Class: Clacky::Tools::Grep

Inherits:
Base
  • Object
show all
Defined in:
lib/clacky/tools/grep.rb

Constant Summary collapse

MAX_FILE_SIZE =

Maximum file size to search (1MB)

1_048_576
MAX_LINE_LENGTH =

Maximum line length to display (to avoid huge outputs)

500

Instance Method Summary collapse

Methods inherited from Base

#category, #description, #name, #parameters, #to_function_definition

Instance Method Details

#execute(pattern:, path: ".", file_pattern: "**/*", case_insensitive: false, context_lines: 0, max_files: 50, max_matches_per_file: 50, max_total_matches: 200, max_file_size: MAX_FILE_SIZE, max_files_to_search: 10000, working_dir: nil) ⇒ Object 



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
 
# File 'lib/clacky/tools/grep.rb', line 66

def execute(
  pattern:,
  path: ".",
  file_pattern: "**/*",
  case_insensitive: false,
  context_lines: 0,
  max_files: 50,
  max_matches_per_file: 50,
  max_total_matches: 200,
  max_file_size: MAX_FILE_SIZE,
  max_files_to_search: 10000,
  working_dir: nil
)
  # Validate pattern
  if pattern.nil? || pattern.strip.empty?
    return { error: "Pattern cannot be empty" }
  end

  # Validate and expand path relative to working_dir when provided
  begin
    expanded_path = expand_path(path, working_dir: working_dir)
  rescue StandardError => e
    return { error: "Invalid path: #{e.message}" }
  end

  unless File.exist?(expanded_path)
    return { error: "Path does not exist: #{path}" }
  end

  if File.directory?(expanded_path) && Clacky::Utils::FileIgnoreHelper.dangerous_root?(expanded_path)
    return {
      error: "Refusing to recursively grep from broad path '#{path}'. " \
             "Narrow the path to a specific subdirectory, " \
             "or use '.' to search the working directory."
    }
  end

  # Limit context_lines
  context_lines = [[context_lines, 0].max, 10].min

  begin
    # Compile regex
    regex_options = case_insensitive ? Regexp::IGNORECASE : 0
    regex = Regexp.new(pattern, regex_options)

    results = []
    total_matches = 0
    files_searched = 0
    skipped = {
      binary: 0,
      too_large: 0,
      ignored: 0
    }
    truncation_reason = nil

    files = if File.file?(expanded_path)
              [expanded_path]
            else
              # Auto-expand bare extension patterns (e.g. "*.ts") to recursive
              # (e.g. "**/*.ts"), matching ripgrep's default behaviour.
              # Patterns already containing "/" are left as-is — the user is
              # intentionally scoping to a specific directory.
              effective_pattern = file_pattern.include?("/") ? file_pattern : "**/#{file_pattern}"
              fnmatch_flags = File::FNM_PATHNAME | File::FNM_DOTMATCH
              collected = []
              walk_status = {}
              Clacky::Utils::FileIgnoreHelper.walk_files(expanded_path, skipped: skipped, status: walk_status) do |f|
                relative = f[(expanded_path.length + 1)..]
                collected << f if File.fnmatch(effective_pattern, relative, fnmatch_flags)
              end
              if walk_status[:truncated]
                truncation_reason ||= "walk #{walk_status[:truncation_reason]}"
              end
              collected
            end

    files.each do |file|
      if files_searched >= max_files_to_search
        truncation_reason ||= "max_files_to_search limit reached"
        break
      end

      # Skip binary files
      if Clacky::Utils::FileProcessor.binary_file_path?(file)
        skipped[:binary] += 1
        next
      end

      # Skip files that are too large
      if File.size(file) > max_file_size
        skipped[:too_large] += 1
        next
      end

      files_searched += 1

      # Check if we've found enough matching files
      if results.length >= max_files
        truncation_reason ||= "max_files limit reached"
        break
      end

      # Check if we've found enough total matches
      if total_matches >= max_total_matches
        truncation_reason ||= "max_total_matches limit reached"
        break
      end

      # Search the file
      matches = search_file(file, regex, context_lines, max_matches_per_file)
      next if matches.empty?

      # Add remaining matches respecting max_total_matches
      remaining_matches = max_total_matches - total_matches
      matches = matches.take(remaining_matches) if remaining_matches < matches.length

      results << {
        file: File.expand_path(file),
        matches: matches
      }
      total_matches += matches.length
    end

    {
      results: results,
      total_matches: total_matches,
      files_searched: files_searched,
      files_with_matches: results.length,
      skipped_files: skipped,
      truncated: !truncation_reason.nil?,
      truncation_reason: truncation_reason,
      error: nil
    }
  rescue RegexpError => e
    { error: "Invalid regex pattern: #{e.message}" }
  rescue StandardError => e
    { error: "Failed to search files: #{e.message}" }
  end
end

#format_call(args) ⇒ Object 



206
207
208
209
210
211
212
213
214
215
 
# File 'lib/clacky/tools/grep.rb', line 206

def format_call(args)
  pattern = args[:pattern] || args['pattern'] || ''
  path = args[:path] || args['path'] || '.'

  # Truncate pattern if too long
  display_pattern = pattern.length > 30 ? "#{pattern[0..27]}..." : pattern
  display_path = path == '.' ? 'current dir' : (path.length > 20 ? "...#{path[-17..]}" : path)

  "grep(\"#{display_pattern}\" in #{display_path})"
end

#format_result(result) ⇒ Object 



217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
 
# File 'lib/clacky/tools/grep.rb', line 217

def format_result(result)
  if result[:error]
    "[Error] #{result[:error]}"
  else
    matches = result[:total_matches] || 0
    files = result[:files_with_matches] || 0
    msg = "[OK] Found #{matches} matches in #{files} files"

    # Add truncation info if present
    if result[:truncated] && result[:truncation_reason]
      msg += " (truncated: #{result[:truncation_reason]})"
    end

    msg
  end
end

#format_result_for_llm(result) ⇒ Object

Format result for LLM consumption - return a compact version to save tokens



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
 
# File 'lib/clacky/tools/grep.rb', line 235

def format_result_for_llm(result)
  # If there's an error, return it as-is
  return result if result[:error]

  # Build a compact summary with file list and sample matches
  compact = {
    summary: {
      total_matches: result[:total_matches],
      files_with_matches: result[:files_with_matches],
      files_searched: result[:files_searched],
      truncated: result[:truncated],
      truncation_reason: result[:truncation_reason]
    }
  }

  # Include list of files with match counts
  if result[:results] && !result[:results].empty?
    compact[:files] = result[:results].map do |file_result|
      {
        file: file_result[:file],
        match_count: file_result[:matches].length
      }
    end

    # Include sample matches (first 2 matches from first 3 files) for context
    sample_results = result[:results].take(3)
    compact[:sample_matches] = sample_results.map do |file_result|
      {
        file: file_result[:file],
        matches: file_result[:matches].take(2).map do |match|
          {
            line_number: match[:line_number],
            line: match[:line]
            # Omit context to save space - it's rarely needed by LLM
          }
        end
      }
    end
  end

  compact
end

#get_line_context(file, match_index, context_lines) ⇒ Object

Get context lines around a match



314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
 
# File 'lib/clacky/tools/grep.rb', line 314

def get_line_context(file, match_index, context_lines)
  lines = File.readlines(file, chomp: true).map! { |l| safe_utf8(l) }
  start_line = [0, match_index - context_lines].max
  end_line = [lines.length - 1, match_index + context_lines].min

  context = []
  (start_line..end_line).each do |i|
    line_content = lines[i]
    # Truncate long lines in context too
    display_content = line_content.length > MAX_LINE_LENGTH ?
                    "#{line_content[0...MAX_LINE_LENGTH]}..." :
                    line_content

    context << {
      line_number: i + 1,
      content: display_content,
      is_match: i == match_index
    }
  end

  context
rescue StandardError
  nil
end

#search_file(file, regex, context_lines, max_matches) ⇒ Object 



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
 
# File 'lib/clacky/tools/grep.rb', line 279

def search_file(file, regex, context_lines, max_matches)
  matches = []

  # Use File.foreach for memory-efficient line-by-line reading.
  # Scrub invalid UTF-8 bytes so results survive JSON encoding.
  File.foreach(file, chomp: true).with_index do |raw_line, index|
    line = safe_utf8(raw_line)
    # Stop if we have enough matches for this file
    break if matches.length >= max_matches

    next unless line.match?(regex)

    # Truncate long lines
    display_line = line.length > MAX_LINE_LENGTH ? "#{line[0...MAX_LINE_LENGTH]}..." : line

    # Get context if requested
    if context_lines > 0
      context = get_line_context(file, index, context_lines)
    else
      context = nil
    end

    matches << {
      line_number: index + 1,
      line: display_line,
      context: context
    }
  end

  matches
rescue StandardError
  []
end