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
|
/*
Ousía
Copyright (C) 2014, 2015 Benjamin Paaßen, Andreas Stöckel
This program is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program. If not, see <http://www.gnu.org/licenses/>.
*/
#include <algorithm>
#include <core/common/CharReader.hpp>
#include <core/common/Utils.hpp>
#include "SourceContextReader.hpp"
namespace ousia {
SourceContextReader::SourceContextReader() : cache{0} {}
SourceContext SourceContextReader::readContext(CharReader &reader,
const SourceRange &range,
size_t maxContextLength,
const std::string &filename)
{
// Abort if the given range is not valid
if (!range.isValid()) { // (I2)
return SourceContext{};
}
// Set the filename and the range
SourceContext ctx;
ctx.startLine = 1;
ctx.startColumn = 1;
ctx.endLine = 1;
ctx.endColumn = 1;
ctx.range = range;
ctx.filename = filename;
// Some constants for convenience
const SourceOffset start = range.getStart();
const SourceOffset end = range.getEnd();
const SourceOffset lastCacheOffs = cache.back();
// Find the entry in the cache that is just below the given start offset
// and jump to this location
size_t offs = 0;
auto it = std::lower_bound(cache.begin(), cache.end(), start);
if (it != cache.begin()) {
it--; // Go to the previous entry
offs = *it; // Read the corresponding byte offset
size_t line = it - cache.begin() + 1;
ctx.startLine = line;
ctx.endLine = line;
}
// Move the char reader to the specified offset, abort if this did not work
// out
if (offs != reader.seek(offs)) {
return SourceContext{};
}
// TODO: Handle skew introduced by linebreak processing \n\r => \n
// Read until the requested byte offset is reached, track linebreaks in the
// linebreak cache
std::vector<char> lineBuf;
size_t lineBufStart = offs;
size_t lastLineStart = offs;
char c;
while (reader.read(c)) {
// Fetch the offset after this character
const size_t nextOffs = reader.getOffset();
// Fetch the current offset, check whether start was reached
const bool reachedStart = offs >= start;
const bool reachedEnd = offs >= end;
// Handle linebreaks and update the linebreak cache
if (c == '\n') {
// Update the linebreak cache if we are in uncached regions
if (offs > lastCacheOffs) {
cache.push_back(nextOffs);
}
if (!reachedStart) {
ctx.startLine++;
ctx.startColumn = 1;
lineBuf.clear();
lineBufStart = nextOffs;
lastLineStart = nextOffs;
} else {
lineBuf.push_back('\n');
}
if (!reachedEnd) {
ctx.endLine++;
ctx.endColumn = 1;
} else {
// This was the last character, abort
break;
}
} else {
// Increment the start and the end column if this is not an
// UTF8-continuation byte (note that we count unicode codepoints not
// actual characters, which may be more than one codepoint)
if (!((c & 0x80) && !(c & 0x40))) {
if (!reachedStart) {
ctx.startColumn++;
}
if (!reachedEnd) {
ctx.endColumn++;
}
}
// Record all characters when start is reached or at least when
// the distance to start is smaller than the maximum context length
// TODO: This is suboptimal as parts of lineBuf are thrown away
// later. If the given range is really large, this will waste huge
// amounts of RAM.
if (reachedStart || (start - offs <= maxContextLength)) {
if (lineBuf.empty()) {
lineBufStart = offs;
}
lineBuf.push_back(c);
}
}
// Set the new offset
offs = nextOffs;
}
// If we did not reach the end or for some reason the lineBufStart is larger
// than start (to assure invariant I1 is fulfilled), abort
offs = reader.getOffset();
if (offs < end || lineBufStart > start) { // (I1)
return SourceContext{};
}
// Calculate a first relative position and length
ctx.relPos = start - lineBufStart; // lineBufStart > start (I1)
ctx.relLen = end - start; // end >= start (I2)
// Remove linebreaks at the beginning and the end
const std::pair<size_t, size_t> b = Utils::trim(
lineBuf,
[&lineBuf](size_t i) { return Utils::isLinebreak(lineBuf[i]); });
ssize_t s = b.first, e = b.second;
s = std::min(s, static_cast<ssize_t>(ctx.relPos));
// Remember the trimmed positions, only continue if the context text did
// not entirely consist of linebreaks
const ssize_t ts = s, te = e; // s >= 0, e >= 0, ts >= 0, te >= 0 (I3)
if (te > ts) {
// Trim the line further if it is longer than the maxContextLength
if (static_cast<size_t>(te - ts) > maxContextLength &&
maxContextLength != MAX_MAX_CONTEXT_LENGTH) {
ssize_t c = (ctx.relPos + ctx.relLen / 2);
s = c - maxContextLength / 2;
e = c + maxContextLength / 2;
// Account for rounding error
if (static_cast<size_t>(e - s) < maxContextLength) {
e++;
}
// Redistribute available characters at the beginning or the end
if (s < ts) {
e = e + (ts - s);
s = ts; // ts >= 0 => s >= 0 (I3)
}
if (e > te) {
s = s - std::min(s - ts, e - te); // ts - s <= s => s >= 0
e = te; // te >= 0 => e >= 0 (I3)
}
}
// Update the relative position and length, set the "truncated" flags
ctx.relPos = std::max<ssize_t>(0, start - lineBufStart - s);
ctx.relLen = std::min<ssize_t>(ctx.relLen, e - s);
ctx.truncatedStart = s > ts || lastLineStart < lineBufStart;
ctx.truncatedEnd = e < te;
// Copy the selected area to the output string
ctx.text = std::string{&lineBuf[s], static_cast<size_t>(e - s)};
}
return ctx;
}
SourceContext SourceContextReader::readContext(CharReader &reader,
const SourceRange &range,
const std::string &filename)
{
return readContext(reader, range, MAX_MAX_CONTEXT_LENGTH, filename);
}
}
|
