jupiter 4.0.0

Jupiter is a library for providing high-throughput ultra low latency services via the RESP protocol as defined by Redis.
Documentation
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
//! Provides a parser and wrapper for handling incoming RESP requests.
//!
//! A RESP request (that is "REdis Serialization Protocol") is quite simple. It starts with a
//! "*" followed by the number of arguments. For each argument, there is a "$" followed by the
//! number of bytes in the argument string followed by "\r\n". After the string data, yet another
//! CRLF (\r\n) is output.
//!
//! Therefore a simple request might look like:
//! * "PING" => `*1\r\n$4\r\nPING\r\n`
//! * "SET my-key 5 => `*3\r\n$3\r\SET\r\n$6\r\nmy-key\r\n$1\r\n5\r\n`
//!
//! As re receive these request via a network interface, which might also provide partial requests
//! we require a very fast and efficient algorithm to detect if the given data is a valid request
//! and what its parameters are.
//!
//! [Request::parse](Request::parse) implements this algorithm which can detect partial requests
//! in less than 100ns and also parse full requests well below 500ns! As internally all results are
//! only indices into the given byte buffer, only a single allocation for the list of offsets
//! is performed and no data is copied whatsoever.
//!
//! # Examples
//!
//! Parsing a simple request:
//! ```
//! # use bytes::BytesMut;
//! # use jupiter::request::Request;
//! let mut bytes = BytesMut::from("*1\r\n$4\r\nPING\r\n");
//! let result = Request::parse(&mut bytes).unwrap().unwrap();
//!
//! assert_eq!(result.command(), "PING");
//! assert_eq!(result.parameter_count(), 0);
//! ```
//!
//! Parsing a partial request:
//! ```
//! # use bytes::BytesMut;
//! # use jupiter::request::Request;
//! let mut bytes = BytesMut::from("*2\r\n$4\r\nPING\r\n$7\r\nTESTP");
//! let result = Request::parse(&mut bytes).unwrap();
//!
//! assert_eq!(result.is_none(), true);
//! ```
//!
//! Parsing an invalid request:
//! ```
//! # use bytes::BytesMut;
//! # use jupiter::request::Request;
//! let mut bytes = BytesMut::from("$4\r\nPING\r\n");
//! let result = Request::parse(&mut bytes);
//!
//! assert_eq!(result.is_err(), true);
//! ```
//!
//! Building a request for test environments:
//! ```
//! # use jupiter::request::Request;
//! let request = Request::example(vec!("PING"));
//! assert_eq!(request.command(), "PING");
//! ```
use std::fmt::{Display, Formatter, Write};

use anyhow::{anyhow, Context, Result};
use bytes::{Bytes, BytesMut};

/// Provides an internal representation of either the command or a parameter.
///
/// We only need to keep the byte offsets around as the underlying byte buffer is tied to
/// the request anyway.
#[derive(Copy, Clone, Debug)]
struct Range {
    start: usize,
    end: usize,
}

impl Range {
    /// Computes the start of the subsequent range by skipping over the CRLF.
    fn next_offset(&self) -> usize {
        self.end + 3
    }
}

impl Display for Range {
    fn fmt(&self, f: &mut Formatter<'_>) -> std::fmt::Result {
        write!(f, "{}..{}", self.start, self.end)
    }
}

/// Represents a parsed RESP request.
///
/// Note that we treat the 1st parameter as "command" and re-number all other parameters
/// accordingly. Therefore "SET x y" will have "SET" as command, "x" as first parameter
/// (index: 0) and "y" as second (index: 1).
pub struct Request {
    len: usize,
    data: Bytes,
    command: Range,
    arguments: Vec<Range>,
}

impl Request {
    const DOLLAR: u8 = b'$';
    const ASTERISK: u8 = b'*';
    const CR: u8 = b'\r';
    const ZERO_DIGIT: u8 = b'0';
    const NINE_DIGIT: u8 = b'9';

    /// Tries to parse a RESP request from the given byte buffer.
    ///
    /// If malformed data is detected, we return an **Err**. Otherwise we either return an
    /// empty optional, in case only a partial request is present or otherwise a full request
    /// which has then the form `Ok(Some(Request))`.
    ///
    ///
    /// # Examples
    ///
    /// Parsing a simple request:
    /// ```
    /// # use bytes::BytesMut;
    /// # use jupiter::request::Request;
    /// let mut bytes = BytesMut::from("*3\r\n$3\r\nSET\r\n$1\r\nx\r\n$1\r\ny\r\n");
    /// let result = Request::parse(&mut bytes).unwrap().unwrap();
    ///
    /// assert_eq!(result.command(), "SET");
    /// assert_eq!(result.parameter_count(), 2);
    /// assert_eq!(result.str_parameter(0).unwrap(), "x");
    /// assert_eq!(result.str_parameter(1).unwrap(), "y");
    /// assert_eq!(result.str_parameter(2).is_err(), true);
    /// ```
    pub fn parse(data: &BytesMut) -> anyhow::Result<Option<Request>> {
        let len = data.len();

        // Abort as early as possible if a partial request is present...
        if len < 4 || data[data.len() - 2] != Request::CR {
            Ok(None)
        } else {
            Request::parse_inner(data)
        }
    }

    /// Provides a helper function to create an example request in test environments.
    ///
    /// # Example
    /// ```
    /// # use jupiter::request::Request;
    /// let request = Request::example(vec!("PING"));
    /// assert_eq!(request.command(), "PING");
    /// ```
    pub fn example(data: Vec<&str>) -> Request {
        let mut input = String::new();
        let _ = write!(input, "*{}\r\n", data.len());
        for param in data {
            let _ = write!(input, "${}\r\n{}\r\n", param.len(), param);
        }

        Request::parse(&BytesMut::from(input.as_str()))
            .unwrap()
            .unwrap()
    }

    fn parse_inner(data: &BytesMut) -> anyhow::Result<Option<Request>> {
        // Check general validity of the request...
        let mut offset = 0;
        if data[0] != Request::ASTERISK {
            return Err(anyhow!("A request must be an array of bulk strings!"));
        } else {
            offset += 1;
        }

        // Parse the number of arguments...
        let (mut num_args, range) = match Request::read_int(data, offset)? {
            Some((num_args, range)) => (num_args - 1, range),
            _ => return Ok(None),
        };
        offset = range.next_offset();

        // Parse the first parameter as "command"...
        let command = match Request::read_bulk_string(data, offset)? {
            Some(range) => range,
            _ => return Ok(None),
        };
        offset = command.next_offset();

        // Parse the remaining arguments...
        let mut arguments = Vec::with_capacity(num_args as usize);
        while num_args > 0 {
            if let Some(range) = Request::read_bulk_string(data, offset)? {
                arguments.push(range);
                num_args -= 1;
                offset = range.next_offset();
            } else {
                return Ok(None);
            }
        }

        Ok(Some(Request {
            len: offset,
            data: data.clone().freeze(),
            command,
            arguments,
        }))
    }

    /// Tries to parse a number.
    ///
    /// This is either the number of arguments or the length of an argument string. Note that
    /// the algorithm and also the return type is a bit more complex as we have to handle the
    /// happy path (valid number being read) as well as an error (invalid number found) and also
    /// the partial request case (we didn't discover the final CR which marks the end of the
    /// number).
    fn read_int(buffer: &BytesMut, offset: usize) -> anyhow::Result<Option<(i32, Range)>> {
        let mut value: i32 = 0;
        let mut index = offset;
        while index < buffer.len() {
            let digit = buffer[index];
            if digit == Request::CR {
                return if buffer.len() > index {
                    Ok(Some((
                        value,
                        Range {
                            start: offset,
                            end: index - 1,
                        },
                    )))
                } else {
                    Ok(None)
                };
            }
            if !(Request::ZERO_DIGIT..=Request::NINE_DIGIT).contains(&digit) {
                return Err(anyhow!("Malformed integer at position {}", index));
            }

            value = value * 10 + (digit - Request::ZERO_DIGIT) as i32;
            index += 1;
        }

        Ok(None)
    }

    fn read_bulk_string(buffer: &BytesMut, offset: usize) -> anyhow::Result<Option<Range>> {
        if offset >= buffer.len() {
            return Ok(None);
        }
        if buffer[offset] != Request::DOLLAR {
            return Err(anyhow!("Expected a bulk string at {}", offset));
        }

        if let Some((length, range)) = Request::read_int(buffer, offset + 1)? {
            let next_offset = range.next_offset();
            if buffer.len() >= next_offset + length as usize + 2 {
                return Ok(Some(Range {
                    start: next_offset,
                    end: next_offset + length as usize - 1,
                }));
            }
        }

        Ok(None)
    }

    /// Returns the command in the request (this is the first parameter).
    pub fn command(&self) -> &str {
        std::str::from_utf8(&self.data[self.command.start..=self.command.end]).unwrap()
    }

    /// Returns the number of parameters (not counting the command itself).
    pub fn parameter_count(&self) -> usize {
        self.arguments.len()
    }

    /// Returns the n-th parameter (not including the command).
    ///
    /// Returns an **Err** if the requested index is outside of the range of detected
    /// parameters.
    pub fn parameter(&self, index: usize) -> Result<Bytes> {
        if index < self.arguments.len() {
            Ok(self
                .data
                .slice(self.arguments[index].start..=self.arguments[index].end))
        } else {
            Err(anyhow!(
                "Invalid parameter index {} (only {} are present)",
                index,
                self.arguments.len()
            ))
        }
    }

    /// Returns the n-th parameter as UTF-8 string.
    ///
    /// Returns an **Err** if either the requested index is out of range or if the parameter
    /// data isn't a valid UTF-8 sequence.
    pub fn str_parameter(&self, index: usize) -> Result<&str> {
        if index < self.arguments.len() {
            let range = self.arguments[index];
            std::str::from_utf8(&self.data[range.start..=range.end]).with_context(|| {
                format!(
                    "Failed to parse parameter {} (range {}) as UTF-8 string!",
                    index, range
                )
            })
        } else {
            Err(anyhow!(
                "Invalid parameter index {} (only {} are present)",
                index,
                self.arguments.len()
            ))
        }
    }

    /// Returns the n-th parameter as integer.
    ///
    /// Returns an **Err** if either the requested index is out of range or if the parameter
    /// data isn't a valid integer number.
    pub fn int_parameter(&self, index: usize) -> Result<i32> {
        let string = self.str_parameter(index)?;
        string.parse().with_context(|| {
            format!(
                "Failed to parse parameter {} ('{}') as integer!",
                index, string
            )
        })
    }

    /// Returns the total length on bytes for this request.
    pub fn len(&self) -> usize {
        self.len
    }

    /// Determines if the request is completely empty.
    pub fn is_empty(&self) -> bool {
        self.len == 0
    }
}

#[cfg(test)]
mod tests {
    use crate::request::Request;
    use bytes::BytesMut;

    #[test]
    fn a_command_is_successfully_parsed() {
        let request = Request::parse(&mut BytesMut::from(
            "*3\r\n$10\r\ntest.hello\r\n$5\r\nWorld\r\n$2\r\n42\r\n",
        ))
        .unwrap()
        .unwrap();

        assert_eq!(request.parameter_count(), 2);
        assert_eq!(request.command(), "test.hello");

        assert_eq!(request.str_parameter(0).unwrap(), "World");
        assert_eq!(
            std::str::from_utf8(request.parameter(0).unwrap().as_ref()).unwrap(),
            "World"
        );
        assert_eq!(request.int_parameter(1).unwrap(), 42);

        assert_eq!(request.str_parameter(2).is_err(), true);
        assert_eq!(request.int_parameter(2).is_err(), true);
        assert_eq!(request.parameter(2).is_err(), true);
    }

    #[test]
    fn missing_array_is_detected() {
        let result = Request::parse(&mut BytesMut::from("+GET\r\n"));
        assert_eq!(result.is_err(), true);
    }

    #[test]
    fn non_bulk_string_is_detected() {
        let result = Request::parse(&mut BytesMut::from("*1\r\n+GET\r\n"));
        assert_eq!(result.is_err(), true);
    }

    #[test]
    fn invalid_number_is_detected() {
        let result = Request::parse(&mut BytesMut::from("*GET\r\n"));
        assert_eq!(result.is_err(), true);
    }

    #[test]
    fn an_incomplete_command_is_skipped() {
        {
            let result = Request::parse(&mut BytesMut::from("")).unwrap();
            assert_eq!(result.is_none(), true);
        }
        {
            let result = Request::parse(&mut BytesMut::from("*")).unwrap();
            assert_eq!(result.is_none(), true);
        }
        {
            let result = Request::parse(&mut BytesMut::from("*1")).unwrap();
            assert_eq!(result.is_none(), true);
        }
        {
            let result = Request::parse(&mut BytesMut::from("*1\r")).unwrap();
            assert_eq!(result.is_none(), true);
        }
        {
            let result = Request::parse(&mut BytesMut::from("*1\r\n")).unwrap();
            assert_eq!(result.is_none(), true);
        }
        {
            let result = Request::parse(&mut BytesMut::from("*2\r\n$10\r\ntest.h")).unwrap();
            assert_eq!(result.is_none(), true);
        }
        {
            let result =
                Request::parse(&mut BytesMut::from("*2\r\n$10\r\ntest.hello\r\n")).unwrap();
            assert_eq!(result.is_none(), true);
        }
    }
}