test: Add 2, update 4 and remove 1 files

src/dshash.cpp

@@ -0,0 +1,324 @@
+#include "dshash.hpp"
+#include <fstream>
+#include <sstream>
+#include <iomanip>
+#include <cstring>
+#include <algorithm>
+
+const uint32_t DSHash::K[64] = {
+    0x428a2f98, 0x71374491, 0xb5c0fbcf, 0xe9b5dba5,
+    0x3956c25b, 0x59f111f1, 0x923f82a4, 0xab1c5ed5,
+    0xd807aa98, 0x12835b01, 0x243185be, 0x550c7dc3,
+    0x72be5d74, 0x80deb1fe, 0x9bdc06a7, 0xc19bf174,
+    0xe49b69c1, 0xefbe4786, 0x0fc19dc6, 0x240ca1cc,
+    0x2de92c6f, 0x4a7484aa, 0x5cb0a9dc, 0x76f988da,
+    0x983e5152, 0xa831c66d, 0xb00327c8, 0xbf597fc7,
+    0xc6e00bf3, 0xd5a79147, 0x06ca6351, 0x14292967,
+    0x27b70a85, 0x2e1b2138, 0x4d2c6dfc, 0x53380d13,
+    0x650a7354, 0x766a0abb, 0x81c2c92e, 0x92722c85,
+    0xa2bfe8a1, 0xa81a664b, 0xc24b8b70, 0xc76c51a3,
+    0xd192e819, 0xd6990624, 0xf40e3585, 0x106aa070,
+    0x19a4c116, 0x1e376c08, 0x2748774c, 0x34b0bcb5,
+    0x391c0cb3, 0x4ed8aa4a, 0x5b9cca4f, 0x682e6ff3,
+    0x748f82ee, 0x78a5636f, 0x84c87814, 0x8cc70208,
+    0x90befffa, 0xa4506ceb, 0xbef9a3f7, 0xc67178f2
+};
+
+static inline uint32_t rotr(uint32_t x, uint32_t n) {
+    return (x >> n) | (x << (32 - n));
+}
+
+static inline uint32_t ch(uint32_t x, uint32_t y, uint32_t z) {
+    return (x & y) ^ (~x & z);
+}
+
+static inline uint32_t maj(uint32_t x, uint32_t y, uint32_t z) {
+    return (x & y) ^ (x & z) ^ (y & z);
+}
+
+static inline uint32_t sigma0(uint32_t x) {
+    return rotr(x, 2) ^ rotr(x, 13) ^ rotr(x, 22);
+}
+
+static inline uint32_t sigma1(uint32_t x) {
+    return rotr(x, 6) ^ rotr(x, 11) ^ rotr(x, 25);
+}
+
+static inline uint32_t gamma0(uint32_t x) {
+    return rotr(x, 7) ^ rotr(x, 18) ^ (x >> 3);
+}
+
+static inline uint32_t gamma1(uint32_t x) {
+    return rotr(x, 17) ^ rotr(x, 19) ^ (x >> 10);
+}
+
+void DSHash::processBlock(const uint8_t* block) {
+    uint32_t w[64];
+    
+    for (int i = 0; i < 16; i++) {
+        w[i] = (block[i * 4] << 24) |
+               (block[i * 4 + 1] << 16) |
+               (block[i * 4 + 2] << 8) |
+               (block[i * 4 + 3]);
+    }
+    
+    for (int i = 16; i < 64; i++) {
+        w[i] = gamma1(w[i - 2]) + w[i - 7] + gamma0(w[i - 15]) + w[i - 16];
+    }
+    
+    uint32_t a = h[0];
+    uint32_t b = h[1];
+    uint32_t c = h[2];
+    uint32_t d = h[3];
+    uint32_t e = h[4];
+    uint32_t f = h[5];
+    uint32_t g = h[6];
+    uint32_t hh = h[7];
+    
+    for (int i = 0; i < 64; i++) {
+        uint32_t t1 = hh + sigma1(e) + ch(e, f, g) + K[i] + w[i];
+        uint32_t t2 = sigma0(a) + maj(a, b, c);
+        hh = g;
+        g = f;
+        f = e;
+        e = d + t1;
+        d = c;
+        c = b;
+        b = a;
+        a = t1 + t2;
+    }
+    
+    h[0] += a;
+    h[1] += b;
+    h[2] += c;
+    h[3] += d;
+    h[4] += e;
+    h[5] += f;
+    h[6] += g;
+    h[7] += hh;
+}
+
+void DSHash::update(const uint8_t* data, size_t length) {
+    if (finalized) return;
+    
+    total_length += length;
+    
+    buffer.insert(buffer.end(), data, data + length);
+    
+    while (buffer.size() >= BLOCK_SIZE) {
+        processBlock(buffer.data());
+        buffer.erase(buffer.begin(), buffer.begin() + BLOCK_SIZE);
+    }
+}
+
+void DSHash::update(const std::string& str) {
+    update(reinterpret_cast<const uint8_t*>(str.data()), str.size());
+}
+
+void DSHash::padMessage() {
+    uint64_t bit_length = total_length * 8;
+    
+    buffer.push_back(0x80);
+    
+    while ((buffer.size() % 64) != 56) {
+        buffer.push_back(0x00);
+    }
+    
+    for (int i = 7; i >= 0; i--) {
+        buffer.push_back((bit_length >> (i * 8)) & 0xff);
+    }
+}
+
+tHash DSHash::finalize() {
+    if (finalized) {
+        tHash result;
+        for (int i = 0; i < 8; i++) {
+            result[i * 4] = (h[i] >> 24) & 0xff;
+            result[i * 4 + 1] = (h[i] >> 16) & 0xff;
+            result[i * 4 + 2] = (h[i] >> 8) & 0xff;
+            result[i * 4 + 3] = h[i] & 0xff;
+        }
+        return result;
+    }
+    
+    padMessage();
+    
+    while (!buffer.empty()) {
+        processBlock(buffer.data());
+        buffer.erase(buffer.begin(), buffer.begin() + BLOCK_SIZE);
+    }
+    
+    finalized = true;
+    
+    tHash result;
+    for (int i = 0; i < 8; i++) {
+        result[i * 4] = (h[i] >> 24) & 0xff;
+        result[i * 4 + 1] = (h[i] >> 16) & 0xff;
+        result[i * 4 + 2] = (h[i] >> 8) & 0xff;
+        result[i * 4 + 3] = h[i] & 0xff;
+    }
+    
+    return result;
+}
+
+std::string DSHash::hashString(const std::string& str) {
+    h[0] = 0x6a09e667;
+    h[1] = 0xbb67ae85;
+    h[2] = 0x3c6ef372;
+    h[3] = 0xa54ff53a;
+    h[4] = 0x510e527f;
+    h[5] = 0x9b05688c;
+    h[6] = 0x1f83d9ab;
+    h[7] = 0x5be0cd19;
+    buffer.clear();
+    total_length = 0;
+    finalized = false;
+    
+    update(str);
+    tHash hash = finalize();
+    
+    std::stringstream ss;
+    for (uint8_t byte : hash) {
+        ss << std::hex << std::setfill('0') << std::setw(2) << static_cast<int>(byte);
+    }
+    return ss.str();
+}
+
+std::string DSHash::hashFile(const std::filesystem::path& filepath) {
+    std::ifstream file(filepath, std::ios::binary);
+    if (!file) {
+        throw std::runtime_error("Cannot open file: " + filepath.string());
+    }
+    
+    h[0] = 0x6a09e667;
+    h[1] = 0xbb67ae85;
+    h[2] = 0x3c6ef372;
+    h[3] = 0xa54ff53a;
+    h[4] = 0x510e527f;
+    h[5] = 0x9b05688c;
+    h[6] = 0x1f83d9ab;
+    h[7] = 0x5be0cd19;
+    buffer.clear();
+    total_length = 0;
+    finalized = false;
+    
+    constexpr size_t BUFFER_SIZE = 8192;
+    char buffer[BUFFER_SIZE];
+    
+    while (file.read(buffer, BUFFER_SIZE) || file.gcount() > 0) {
+        update(reinterpret_cast<const uint8_t*>(buffer), file.gcount());
+    }
+    
+    tHash hash = finalize();
+    
+    std::stringstream ss;
+    for (uint8_t byte : hash) {
+        ss << std::hex << std::setfill('0') << std::setw(2) << static_cast<int>(byte);
+    }
+    return ss.str();
+}
+
+std::string DSHash::hashDirectory(const std::filesystem::path& dirpath) {
+    if (!std::filesystem::is_directory(dirpath)) {
+        throw std::runtime_error("Not a directory: " + dirpath.string());
+    }
+    
+    h[0] = 0x6a09e667;
+    h[1] = 0xbb67ae85;
+    h[2] = 0x3c6ef372;
+    h[3] = 0xa54ff53a;
+    h[4] = 0x510e527f;
+    h[5] = 0x9b05688c;
+    h[6] = 0x1f83d9ab;
+    h[7] = 0x5be0cd19;
+    buffer.clear();
+    total_length = 0;
+    finalized = false;
+    
+    std::vector<std::filesystem::path> paths;
+    
+    for (const auto& entry : std::filesystem::recursive_directory_iterator(dirpath)) {
+        if (entry.is_regular_file()) {
+            paths.push_back(entry.path());
+        }
+    }
+    
+    std::sort(paths.begin(), paths.end());
+    
+    for (const auto& path : paths) {
+        std::string relative = std::filesystem::relative(path, dirpath).string();
+        update(relative);
+        
+        DSHash fileHasher(path);
+        std::string fileHashStr = fileHasher.toString();
+        update(fileHashStr);
+    }
+    
+    tHash hash = finalize();
+    
+    std::stringstream ss;
+    for (uint8_t byte : hash) {
+        ss << std::hex << std::setfill('0') << std::setw(2) << static_cast<int>(byte);
+    }
+    return ss.str();
+}
+
+DSHash::DSHash(const std::filesystem::path& path) {
+    h[0] = 0x6a09e667;
+    h[1] = 0xbb67ae85;
+    h[2] = 0x3c6ef372;
+    h[3] = 0xa54ff53a;
+    h[4] = 0x510e527f;
+    h[5] = 0x9b05688c;
+    h[6] = 0x1f83d9ab;
+    h[7] = 0x5be0cd19;
+    
+    if (std::filesystem::is_regular_file(path)) {
+        hashFile(path);
+    } else if (std::filesystem::is_directory(path)) {
+        hashDirectory(path);
+    } else {
+        throw std::runtime_error("Path is neither file nor directory: " + path.string());
+    }
+}
+
+DSHash::DSHash(const std::string& str) {
+    h[0] = 0x6a09e667;
+    h[1] = 0xbb67ae85;
+    h[2] = 0x3c6ef372;
+    h[3] = 0xa54ff53a;
+    h[4] = 0x510e527f;
+    h[5] = 0x9b05688c;
+    h[6] = 0x1f83d9ab;
+    h[7] = 0x5be0cd19;
+    
+    hashString(str);
+}
+
+std::string DSHash::toString() {
+    if (!finalized) {
+        finalize();
+    }
+    
+    tHash hash = get();
+    std::stringstream ss;
+    for (uint8_t byte : hash) {
+        ss << std::hex << std::setfill('0') << std::setw(2) << static_cast<int>(byte);
+    }
+    return ss.str();
+}
+
+tHash DSHash::get() {
+    if (!finalized) {
+        return finalize();
+    }
+    
+    tHash result;
+    for (int i = 0; i < 8; i++) {
+        result[i * 4] = (h[i] >> 24) & 0xff;
+        result[i * 4 + 1] = (h[i] >> 16) & 0xff;
+        result[i * 4 + 2] = (h[i] >> 8) & 0xff;
+        result[i * 4 + 3] = h[i] & 0xff;
+    }
+    return result;
+}

src/dshash.hpp

@@ -0,0 +1,43 @@
+#ifndef DSHASH_HPP
+#define DSHASH_HPP
+
+#include <array>
+#include <string>
+#include <vector>
+#include <cstdint>
+#include <filesystem>
+
+typedef std::array<uint8_t, 32> tHash;
+
+class DSHash {
+public:    
+    explicit DSHash(const std::filesystem::path& path);
+    explicit DSHash(const std::string& str);
+    
+    std::string toString();
+    tHash get();
+    
+private:
+    std::string hashString(const std::string& str);
+    std::string hashFile(const std::filesystem::path& filepath);
+    std::string hashDirectory(const std::filesystem::path& dirpath);
+
+    void update(const uint8_t* data, size_t length);
+    void update(const std::string& str);
+    tHash finalize();
+
+private:
+    void processBlock(const uint8_t* block);
+    void padMessage();
+    
+    static constexpr size_t BLOCK_SIZE = 64;
+    
+    std::vector<uint8_t> buffer;
+    uint64_t total_length = 0;
+    uint32_t h[8];
+    bool finalized = false;
+    
+    static const uint32_t K[64];
+};
+
+#endif

src/generator.cpp

@@ -5,7 +5,6 @@
 #include <filesystem>
 #include <sstream>
 #include <vector>
-#include "xxhash.hpp"
 #include <sys/stat.h>  // For file permissions
 #include <cstring>     // For strlen
 

src/hash.cpp

@@ -1,33 +1,21 @@
 #include "hash.hpp"
-#define XXH_INLINE_ALL
+#include "dshash.hpp"
-#include "xxhash.hpp"
+#include <filesystem>
-#include <fstream>
-#include <sstream>
-#include <iomanip>
-
-static std::string to_hex64(uint64_t value) {
-    std::ostringstream oss;
-    oss << std::hex << std::setw(16) << std::setfill('0') << value;
-    return oss.str();
-}
 
 std::string hash_data(const std::string& data) {
-    uint64_t h = XXH3_64bits(data.data(), data.size());
+    // Create a DSHash object from string data
-    return to_hex64(h);
+    DSHash hasher(data);
+    return hasher.toString();
 }
 
 std::string hash_file(const std::string& path) {
-    std::ifstream file(path, std::ios::binary);
+    // Check if file exists
-    if (!file) return "";
+    if (!std::filesystem::exists(path)) {
-    XXH64_state_t* state = XXH64_createState();
+        return "";
-    XXH64_reset(state, 0);
-    char buf[4096];
-    while (file) {
-        file.read(buf, sizeof(buf));
-        std::streamsize n = file.gcount();
-        if (n > 0) XXH64_update(state, buf, static_cast<size_t>(n));
     }
-    uint64_t h = XXH64_digest(state);
+    
-    XXH64_freeState(state);
+    // Create a DSHash object from file path
-    return to_hex64(h);
+    std::filesystem::path filepath(path);
+    DSHash hasher(filepath);
+    return hasher.toString();
 } 

test/dehydrate_test_data/generated/_original.cpp

@@ -107,21 +107,6 @@ static bool _recreate_file_(const std::filesystem::path& outpath, uint64_t file_
 bool recreate_tree(std::string destination_folder) {
     namespace fs = std::filesystem;
     bool any_written = false;
-    {
-        // File: subdir/nested.txt
-        fs::path outpath = fs::path(destination_folder) / "subdir/nested.txt";
-        static const char filedata_base64[] = "VGhpcyBmaWxlIGlzIGluIGEgc3ViZGlyZWN0b3J5Lg==";
-
-        // Decode Base64 data
-        size_t decoded_size = (strlen(filedata_base64) * 3) / 4;
-        unsigned char* decoded_data = new unsigned char[decoded_size];
-        size_t actual_size;
-        base64_decode(filedata_base64, strlen(filedata_base64), decoded_data, &actual_size);
-
-        bool file_written = _recreate_file_(outpath, 14153000318456068100ULL, std::filesystem::perms(256), decoded_data, actual_size);
-        delete[] decoded_data;
-        any_written = any_written || file_written;
-    }
     {
         // File: small.txt
         fs::path outpath = fs::path(destination_folder) / "small.txt";
@@ -152,6 +137,21 @@ bool recreate_tree(std::string destination_folder) {
         delete[] decoded_data;
         any_written = any_written || file_written;
     }
+    {
+        // File: test3.sh
+        fs::path outpath = fs::path(destination_folder) / "test3.sh";
+        static const char filedata_base64[] = "IyEvYmluL2Jhc2gKZWNobyAnSGVsbG8gZnJvbSB0ZXN0IHNjcmlwdCcKZXhpdCAw";
+
+        // Decode Base64 data
+        size_t decoded_size = (strlen(filedata_base64) * 3) / 4;
+        unsigned char* decoded_data = new unsigned char[decoded_size];
+        size_t actual_size;
+        base64_decode(filedata_base64, strlen(filedata_base64), decoded_data, &actual_size);
+
+        bool file_written = _recreate_file_(outpath, 14335927320996074478ULL, std::filesystem::perms(488), decoded_data, actual_size);
+        delete[] decoded_data;
+        any_written = any_written || file_written;
+    }
     {
         // File: test1.txt
         fs::path outpath = fs::path(destination_folder) / "test1.txt";
@@ -169,9 +169,9 @@ bool recreate_tree(std::string destination_folder) {
         any_written = any_written || file_written;
     }
     {
-        // File: test3.sh
+        // File: subdir/nested.txt
-        fs::path outpath = fs::path(destination_folder) / "test3.sh";
+        fs::path outpath = fs::path(destination_folder) / "subdir/nested.txt";
-        static const char filedata_base64[] = "IyEvYmluL2Jhc2gKZWNobyAnSGVsbG8gZnJvbSB0ZXN0IHNjcmlwdCcKZXhpdCAw";
+        static const char filedata_base64[] = "VGhpcyBmaWxlIGlzIGluIGEgc3ViZGlyZWN0b3J5Lg==";
 
         // Decode Base64 data
         size_t decoded_size = (strlen(filedata_base64) * 3) / 4;
@@ -179,7 +179,7 @@ bool recreate_tree(std::string destination_folder) {
         size_t actual_size;
         base64_decode(filedata_base64, strlen(filedata_base64), decoded_data, &actual_size);
 
-        bool file_written = _recreate_file_(outpath, 14335927320996074478ULL, std::filesystem::perms(488), decoded_data, actual_size);
+        bool file_written = _recreate_file_(outpath, 14153000318456068100ULL, std::filesystem::perms(256), decoded_data, actual_size);
         delete[] decoded_data;
         any_written = any_written || file_written;
     }