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

#include <cstring>

#include <openssl/evp.h>

#include "lix/libutil/args.hh"
#include "lix/libutil/hash.hh"
#include "lix/libutil/archive.hh"
#include "lix/libutil/charptr-cast.hh"
#include "lix/libutil/logging.hh"
#include "lix/libutil/split.hh"
#include "lix/libutil/strings.hh"

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

namespace nix {

const std::set<std::string> hashTypes = {"md5", "sha1", "sha256", "sha512"};

const std::string base16Chars = "0123456789abcdef";


static std::string printHash16(const Hash & hash)
{
    std::string buf;
    buf.reserve(hash.hashSize * 2);
    for (unsigned int i = 0; i < hash.hashSize; i++) {
        buf.push_back(base16Chars[hash.hash[i] >> 4]);
        buf.push_back(base16Chars[hash.hash[i] & 0x0f]);
    }
    return buf;
}

std::string Hash::to_string(HashFormat format, bool includeType) const
{
    std::string s;
    if (format == HashFormat::SRI || includeType) {
        s += printHashType(type);
        s += format == HashFormat::SRI ? '-' : ':';
    }
    switch (format) {
    case HashFormat::Base16:
        s += printHash16(*this);
        break;
    case HashFormat::Base32:
        s += base32EncodeStr(std::string_view(charptr_cast<const char *>(hash), hashSize));
        break;
    case HashFormat::Base64:
    case HashFormat::SRI:
        s += base64Encode(std::string_view(charptr_cast<const char *>(hash), hashSize));
        break;
    }
    return s;
}

Hash Hash::dummy(HashType::SHA256);

Hash Hash::parseSRI(std::string_view original) {
    auto rest = original;

    // Parse the has type before the separater, if there was one.
    auto hashRaw = splitPrefixTo(rest, '-');
    if (!hashRaw)
        throw BadHash("hash '%s' is not SRI", original);
    HashType parsedType = parseHashType(*hashRaw);

    return Hash(rest, parsedType, true);
}

// Mutates the string to eliminate the prefixes when found
static std::pair<std::optional<HashType>, bool> getParsedTypeAndSRI(std::string_view & rest)
{
    bool isSRI = false;

    // Parse the hash type before the separator, if there was one.
    std::optional<HashType> optParsedType;
    {
        auto hashRaw = splitPrefixTo(rest, ':');

        if (!hashRaw) {
            hashRaw = splitPrefixTo(rest, '-');
            if (hashRaw)
                isSRI = true;
        }
        if (hashRaw)
            optParsedType = parseHashType(*hashRaw);
    }

    return {optParsedType, isSRI};
}

Hash Hash::parseAnyPrefixed(std::string_view original)
{
    auto rest = original;
    auto [optParsedType, isSRI] = getParsedTypeAndSRI(rest);

    // Either the string or user must provide the type, if they both do they
    // must agree.
    if (!optParsedType)
        throw BadHash("hash '%s' does not include a type", rest);

    return Hash(rest, *optParsedType, isSRI);
}

Hash Hash::parseAny(std::string_view original, std::optional<HashType> optType)
{
    auto rest = original;
    auto [optParsedType, isSRI] = getParsedTypeAndSRI(rest);

    // Either the string or user must provide the type, if they both do they
    // must agree.
    if (!optParsedType && !optType)
        throw BadHash("hash '%s' does not include a type, nor is the type otherwise known from context", rest);
    else if (optParsedType && optType && *optParsedType != *optType)
        throw BadHash("hash '%s' should have type '%s'", original, printHashType(*optType));

    HashType hashType = optParsedType ? *optParsedType : *optType;
    return Hash(rest, hashType, isSRI);
}

Hash Hash::parseNonSRIUnprefixed(std::string_view s, HashType type)
{
    return Hash(s, type, false);
}

Hash::Hash(std::string_view rest, HashType type, bool isSRI)
    : Hash(type)
{
    if (!isSRI && rest.size() == base16Len()) {

        auto parseHexDigit = [&](char c) {
            if (c >= '0' && c <= '9') return c - '0';
            if (c >= 'A' && c <= 'F') return c - 'A' + 10;
            if (c >= 'a' && c <= 'f') return c - 'a' + 10;
            throw BadHash("invalid base-16 hash '%s'", rest);
        };

        for (unsigned int i = 0; i < hashSize; i++) {
            const size_t j = i << 1;
            hash[i] = parseHexDigit(rest[j]) << 4 | parseHexDigit(rest[j + 1]);
        }
    }

    else if (!isSRI && rest.size() == base32Len()) {
        auto d = base32Decode(rest);
        memcpy(hash, d.data(), hashSize);
    }

    else if (isSRI || rest.size() == base64Len()) {
        auto d = base64Decode(rest);
        if (d.size() != hashSize)
            throw BadHash("invalid %s hash '%s'", isSRI ? "SRI" : "base-64", rest);
        assert(hashSize);
        memcpy(hash, d.data(), hashSize);
    }

    else
        throw BadHash("hash '%s' has wrong length for hash type '%s'", rest, printHashType(this->type));
}

Hash newHashAllowEmpty(std::string_view hashStr, std::optional<HashType> ht)
{
    if (hashStr.empty()) {
        if (!ht)
            throw BadHash("empty hash requires explicit hash type");
        Hash h(*ht);
        printTaggedWarning("found empty hash, assuming '%s'", h.to_string(HashFormat::SRI, true));
        return h;
    } else
        return Hash::parseAny(hashStr, ht);
}


static detail::EvpMdCtxPtr start(HashType ht)
{
    detail::EvpMdCtxPtr ctx(EVP_MD_CTX_new(), &EVP_MD_CTX_free);
    if (!ctx) {
        throw Error("failed to create message digest context");
    }

    int ok;
    switch (ht) {
    case HashType::MD5: ok = EVP_DigestInit_ex(ctx.get(), EVP_md5(), NULL); break;
    case HashType::SHA1: ok = EVP_DigestInit_ex(ctx.get(), EVP_sha1(), NULL); break;
    case HashType::SHA256: ok = EVP_DigestInit_ex(ctx.get(), EVP_sha256(), NULL); break;
    case HashType::SHA512: ok = EVP_DigestInit_ex(ctx.get(), EVP_sha512(), NULL); break;
    }
    if (!ok) {
        throw Error("failed to initialize message digest");
    }

    return ctx;
}


static void update(detail::EvpMdCtxPtr & ctx, std::string_view data)
{
    if (!EVP_DigestUpdate(ctx.get(), data.data(), data.size())) {
        throw Error("failed to update message digest with %zu bytes", data.size());
    }
}

static void finish(detail::EvpMdCtxPtr & ctx, unsigned char * hash)
{
    if (!EVP_DigestFinal_ex(ctx.get(), hash, NULL)) {
        throw Error("failed to finalize message digest");
    }
}

Hash hashString(HashType ht, std::string_view s)
{
    Hash hash(ht);
    detail::EvpMdCtxPtr ctx = start(ht);
    update(ctx, s);
    finish(ctx, hash.hash);
    return hash;
}

Hash hashFile(HashType ht, const Path & path)
{
    HashSink sink(ht);
    sink << readFileSource(path);
    return sink.finish().first;
}

HashSink::HashSink(HashType ht) : ht(ht), ctx(start(ht)), bytes(0) {}

void HashSink::writeUnbuffered(std::string_view data)
{
    bytes += data.size();
    update(ctx, data);
}

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

HashResult HashSink::currentHash()
{
    flush();
    detail::EvpMdCtxPtr ctx2 = start(ht);
    if (!EVP_MD_CTX_copy(ctx2.get(), ctx.get())) {
        throw Error("failed to copy message digest");
    }
    Hash hash(ht);
    nix::finish(ctx2, hash.hash);
    return HashResult(hash, bytes);
}

HashResult hashPath(HashType ht, const PreparedDump & path)
{
    HashSink sink(ht);
    sink << path.dump();
    return sink.finish();
}

Hash compressHash(const Hash & hash, size_t newSize)
{
    Hash h(newSize, hash.type);

    for (const auto [idx, c] : enumerate(hash.as_span())) {
        h.hash[idx % newSize] ^= c;
    }

    return h;
}

std::optional<HashType> parseHashTypeOpt(std::string_view s)
{
    if (s == "md5") {
        return HashType::MD5;
    }
    if (s == "sha1") {
        return HashType::SHA1;
    }
    if (s == "sha256") {
        return HashType::SHA256;
    }
    if (s == "sha512") {
        return HashType::SHA512;
    }

    return std::nullopt;
}

HashType parseHashType(std::string_view s)
{
    if (auto opt_h = parseHashTypeOpt(s)) {
        return *opt_h;
    }

    throw UsageError("unknown hash algorithm '%1%'", s);
}

std::string_view printHashType(HashType type)
{
    switch (type) {
    case HashType::MD5:
        return "md5";
    case HashType::SHA1:
        return "sha1";
    case HashType::SHA256:
        return "sha256";
    case HashType::SHA512:
        return "sha512";
    default:
        // illegal hash type enum value internally, as opposed to external input
        // which should be validated with nice error message.
        assert(false);
    }
}

}