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
use std::hash::Hash;

use percent_encoding::{percent_decode, AsciiSet, CONTROLS};

use crate::{
    dmp::{Diff, DiffMatchPatch, Time},
    Ops,
};

pub type Efficient = u8;
pub type Compat = char;

// Appending controls to ensure exact same encoding as cpp variant
const ENCODE_SET: &AsciiSet = &CONTROLS
    .add(b'"')
    .add(b'<')
    .add(b'>')
    .add(b'`')
    .add(b'{')
    .add(b'}')
    .add(b'%')
    .add(b'[')
    .add(b'\\')
    .add(b']')
    .add(b'^')
    .add(b'|');

pub trait DType: Copy + Ord + Eq + Hash {
    // fn differ(dmp: &DiffMatchPatch, txt_old: &str, txt_new: &str) -> Result<Vec<Diff<Self>>, crate::errors::Error>;
    fn bisect_split(
        dmp: &DiffMatchPatch,
        old: &[Self],
        new: &[Self],
        x: usize,
        y: usize,
        deadline: Option<Time>,
    ) -> Result<Vec<Diff<Self>>, crate::errors::Error>;

    fn from_char(c: char) -> Self;
    fn as_char(&self) -> Option<char>;
    fn from_str(str: &str) -> Vec<Self>;
    fn to_string(data: &[Self]) -> Result<String, crate::Error>;

    fn is_linebreak_end(input: &[Self]) -> bool;
    fn is_linebreak_start(input: &[Self]) -> bool;

    fn percent_encode(input: &[Self]) -> Vec<Self>;
    fn percent_decode(input: &[Self]) -> Vec<Self>;

    fn humanize(_diffs: &mut Vec<Diff<Self>>) -> Result<(), crate::Error> {
        Ok(())
    }
}

impl DType for u8 {
    fn bisect_split(
        dmp: &DiffMatchPatch,
        old: &[u8],
        new: &[u8],
        x: usize,
        y: usize,
        deadline: Option<Time>,
    ) -> Result<Vec<Diff<u8>>, crate::errors::Error> {
        let old_a = &old[..x];
        let new_a = &new[..y];

        let old_b = &old[x..];
        let new_b = &new[y..];

        // Compute both diffs serially.
        let mut diffs_a = dmp.diff_internal(old_a, new_a, false, deadline)?;
        diffs_a.append(&mut dmp.diff_internal(old_b, new_b, false, deadline)?);

        Ok(diffs_a)
    }

    fn from_char(c: char) -> Self {
        c as u8
    }

    fn as_char(&self) -> Option<char> {
        Some(*self as char)
    }

    fn from_str(str: &str) -> Vec<Self> {
        str.as_bytes().to_vec()
    }

    #[inline]
    fn to_string(data: &[Self]) -> Result<String, crate::Error> {
        std::str::from_utf8(data)
            .map_err(|_| crate::Error::Utf8Error)
            .map(|s| s.to_string())
    }

    #[inline]
    fn is_linebreak_end(input: &[Self]) -> bool {
        input.ends_with(b"\n\n") || input.ends_with(b"\n\r\n")
    }

    #[inline]
    fn is_linebreak_start(input: &[Self]) -> bool {
        input.starts_with(b"\r\n\n")
            || input.starts_with(b"\r\n\r\n")
            || input.starts_with(b"\n\r\n")
            || input.starts_with(b"\n\n")
    }

    #[inline]
    fn percent_encode(input: &[Self]) -> Vec<Self> {
        percent_encoding::percent_encode(input, ENCODE_SET)
            .collect::<String>()
            .as_bytes()
            .to_vec()
    }

    #[inline]
    fn percent_decode(input: &[Self]) -> Vec<Self> {
        percent_decode(input).collect()
    }

    #[inline]
    fn humanize(diffs: &mut Vec<Diff<Self>>) -> Result<(), crate::Error> {
        let mut idx = 0_usize;
        let mut err_prefix = vec![];

        let mut err_start = None;

        // First pass, we'll chomp of errors in the diffs?
        // The pattern we have seen is that
        while idx < diffs.len() {
            let diff = &mut diffs[idx];

            if let Err(e) = std::str::from_utf8(diff.data()) {
                // Errors can come in 2 forms
                // 1. error at the end of bytes - we'll keep prefixing the error bytes to all non equalities that follow
                // 2. error at the begining of bytes - this one is tricky - we'll need to figure out the suffix at which the rest of the string is valid
                if e.error_len().is_none() && err_start.is_none() {
                    err_start = Some(idx);

                    if diff.op() == Ops::Equal {
                        err_prefix = diff.data()[e.valid_up_to()..].to_vec();
                        diff.1 = if e.valid_up_to() > 0 {
                            diff.data()[..e.valid_up_to()].to_vec()
                        } else {
                            vec![]
                        };

                        idx += 1;
                        continue;
                    }
                }

                if let Some(err_start_idx) = err_start {
                    // For insert and delete add the prefix collected earlier (end error bytes)
                    if diff.op() == Ops::Delete || diff.op() == Ops::Insert {
                        diff.1 = [&err_prefix, diff.data()].concat();
                    } else {
                        if let Some(err_len) = e.error_len() {
                            // Iteratively figure out at what point does the error go away if at-all
                            let mut suffix = diff.data()[..err_len].to_vec();
                            let mut data = diff.data()[err_len..].to_vec();

                            while let Err(e) = std::str::from_utf8(&data) {
                                if e.error_len().is_none() {
                                    break;
                                }

                                // should never panic cos empty data is also a valid utf8
                                let first_byte = data.remove(0);
                                suffix.push(first_byte);
                            }

                            // here, we have a suffix to be added to all previous cases and a data that might be good string or error at the end of bytes
                            // which is a separate cycle

                            // Let's add the suffix to all the intermediate steps
                            diff.1 = data.to_vec();
                            diffs
                                .iter_mut()
                                .take(idx)
                                .skip(err_start_idx)
                                .for_each(|d| {
                                    if d.op() == Ops::Equal {
                                        return;
                                    }
                                    d.1 = [d.data(), &suffix[..]].concat();
                                });

                            // An equality within edits, lets seek the next one and update this suffix too
                            if data.is_empty() {
                                if idx < diffs.len() - 1 && diffs[idx + 1].op() != Ops::Equal {
                                    diffs[idx + 1].1 =
                                        [&err_prefix[..], &suffix, diffs[idx + 1].data()].concat();
                                }

                                diffs.remove(idx);
                            }
                        }

                        // Move back to where all of this started
                        idx = err_start_idx;
                        err_start = None;
                        err_prefix = vec![];
                        continue;
                    }
                }
            }
            idx += 1;
        }

        Ok(())
    }
}

impl DType for char {
    fn bisect_split(
        dmp: &DiffMatchPatch,
        old: &[char],
        new: &[char],
        x: usize,
        y: usize,
        deadline: Option<Time>,
    ) -> Result<Vec<Diff<char>>, crate::errors::Error> {
        let old_a = &old[..x];
        let new_a = &new[..y];

        let old_b = &old[x..];
        let new_b = &new[y..];

        // Compute both diffs serially.
        let mut diffs_a = dmp.diff_internal(old_a, new_a, false, deadline)?;
        diffs_a.append(&mut dmp.diff_internal(old_b, new_b, false, deadline)?);

        Ok(diffs_a)
    }

    fn from_char(c: char) -> Self {
        c
    }

    fn as_char(&self) -> Option<char> {
        Some(*self)
    }

    fn from_str(str: &str) -> Vec<Self> {
        str.chars().collect::<Vec<_>>()
    }

    #[inline]
    fn to_string(data: &[Self]) -> Result<String, crate::Error> {
        Ok(data.iter().collect::<String>())
    }

    #[inline]
    fn is_linebreak_end(input: &[Self]) -> bool {
        input.ends_with(&['\n', '\n']) || input.ends_with(&['\n', '\r', '\n'])
    }

    #[inline]
    fn is_linebreak_start(input: &[Self]) -> bool {
        input.starts_with(&['\r', '\n', '\n'])
            || input.starts_with(&['\r', '\n', '\r', '\n'])
            || input.starts_with(&['\n', '\r', '\n'])
            || input.starts_with(&['\n', '\n'])
    }

    #[inline]
    fn percent_encode(input: &[Self]) -> Vec<Self> {
        let d = input
            .iter()
            .map(|c| {
                let mut b = vec![0; c.len_utf8()];
                c.encode_utf8(&mut b);

                b
            })
            .collect::<Vec<_>>()
            .concat();

        let encoded = percent_encoding::percent_encode(&d[..], ENCODE_SET).collect::<String>();

        Self::from_str(&encoded)
    }

    #[inline]
    fn percent_decode(input: &[Self]) -> Vec<Self> {
        let ip = input.iter().collect::<String>();
        percent_decode(ip.as_bytes())
            .decode_utf8()
            .unwrap()
            .chars()
            .collect()
    }
}

impl DType for usize {
    fn bisect_split(
        dmp: &DiffMatchPatch,
        old: &[usize],
        new: &[usize],
        x: usize,
        y: usize,
        deadline: Option<Time>,
    ) -> Result<Vec<Diff<usize>>, crate::errors::Error> {
        let old_a = &old[..x];
        let new_a = &new[..y];

        let old_b = &old[x..];
        let new_b = &new[y..];

        // Compute both diffs serially.
        let mut diffs_a = dmp.diff_lines(old_a, new_a, deadline)?;
        diffs_a.append(&mut dmp.diff_lines(old_b, new_b, deadline)?);

        Ok(diffs_a)
    }

    fn from_char(c: char) -> Self {
        (c as u8) as usize
    }

    fn as_char(&self) -> Option<char> {
        char::from_digit(*self as u32, 10)
    }

    fn from_str(_: &str) -> Vec<Self> {
        unimplemented!()
    }

    fn to_string(_: &[Self]) -> Result<String, crate::Error> {
        unimplemented!()
    }

    fn is_linebreak_end(_: &[Self]) -> bool {
        unimplemented!()
    }

    #[inline]
    fn is_linebreak_start(_: &[Self]) -> bool {
        unimplemented!()
    }

    #[inline]
    fn percent_encode(_: &[Self]) -> Vec<Self> {
        unimplemented!()
    }

    #[inline]
    fn percent_decode(_: &[Self]) -> Vec<Self> {
        unimplemented!()
    }
}