mame/src/tools/split.c

/***************************************************************************

    split.c

    Simple file splitter/joiner with hashes.

****************************************************************************

    Copyright Aaron Giles
    All rights reserved.

    Redistribution and use in source and binary forms, with or without
    modification, are permitted provided that the following conditions are
    met:

        * Redistributions of source code must retain the above copyright
          notice, this list of conditions and the following disclaimer.
        * Redistributions in binary form must reproduce the above copyright
          notice, this list of conditions and the following disclaimer in
          the documentation and/or other materials provided with the
          distribution.
        * Neither the name 'MAME' nor the names of its contributors may be
          used to endorse or promote products derived from this software
          without specific prior written permission.

    THIS SOFTWARE IS PROVIDED BY AARON GILES ''AS IS'' AND ANY EXPRESS OR
    IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
    WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
    DISCLAIMED. IN NO EVENT SHALL AARON GILES BE LIABLE FOR ANY DIRECT,
    INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
    (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
    SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
    HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
    STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING
    IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
    POSSIBILITY OF SUCH DAMAGE.

****************************************************************************/

#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <ctype.h>
#include "astring.h"
#include "corefile.h"
#include "corestr.h"
#include "sha1.h"

#define DEFAULT_SPLIT_SIZE      100
#define MAX_PARTS               1000



/***************************************************************************
    CORE IMPLEMENTATION
***************************************************************************/


/*-------------------------------------------------
    compute_hash_as_string - compute an SHA1
    hash over a buffer and return a string
-------------------------------------------------*/

static void compute_hash_as_string(astring &buffer, void *data, UINT32 length)
{
	char expanded[SHA1_DIGEST_SIZE * 2];
	UINT8 sha1digest[SHA1_DIGEST_SIZE];
	struct sha1_ctx sha1;
	int ch;

	// compute the SHA1
	sha1_init(&sha1);
	sha1_update(&sha1, length, (const UINT8 *)data);
	sha1_final(&sha1);
	sha1_digest(&sha1, sizeof(sha1digest), sha1digest);

	// expand the digest to a string
	for (ch = 0; ch < SHA1_DIGEST_SIZE; ch++)
	{
		expanded[ch * 2 + 0] = "0123456789ABCDEF"[sha1digest[ch] >> 4];
		expanded[ch * 2 + 1] = "0123456789ABCDEF"[sha1digest[ch] & 15];
	}

	// copy it to the buffer
	buffer.cpy(expanded, sizeof(expanded));
}


/*-------------------------------------------------
    split_file - split a file into multiple parts
-------------------------------------------------*/

static int split_file(const char *filename, const char *basename, UINT32 splitsize)
{
	astring outfilename, basefilename, splitfilename;
	core_file *outfile = NULL, *infile = NULL, *splitfile = NULL;
	astring computedhash;
	void *splitbuffer = NULL;
	int index, partnum;
	UINT64 totallength;
	file_error filerr;
	int error = 1;

	// convert split size to MB
	if (splitsize > 500)
	{
		fprintf(stderr, "Fatal error: maximum split size is 500MB (even that is way huge!)\n");
		goto cleanup;
	}
	splitsize *= 1024 * 1024;

	// open the file for read
	filerr = core_fopen(filename, OPEN_FLAG_READ, &infile);
	if (filerr != FILERR_NONE)
	{
		fprintf(stderr, "Fatal error: unable to open file '%s'\n", filename);
		goto cleanup;
	}

	// get the total length
	totallength = core_fsize(infile);
	if (totallength < splitsize)
	{
		fprintf(stderr, "Fatal error: file is smaller than the split size\n");
		goto cleanup;
	}
	if ((UINT64)splitsize * MAX_PARTS < totallength)
	{
		fprintf(stderr, "Fatal error: too many splits (maximum is %d)\n", MAX_PARTS);
		goto cleanup;
	}

	// allocate a buffer for reading
	splitbuffer = malloc(splitsize);
	if (splitbuffer == NULL)
	{
		fprintf(stderr, "Fatal error: unable to allocate memory for the split\n");
		goto cleanup;
	}

	// find the base name of the file
	basefilename.cpy(basename);
	index = basefilename.rchr(0, PATH_SEPARATOR[0]);
	if (index != -1)
		basefilename.del(0, index + 1);

	// compute the split filename
	splitfilename.cpy(basename).cat(".split");

	// create the split file
	filerr = core_fopen(splitfilename, OPEN_FLAG_WRITE | OPEN_FLAG_CREATE | OPEN_FLAG_NO_BOM, &splitfile);
	if (filerr != FILERR_NONE)
	{
		fprintf(stderr, "Fatal error: unable to create split file '%s'\n", splitfilename.cstr());
		goto cleanup;
	}

	// write the basics out
	core_fprintf(splitfile, "splitfile=%s\n", basefilename.cstr());
	core_fprintf(splitfile, "splitsize=%d\n", splitsize);

	printf("Split file is '%s'\n", splitfilename.cstr());
	printf("Splitting file %s into chunks of %dMB...\n", basefilename.cstr(), splitsize / (1024 * 1024));

	// now iterate until done
	for (partnum = 0; partnum < 1000; partnum++)
	{
		UINT32 actual, length;

		printf("Reading part %d...", partnum);

		// read as much as we can from the file
		length = core_fread(infile, splitbuffer, splitsize);
		if (length == 0)
			break;

		// hash what we have
		compute_hash_as_string(computedhash, splitbuffer, length);

		// write that info to the split file
		core_fprintf(splitfile, "hash=%s file=%s.%03d\n", computedhash.cstr(), basefilename.cstr(), partnum);

		// compute the full filename for this guy
		outfilename.printf("%s.%03d", basename, partnum);

		// create it
		filerr = core_fopen(outfilename, OPEN_FLAG_WRITE | OPEN_FLAG_CREATE, &outfile);
		if (filerr != FILERR_NONE)
		{
			printf("\n");
			fprintf(stderr, "Fatal error: unable to create output file '%s'\n", outfilename.cstr());
			goto cleanup;
		}

		printf(" writing %s.%03d...", basefilename.cstr(), partnum);

		// write the data
		actual = core_fwrite(outfile, splitbuffer, length);
		if (actual != length)
		{
			printf("\n");
			fprintf(stderr, "Fatal error: Error writing output file (out of space?)\n");
			goto cleanup;
		}
		core_fclose(outfile);
		outfile = NULL;

		printf(" done\n");

		// stop if this is the end
		if (length < splitsize)
			break;
	}
	printf("File split successfully\n");

	// if we get here, clear the errors
	error = 0;

cleanup:
	if (splitfile != NULL)
	{
		core_fclose(splitfile);
		if (error != 0)
			remove(splitfilename);
	}
	if (infile != NULL)
		core_fclose(infile);
	if (outfile != NULL)
	{
		core_fclose(outfile);
		if (error != 0)
			remove(outfilename);
	}
	if (splitbuffer != NULL)
		free(splitbuffer);
	return error;
}


/*-------------------------------------------------
    join_file - rejoin a file from its split
    parts
-------------------------------------------------*/

static int join_file(const char *filename, const char *outname, int write_output)
{
	astring expectedhash, computedhash;
	astring outfilename, infilename;
	astring basepath;
	core_file *outfile = NULL, *infile = NULL, *splitfile = NULL;
	void *splitbuffer = NULL;
	file_error filerr;
	UINT32 splitsize;
	char buffer[256];
	int error = 1;
	int index;

	// open the file for read
	filerr = core_fopen(filename, OPEN_FLAG_READ, &splitfile);
	if (filerr != FILERR_NONE)
	{
		fprintf(stderr, "Fatal error: unable to open file '%s'\n", filename);
		goto cleanup;
	}

	// read the first line and verify this is a split file
	if (!core_fgets(buffer, sizeof(buffer), splitfile) || strncmp(buffer, "splitfile=", 10) != 0)
	{
		fprintf(stderr, "Fatal error: corrupt or incomplete split file at line:\n%s\n", buffer);
		goto cleanup;
	}
	outfilename.cpy(buffer + 10).trimspace();

	// compute the base path
	basepath.cpy(filename);
	index = basepath.rchr(0, PATH_SEPARATOR[0]);
	if (index != -1)
		basepath.del(index + 1);
	else
		basepath.reset();

	// override the output filename if specified, otherwise prepend the path
	if (outname != NULL)
		outfilename.cpy(outname);
	else
		outfilename.ins(0, basepath);

	// read the split size
	if (!core_fgets(buffer, sizeof(buffer), splitfile) || sscanf(buffer, "splitsize=%d", &splitsize) != 1)
	{
		fprintf(stderr, "Fatal error: corrupt or incomplete split file at line:\n%s\n", buffer);
		goto cleanup;
	}

	// attempt to open the new file
	if (write_output)
	{
		// don't overwrite the original!
		filerr = core_fopen(outfilename, OPEN_FLAG_READ, &outfile);
		if (filerr == FILERR_NONE)
		{
			core_fclose(outfile);
			outfile = NULL;
			fprintf(stderr, "Fatal error: output file '%s' already exists\n", outfilename.cstr());
			goto cleanup;
		}

		// open the output for write
		filerr = core_fopen(outfilename, OPEN_FLAG_WRITE | OPEN_FLAG_CREATE, &outfile);
		if (filerr != FILERR_NONE)
		{
			fprintf(stderr, "Fatal error: unable to create file '%s'\n", outfilename.cstr());
			goto cleanup;
		}
	}

	printf("%s file '%s'...\n", write_output ? "Joining" : "Verifying", outfilename.cstr());

	// now iterate through each file
	while (core_fgets(buffer, sizeof(buffer), splitfile))
	{
		UINT32 length, actual;

		// make sure the hash and filename are in the right place
		if (strncmp(buffer, "hash=", 5) != 0 || strncmp(buffer + 5 + SHA1_DIGEST_SIZE * 2, " file=", 6) != 0)
		{
			fprintf(stderr, "Fatal error: corrupt or incomplete split file at line:\n%s\n", buffer);
			goto cleanup;
		}
		expectedhash.cpy(buffer + 5, SHA1_DIGEST_SIZE * 2);
		infilename.cpy(buffer + 5 + SHA1_DIGEST_SIZE * 2 + 6).trimspace();

		printf("  Reading file '%s'...", infilename.cstr());

		// read the file's contents
		infilename.ins(0, basepath);
		filerr = core_fload(infilename, &splitbuffer, &length);
		if (filerr != FILERR_NONE)
		{
			printf("\n");
			fprintf(stderr, "Fatal error: unable to load file '%s'\n", infilename.cstr());
			goto cleanup;
		}

		// hash the contents
		compute_hash_as_string(computedhash, splitbuffer, length);

		// compare
		if (computedhash != expectedhash)
		{
			printf("\n");
			fprintf(stderr, "Fatal error: file '%s' has incorrect hash\n  Expected: %s\n  Computed: %s\n", infilename.cstr(), expectedhash.cstr(), computedhash.cstr());
			goto cleanup;
		}

		// append to the file
		if (write_output)
		{
			printf(" writing...");

			actual = core_fwrite(outfile, splitbuffer, length);
			if (actual != length)
			{
				printf("\n");
				fprintf(stderr, "Fatal error: Error writing output file (out of space?)\n");
				goto cleanup;
			}

			printf(" done\n");
		}
		else
			printf(" verified\n");

		// release allocated memory
		osd_free(splitbuffer);
		splitbuffer = NULL;
	}
	if (write_output)
		printf("File re-created successfully\n");
	else
		printf("File verified successfully\n");

	// if we get here, clear the errors
	error = 0;

cleanup:
	if (splitfile != NULL)
		core_fclose(splitfile);
	if (infile != NULL)
		core_fclose(infile);
	if (outfile != NULL)
	{
		core_fclose(outfile);
		if (error != 0)
			remove(outfilename);
	}
	if (splitbuffer != NULL)
		osd_free(splitbuffer);
	return error;
}


/*-------------------------------------------------
    main - primary entry point
-------------------------------------------------*/

int main(int argc, char *argv[])
{
	int result;

	/* skip if no command */
	if (argc < 2)
		goto usage;

	/* split command */
	if (core_stricmp(argv[1], "-split") == 0)
	{
		if (argc != 4 && argc != 5)
			goto usage;
		result = split_file(argv[2], argv[3], (argc < 5) ? DEFAULT_SPLIT_SIZE : atoi(argv[4]));
	}

	/* join command */
	else if (core_stricmp(argv[1], "-join") == 0)
	{
		if (argc != 3 && argc != 4)
			goto usage;
		result = join_file(argv[2], (argc >= 4) ? argv[3] : NULL, TRUE);
	}

	/* verify command */
	else if (core_stricmp(argv[1], "-verify") == 0)
	{
		if (argc != 3)
			goto usage;
		result = join_file(argv[2], NULL, FALSE);
	}
	else
		goto usage;

	return result;

usage:
	fprintf(stderr,
		"Usage:\n"
		"  split -split <bigfile> <basename> [<size>] -- split file into parts\n"
		"  split -join <splitfile> [<outputfile>] -- join file parts into original file\n"
		"  split -verify <splitfile> -- verify a split file\n"
		"\n"
		"Where:\n"
		"  <bigfile> is the large file you wish to split\n"
		"  <basename> is the base path and name to assign to the split files\n"
		"  <size> is the optional split size, in MB (100MB is the default)\n"
		"  <splitfile> is the name of the <basename>.split generated with -split\n"
		"  <outputfile> is the name of the file to output (defaults to original name)\n"
	);
	return 0;
}