Merge pull request 'feat: implement chunk serialization protocol with -s flag' (#5) from chunk-serialization into main

Reviewed-on: #5
This commit is contained in:
2026-07-04 21:49:55 +02:00
9 changed files with 195 additions and 140 deletions
+11 -2
View File
@@ -17,9 +17,18 @@
#include <dirent.h>
int send_chunk(Client *client, Chunk *chunk, Config *config) {
if (config->use_compression && config->use_chunk_serialization) {
Data *data = chunk_compress(chunk, config->compression_level);
if (config->use_chunk_serialization) {
send_status(client->file_descriptor, STATUS_CHUNK);
Data *data;
if (config->use_compression) {
data = chunk_compress(chunk, config->compression_level);
} else {
for (int i = 0; i < chunk->element_count; i++)
file_load_data(chunk->items[i]);
data = chunk_serialize(chunk);
}
send_data(client->file_descriptor, data->data, data->size);
data_destroy(data);
} else {
for (int i = 0; i < chunk->element_count; i++) {
send_status(client->file_descriptor, STATUS_NEXT);
+61 -14
View File
@@ -1,3 +1,4 @@
#include "chunk.h"
#include "config.h"
#include "data.h"
#include "file.h"
@@ -16,13 +17,36 @@ FileReceive *receive_file_receive(Config *config, int file_descriptor) {
Data *file_data = receive_data(file_descriptor);
if (config->use_compression) {
Data *file_data_uncompressed = data_decompress(file_data);
free(file_data);
data_destroy(file_data);
file_data = file_data_uncompressed;
}
FileReceive *file = file_receive_create(path, file_data);
return file;
}
static void receive_chunk_enqueue(int file_descriptor, Config *config,
PipelineContextReceiver *context) {
Data *chunk_data = receive_data(file_descriptor);
Data *data_to_process = chunk_data;
if (config->use_compression) {
data_to_process = data_decompress(chunk_data);
data_destroy(chunk_data);
}
Chunk *chunk = chunk_deserialize(data_to_process);
data_destroy(data_to_process);
for (int i = 0; i < chunk->element_count; i++) {
FileReceive *file = file_receive_create(chunk->items[i]->path,
chunk->items[i]->data);
chunk->items[i]->path = NULL;
chunk->items[i]->data = NULL;
queue_enqueue_multithreaded(context->queue, file, &context->mutex,
&context->condition_not_empty,
&context->condition_not_full);
}
chunk_destroy(chunk);
}
int receive_thread(void *pipeline_context) {
PipelineContextReceiver *context =
(PipelineContextReceiver *)pipeline_context;
@@ -31,11 +55,17 @@ int receive_thread(void *pipeline_context) {
Config *config = context->config;
mtx_unlock(&context->mutex);
while (receive_status(file_descriptor) == STATUS_NEXT) {
FileReceive *file = receive_file_receive(config, file_descriptor);
queue_enqueue_multithreaded(context->queue, file, &context->mutex,
&context->condition_not_empty,
&context->condition_not_full);
Status status = receive_status(file_descriptor);
while (status == STATUS_NEXT || status == STATUS_CHUNK) {
if (status == STATUS_CHUNK) {
receive_chunk_enqueue(file_descriptor, config, context);
} else {
FileReceive *file = receive_file_receive(config, file_descriptor);
queue_enqueue_multithreaded(context->queue, file, &context->mutex,
&context->condition_not_empty,
&context->condition_not_full);
}
status = receive_status(file_descriptor);
}
mtx_lock(&context->mutex);
context->receiver_done = true;
@@ -68,17 +98,34 @@ int write_thread(void *pipeline_context) {
int receive_files(Config *config, int file_descriptor) {
Status status = receive_status(file_descriptor);
while (status == STATUS_NEXT) {
FileReceive *file = receive_file_receive(config, file_descriptor);
if (config->save_to_disk)
to_disk(path_cat(config->receive_root_directory, file->path),
file->data->data, file->data->size);
file_receive_destroy(file);
// send_status(file_descriptor, STATUS_OK);
while (status == STATUS_NEXT || status == STATUS_CHUNK) {
if (status == STATUS_CHUNK) {
Data *chunk_data = receive_data(file_descriptor);
Data *data_to_process = chunk_data;
if (config->use_compression) {
data_to_process = data_decompress(chunk_data);
data_destroy(chunk_data);
}
Chunk *chunk = chunk_deserialize(data_to_process);
data_destroy(data_to_process);
for (int i = 0; i < chunk->element_count; i++) {
if (config->save_to_disk)
to_disk(path_cat(config->receive_root_directory, chunk->items[i]->path),
chunk->items[i]->data->data, chunk->items[i]->data->size);
}
chunk_destroy(chunk);
} else {
FileReceive *file = receive_file_receive(config, file_descriptor);
if (config->save_to_disk)
to_disk(path_cat(config->receive_root_directory, file->path),
file->data->data, file->data->size);
file_receive_destroy(file);
}
status = receive_status(file_descriptor);
}
if (status != STATUS_FINISHED) {
log_message(LOG_LEVEL_ERROR, "Did not receive FINISHED or NEXT Status");
log_message(LOG_LEVEL_ERROR, "Did not receive FINISHED Status");
send_status(file_descriptor, STATUS_ERROR);
return -1;
}
+110 -106
View File
@@ -4,7 +4,6 @@
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <zstd.h>
#include "chunk.h"
#include "array_list.h"
@@ -99,40 +98,129 @@ Data *chunk_format(Chunk *chunk) {
return chunk_data_create(data, buffer_size);
}
Data *chunk_compress(Chunk *chunk, int compression_level) {
log_message(LOG_LEVEL_DEBUG, "Starting to gather data for chunk compression");
Data *chunk_serialize(Chunk *chunk) {
unsigned long long data_size = 0;
for (int i = 0; i < chunk->element_count; i++) {
data_size += sizeof(unsigned long long);
data_size += sizeof(size_t);
data_size += strlen(chunk->items[i]->path);
data_size += sizeof(unsigned long long);
data_size += sizeof(size_t);
data_size += chunk->items[i]->stats.st_size;
}
Data *data = data_create_empty(data_size);
if (data == NULL) {
log_message(LOG_LEVEL_ERROR,
"Could not allocate memory for chunk compression");
"Could not allocate memory for chunk serialization");
exit(EXIT_FAILURE);
}
char *data_pointer = data->data;
for (int i = 0; i < chunk->element_count; i++) {
// path length
size_t path_len = strlen(chunk->items[i]->path);
memcpy(data_pointer, &path_len, sizeof(size_t));
data_pointer += sizeof(size_t);
memcpy(data_pointer, chunk->items[i]->path, path_len);
data_pointer += path_len;
// file data
unsigned long long data_size = chunk->items[i]->stats.st_size;
memcpy(data_pointer, &data_size, sizeof(size_t));
size_t file_data_size = chunk->items[i]->stats.st_size;
memcpy(data_pointer, &file_data_size, sizeof(size_t));
data_pointer += sizeof(size_t);
memcpy(data_pointer, chunk->items[i]->data->data, data_size);
data_pointer += data_size;
memcpy(data_pointer, chunk->items[i]->data->data, file_data_size);
data_pointer += file_data_size;
}
return data;
}
Chunk *chunk_deserialize(Data *data) {
ArrayList *files = array_list_create(file_destroy);
char *data_pointer = data->data;
size_t remaining_size = data->size;
while (remaining_size > 0) {
if (remaining_size < sizeof(size_t)) {
log_message(LOG_LEVEL_ERROR, "Invalid chunk format: not enough data for path length");
array_list_delete(files);
return NULL;
}
size_t path_len = *(size_t *)data_pointer;
data_pointer += sizeof(size_t);
remaining_size -= sizeof(size_t);
if (remaining_size < path_len) {
log_message(LOG_LEVEL_ERROR, "Invalid chunk format: not enough data for path");
array_list_delete(files);
return NULL;
}
char *path = malloc(path_len + 1);
if (path == NULL) {
perror("Could not allocate memory for file path");
array_list_delete(files);
return NULL;
}
memcpy(path, data_pointer, path_len);
path[path_len] = '\0';
data_pointer += path_len;
remaining_size -= path_len;
if (remaining_size < sizeof(size_t)) {
log_message(LOG_LEVEL_ERROR, "Invalid chunk format: not enough data for data size");
free(path);
array_list_delete(files);
return NULL;
}
size_t file_data_size = *(size_t *)data_pointer;
data_pointer += sizeof(size_t);
remaining_size -= sizeof(size_t);
if (remaining_size < file_data_size) {
log_message(LOG_LEVEL_ERROR, "Invalid chunk format: not enough data for file content");
free(path);
array_list_delete(files);
return NULL;
}
struct stat st = {0};
st.st_size = file_data_size;
File *file = file_create(path, &st);
if (file == NULL) {
free(path);
array_list_delete(files);
return NULL;
}
void *file_data = malloc(file_data_size);
if (file_data == NULL) {
perror("Could not allocate memory for file data");
free(path);
array_list_delete(files);
return NULL;
}
memcpy(file_data, data_pointer, file_data_size);
file->data = data_create(file_data, file_data_size);
data_pointer += file_data_size;
remaining_size -= file_data_size;
array_list_add(files, file);
free(path);
}
log_message(LOG_LEVEL_DEBUG, "Chunk succesfully compressed");
Data *compressed = data_compress(data, compression_level);
data_destroy(data);
File **file_array = (File **)array_list_to_array(files);
Chunk *chunk = chunk_create(file_array, files->size);
free(file_array);
files->item_destroyer = NULL;
array_list_delete(files);
return chunk;
}
Data *chunk_compress(Chunk *chunk, int compression_level) {
log_message(LOG_LEVEL_DEBUG, "Starting to compress chunk");
Data *serialized = chunk_serialize(chunk);
Data *compressed = data_compress(serialized, compression_level);
data_destroy(serialized);
log_message(LOG_LEVEL_DEBUG, "Chunk successfully compressed");
return compressed;
}
@@ -143,99 +231,15 @@ Chunk *chunk_decompress(Data *compressed_data) {
log_message(LOG_LEVEL_ERROR, "Failed to decompress chunk data");
return NULL;
}
ArrayList *files = array_list_create(file_destroy);
char *data_pointer = uncompressed_data->data;
size_t remaining_size = uncompressed_data->size;
while (remaining_size > 0) {
if (remaining_size < sizeof(size_t)) {
log_message(LOG_LEVEL_ERROR, "Invalid chunk format: not enough data for path length");
array_list_delete(files);
data_destroy(uncompressed_data);
return NULL;
}
size_t path_len = *(size_t *)data_pointer;
data_pointer += sizeof(size_t);
remaining_size -= sizeof(size_t);
if (remaining_size < path_len) {
log_message(LOG_LEVEL_ERROR, "Invalid chunk format: not enough data for path");
array_list_delete(files);
data_destroy(uncompressed_data);
return NULL;
}
char *path = malloc(path_len + 1);
if (path == NULL) {
perror("Could not allocate memory for file path");
array_list_delete(files);
data_destroy(uncompressed_data);
return NULL;
}
memcpy(path, data_pointer, path_len);
path[path_len] = '\0';
data_pointer += path_len;
remaining_size -= path_len;
if (remaining_size < sizeof(size_t)) {
log_message(LOG_LEVEL_ERROR, "Invalid chunk format: not enough data for data size");
free(path);
array_list_delete(files);
data_destroy(uncompressed_data);
return NULL;
}
size_t data_size = *(size_t *)data_pointer;
data_pointer += sizeof(size_t);
remaining_size -= sizeof(size_t);
if (remaining_size < data_size) {
log_message(LOG_LEVEL_ERROR, "Invalid chunk format: not enough data for file content");
free(path);
array_list_delete(files);
data_destroy(uncompressed_data);
return NULL;
}
struct stat st = {0};
st.st_size = data_size;
File *file = file_create(path, &st);
if (file == NULL) {
free(path);
array_list_delete(files);
data_destroy(uncompressed_data);
return NULL;
}
void *file_data = malloc(data_size);
if (file_data == NULL) {
perror("Could not allocate memory for file data");
free(path);
array_list_delete(files);
data_destroy(uncompressed_data);
return NULL;
}
memcpy(file_data, data_pointer, data_size);
file->data = data_create(file_data, data_size);
data_pointer += data_size;
remaining_size -= data_size;
array_list_add(files, file);
free(path);
Chunk *chunk = chunk_deserialize(uncompressed_data);
if (chunk == NULL) {
log_message(LOG_LEVEL_ERROR, "Failed to deserialize chunk data");
data_destroy(uncompressed_data);
return NULL;
}
// Create the chunk from the files
File **file_array = (File **)array_list_to_array(files);
Chunk *chunk = chunk_create(file_array, files->size);
// Clean up - files are now owned by the chunk
free(file_array);
files->item_destroyer = NULL;
array_list_delete(files);
data_destroy(uncompressed_data);
log_message(LOG_LEVEL_DEBUG, "Chunk successfully decompressed");
return chunk;
}
+2 -3
View File
@@ -6,8 +6,6 @@
#include <sys/stat.h>
#define DESIRED_CHUNK_SIZE 10 * 1024 * 1024
#define FILE_PATH_SEPERATOR "#&&SEPP&&#"
#define FILE_PATH_DATA_SEPERATOR "#&&SEPD&&#"
typedef struct {
File **items;
@@ -18,10 +16,11 @@ Chunk *chunk_create(File **items, int element_count);
void chunk_destroy(void *chunk);
void chunk_print(void *chunk);
Data *chunk_format(Chunk *chunk);
Data *chunk_serialize(Chunk *chunk);
Chunk *chunk_deserialize(Data *data);
Data *chunk_compress(Chunk *chunk, int compression_level);
Chunk *chunk_decompress(Data *compressed_data);
Data *chunk_data_create(void *data, unsigned long long data_size);
void chunk_data_delete(void *chunk);
void chunk_data_to_disk(Data *chunk, char *root_directory);
#endif
+1 -1
View File
@@ -100,4 +100,4 @@ void file_receive_destroy(void *file_receive) {
free(file);
}
FileReceive *file_receive_from_buffer(void *buffer) {}
-3
View File
@@ -21,11 +21,8 @@ void file_load_data(File *file);
void file_print(void *item);
void file_send_single_calls(File *file, int file_descriptor);
size_t file_content_to_buffer(File *file);
Data *file_compress(File *file);
FileReceive *file_receive_create(char *path, Data *data);
void file_receive_destroy(void *file_receive);
FileReceive *file_receive_from_buffer(void *buffer);
FileReceive *file_receive_decompress(void *FileReceive);
#endif
+2
View File
@@ -195,6 +195,8 @@ const char *status_to_string(Status status) {
return "FINISHED";
case STATUS_NEXT:
return "NEXT";
case STATUS_CHUNK:
return "CHUNK";
default:
return "UNKNOWN";
}
+1 -1
View File
@@ -5,7 +5,7 @@
#include <netinet/in.h>
typedef int Status;
enum NET_STATUS { STATUS_OK, STATUS_ERROR, STATUS_FINISHED, STATUS_NEXT };
enum NET_STATUS { STATUS_OK, STATUS_ERROR, STATUS_FINISHED, STATUS_NEXT, STATUS_CHUNK };
typedef struct Server {
struct sockaddr_in address;
+7 -10
View File
@@ -36,18 +36,15 @@ CLIENT_CMD_PREFIX = [
TEST_CASES = [
{"name": "Standard (Single-threaded)", "flags": []},
{"name": "Multithreading (-m)", "flags": ["-m"]},
# {"name": "Compression (-c -5)", "flags": ["-c -5"]},
{"name": "Compression (-c 0)", "flags": ["-c 0"]},
# {"name": "Compression (-c 10)", "flags": ["-c 10"]},
# {"name": "Compression (-c 20)", "flags": ["-c 20"]},
# {"name": "Chunk Serialization (-s)", "flags": ["-s"]},
{"name": "Chunk Serialization (-s)", "flags": ["-s"]},
{"name": "Compression + Chunk Serialization (-c -s)", "flags": ["-c", "-s"]},
{"name": "Multithreading + Compression (-m -c)", "flags": ["-m", "-c"]},
# {"name": "Multithreading + Chunk Serialization (-m -s)", "flags": ["-m", "-s"]},
# {"name": "Compression + Chunk Serialization (-c -s)", "flags": ["-c", "-s"]},
# {
# "name": "Multithreading + Compression + Chunk Serialization (-m -c -s)",
# "flags": ["-m", "-c", "-s"],
# },
{"name": "Multithreading + Chunk Serialization (-m -s)", "flags": ["-m", "-s"]},
{
"name": "Multithreading + Compression + Chunk Serialization (-m -c -s)",
"flags": ["-m", "-c", "-s"],
},
]