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
use std::borrow::{Borrow, Cow};
use std::fmt::{self, Debug, Formatter};
use std::io;
use std::ops::Deref;

#[cfg(feature = "async-tokio")]
use std::{
    pin::Pin,
    task::{Context, Poll},
};

#[cfg(feature = "serialize")]
use serde::de::{Deserialize, Deserializer, Error, Visitor};
#[cfg(feature = "serialize")]
use serde::ser::{Serialize, Serializer};

#[allow(clippy::ptr_arg)]
pub fn write_cow_string(f: &mut Formatter, cow_string: &Cow<[u8]>) -> fmt::Result {
    match cow_string {
        Cow::Owned(s) => {
            write!(f, "Owned(")?;
            write_byte_string(f, s)?;
        }
        Cow::Borrowed(s) => {
            write!(f, "Borrowed(")?;
            write_byte_string(f, s)?;
        }
    }
    write!(f, ")")
}

pub fn write_byte_string(f: &mut Formatter, byte_string: &[u8]) -> fmt::Result {
    write!(f, "\"")?;
    for b in byte_string {
        match *b {
            32..=33 | 35..=126 => write!(f, "{}", *b as char)?,
            34 => write!(f, "\\\"")?,
            _ => write!(f, "{:#02X}", b)?,
        }
    }
    write!(f, "\"")?;
    Ok(())
}

////////////////////////////////////////////////////////////////////////////////////////////////////

/// A version of [`Cow`] that can borrow from two different buffers, one of them
/// is a deserializer input.
///
/// # Lifetimes
///
/// - `'i`: lifetime of the data that deserializer borrow from the parsed input
/// - `'s`: lifetime of the data that owned by a deserializer
pub enum CowRef<'i, 's, B>
where
    B: ToOwned + ?Sized,
{
    /// An input borrowed from the parsed data
    Input(&'i B),
    /// An input borrowed from the buffer owned by another deserializer
    Slice(&'s B),
    /// An input taken from an external deserializer, owned by that deserializer
    Owned(<B as ToOwned>::Owned),
}
impl<'i, 's, B> Deref for CowRef<'i, 's, B>
where
    B: ToOwned + ?Sized,
    B::Owned: Borrow<B>,
{
    type Target = B;

    fn deref(&self) -> &B {
        match *self {
            Self::Input(borrowed) => borrowed,
            Self::Slice(borrowed) => borrowed,
            Self::Owned(ref owned) => owned.borrow(),
        }
    }
}

impl<'i, 's, B> Debug for CowRef<'i, 's, B>
where
    B: ToOwned + ?Sized + Debug,
    B::Owned: Debug,
{
    fn fmt(&self, f: &mut Formatter) -> fmt::Result {
        match *self {
            Self::Input(borrowed) => Debug::fmt(borrowed, f),
            Self::Slice(borrowed) => Debug::fmt(borrowed, f),
            Self::Owned(ref owned) => Debug::fmt(owned, f),
        }
    }
}

////////////////////////////////////////////////////////////////////////////////////////////////////

/// Wrapper around `Vec<u8>` that has a human-readable debug representation:
/// printable ASCII symbols output as is, all other output in HEX notation.
///
/// Also, when [`serialize`] feature is on, this type deserialized using
/// [`deserialize_byte_buf`](serde::Deserializer::deserialize_byte_buf) instead
/// of vector's generic [`deserialize_seq`](serde::Deserializer::deserialize_seq)
///
/// [`serialize`]: ../index.html#serialize
#[derive(PartialEq, Eq)]
pub struct ByteBuf(pub Vec<u8>);

impl Debug for ByteBuf {
    fn fmt(&self, f: &mut Formatter) -> fmt::Result {
        write_byte_string(f, &self.0)
    }
}

#[cfg(feature = "serialize")]
impl<'de> Deserialize<'de> for ByteBuf {
    fn deserialize<D>(d: D) -> Result<Self, D::Error>
    where
        D: Deserializer<'de>,
    {
        struct ValueVisitor;

        impl<'de> Visitor<'de> for ValueVisitor {
            type Value = ByteBuf;

            fn expecting(&self, f: &mut Formatter) -> fmt::Result {
                f.write_str("byte data")
            }

            fn visit_bytes<E: Error>(self, v: &[u8]) -> Result<Self::Value, E> {
                Ok(ByteBuf(v.to_vec()))
            }

            fn visit_byte_buf<E: Error>(self, v: Vec<u8>) -> Result<Self::Value, E> {
                Ok(ByteBuf(v))
            }
        }

        d.deserialize_byte_buf(ValueVisitor)
    }
}

#[cfg(feature = "serialize")]
impl Serialize for ByteBuf {
    fn serialize<S>(&self, serializer: S) -> Result<S::Ok, S::Error>
    where
        S: Serializer,
    {
        serializer.serialize_bytes(&self.0)
    }
}

////////////////////////////////////////////////////////////////////////////////////////////////////

/// Wrapper around `&[u8]` that has a human-readable debug representation:
/// printable ASCII symbols output as is, all other output in HEX notation.
///
/// Also, when [`serialize`] feature is on, this type deserialized using
/// [`deserialize_bytes`](serde::Deserializer::deserialize_bytes) instead
/// of vector's generic [`deserialize_seq`](serde::Deserializer::deserialize_seq)
///
/// [`serialize`]: ../index.html#serialize
#[derive(PartialEq, Eq)]
pub struct Bytes<'de>(pub &'de [u8]);

impl<'de> Debug for Bytes<'de> {
    fn fmt(&self, f: &mut Formatter) -> fmt::Result {
        write_byte_string(f, self.0)
    }
}

#[cfg(feature = "serialize")]
impl<'de> Deserialize<'de> for Bytes<'de> {
    fn deserialize<D>(d: D) -> Result<Self, D::Error>
    where
        D: Deserializer<'de>,
    {
        struct ValueVisitor;

        impl<'de> Visitor<'de> for ValueVisitor {
            type Value = Bytes<'de>;

            fn expecting(&self, f: &mut Formatter) -> fmt::Result {
                f.write_str("borrowed bytes")
            }

            fn visit_borrowed_bytes<E: Error>(self, v: &'de [u8]) -> Result<Self::Value, E> {
                Ok(Bytes(v))
            }
        }

        d.deserialize_bytes(ValueVisitor)
    }
}

#[cfg(feature = "serialize")]
impl<'de> Serialize for Bytes<'de> {
    fn serialize<S>(&self, serializer: S) -> Result<S::Ok, S::Error>
    where
        S: Serializer,
    {
        serializer.serialize_bytes(self.0)
    }
}

////////////////////////////////////////////////////////////////////////////////////////////////////

/// A simple producer of infinite stream of bytes, useful in tests.
///
/// Will repeat `chunk` field indefinitely.
pub struct Fountain<'a> {
    /// That piece of data repeated infinitely...
    pub chunk: &'a [u8],
    /// Part of `chunk` that was consumed by BufRead impl
    pub consumed: usize,
    /// The overall count of read bytes
    pub overall_read: u64,
}

impl<'a> io::Read for Fountain<'a> {
    fn read(&mut self, buf: &mut [u8]) -> io::Result<usize> {
        let available = &self.chunk[self.consumed..];
        let len = buf.len().min(available.len());
        let (portion, _) = available.split_at(len);

        buf.copy_from_slice(portion);
        Ok(len)
    }
}

impl<'a> io::BufRead for Fountain<'a> {
    #[inline]
    fn fill_buf(&mut self) -> io::Result<&[u8]> {
        Ok(&self.chunk[self.consumed..])
    }

    fn consume(&mut self, amt: usize) {
        self.consumed += amt;
        if self.consumed == self.chunk.len() {
            self.consumed = 0;
        }
        self.overall_read += amt as u64;
    }
}

#[cfg(feature = "async-tokio")]
impl<'a> tokio::io::AsyncRead for Fountain<'a> {
    fn poll_read(
        self: Pin<&mut Self>,
        _cx: &mut Context<'_>,
        buf: &mut tokio::io::ReadBuf<'_>,
    ) -> Poll<io::Result<()>> {
        let available = &self.chunk[self.consumed..];
        let len = buf.remaining().min(available.len());
        let (portion, _) = available.split_at(len);

        buf.put_slice(portion);
        Poll::Ready(Ok(()))
    }
}

#[cfg(feature = "async-tokio")]
impl<'a> tokio::io::AsyncBufRead for Fountain<'a> {
    #[inline]
    fn poll_fill_buf(self: Pin<&mut Self>, _cx: &mut Context<'_>) -> Poll<io::Result<&[u8]>> {
        Poll::Ready(io::BufRead::fill_buf(self.get_mut()))
    }

    #[inline]
    fn consume(self: Pin<&mut Self>, amt: usize) {
        io::BufRead::consume(self.get_mut(), amt);
    }
}

////////////////////////////////////////////////////////////////////////////////////////////////////

/// A function to check whether the byte is a whitespace (blank, new line, carriage return or tab).
#[inline]
pub const fn is_whitespace(b: u8) -> bool {
    matches!(b, b' ' | b'\r' | b'\n' | b'\t')
}

/// Calculates name from an element-like content. Name is the first word in `content`,
/// where word boundaries is XML whitespace characters.
///
/// 'Whitespace' refers to the definition used by [`is_whitespace`].
#[inline]
pub const fn name_len(mut bytes: &[u8]) -> usize {
    // Note: A pattern matching based approach (instead of indexing) allows
    // making the function const.
    let mut len = 0;
    while let [first, rest @ ..] = bytes {
        if is_whitespace(*first) {
            break;
        }
        len += 1;
        bytes = rest;
    }
    len
}

/// Returns a byte slice with leading XML whitespace bytes removed.
///
/// 'Whitespace' refers to the definition used by [`is_whitespace`].
#[inline]
pub const fn trim_xml_start(mut bytes: &[u8]) -> &[u8] {
    // Note: A pattern matching based approach (instead of indexing) allows
    // making the function const.
    while let [first, rest @ ..] = bytes {
        if is_whitespace(*first) {
            bytes = rest;
        } else {
            break;
        }
    }
    bytes
}

/// Returns a byte slice with trailing XML whitespace bytes removed.
///
/// 'Whitespace' refers to the definition used by [`is_whitespace`].
#[inline]
pub const fn trim_xml_end(mut bytes: &[u8]) -> &[u8] {
    // Note: A pattern matching based approach (instead of indexing) allows
    // making the function const.
    while let [rest @ .., last] = bytes {
        if is_whitespace(*last) {
            bytes = rest;
        } else {
            break;
        }
    }
    bytes
}

////////////////////////////////////////////////////////////////////////////////////////////////////

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

    #[test]
    fn write_byte_string0() {
        let bytes = ByteBuf(vec![10, 32, 32, 32, 32, 32, 32, 32, 32]);
        assert_eq!(format!("{:?}", bytes), "\"0xA        \"");
    }

    #[test]
    fn write_byte_string1() {
        let bytes = ByteBuf(vec![
            104, 116, 116, 112, 58, 47, 47, 119, 119, 119, 46, 119, 51, 46, 111, 114, 103, 47, 50,
            48, 48, 50, 47, 48, 55, 47, 111, 119, 108, 35,
        ]);
        assert_eq!(
            format!("{:?}", bytes),
            r##""http://www.w3.org/2002/07/owl#""##
        );
    }

    #[test]
    fn write_byte_string3() {
        let bytes = ByteBuf(vec![
            67, 108, 97, 115, 115, 32, 73, 82, 73, 61, 34, 35, 66, 34,
        ]);
        assert_eq!(format!("{:?}", bytes), r##""Class IRI=\"#B\"""##);
    }

    #[test]
    fn name_len() {
        assert_eq!(super::name_len(b""), 0);
        assert_eq!(super::name_len(b" abc"), 0);
        assert_eq!(super::name_len(b" \t\r\n"), 0);

        assert_eq!(super::name_len(b"abc"), 3);
        assert_eq!(super::name_len(b"abc "), 3);

        assert_eq!(super::name_len(b"a bc"), 1);
        assert_eq!(super::name_len(b"ab\tc"), 2);
        assert_eq!(super::name_len(b"ab\rc"), 2);
        assert_eq!(super::name_len(b"ab\nc"), 2);
    }

    #[test]
    fn trim_xml_start() {
        assert_eq!(Bytes(super::trim_xml_start(b"")), Bytes(b""));
        assert_eq!(Bytes(super::trim_xml_start(b"abc")), Bytes(b"abc"));
        assert_eq!(
            Bytes(super::trim_xml_start(b"\r\n\t ab \t\r\nc \t\r\n")),
            Bytes(b"ab \t\r\nc \t\r\n")
        );
    }

    #[test]
    fn trim_xml_end() {
        assert_eq!(Bytes(super::trim_xml_end(b"")), Bytes(b""));
        assert_eq!(Bytes(super::trim_xml_end(b"abc")), Bytes(b"abc"));
        assert_eq!(
            Bytes(super::trim_xml_end(b"\r\n\t ab \t\r\nc \t\r\n")),
            Bytes(b"\r\n\t ab \t\r\nc")
        );
    }
}