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
/*
 * Copyright Amazon.com, Inc. or its affiliates. All Rights Reserved.
 * SPDX-License-Identifier: Apache-2.0
 */

use std::borrow::Cow;

use aws_smithy_types::{Blob, DateTime};
use minicbor::decode::Error;

use crate::data::Type;

/// Provides functions for decoding a CBOR object with a known schema.
///
/// Although CBOR is a self-describing format, this decoder is tailored for cases where the schema
/// is known in advance. Therefore, the caller can determine which object key exists at the current
/// position by calling `str` method, and call the relevant function based on the predetermined schema
/// for that key. If an unexpected key is encountered, the caller can use the `skip` method to skip
/// over the element.
#[derive(Debug, Clone)]
pub struct Decoder<'b> {
    decoder: minicbor::Decoder<'b>,
}

/// When any of the decode methods are called they look for that particular data type at the current
/// position. If the CBOR data tag does not match the type, a `DeserializeError` is returned.
#[derive(Debug)]
pub struct DeserializeError {
    #[allow(dead_code)]
    _inner: Error,
}

impl std::fmt::Display for DeserializeError {
    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
        self._inner.fmt(f)
    }
}

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

impl DeserializeError {
    pub(crate) fn new(inner: Error) -> Self {
        Self { _inner: inner }
    }

    /// More than one union variant was detected: `unexpected_type` was unexpected.
    pub fn unexpected_union_variant(unexpected_type: Type, at: usize) -> Self {
        Self {
            _inner: Error::type_mismatch(unexpected_type.into_minicbor_type())
                .with_message("encountered unexpected union variant; expected end of union")
                .at(at),
        }
    }

    /// Unknown union variant was detected. Servers reject unknown union varaints.
    pub fn unknown_union_variant(variant_name: &str, at: usize) -> Self {
        Self {
            _inner: Error::message(format!(
                "encountered unknown union variant {}",
                variant_name
            ))
            .at(at),
        }
    }

    /// More than one union variant was detected, but we never even got to parse the first one.
    /// We immediately raise this error when detecting a union serialized as a fixed-length CBOR
    /// map whose length (specified upfront) is a value different than 1.
    pub fn mixed_union_variants(at: usize) -> Self {
        Self {
            _inner: Error::message(
                "encountered mixed variants in union; expected a single union variant to be set",
            )
            .at(at),
        }
    }

    /// Expected end of stream but more data is available.
    pub fn expected_end_of_stream(at: usize) -> Self {
        Self {
            _inner: Error::message("encountered additional data; expected end of stream").at(at),
        }
    }

    /// Returns a custom error with an offset.
    pub fn custom(message: impl Into<Cow<'static, str>>, at: usize) -> Self {
        Self {
            _inner: Error::message(message.into()).at(at),
        }
    }

    /// An unexpected type was encountered.
    // We handle this one when decoding sparse collections: we have to expect either a `null` or an
    // item, so we try decoding both.
    pub fn is_type_mismatch(&self) -> bool {
        self._inner.is_type_mismatch()
    }
}

/// Macro for delegating method calls to the decoder.
///
/// This macro generates wrapper methods for calling specific methods on the decoder and returning
/// the result with error handling.
///
/// # Example
///
/// ```ignore
/// delegate_method! {
///     /// Wrapper method for encoding method `encode_str` on the decoder.
///     encode_str_wrapper => encode_str(String);
///     /// Wrapper method for encoding method `encode_int` on the decoder.
///     encode_int_wrapper => encode_int(i32);
/// }
/// ```
macro_rules! delegate_method {
    ($($(#[$meta:meta])* $wrapper_name:ident => $encoder_name:ident($result_type:ty);)+) => {
        $(
            pub fn $wrapper_name(&mut self) -> Result<$result_type, DeserializeError> {
                self.decoder.$encoder_name().map_err(DeserializeError::new)
            }
        )+
    };
}

impl<'b> Decoder<'b> {
    pub fn new(bytes: &'b [u8]) -> Self {
        Self {
            decoder: minicbor::Decoder::new(bytes),
        }
    }

    pub fn datatype(&self) -> Result<Type, DeserializeError> {
        self.decoder
            .datatype()
            .map(Type::new)
            .map_err(DeserializeError::new)
    }

    delegate_method! {
        /// Skips the current CBOR element.
        skip => skip(());
        /// Reads a boolean at the current position.
        boolean => bool(bool);
        /// Reads a byte at the current position.
        byte => i8(i8);
        /// Reads a short at the current position.
        short => i16(i16);
        /// Reads a integer at the current position.
        integer => i32(i32);
        /// Reads a long at the current position.
        long => i64(i64);
        /// Reads a float at the current position.
        float => f32(f32);
        /// Reads a double at the current position.
        double => f64(f64);
        /// Reads a null CBOR element at the current position.
        null => null(());
        /// Returns the number of elements in a definite list. For indefinite lists it returns a `None`.
        list => array(Option<u64>);
        /// Returns the number of elements in a definite map. For indefinite map it returns a `None`.
        map => map(Option<u64>);
    }

    /// Returns the current position of the buffer, which will be decoded when any of the methods is called.
    pub fn position(&self) -> usize {
        self.decoder.position()
    }

    /// Returns a `Cow::Borrowed(&str)` if the element at the current position in the buffer is a definite
    /// length string. Otherwise, it returns a `Cow::Owned(String)` if the element at the current position is an
    /// indefinite-length string. An error is returned if the element is neither a definite length nor an
    /// indefinite-length string.
    pub fn str(&mut self) -> Result<Cow<'b, str>, DeserializeError> {
        let bookmark = self.decoder.position();
        match self.decoder.str() {
            Ok(str_value) => Ok(Cow::Borrowed(str_value)),
            Err(e) if e.is_type_mismatch() => {
                // Move the position back to the start of the CBOR element and then try
                // decoding it as an indefinite length string.
                self.decoder.set_position(bookmark);
                Ok(Cow::Owned(self.string()?))
            }
            Err(e) => Err(DeserializeError::new(e)),
        }
    }

    /// Allocates and returns a `String` if the element at the current position in the buffer is either a
    /// definite-length or an indefinite-length string. Otherwise, an error is returned if the element is not a string type.
    pub fn string(&mut self) -> Result<String, DeserializeError> {
        let mut iter = self.decoder.str_iter().map_err(DeserializeError::new)?;
        let head = iter.next();

        let decoded_string = match head {
            None => String::new(),
            Some(head) => {
                let mut combined_chunks = String::from(head.map_err(DeserializeError::new)?);
                for chunk in iter {
                    combined_chunks.push_str(chunk.map_err(DeserializeError::new)?);
                }
                combined_chunks
            }
        };

        Ok(decoded_string)
    }

    /// Returns a `blob` if the element at the current position in the buffer is a byte string. Otherwise,
    /// a `DeserializeError` error is returned.
    pub fn blob(&mut self) -> Result<Blob, DeserializeError> {
        let iter = self.decoder.bytes_iter().map_err(DeserializeError::new)?;
        let parts: Vec<&[u8]> = iter
            .collect::<Result<_, _>>()
            .map_err(DeserializeError::new)?;

        Ok(if parts.len() == 1 {
            Blob::new(parts[0]) // Directly convert &[u8] to Blob if there's only one part.
        } else {
            Blob::new(parts.concat()) // Concatenate all parts into a single Blob.
        })
    }

    /// Returns a `DateTime` if the element at the current position in the buffer is a `timestamp`. Otherwise,
    /// a `DeserializeError` error is returned.
    pub fn timestamp(&mut self) -> Result<DateTime, DeserializeError> {
        let tag = self.decoder.tag().map_err(DeserializeError::new)?;
        let timestamp_tag = minicbor::data::Tag::from(minicbor::data::IanaTag::Timestamp);

        if tag != timestamp_tag {
            Err(DeserializeError::new(Error::message(
                "expected timestamp tag",
            )))
        } else {
            // Values that are more granular than millisecond precision SHOULD be truncated to fit
            // millisecond precision for epoch-seconds:
            // https://smithy.io/2.0/spec/protocol-traits.html#timestamp-formats
            //
            // Without truncation, the `RpcV2CborDateTimeWithFractionalSeconds` protocol test would
            // fail since the upstream test expect `123000000` in subsec but the decoded actual
            // subsec would be `123000025`.
            // https://github.com/smithy-lang/smithy/blob/6466fe77c65b8a17b219f0b0a60c767915205f95/smithy-protocol-tests/model/rpcv2Cbor/fractional-seconds.smithy#L17
            let epoch_seconds = self.decoder.f64().map_err(DeserializeError::new)?;
            let mut result = DateTime::from_secs_f64(epoch_seconds);
            let subsec_nanos = result.subsec_nanos();
            result.set_subsec_nanos((subsec_nanos / 1_000_000) * 1_000_000);
            Ok(result)
        }
    }
}

#[allow(dead_code)] // to avoid `never constructed` warning
#[derive(Debug)]
pub struct ArrayIter<'a, 'b, T> {
    inner: minicbor::decode::ArrayIter<'a, 'b, T>,
}

impl<'a, 'b, T: minicbor::Decode<'b, ()>> Iterator for ArrayIter<'a, 'b, T> {
    type Item = Result<T, DeserializeError>;

    fn next(&mut self) -> Option<Self::Item> {
        self.inner
            .next()
            .map(|opt| opt.map_err(DeserializeError::new))
    }
}

#[allow(dead_code)] // to avoid `never constructed` warning
#[derive(Debug)]
pub struct MapIter<'a, 'b, K, V> {
    inner: minicbor::decode::MapIter<'a, 'b, K, V>,
}

impl<'a, 'b, K, V> Iterator for MapIter<'a, 'b, K, V>
where
    K: minicbor::Decode<'b, ()>,
    V: minicbor::Decode<'b, ()>,
{
    type Item = Result<(K, V), DeserializeError>;

    fn next(&mut self) -> Option<Self::Item> {
        self.inner
            .next()
            .map(|opt| opt.map_err(DeserializeError::new))
    }
}

pub fn set_optional<B, F>(builder: B, decoder: &mut Decoder, f: F) -> Result<B, DeserializeError>
where
    F: Fn(B, &mut Decoder) -> Result<B, DeserializeError>,
{
    match decoder.datatype()? {
        crate::data::Type::Null => {
            decoder.null()?;
            Ok(builder)
        }
        _ => f(builder, decoder),
    }
}

#[cfg(test)]
mod tests {
    use crate::Decoder;
    use aws_smithy_types::date_time::Format;

    #[test]
    fn test_definite_str_is_cow_borrowed() {
        // Definite length key `thisIsAKey`.
        let definite_bytes = [
            0x6a, 0x74, 0x68, 0x69, 0x73, 0x49, 0x73, 0x41, 0x4b, 0x65, 0x79,
        ];
        let mut decoder = Decoder::new(&definite_bytes);
        let member = decoder.str().expect("could not decode str");
        assert_eq!(member, "thisIsAKey");
        assert!(matches!(member, std::borrow::Cow::Borrowed(_)));
    }

    #[test]
    fn test_indefinite_str_is_cow_owned() {
        // Indefinite length key `this`, `Is`, `A` and `Key`.
        let indefinite_bytes = [
            0x7f, 0x64, 0x74, 0x68, 0x69, 0x73, 0x62, 0x49, 0x73, 0x61, 0x41, 0x63, 0x4b, 0x65,
            0x79, 0xff,
        ];
        let mut decoder = Decoder::new(&indefinite_bytes);
        let member = decoder.str().expect("could not decode str");
        assert_eq!(member, "thisIsAKey");
        assert!(matches!(member, std::borrow::Cow::Owned(_)));
    }

    #[test]
    fn test_empty_str_works() {
        let bytes = [0x60];
        let mut decoder = Decoder::new(&bytes);
        let member = decoder.str().expect("could not decode empty str");
        assert_eq!(member, "");
    }

    #[test]
    fn test_empty_blob_works() {
        let bytes = [0x40];
        let mut decoder = Decoder::new(&bytes);
        let member = decoder.blob().expect("could not decode an empty blob");
        assert_eq!(member, aws_smithy_types::Blob::new(&[]));
    }

    #[test]
    fn test_indefinite_length_blob() {
        // Indefinite length blob containing bytes corresponding to `indefinite-byte, chunked, on each comma`.
        // https://cbor.nemo157.com/#type=hex&value=bf69626c6f6256616c75655f50696e646566696e6974652d627974652c49206368756e6b65642c4e206f6e206561636820636f6d6d61ffff
        let indefinite_bytes = [
            0x5f, 0x50, 0x69, 0x6e, 0x64, 0x65, 0x66, 0x69, 0x6e, 0x69, 0x74, 0x65, 0x2d, 0x62,
            0x79, 0x74, 0x65, 0x2c, 0x49, 0x20, 0x63, 0x68, 0x75, 0x6e, 0x6b, 0x65, 0x64, 0x2c,
            0x4e, 0x20, 0x6f, 0x6e, 0x20, 0x65, 0x61, 0x63, 0x68, 0x20, 0x63, 0x6f, 0x6d, 0x6d,
            0x61, 0xff,
        ];
        let mut decoder = Decoder::new(&indefinite_bytes);
        let member = decoder.blob().expect("could not decode blob");
        assert_eq!(
            member,
            aws_smithy_types::Blob::new("indefinite-byte, chunked, on each comma".as_bytes())
        );
    }

    #[test]
    fn test_timestamp_should_be_truncated_to_fit_millisecond_precision() {
        // Input bytes are derived from the `RpcV2CborDateTimeWithFractionalSeconds` protocol test,
        // extracting portion representing a timestamp value.
        let bytes = [
            0xc1, 0xfb, 0x41, 0xcc, 0x37, 0xdb, 0x38, 0x0f, 0xbe, 0x77, 0xff,
        ];
        let mut decoder = Decoder::new(&bytes);
        let timestamp = decoder.timestamp().expect("should decode timestamp");
        assert_eq!(
            timestamp,
            aws_smithy_types::date_time::DateTime::from_str(
                "2000-01-02T20:34:56.123Z",
                Format::DateTime
            )
            .unwrap()
        );
    }
}