming finishes so custom readFile() backends can be async.
							for (const match of matches) {
								if (contextValue === 0 && match.lineText !== undefined) {
									const relativePath = formatPath(match.filePath);
									const sanitized = match.lineText
										.replace(/\r\n/g, "\n")
										.replace(/\r/g, "")
										.replace(/\n$/, "");
									const { text: truncatedText, wasTruncated } = truncateLine(sanitized);
									if (wasTruncated) linesTruncated = true;
									outputLines.push(`${relativePath}:${match.lineNumber}: ${truncatedText}`);
								} else {
									const block = await formatBlock(match.filePath, match.lineNumber);
									outputLines.push(...block);
								}
							}

							const rawOutput = outputLines.join("\n");
							// Apply byte truncation. There is no line limit here because the match limit already capped rows.
							const truncation = truncateHead(rawOutput, { maxLines: Number.MAX_SAFE_INTEGER });
							let output = truncation.content;
							const details: GrepToolDetails = {};
							// Build actionable notices for truncation and match limits.
							const notices: string[] = [];
							if (matchLimitReached) {
								notices.push(
									`${effectiveLimit} matches limit reached. Use limit=${effectiveLimit * 2} for more, or refine pattern`,
								);
								details.matchLimitReached = effectiveLimit;
							}
							if (truncation.truncated) {
								notices.push(`${formatSize(DEFAULT_MAX_BYTES)} limit reached`);
								details.truncation = truncation;
							}
							if (linesTruncated) {
								notices.push(
									`Some lines truncated to ${GREP_MAX_LINE_LENGTH} chars. Use read tool to see full lines`,
								);
								details.linesTruncated = true;
							}
							if (notices.length > 0) output += `\n\n[${notices.join(". ")}]`;
							settle(() =>
								resolve({
									content: [{ type: "text", text: output }],
									details: Object.keys(details).length > 0 ? details : undefined,
								}),
							);
						});
					} catch (err) {
						settle(() => reject(err as Error));
					}
				})();
			});
		},
		...grepRenderers,
	};
}

export function createGrepTool(cwd: string, options?: GrepToolOptions): AgentTool<typeof grepSchema> {
	return wrapToolDefinition(createGrepToolDefinition(cwd, options));
}