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
/*! Provides byte stream parsing utilities. ```rust,ignore let reader = File::open("..."); // reader is anything that implements io::Read let alloc_size = 1000; // Set custom allocation size let muncher = ReadMuncher::<DataItem, _>::new(&reader, alloc_size, |bytes, is_eof| { // parse function here }); for packet in &muncher { // packet is DataItem } ``` */ use anyhow::Error; use std::io::Read; /// Output type for `ReadMuncher` parse function /// /// `T` is the iterator output type (e.g. `Packet` or `Frame` structs). /// /// `usize` is the number of bytes consumed to created the output type. pub type MunchOutput<T> = Option<(T, usize)>; /** Continuously reads bytes from a `Read` implementor, parses that byte stream with the provided parser function, and provides an iterator over the parsed slices/packets/frames. ```rust # use noodle::ReadMuncher; # fn main() { use std::io::Cursor; let mut read = Cursor::new(vec![0xff, 1, 0xff, 2, 2, 0xff, 3, 3, 3]); let munched: Vec<Vec<u8>> = ReadMuncher::new(&mut read, 5, |b, _| { if b.len() >= 2 { let skip = b[1] as usize + 1; if b.len() > skip { let blob = &b[..skip]; Ok(Some((blob.to_owned(), skip))) } else { Ok(None) } } else { Ok(None) } }) .collect(); assert_eq!(munched.len(), 3); assert_eq!(munched[0], vec![0xff, 1]); assert_eq!(munched[1], vec![0xff, 2, 2]); assert_eq!(munched[2], vec![0xff, 3, 3, 3]); # } ``` */ pub struct ReadMuncher<'a, T, F> where F: Fn(&[u8], bool) -> Result<MunchOutput<T>, Error>, { reader: &'a mut dyn Read, buffer: Vec<u8>, parse_location: usize, alloc_size: usize, // parse_fn: &'b ParseFn<T>, parse_fn: F, complete: bool, read_end_location: usize, } impl<'a, T, F> ReadMuncher<'a, T, F> where F: Fn(&[u8], bool) -> Result<MunchOutput<T>, Error>, { /// Starts a new `ReadMuncher` instance. /// /// ### reader /// /// The `Read` implementor to be read for bytes. /// /// ### alloc_size /// /// Internal buffer allocation increment size. /// /// This sets the initial buffer size, and size increase increments when necessary. /// /// ### parse_fn /// /// The parse function called against the read buffer. This can be a static function or a closure, with signature `Fn(&[u8], bool) -> Result<MunchOutput<T>, Error>`. /// /// The first parameter is a reference to the unconsumed slice of the read buffer. /// The second parameter is a boolean, which signals EOF (no more bytes available to read). /// /// The return type is used internally to consume/step forward. Below is some example return values and what they do: /// /// ```ignore /// /// Ok(Some((item, 12))) // 'item' is returned, buffer drains 12 bytes /// /// Ok(None) // Not enough bytes, keep trying /// /// Err(...) // Error occurred. Iterator panics and prints to stderr /// /// ``` pub fn new(reader: &'a mut dyn Read, alloc_size: usize, parse_fn: F) -> Self { Self { reader, alloc_size, buffer: Vec::with_capacity(alloc_size), parse_location: 0, parse_fn, complete: false, read_end_location: 0, } } /// Removes used bytes from buffer, and shifts everything to start. /// Returns remaining buffer bytes when at EOF. fn resize_no_alloc(&mut self) -> Result<Option<Vec<u8>>, Error> { let full_len = self.buffer.len(); self.buffer.drain(0..self.parse_location); self.buffer.resize_with(full_len, || 0); self.read_end_location -= self.parse_location; let num_new_bytes = self .reader .read(&mut self.buffer[(full_len - self.parse_location)..])?; self.read_end_location += num_new_bytes; if num_new_bytes == 0 { let blob = &self.buffer[..self.read_end_location]; self.complete = true; Ok(Some(blob.to_owned())) } else { self.parse_location = 0; Ok(None) } } /// Extends the buffer to allow storing more bytes. /// Returns remaining buffer bytes when at EOF. fn resize_alloc(&mut self) -> Result<Option<Vec<u8>>, Error> { let old_len = self.buffer.len(); self.buffer.resize_with(old_len + self.alloc_size, || 0); let num_new_bytes = self.reader.read(&mut self.buffer[old_len..])?; self.read_end_location += num_new_bytes; if num_new_bytes == 0 { let blob = &self.buffer[..self.read_end_location]; self.complete = true; Ok(Some(blob.to_owned())) } else { Ok(None) } } } impl<'a, T, F> Iterator for ReadMuncher<'a, T, F> where F: Fn(&[u8], bool) -> Result<MunchOutput<T>, Error>, { type Item = T; fn next(&mut self) -> Option<Self::Item> { if self.complete == true { return None; } loop { // let buf: &'b [u8] = &self.buffer[self.parse_location..]; let parse_result = (self.parse_fn)(&self.buffer[self.parse_location..], false); match parse_result { // Complete parse Ok(Some((item, n))) => { self.parse_location += n; return Some(item); } // Handle incomplete parse Ok(None) => { // Partial buffer not enough, shift existing and fill if self.parse_location != 0 { match self.resize_no_alloc() { Ok(r) => { // EOF if let Some(last) = r { match (self.parse_fn)(&last, true) { Ok(Some((item, _))) => return Some(item), _ => return None, } } } Err(e) => eprintln!("Error while resizing: {:?}", e), } // Entire buffer not enough, increase buffer size and refill } else { match self.resize_alloc() { Ok(r) => { // EOF if let Some(last) = r { match (self.parse_fn)(&last, true) { Ok(Some((item, _))) => return Some(item), _ => return None, } } } Err(e) => eprintln!("Error while resizing: {:?}", e), } } } Err(e) => { eprintln!("Error while parsing: {:?}", e); return None; } } } } } #[cfg(test)] mod test { use super::*; /// Tests passing static function as parser. #[test] pub fn static_fn() { use std::io::Cursor; use std::result::Result; fn my_parse_function(bytes: &[u8], _: bool) -> Result<MunchOutput<Vec<u8>>, Error> { if bytes.len() >= 2 { let skip = bytes[1] as usize + 1; if bytes.len() > skip { let blob = &bytes[..skip]; Ok(Some((blob.to_owned(), skip))) } else { Ok(None) } } else { Ok(None) } } let mut read = Cursor::new(vec![0xff, 1, 0xff, 2, 2, 0xff, 3, 3, 3]); let munched: Vec<Vec<u8>> = ReadMuncher::new(&mut read, 5, my_parse_function).collect(); assert_eq!(munched.len(), 3); assert_eq!(munched[0], vec![0xff, 1]); assert_eq!(munched[1], vec![0xff, 2, 2]); assert_eq!(munched[2], vec![0xff, 3, 3, 3]); } /// Tests passing closure as parser function. #[test] pub fn closure() { use std::io::Cursor; let mut read = Cursor::new(vec![0xff, 1, 0xff, 2, 2, 0xff, 3, 3, 3]); let munched: Vec<Vec<u8>> = ReadMuncher::new(&mut read, 5, |b, _| { if b.len() >= 2 { let skip = b[1] as usize + 1; if b.len() > skip { let blob = &b[..skip]; Ok(Some((blob.to_owned(), skip))) } else { Ok(None) } } else { Ok(None) } }) .collect(); assert_eq!(munched.len(), 3); assert_eq!(munched[0], vec![0xff, 1]); assert_eq!(munched[1], vec![0xff, 2, 2]); assert_eq!(munched[2], vec![0xff, 3, 3, 3]); } }