summaryrefslogtreecommitdiff
path: root/tools/quickbook/src/files.cpp
blob: 9999b0f220165560e148093a2593b5d9be89a6c5 (plain)
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
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
/*=============================================================================
    Copyright (c) 2002 2004 2006 Joel de Guzman
    Copyright (c) 2004 Eric Niebler
    http://spirit.sourceforge.net/

    Use, modification and distribution is subject to the Boost Software
    License, Version 1.0. (See accompanying file LICENSE_1_0.txt or copy at
    http://www.boost.org/LICENSE_1_0.txt)
=============================================================================*/
#include "files.hpp"
#include <boost/filesystem/fstream.hpp>
#include <boost/unordered_map.hpp>
#include <boost/range/algorithm/upper_bound.hpp>
#include <boost/range/algorithm/transform.hpp>
#include <boost/foreach.hpp>
#include <fstream>
#include <iterator>
#include <vector>

namespace quickbook
{
    namespace
    {
        boost::unordered_map<fs::path, file_ptr> files;
    }

    // Read the first few bytes in a file to see it starts with a byte order
    // mark. If it doesn't, then write the characters we've already read in.
    // Although, given how UTF-8 works, if we've read anything in, the files
    // probably broken.

    template <typename InputIterator, typename OutputIterator>
    bool check_bom(InputIterator& begin, InputIterator end,
            OutputIterator out, char const* chars, int length)
    {
        char const* ptr = chars;

        while(begin != end && *begin == *ptr) {
            ++begin;
            ++ptr;
            --length;
            if(length == 0) return true;
        }

        // Failed to match, so write the skipped characters to storage:
        while(chars != ptr) *out++ = *chars++;

        return false;
    }

    template <typename InputIterator, typename OutputIterator>
    std::string read_bom(InputIterator& begin, InputIterator end,
            OutputIterator out)
    {
        if(begin == end) return "";

        const char* utf8 = "\xef\xbb\xbf" ;
        const char* utf32be = "\0\0\xfe\xff";
        const char* utf32le = "\xff\xfe\0\0";

        unsigned char c = *begin;
        switch(c)
        {
        case 0xEF: { // UTF-8
            return check_bom(begin, end, out, utf8, 3) ? "UTF-8" : "";
        }
        case 0xFF: // UTF-16/UTF-32 little endian
            return !check_bom(begin, end, out, utf32le, 2) ? "" :
                check_bom(begin, end, out, utf32le + 2, 2) ? "UTF-32" : "UTF-16";
        case 0: // UTF-32 big endian
            return check_bom(begin, end, out, utf32be, 4) ? "UTF-32" : "";
        case 0xFE: // UTF-16 big endian
            return check_bom(begin, end, out, utf32be + 2, 2) ? "UTF-16" : "";
        default:
            return "";
        }
    }

    // Copy a string, converting mac and windows style newlines to unix
    // newlines.

    template <typename InputIterator, typename OutputIterator>
    void normalize(InputIterator begin, InputIterator end,
            OutputIterator out)
    {
        std::string encoding = read_bom(begin, end, out);

        if(encoding != "UTF-8" && encoding != "")
        throw load_error(encoding +
            " is not supported. Please use UTF-8.");

        while(begin != end) {
            if(*begin == '\r') {
                *out++ = '\n';
                ++begin;
                if(begin != end && *begin == '\n') ++begin;
            }
            else {
                *out++ = *begin++;
            }
        }
    }

    file_ptr load(fs::path const& filename, unsigned qbk_version)
    {
        boost::unordered_map<fs::path, file_ptr>::iterator pos
            = files.find(filename);

        if (pos == files.end())
        {
            fs::ifstream in(filename, std::ios_base::in);

            if (!in)
                throw load_error("Could not open input file.");

            // Turn off white space skipping on the stream
            in.unsetf(std::ios::skipws);

            std::string source;
            normalize(
                std::istream_iterator<char>(in),
                std::istream_iterator<char>(),
                std::back_inserter(source));

            if (in.bad())
                throw load_error("Error reading input file.");

            bool inserted;

            boost::tie(pos, inserted) = files.emplace(
                filename, new file(filename, source, qbk_version));

            assert(inserted);
        }

        return pos->second;
    }

    std::ostream& operator<<(std::ostream& out, file_position const& x)
    {
        return out << "line: " << x.line << ", column: " << x.column;
    }

    file_position relative_position(
        boost::string_ref::const_iterator begin,
        boost::string_ref::const_iterator iterator)
    {
        file_position pos;
        boost::string_ref::const_iterator line_begin = begin;

        while (begin != iterator)
        {
            if (*begin == '\r')
            {
                ++begin;
                ++pos.line;
                line_begin = begin;
            }
            else if (*begin == '\n')
            {
                ++begin;
                ++pos.line;
                line_begin = begin;
                if (begin == iterator) break;
                if (*begin == '\r')
                {
                    ++begin;
                    line_begin = begin;
                }
            }
            else
            {
                ++begin;
            }
        }

        pos.column = iterator - line_begin + 1;
        return pos;
    }

    file_position file::position_of(boost::string_ref::const_iterator iterator) const
    {
        return relative_position(source().begin(), iterator);
    }

    // Mapped files.

    struct mapped_file_section
    {
        enum section_types {
            normal,
            empty,
            indented
        };
    
        std::string::size_type original_pos;
        std::string::size_type our_pos;
        section_types section_type;

        mapped_file_section(
                std::string::size_type original_pos,
                std::string::size_type our_pos,
                section_types section_type = normal) :
            original_pos(original_pos), our_pos(our_pos),
            section_type(section_type) {}
    };

    struct mapped_section_original_cmp
    {
        bool operator()(mapped_file_section const& x,
                mapped_file_section const& y)
        {
            return x.original_pos < y.original_pos;
        }

        bool operator()(mapped_file_section const& x,
                std::string::size_type const& y)
        {
            return x.original_pos < y;
        }

        bool operator()(std::string::size_type const& x,
                mapped_file_section const& y)
        {
            return x < y.original_pos;
        }
    };

    struct mapped_section_pos_cmp
    {
        bool operator()(mapped_file_section const& x,
                mapped_file_section const& y)
        {
            return x.our_pos < y.our_pos;
        }

        bool operator()(mapped_file_section const& x,
                std::string::size_type const& y)
        {
            return x.our_pos < y;
        }

        bool operator()(std::string::size_type const& x,
                mapped_file_section const& y)
        {
            return x < y.our_pos;
        }
    };
    
    struct mapped_file : file
    {
        mapped_file(file_ptr original) :
            file(*original, std::string()),
            original(original), mapped_sections()
        {}

        file_ptr original;
        std::vector<mapped_file_section> mapped_sections;
        
        void add_empty_mapped_file_section(boost::string_ref::const_iterator pos) {
            std::string::size_type original_pos =
                pos - original->source().begin();
        
            if (mapped_sections.empty() ||
                    mapped_sections.back().section_type !=
                        mapped_file_section::empty ||
                    mapped_sections.back().original_pos != original_pos)
            {
                mapped_sections.push_back(mapped_file_section(
                        original_pos, source().size(),
                        mapped_file_section::empty));
            }
        }

        void add_mapped_file_section(boost::string_ref::const_iterator pos) {
            mapped_sections.push_back(mapped_file_section(
                pos - original->source().begin(), source().size()));
        }

        void add_indented_mapped_file_section(boost::string_ref::const_iterator pos)
        {
            mapped_sections.push_back(mapped_file_section(
                pos - original->source().begin(), source().size(),
                mapped_file_section::indented));
        }

        std::string::size_type to_original_pos(
            std::vector<mapped_file_section>::const_iterator section,
            std::string::size_type pos) const
        {
            switch (section->section_type) {
                case mapped_file_section::normal:
                    return pos - section->our_pos + section->original_pos;

                case mapped_file_section::empty:
                    return section->original_pos;

                case mapped_file_section::indented: {
                    // Will contain the start of the current line.
                    boost::string_ref::size_type our_line = section->our_pos;

                    // Will contain the number of lines in the block before
                    // the current line.
                    unsigned newline_count = 0;

                    for(boost::string_ref::size_type i = section->our_pos;
                        i != pos; ++i)
                    {
                        if (source()[i] == '\n') {
                            our_line = i + 1;
                            ++newline_count;
                        }
                    }

                    // The start of the line in the original source.
                    boost::string_ref::size_type original_line =
                        section->original_pos;
                    
                    while(newline_count > 0) {
                        if (original->source()[original_line] == '\n')
                            --newline_count;
                        ++original_line;
                    }

                    // The start of line content (i.e. after indentation).
                    our_line = skip_indentation(source(), our_line);

                    // The position is in the middle of indentation, so
                    // just return the start of the whitespace, which should
                    // be good enough.
                    if (our_line > pos) return original_line;

                    original_line =
                        skip_indentation(original->source(), original_line);

                    // Confirm that we are actually in the same position.
                    assert(original->source()[original_line] ==
                        source()[our_line]);

                    // Calculate the position
                    return original_line + (pos - our_line);
                }
                default:
                    assert(false);
                    return section->original_pos;
            }
        }
        
        std::vector<mapped_file_section>::const_iterator find_section(
            boost::string_ref::const_iterator pos) const
        {
            std::vector<mapped_file_section>::const_iterator section =
                boost::upper_bound(mapped_sections,
                    std::string::size_type(pos - source().begin()),
                    mapped_section_pos_cmp());
            assert(section != mapped_sections.begin());
            --section;

            return section;
        }

        virtual file_position position_of(boost::string_ref::const_iterator) const;

    private:

        static std::string::size_type skip_indentation(
                boost::string_ref src, std::string::size_type i)
        {
            while (i != src.size() && (src[i] == ' ' || src[i] == '\t')) ++i;
            return i;
        }

    };

    namespace {
        std::list<mapped_file> mapped_files;
    }

    struct mapped_file_builder_data
    {
        mapped_file_builder_data() { reset(); }
        void reset() { new_file.reset(); }
    
        boost::intrusive_ptr<mapped_file> new_file;
    };

    mapped_file_builder::mapped_file_builder() : data(0) {}
    mapped_file_builder::~mapped_file_builder() { delete data; }

    void mapped_file_builder::start(file_ptr f)
    {
        if (!data) {
            data = new mapped_file_builder_data;
        }

        assert(!data->new_file);
        data->new_file = new mapped_file(f);
    }

    file_ptr mapped_file_builder::release()
    {
        file_ptr r = data->new_file;
        data->reset();
        return r;
    }

    void mapped_file_builder::clear()
    {
        data->reset();
    }
    
    bool mapped_file_builder::empty() const
    {
        return data->new_file->source().empty();
    }

    mapped_file_builder::pos mapped_file_builder::get_pos() const
    {
        return data->new_file->source().size();
    }
    
    void mapped_file_builder::add_at_pos(boost::string_ref x, iterator pos)
    {
        data->new_file->add_empty_mapped_file_section(pos);
        data->new_file->source_.append(x.begin(), x.end());
    }

    void mapped_file_builder::add(boost::string_ref x)
    {
        data->new_file->add_mapped_file_section(x.begin());
        data->new_file->source_.append(x.begin(), x.end());
    }

    void mapped_file_builder::add(mapped_file_builder const& x)
    {
        add(x, 0, x.data->new_file->source_.size());
    }

    void mapped_file_builder::add(mapped_file_builder const& x,
            pos begin, pos end)
    {
        assert(data->new_file->original == x.data->new_file->original);
        assert(begin <= x.data->new_file->source_.size());
        assert(end <= x.data->new_file->source_.size());

        if (begin != end)
        {
            std::vector<mapped_file_section>::const_iterator start =
                x.data->new_file->find_section(
                    x.data->new_file->source().begin() + begin);
    
            std::string::size_type size = data->new_file->source_.size();
    
            data->new_file->mapped_sections.push_back(mapped_file_section(
                    x.data->new_file->to_original_pos(start, begin),
                    size, start->section_type));
    
            for (++start; start != x.data->new_file->mapped_sections.end() &&
                    start->our_pos < end; ++start)
            {
                data->new_file->mapped_sections.push_back(mapped_file_section(
                    start->original_pos, start->our_pos - begin + size,
                    start->section_type));
            }
    
            data->new_file->source_.append(
                x.data->new_file->source_.begin() + begin,
            x.data->new_file->source_.begin() + end);
        }
    }

    boost::string_ref::size_type indentation_count(boost::string_ref x)
    {
        unsigned count = 0;

        for(boost::string_ref::const_iterator begin = x.begin(), end = x.end();
            begin != end; ++begin)
        {
            switch(*begin)
            {
            case ' ':
                ++count;
                break;
            case '\t':
                // hardcoded tab to 4 for now
                count = count - (count % 4) + 4;
                break;
            default:
                assert(false);
            }
        }

        return count;
    }

    void mapped_file_builder::unindent_and_add(boost::string_ref x)
    {
        // I wanted to do everything using a string_ref, but unfortunately
        // they don't have all the overloads used in here. So...
        std::string const program(x.begin(), x.end());

        // Erase leading blank lines and newlines:
        std::string::size_type start = program.find_first_not_of(" \t\r\n");
        if (start == std::string::npos) return;

        start = program.find_last_of("\r\n", start);
        start = start == std::string::npos ? 0 : start + 1;

        assert(start < program.size());

        // Get the first line indentation
        std::string::size_type indent = program.find_first_not_of(" \t", start) - start;
        boost::string_ref::size_type full_indent = indentation_count(
            boost::string_ref(&program[start], indent));

        std::string::size_type pos = start;

        // Calculate the minimum indent from the rest of the lines
        // Detecting a mix of spaces and tabs.
        while (std::string::npos != (pos = program.find_first_of("\r\n", pos)))
        {
            pos = program.find_first_not_of("\r\n", pos);
            if (std::string::npos == pos) break;

            std::string::size_type n = program.find_first_not_of(" \t", pos);
            if (n == std::string::npos) break;

            char ch = program[n];
            if (ch == '\r' || ch == '\n') continue; // ignore empty lines

            indent = (std::min)(indent, n-pos);
            full_indent = (std::min)(full_indent, indentation_count(
                boost::string_ref(&program[pos], n-pos)));
        }

        // Detect if indentation is mixed.
        bool mixed_indentation = false;
        boost::string_ref first_indent(&program[start], indent);
        pos = start;

        while (std::string::npos != (pos = program.find_first_of("\r\n", pos)))
        {
            pos = program.find_first_not_of("\r\n", pos);
            if (std::string::npos == pos) break;

            std::string::size_type n = program.find_first_not_of(" \t", pos);
            if (n == std::string::npos || n-pos < indent) continue;

            if (boost::string_ref(&program[pos], indent) != first_indent) {
                mixed_indentation = true;
                break;
            }
        }

        // Trim white spaces from column 0..indent
        std::string unindented_program;
        std::string::size_type copy_start = start;
        pos = start;

        do {
            if (std::string::npos == (pos = program.find_first_not_of("\r\n", pos)))
                break;

            unindented_program.append(program.begin() + copy_start, program.begin() + pos);
            copy_start = pos;

            // Find the end of the indentation.
            std::string::size_type next = program.find_first_not_of(" \t", pos);
            if (next == std::string::npos) next = program.size();

            if (mixed_indentation)
            {
                unsigned length = indentation_count(boost::string_ref(
                    &program[pos], next - pos));

                if (length > full_indent) {
                    std::string new_indentation(length - full_indent, ' ');
                    unindented_program.append(new_indentation);
                }

                copy_start = next;
            }
            else
            {
                copy_start = (std::min)(pos + indent, next);
            }

            pos = next;
        } while (std::string::npos !=
            (pos = program.find_first_of("\r\n", pos)));

        unindented_program.append(program.begin() + copy_start, program.end());

        data->new_file->add_indented_mapped_file_section(x.begin());
        data->new_file->source_.append(unindented_program);
    }

    file_position mapped_file::position_of(boost::string_ref::const_iterator pos) const
    {
        return original->position_of(original->source().begin() +
            to_original_pos(find_section(pos), pos - source().begin()));
    }
}