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
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
//! Searches for a contiguous array of bytes determined by a given pattern. The pattern can include
//! supported wildcard characters, as seen below.
//!
//! ## Wildcards
//! * `?` match any byte
//!
//! ## Example Patterns
//! * `fe 00 68 98` - matches only `fe 00 68 98`
//! * `8d 11 ? ? 8f` - could match `8d 11 9e ef 8f` or `8d 11 0 0 8f` for example
//!
//! ## Example Usage
//! The [`scan`] function is used to scan for a pattern within the output of a [`Read`]. Using a
//! [`Cursor`](std::io::Cursor) to scan within a byte array in memory could look as follows:
//!
//! ```rust
//! use patternscan::scan;
//! use std::io::Cursor;
//!
//! let bytes = [0x10, 0x20, 0x30, 0x40, 0x50];
//! let pattern = "20 30 40";
//! let locs = scan(Cursor::new(bytes), &pattern).unwrap(); // Will equal vec![1], the index of
//!                                                         // the pattern
//! ```
//!
//! Any struct implementing [`Read`] can be passed as the reader which should be scanned for
//! ocurrences of a pattern, so one could scan for a byte sequence within an executable as follows:
//!
//! ```ignore
//! use patternscan::scan;
//! use std::fs::File;
//!
//! let reader = File::open("somebinary.exe").unwrap();
//! let instruction = "A3 ? ? ? ?";
//! let locs = scan(reader, &instruction).unwrap();
//! ```
//!
//! For more example uses of this module, see the
//! [tests](https://github.com/lewisclark/patternscan/blob/master/src/lib.rs#L128)
use std::fmt::{self, Display};
use std::io::Read;
use std::str::FromStr;

/// Size of chunks to be read from `reader` when looking for patterns.
///
/// In [`Matches`] (which in turn is used in [`scan`] and [`scan_first_match`]), bytes are read
/// from the provided [`Read`] type into a fixed-size internal buffer. The length of this buffer is
/// given by `CHUNK_SIZE`.
pub const CHUNK_SIZE: usize = 0x800;

/// Scan for any instances of `pattern` in the bytes read by `reader`.
///
/// Returns a [`Result`] containing a vector of indices of the start of each match within the
/// bytes. If no matches are found, this vector will be empty. Returns an [`Error`] if an error was
/// encountered while scanning, which could occur if the pattern is invalid (i.e: contains
/// something other than 8-bit hex values and wildcards), or if the reader encounters an error.
pub fn scan(reader: impl Read, pattern: &str) -> Result<Vec<usize>, Error> {
    let matches = Matches::from_pattern_str(reader, pattern)?;
    matches.collect()
}

/// Scan for the first instance of `pattern` in the bytes read by `reader`.
///
/// This function should be used instead of [`scan`] if you just want to test whether a byte string
/// contains a given pattern, or just find the first instance, as it returns as soon as the first
/// match is found, and will therefore be more efficient for these purposes in long strings where
/// the pattern occurs early.
///
/// Returns a [`Result`] containing an [`Option`]. This Option will contain `Some(index)` if the
/// pattern was found, where `index` is the index of the first match, and `None` if the pattern
/// was not found. Returns an [`Error`] if an error was encountered while scanning, which could
/// occur if the pattern is invalid (i.e: contains something other than 8-bit hex values and
/// wildcards), or if the reader encounters an error.
pub fn scan_first_match(reader: impl Read, pattern: &str) -> Result<Option<usize>, Error> {
    let mut matches = Matches::from_pattern_str(reader, pattern)?;
    matches.next().transpose()
}

/// Determine whether a byte slice matches a pattern.
pub fn pattern_matches(bytes: &[u8], pattern: &Pattern) -> bool {
    if bytes.len() < pattern.len() {
        false
    } else {
        pattern == bytes
    }
}

/// Represents an error which occurred while scanning for a pattern.
#[derive(Debug)]
pub struct Error {
    /// String detailing the error
    e: String,
}

impl Error {
    pub fn new(e: String) -> Self {
        Self { e }
    }
}

impl Display for Error {
    fn fmt(&self, fmt: &mut fmt::Formatter) -> fmt::Result {
        write!(fmt, "Pattern scan error: {}", self.e)
    }
}

impl std::error::Error for Error {}

/// Represents a single byte in a search pattern.
#[derive(PartialEq, Eq)]
pub enum PatternByte {
    Byte(u8),
    Any,
}

impl FromStr for PatternByte {
    type Err = Error;

    /// Create an instance of [`PatternByte`] from a string.
    ///
    /// This string should either be a hexadecimal byte, or a "?". Will return an error if the
    /// string is not a "?", or it cannot be converted into an 8-bit integer when interpreted as
    /// hexadecimal.
    fn from_str(s: &str) -> Result<Self, Error> {
        if s == "?" {
            Ok(Self::Any)
        } else {
            let n = match u8::from_str_radix(s, 16) {
                Ok(n) => Ok(n),
                Err(e) => Err(Error::new(format!("from_str_radix failed: {}", e))),
            }?;

            Ok(Self::Byte(n))
        }
    }
}

impl PartialEq<u8> for PatternByte {
    fn eq(&self, other: &u8) -> bool {
        match self {
            PatternByte::Any => true,
            PatternByte::Byte(b) => b == other,
        }
    }
}

/// Represents a pattern to search for in a byte string.
#[derive(PartialEq, Eq)]
pub struct Pattern {
    bytes: Vec<PatternByte>,
}

impl Pattern {
    fn new(bytes: Vec<PatternByte>) -> Self {
        Self { bytes }
    }

    fn len(&self) -> usize {
        self.bytes.len()
    }
}

impl FromStr for Pattern {
    type Err = Error;

    fn from_str(s: &str) -> Result<Self, Error> {
        let mut bytes = Vec::new();

        for segment in s.split_ascii_whitespace() {
            bytes.push(PatternByte::from_str(segment)?);
        }

        Ok(Self::new(bytes))
    }
}

impl PartialEq<[u8]> for Pattern {
    fn eq(&self, other: &[u8]) -> bool {
        Iterator::zip(self.bytes.iter(), other.iter()).all(|(pb, b)| pb == b)
    }
}

/// Iterator over locations of matches for a pattern found within a byte string.
///
/// This struct implements the actual logic for pattern matching, and is used by the [`scan`] and
/// [`scan_first_match`] functions to locate matches. The values returned by the iterator are
/// indices of locations of the pattern matches within the byte string produced by `reader`.
///
/// The byte string which the pattern should be searched against is read from `reader` in
/// [`CHUNK_SIZE`] chunks, when required by the iterator.
///
/// ## Example Usage
/// ```rust
/// use patternscan;
/// use std::io::Cursor;
///
/// let bytes = [0x10, 0x20, 0x30, 0x40];
/// let reader = Cursor::new(bytes);
/// let pattern = patternscan::Matches::from_pattern_str(reader, "20 30").unwrap();
/// let match_indices: Result<Vec<usize>, _> = pattern.collect();
/// let match_indices = match_indices.unwrap();
/// ```
pub struct Matches<R: Read> {
    /// Reader from which the byte string to search will be read.
    pub reader: R,
    /// Pattern to search for in the byte string.
    pub pattern: Pattern,

    // Internal state, would be nice to reduce this somehow
    bytes_buf: [u8; CHUNK_SIZE],
    last_bytes_read: usize,
    abs_position: usize,
    rel_position: usize,
}

impl<R: Read> Matches<R> {
    /// Create a new instance of [`Matches`] from an instance of [`Pattern`].
    ///
    /// `reader` should be some [`Read`] type which will produce a byte string to search. `pattern`
    /// should be a [`Pattern`] to search for.
    pub fn from_pattern(mut reader: R, pattern: Pattern) -> Result<Self, Error> {
        // Constraint imposed due to the method used to detect matches over chunk boundaries. We
        // might want to increase the chunk size to account for this?
        if 2 * pattern.len() > CHUNK_SIZE {
            return Err(Error::new(format!(
                "Pattern too long: It can be at most {} bytes",
                CHUNK_SIZE / 2
            )));
        }

        // Perform initial read into the bytes buffer on creation
        // Might be more idiomatic to only perform a read once we're stepping through the iterator,
        // I'm not sure, but this ensures that the state of the struct when an instance is created
        // is reasonable.
        let mut bytes_buf = [0; CHUNK_SIZE];
        let bytes_read = reader
            .read(&mut bytes_buf)
            .map_err(|e| Error::new(format!("Failed to read bytes: {}", e)))?;

        Ok(Self {
            reader,
            pattern,
            bytes_buf,
            last_bytes_read: bytes_read,
            abs_position: 0,
            rel_position: 0,
        })
    }

    /// Create a new instance of [`Matches`] from a string pattern.
    pub fn from_pattern_str(reader: R, pattern: &str) -> Result<Self, Error> {
        let pattern = Pattern::from_str(pattern)?;
        Self::from_pattern(reader, pattern)
    }
}

impl<R: Read> Iterator for Matches<R> {
    type Item = Result<usize, Error>;

    fn next(&mut self) -> Option<Self::Item> {
        loop {
            if self.rel_position == CHUNK_SIZE - self.pattern.len() {
                // This block is what allows us to detect matches over chunk boundaries.
                // When we're close enough to a boundary that a pattern match could overrun, we
                // copy the final bytes in the buffer to the start of the buffer, then read into
                // the rest of the buffer.
                let len = self.pattern.len();

                let boundary_bytes = &self.bytes_buf[CHUNK_SIZE - len..].to_owned();
                self.bytes_buf[..len].copy_from_slice(&boundary_bytes);

                self.last_bytes_read = match self.reader.read(&mut self.bytes_buf[len..]) {
                    Ok(b) => b,
                    Err(e) => return Some(Err(Error::new(format!("Failed to read bytes: {}", e)))),
                };

                self.rel_position = 0;
            }

            if self.rel_position == self.last_bytes_read + self.pattern.len() {
                break;
            }

            for i in self.rel_position..self.last_bytes_read + self.pattern.len() {
                if i == CHUNK_SIZE - self.pattern.len() {
                    break;
                }

                self.abs_position += 1;
                self.rel_position += 1;
                if pattern_matches(&self.bytes_buf[i..], &self.pattern) {
                    return Some(Ok(self.abs_position - 1));
                }
            }

            if self.last_bytes_read == 0 {
                break;
            }
        }

        None
    }
}

// Tests
#[cfg(test)]
mod tests {
    use std::io::Cursor;

    #[test]
    fn simple_scan_start() {
        let bytes = [0x10, 0x20, 0x30, 0x40, 0x50];
        let pattern = "10 20 30";

        assert_eq!(crate::scan(Cursor::new(bytes), &pattern).unwrap(), vec![0]);
    }

    #[test]
    fn simple_scan_middle() {
        let bytes = [0x10, 0x20, 0x30, 0x40, 0x50];
        let pattern = "20 30 40";

        assert_eq!(crate::scan(Cursor::new(bytes), &pattern).unwrap(), vec![1]);
    }

    #[test]
    fn scan_bad_exceeds() {
        let bytes = [0x10, 0x20, 0x30, 0x40, 0x50];
        let pattern = "40 50 60";

        assert_eq!(crate::scan(Cursor::new(bytes), &pattern).unwrap(), vec![]);
    }

    #[test]
    fn scan_exists() {
        let bytes = [0xff, 0xfe, 0x7c, 0x88, 0xfd, 0x90, 0x00];
        let pattern = "fe 7c 88 fd 90 0";

        assert_eq!(crate::scan(Cursor::new(bytes), &pattern).unwrap(), vec![1]);
    }

    #[test]
    fn scan_exists_multiple_q() {
        let bytes = [0xff, 0xfe, 0x7c, 0x88, 0xfd, 0x90, 0x00];
        let pattern = "fe ? ? ? 90";

        assert_eq!(crate::scan(Cursor::new(bytes), &pattern).unwrap(), vec![1]);
    }

    #[test]
    fn scan_exists_multiple_q_starts() {
        let bytes = [0xff, 0xfe, 0x7c, 0x88, 0xfd, 0x90, 0x00];
        let pattern = "? ? ? ? fd";

        assert_eq!(crate::scan(Cursor::new(bytes), &pattern).unwrap(), vec![0]);
    }

    #[test]
    fn scan_nexists_1() {
        let bytes = [0xff, 0xfe, 0x7c, 0x88, 0xfd, 0x90, 0x00];
        let pattern = "78 90 cc dd fe";

        assert_eq!(crate::scan(Cursor::new(bytes), &pattern).unwrap(), vec![]);
    }

    #[test]
    fn scan_nexists_2() {
        let bytes = [0xff, 0xfe, 0x7c, 0x88, 0xfd, 0x90, 0x00];
        let pattern = "fe 7c 88 fd 90 1";

        assert_eq!(crate::scan(Cursor::new(bytes), &pattern).unwrap(), vec![]);
    }

    #[test]
    fn scan_pattern_larger_than_bytes() {
        let bytes = [0xff, 0xfe, 0x7c, 0x88, 0xfd, 0x90, 0x00];
        let pattern = "fe 7c 88 fd 90 0 1";

        assert_eq!(crate::scan(Cursor::new(bytes), &pattern).unwrap(), vec![]);
    }

    #[test]
    fn scan_multiple_instances_of_pattern() {
        let bytes = [0x10, 0x20, 0x30, 0x10, 0x20, 0x30];
        let pattern = "10 20 30";

        assert_eq!(
            crate::scan(Cursor::new(bytes), &pattern).unwrap(),
            vec![0, 3]
        );
    }

    #[test]
    fn scan_multiple_instances_q() {
        let bytes = [0x10, 0x20, 0x30, 0x10, 0x40, 0x30];
        let pattern = "10 ? 30";

        assert_eq!(
            crate::scan(Cursor::new(bytes), &pattern).unwrap(),
            vec![0, 3]
        );
    }

    #[test]
    fn scan_rejects_invalid_pattern() {
        let bytes = [0x10, 0x20, 0x30];
        let pattern = "10 fff 20";

        assert!(crate::scan(Cursor::new(bytes), &pattern).is_err());
    }

    #[test]
    fn scan_first_match_simple_start() {
        let bytes = [0x10, 0x20, 0x30, 0x40, 0x50];
        let pattern = "10 20 30";

        assert_eq!(
            crate::scan_first_match(Cursor::new(bytes), &pattern)
                .unwrap()
                .unwrap(),
            0
        );
    }

    #[test]
    fn scan_first_match_simple_middle() {
        let bytes = [0x10, 0x20, 0x30, 0x40, 0x50];
        let pattern = "20 30 40";

        assert_eq!(
            crate::scan_first_match(Cursor::new(bytes), &pattern)
                .unwrap()
                .unwrap(),
            1
        );
    }

    #[test]
    fn scan_first_match_no_match() {
        let bytes = [0x10, 0x20, 0x30, 0x40, 0x50];
        let pattern = "10 11 12";

        assert!(crate::scan_first_match(Cursor::new(bytes), &pattern)
            .unwrap()
            .is_none());
    }

    #[test]
    fn find_across_chunk_boundary() {
        let mut bytes = vec![0; super::CHUNK_SIZE - 2];
        bytes.push(0xaa);
        bytes.push(0xbb);
        bytes.push(0xcc);
        bytes.push(0xdd);
        let pattern = "aa bb cc dd";

        assert!(crate::scan_first_match(Cursor::new(bytes), &pattern)
            .unwrap()
            .is_some())
    }

    #[test]
    fn correct_index_noninitial_chunk() {
        let mut bytes = vec![0; super::CHUNK_SIZE];
        bytes.push(0xaa);
        bytes.push(0xbb);
        bytes.push(0xcc);
        bytes.push(0xdd);
        let pattern = "aa bb cc dd";

        assert_eq!(
            crate::scan_first_match(Cursor::new(bytes), &pattern)
                .unwrap()
                .unwrap(),
            super::CHUNK_SIZE
        );
    }
}