summary refs log tree commit diff
path: root/nix/libutil/hash.cc
blob: 20d2e4b7240f5cc0dc65f5510c6fcfdfe32c95d6 (plain) (blame)
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
#include "config.h"

#include <iostream>
#include <cstring>

#include "hash.hh"
#include "archive.hh"
#include "util.hh"

#include <sys/types.h>
#include <sys/stat.h>
#include <fcntl.h>


namespace nix {


Hash::Hash()
{
    type = htUnknown;
    hashSize = 0;
    memset(hash, 0, maxHashSize);
}


Hash::Hash(HashType type)
{
    this->type = type;
    hashSize = gcry_md_get_algo_dlen(type);

    if (hashSize == 0) throw Error("unknown hash type");
    assert(hashSize <= maxHashSize);
    memset(hash, 0, maxHashSize);
}


bool Hash::operator == (const Hash & h2) const
{
    if (hashSize != h2.hashSize) return false;
    for (unsigned int i = 0; i < hashSize; i++)
        if (hash[i] != h2.hash[i]) return false;
    return true;
}


bool Hash::operator != (const Hash & h2) const
{
    return !(*this == h2);
}


bool Hash::operator < (const Hash & h) const
{
    for (unsigned int i = 0; i < hashSize; i++) {
        if (hash[i] < h.hash[i]) return true;
        if (hash[i] > h.hash[i]) return false;
    }
    return false;
}


const string base16Chars = "0123456789abcdef";


string printHash(const Hash & hash)
{
    char buf[hash.hashSize * 2];
    for (unsigned int i = 0; i < hash.hashSize; i++) {
        buf[i * 2] = base16Chars[hash.hash[i] >> 4];
        buf[i * 2 + 1] = base16Chars[hash.hash[i] & 0x0f];
    }
    return string(buf, hash.hashSize * 2);
}


Hash parseHash(HashType ht, const string & s)
{
    Hash hash(ht);
    if (s.length() != hash.hashSize * 2)
        throw Error(format("invalid hash `%1%'") % s);
    for (unsigned int i = 0; i < hash.hashSize; i++) {
        string s2(s, i * 2, 2);
        if (!isxdigit(s2[0]) || !isxdigit(s2[1]))
            throw Error(format("invalid hash `%1%'") % s);
        std::istringstream str(s2);
        int n;
        str >> std::hex >> n;
        hash.hash[i] = n;
    }
    return hash;
}


unsigned int hashLength32(const Hash & hash)
{
    return (hash.hashSize * 8 - 1) / 5 + 1;
}


// omitted: E O U T
const string base32Chars = "0123456789abcdfghijklmnpqrsvwxyz";


string printHash32(const Hash & hash)
{
    Hash hash2(hash);
    unsigned int len = hashLength32(hash);

    string s;
    s.reserve(len);

    for (int n = len - 1; n >= 0; n--) {
        unsigned int b = n * 5;
        unsigned int i = b / 8;
        unsigned int j = b % 8;
        unsigned char c =
            (hash.hash[i] >> j)
            | (i >= hash.hashSize - 1 ? 0 : hash.hash[i + 1] << (8 - j));
        s.push_back(base32Chars[c & 0x1f]);
    }

    return s;
}


string printHash16or32(const Hash & hash)
{
    return hash.type == htMD5 ? printHash(hash) : printHash32(hash);
}


Hash parseHash32(HashType ht, const string & s)
{
    Hash hash(ht);
    unsigned int len = hashLength32(ht);
    assert(s.size() == len);

    for (unsigned int n = 0; n < len; ++n) {
        char c = s[len - n - 1];
        unsigned char digit;
        for (digit = 0; digit < base32Chars.size(); ++digit) /* !!! slow */
            if (base32Chars[digit] == c) break;
        if (digit >= 32)
            throw Error(format("invalid base-32 hash '%1%'") % s);
        unsigned int b = n * 5;
        unsigned int i = b / 8;
        unsigned int j = b % 8;
        hash.hash[i] |= digit << j;
        if (i < hash.hashSize - 1) hash.hash[i + 1] |= digit >> (8 - j);
    }

    return hash;
}


Hash parseHash16or32(HashType ht, const string & s)
{
    Hash hash(ht);
    if (s.size() == hash.hashSize * 2)
        /* hexadecimal representation */
        hash = parseHash(ht, s);
    else if (s.size() == hashLength32(hash))
        /* base-32 representation */
        hash = parseHash32(ht, s);
    else
        throw Error(format("hash `%1%' has wrong length for hash type `%2%'")
            % s % printHashType(ht));
    return hash;
}


bool isHash(const string & s)
{
    if (s.length() != 32) return false;
    for (int i = 0; i < 32; i++) {
        char c = s[i];
        if (!((c >= '0' && c <= '9') ||
              (c >= 'a' && c <= 'f')))
            return false;
    }
    return true;
}

/* The "hash context".  */
struct Ctx
{
  /* This copy constructor is needed in 'HashSink::currentHash()' where we
     expect the copy of a 'Ctx' object to yield a truly different context.  */
  Ctx(Ctx &ref)
  {
    if (ref.md_handle == NULL)
      md_handle = NULL;
    else
      gcry_md_copy (&md_handle, ref.md_handle);
  }

  /* Make sure 'md_handle' is always initialized.  */
  Ctx(): md_handle (NULL) { };

  gcry_md_hd_t md_handle;
};


static void start(HashType ht, Ctx & ctx)
{
    gcry_error_t err;

    err = gcry_md_open (&ctx.md_handle, ht, 0);
    assert (err == GPG_ERR_NO_ERROR);
}


static void update(HashType ht, Ctx & ctx,
    const unsigned char * bytes, unsigned int len)
{
    gcry_md_write (ctx.md_handle, bytes, len);
}


static void finish(HashType ht, Ctx & ctx, unsigned char * hash)
{
    memcpy (hash, gcry_md_read (ctx.md_handle, ht),
	    gcry_md_get_algo_dlen (ht));
    gcry_md_close (ctx.md_handle);
    ctx.md_handle = NULL;
}


Hash hashString(HashType ht, const string & s)
{
    Ctx ctx;
    Hash hash(ht);
    start(ht, ctx);
    update(ht, ctx, (const unsigned char *) s.data(), s.length());
    finish(ht, ctx, hash.hash);
    return hash;
}


Hash hashFile(HashType ht, const Path & path)
{
    Ctx ctx;
    Hash hash(ht);
    start(ht, ctx);

    AutoCloseFD fd = open(path.c_str(), O_RDONLY);
    if (fd == -1) throw SysError(format("opening file `%1%'") % path);

    unsigned char buf[8192];
    ssize_t n;
    while ((n = read(fd, buf, sizeof(buf)))) {
        checkInterrupt();
        if (n == -1) throw SysError(format("reading file `%1%'") % path);
        update(ht, ctx, buf, n);
    }

    finish(ht, ctx, hash.hash);
    return hash;
}


HashSink::HashSink(HashType ht) : ht(ht)
{
    ctx = new Ctx;
    bytes = 0;
    start(ht, *ctx);
}

HashSink::~HashSink()
{
    bufPos = 0;
    delete ctx;
}

void HashSink::write(const unsigned char * data, size_t len)
{
    bytes += len;
    update(ht, *ctx, data, len);
}

HashResult HashSink::finish()
{
    flush();
    Hash hash(ht);
    nix::finish(ht, *ctx, hash.hash);
    return HashResult(hash, bytes);
}

HashResult HashSink::currentHash()
{
    flush();
    Ctx ctx2 = *ctx;
    Hash hash(ht);
    nix::finish(ht, ctx2, hash.hash);
    return HashResult(hash, bytes);
}


HashResult hashPath(
    HashType ht, const Path & path, PathFilter & filter)
{
    HashSink sink(ht);
    dumpPath(path, sink, filter);
    return sink.finish();
}


Hash compressHash(const Hash & hash, unsigned int newSize)
{
    Hash h;
    h.hashSize = newSize;
    for (unsigned int i = 0; i < hash.hashSize; ++i)
        h.hash[i % newSize] ^= hash.hash[i];
    return h;
}


HashType parseHashType(const string & s)
{
    if (s == "md5") return htMD5;
    else if (s == "sha1") return htSHA1;
    else if (s == "sha256") return htSHA256;
    else if (s == "sha512") return htSHA512;
    else return htUnknown;
}


string printHashType(HashType ht)
{
    if (ht == htMD5) return "md5";
    else if (ht == htSHA1) return "sha1";
    else if (ht == htSHA256) return "sha256";
    else if (ht == htSHA512) return "sha512";
    else throw Error("cannot print unknown hash type");
}


}