octofs 0.4.3

Standalone MCP filesystem tools server — view, edit, shell, workdir
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

// Copyright 2026 Muvon Un Limited
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
//     http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.

// Shared truncation utilities for smart content display across MCP tools

/// Rough token estimate: ~4 chars per token (good enough for truncation decisions).
pub fn estimate_tokens(content: &str) -> usize {
	content.len().div_ceil(4)
}

/// Truncate content to approximately `max_tokens` tokens, preserving whole lines.
pub fn truncate_to_tokens(content: &str, max_tokens: usize) -> String {
	let max_chars = max_tokens * 4;
	if content.len() <= max_chars {
		return content.to_string();
	}
	// Find a clean line boundary near the limit
	let truncated = &content[..max_chars.min(content.len())];
	if let Some(last_newline) = truncated.rfind('\n') {
		content[..last_newline].to_string()
	} else {
		truncated.to_string()
	}
}

/// Format content with line identifiers (numbers or hashes) and smart elision for display.
pub fn format_content_with_line_numbers(
	lines: &[&str],
	start_line_number: usize,
	view_range: Option<(usize, i64)>,
) -> String {
	// Compute prefixes based on active line mode
	let prefixes: Vec<String> = if super::line_hash::is_hash_mode() {
		super::line_hash::compute_line_hashes(lines)
	} else {
		lines
			.iter()
			.enumerate()
			.map(|(i, _)| format!("{}", start_line_number + i))
			.collect()
	};

	if let Some((start, end)) = view_range {
		let start_idx = if start == 0 {
			0
		} else {
			start.saturating_sub(1)
		};
		let end_idx = if end == -1 {
			lines.len()
		} else {
			(end as usize).min(lines.len())
		};

		if start_idx >= lines.len() || start_idx > end_idx {
			return if start_idx >= lines.len() {
				format!(
					"Start line {} exceeds content length ({} lines)",
					start,
					lines.len()
				)
			} else {
				format!(
					"Start line {} must be less than or equal to end line {}",
					start, end
				)
			};
		}

		let mut result_lines = Vec::new();

		if start_idx > 3 {
			for (i, line) in lines.iter().enumerate().take(2) {
				result_lines.push(format!("{}:{}", prefixes[i], line));
			}
			if start_idx > 5 {
				result_lines.push(format!("[...{} lines more]", start_idx - 2));
			} else {
				for (i, line) in lines.iter().enumerate().take(start_idx).skip(2) {
					result_lines.push(format!("{}:{}", prefixes[i], line));
				}
			}
		} else {
			for (i, line) in lines.iter().enumerate().take(start_idx) {
				result_lines.push(format!("{}:{}", prefixes[i], line));
			}
		}

		for (i, line) in lines.iter().enumerate().take(end_idx).skip(start_idx) {
			result_lines.push(format!("{}:{}", prefixes[i], line));
		}

		let remaining_lines = lines.len() - end_idx;
		if remaining_lines > 3 {
			if remaining_lines > 5 {
				result_lines.push(format!("[...{} lines more]", remaining_lines - 2));
				for (i, line) in lines.iter().enumerate().skip(lines.len() - 2) {
					result_lines.push(format!("{}:{}", prefixes[i], line));
				}
			} else {
				for (i, line) in lines.iter().enumerate().skip(end_idx) {
					result_lines.push(format!("{}:{}", prefixes[i], line));
				}
			}
		} else {
			for (i, line) in lines.iter().enumerate().skip(end_idx) {
				result_lines.push(format!("{}:{}", prefixes[i], line));
			}
		}

		result_lines.join("\n")
	} else {
		lines
			.iter()
			.enumerate()
			.map(|(i, line)| format!("{}:{}", prefixes[i], line))
			.collect::<Vec<_>>()
			.join("\n")
	}
}

/// Format extracted content with proper line identifiers and smart truncation.
/// When `hashes` is provided and hash mode is active, uses those as prefixes.
/// Otherwise falls back to sequential line numbers or auto-computed hashes.
pub fn format_extracted_content_smart(
	lines: &[&str],
	start_line: usize,
	max_display_lines: Option<usize>,
	hashes: Option<&[String]>,
) -> String {
	let prefixes: Vec<String> = if let Some(h) = hashes {
		h.to_vec()
	} else if super::line_hash::is_hash_mode() {
		super::line_hash::compute_line_hashes(lines)
	} else {
		lines
			.iter()
			.enumerate()
			.map(|(i, _)| format!("{}", start_line + i))
			.collect()
	};

	let max_lines = max_display_lines.unwrap_or(50);

	if lines.len() <= max_lines {
		lines
			.iter()
			.enumerate()
			.map(|(i, line)| format!("{}:{}", prefixes[i], line))
			.collect::<Vec<_>>()
			.join("\n")
	} else {
		let show_first = (max_lines * 2) / 3;
		let show_last = max_lines - show_first - 1;

		let mut result_lines = Vec::new();

		for (i, line) in lines.iter().enumerate().take(show_first) {
			result_lines.push(format!("{}:{}", prefixes[i], line));
		}

		let hidden_lines = lines.len() - show_first - show_last;
		result_lines.push(format!("[...{} lines more]", hidden_lines));

		let skip_count = lines.len() - show_last;
		for (i, line) in lines.iter().enumerate().skip(skip_count) {
			result_lines.push(format!("{}:{}", prefixes[i], line));
		}

		result_lines.join("\n")
	}
}

/// Truncate content based on token count with smart boundary detection.
pub fn truncate_content_smart(content: &str, max_tokens: usize) -> String {
	let token_count = estimate_tokens(content);
	if token_count <= max_tokens {
		return content.to_string();
	}
	let truncated = truncate_to_tokens(content, max_tokens);
	format!(
		"{truncated}\n\n⚠️ **MCP RESPONSE TRUNCATED** - Original: {token_count} tokens estimated, max {max_tokens} allowed. Use more specific commands to reduce output size]"
	)
}

/// Global MCP response truncation.
/// Returns `(content, was_truncated)`.
pub fn truncate_mcp_response_global(content: &str, max_tokens: usize) -> (String, bool) {
	if max_tokens == 0 {
		return (content.to_string(), false);
	}

	let token_count = estimate_tokens(content);
	if token_count <= max_tokens {
		return (content.to_string(), false);
	}

	let truncated = truncate_content_smart(content, max_tokens);
	(truncated, true)
}

#[cfg(test)]
mod tests {
	use super::*;

	#[test]
	fn test_mcp_truncation_unlimited() {
		let content = "This is a test content";
		let (result, was_truncated) = truncate_mcp_response_global(content, 0);
		assert_eq!(result, content);
		assert!(!was_truncated);
	}

	#[test]
	fn test_mcp_truncation_under_limit() {
		let content = "Short content";
		let (result, was_truncated) = truncate_mcp_response_global(content, 1000);
		assert_eq!(result, content);
		assert!(!was_truncated);
	}

	#[test]
	fn test_mcp_truncation_over_limit() {
		let content = "This is a very long content that should be truncated when it exceeds the token limit. ".repeat(100);
		let (result, was_truncated) = truncate_mcp_response_global(&content, 50);
		assert!(result.contains("⚠️ **MCP RESPONSE TRUNCATED**"));
		assert!(result.len() < content.len());
		assert!(was_truncated);
	}
}