HuntnGather – Diff between revs 12 and 13

Subversion Repositories:

?pathlinks?

Rev 12 – Rev 14 – Go to most recent revision – Details – Blame – Last modification – View Log

Only display areas with differences – Ignore whitespace

 ///////////////////////////////////////////////////////////////////////////
 //    Copyright (C) 2021 Wizardry and Steamworks - License: MIT          //
 ///////////////////////////////////////////////////////////////////////////
 #include <stdio.h>
 #include <stdlib.h>
 #include <string.h>
 #include <dirent.h>
 #include <signal.h>
 #include <sys/types.h>
 #include <sys/stat.h>
 #include <proto/dos.h>
 #include <proto/exec.h>
 #include "StringStack.h"
 #if !defined ___HAVE_GETOPT___
 #include "getopt.h"
 #endif
 #if defined ___AmigaOS___
 /*************************************************************************/
 /*        Version string used for querrying the program version.         */
 /*************************************************************************/
 TEXT version_string[] =
 	"\0$VER: Gather 1.7 "__DATE__" by Wizardry and Steamworks";
 #endif
 #if !defined TRUE
 #define TRUE 1;
 #endif
 #if !defined FALSE
 #define FALSE 0;
 #endif
 #define MAX_MEM 262144
 #define NAME_BUF 32
 #define PATH_BUF 128
 #define LINE_BUF 256
 #define DEFAULT_DATABASE_FILE "S:gather.db"
 typedef struct {
 	unsigned int dirs;
 	unsigned int files;
 } stats;
 int run = TRUE;
 int verbose = TRUE;
 void SignalHandler(int sig) {
 	// Toggle the run flag to stop execution.
 	run = FALSE;
+}
 int compare(const void *a, const void *b) {
 	const char **p = (const char **)a;
 	const char **q = (const char **)b;
-	return strcmp(*p, *q);
+	return strncmp(*p, *q, strlen(*p));
+}
 /*
+	*
 	* Sorts a database file lexicographically.
 	*/
 void SortDatabase(char *dbFile) {
 	FILE *fp;
 	char *name = NULL;
 	char *path = NULL;
 	char **database;
 	char c;
 	int i;
 	int side;
 	unsigned int line;
 	int name_size;
 	int path_size;
 	// Open database file for reading.
 	if((fp = fopen(dbFile, "r")) == NULL) {
 		fprintf(stderr, "Unable to open gather database for reading.\n");
 		return;
+	}
 	database = malloc(sizeof(char *));
 	name_size = NAME_BUF;
 	name = malloc(name_size * sizeof(char));
 	path_size = PATH_BUF;
 	path = malloc(path_size * sizeof(char));
 	line = 0;
 	side = 0;
 	i = 0;
 	if(verbose) {
 		fprintf(stdout, "Sorting database: '%s'\n", dbFile);
+	}
 	while(run && fscanf(fp, "%c", &c) == 1) {
 #if defined ___AmigaOS___
 		// Check if CTRL+C was pressed and abort the program.
    	if(SetSignal(0L, SIGBREAKF_CTRL_C) & SIGBREAKF_CTRL_C) {
 			run = FALSE;
 			continue;
+		}
 #endif
 		switch(c) {
 			case '\n':
 				// Load up the name and path into the database variable.
 				database = realloc(database, (line + 1) * sizeof(char *));
 				database[line] = malloc((strlen(name) + strlen(path) + 1 + 1) * sizeof(char));
 				sprintf(database[line], "%s\t%s", name, path);
 				++line;
 				free(name);
 				name_size = NAME_BUF;
 				name = malloc(name_size * sizeof(char));
 				--side;
 				i = 0;
 				break;
 			case '\t':
 				free(path);
 				path_size = PATH_BUF;
 				path = malloc(path_size * sizeof(char));
 				++side;
 				i = 0;
 				break;
 			default:
 				switch(side) {
 					case 0:
 					   if(strlen(name) == name_size) {
 					   	name_size = name_size * 1.5;
 					   	name = realloc(name, name_size * sizeof(char));
+					   }
 						//name = realloc(name, (i + 1 + 1) * sizeof(char));
 						name[i] = c;
 						name[i + 1] = '\0';
 						break;
 					case 1:
 						if(strlen(path) == path_size) {
 							path_size = path_size * 1.5;
 							path = realloc(path, path_size * sizeof(char));
+						}
 						//path = realloc(path, (i + 1 + 1) * sizeof(char));
 						path[i] = c;
 						path[i + 1] = '\0';
 						break;
 					default:
 						fprintf(stderr, "Database corrupted.\n");
 						break;
+				}
 				++i;
 				break;
+		}
+	}
 	fclose(fp);
 	// Sort the database.
 	qsort(database, line, sizeof(char *), compare);
 	// Write the database lines back to the database.
 	if((fp = fopen(dbFile, "w+")) == NULL) {
 		fprintf(stderr, "Unable to open gather database for writing.\n");
 		return;
+	}
 	for(i = 0; i < line; ++i) {
 		fprintf(fp, "%s\n", database[i]);
+	}
 	free(database);
 	fclose(fp);
+}
 /*
+	*
 	* Updates a database file "dbFile".
 	*/
 void UpdateDatabase(char *dbFile, stringStack *dirStack, stats *stats) {
 	FILE *fp;
 	DIR *dir;
 	struct dirent *dirEntry;
 	struct stat dirStat;
 	unsigned int size;
 	char *path;
 	char *subPath;
 	if((fp = fopen(dbFile, "w+")) == NULL) {
 		fprintf(stderr, "Unable to open gather database for writing.\n");
 		return;
+	}
 	while(run && !stringStackIsEmpty(dirStack)) {
 #if defined ___AmigaOS___
 		// Check if CTRL+C was pressed and abort the program.
 		if(SetSignal(0L, SIGBREAKF_CTRL_C) & SIGBREAKF_CTRL_C) {
 			run = FALSE;
+		}
 #endif
 		if((path = stringStackPop(dirStack)) == NULL) {
 			return;
+		}
 		if((dir = opendir(path)) == NULL) {
 			return;
+		}
 		while(run && (dirEntry = readdir(dir)) != NULL) {
 #if defined ___AmigaOS___
 			// Check if CTRL+C was pressed and abort the program.
 			if(SetSignal(0L, SIGBREAKF_CTRL_C) & SIGBREAKF_CTRL_C) {
 				run = FALSE;
+			}
 #endif
 			size = sizeof(path) + sizeof(dirEntry->d_name) + 1;
 			switch(path[strlen(path) - 1]) {
 				case '/':
 	 			case ':': // This is a drive path.
 	 				subPath = malloc(size);
 	 				sprintf(subPath, "%s%s", path, dirEntry->d_name);
 	 				break;
 	 			default:
 	 				subPath = malloc(size + 1);
 	 				sprintf(subPath, "%s/%s", path, dirEntry->d_name);
 	 				break;
+			}
 			stat(subPath, &dirStat);
 			if(S_ISDIR(dirStat.st_mode)) {
 	 			stringStackPush(dirStack, subPath);
 	 			++stats->dirs;
 	 			if(verbose) {
 	 				fprintf(stdout,
 	 					"Gathered %d directories and %d files.\r",
 	 					stats->dirs,
 	 					stats->files);
+	 			}
 	 			free(subPath);
 	 			continue;
+	 		}
 	 		// Write to database file.
 	 		fprintf(fp, "%s\t%s\n", dirEntry->d_name, subPath);
 	 		++stats->files;
 	 		if(verbose) {
 	 			fprintf(stdout,
 	 				"Gathered %d directories and %d files.\r",
 	 				stats->dirs,
 	 				stats->files);
+	 		}
 	 		free(subPath);
+	 	}
 	 	closedir(dir);
 		free(path);
+	}
 	if(verbose) {
 		fprintf(stdout, "\n");
+	}
 	fclose(fp);
+}
 /*
+	*
 	* Gets the size of a database "dbFle".
 	*/
 int GetDatabaseSize(char *dbFile) {
 	FILE *fp;
 	int size;
 	if((fp = fopen(dbFile, "r")) == NULL) {
 		fprintf(stderr, "Unable to open gather database for reading.\n");
 		fclose(fp);
 		return 0;
+	}
 	fseek(fp, 0L, SEEK_END);
 	size = ftell(fp);
 	fclose(fp);
 	return size;
+}
 /*
+	*
 	* Counts the lines in a database file "dbFile".
 	*/
 int CountDatabaseLines(char *dbFile) {
 	FILE *fp;
 	int lines;
 	char c;
 	if((fp = fopen(dbFile, "r")) == NULL) {
 		fprintf(stderr, "Unable to open gather database for reading.\n");
 		fclose(fp);
 		return 0;
+	}
 	lines = 0;
 	while(fscanf(fp, "%c", &c) == 1) {
 		switch(c) {
 			case '\n':
 				++lines;
 				break;
+		}
+	}
 	fclose(fp);
 	return lines;
+}
 /*
+	*
 	* Creates "files" temporary filenames.
 	*/
 char **CreateTempFiles(int files) {
 	char **tmpNames;
 	int count;
 	tmpNames = malloc(files * sizeof(char *));
 	if(verbose) {
 		fprintf(stdout, "Creating temporary files.\r");
+	}
 	count = files;
 	while(--count > -1) {
 #if defined ___AmigaOS___
 		// Check if CTRL+C was pressed and abort the program.
 		if(SetSignal(0L, SIGBREAKF_CTRL_C) & SIGBREAKF_CTRL_C) {
 			run = FALSE;
+		}
 #endif
 		tmpNames[count] = tmpnam(NULL);
 		if(verbose) {
 			fprintf(stdout, "Creating temporary files: %d%%\r", 100 - (int)(((float)count / files) * 100.0));
+		}
+	}
 	if(verbose) {
 		fprintf(stdout, "\n");
+	}
 	return tmpNames;
+}
 /*
+	*
 	* Writes lines from the database "dbFile" to temporary filenames "tmpNames".
 	*/
 void WriteTempFiles(char *dbFile, char **tmpNames, int tmpFiles, int tmpLines, int total) {
 	FILE *fp, *tp;
 	char c;
 	int lines;
 	int linesWritten;
 	if((fp = fopen(dbFile, "r")) == NULL) {
 		fprintf(stderr, "Unable to open gather database for reading.\n");
 		return;
+	}
 	if((tp = fopen(tmpNames[--tmpFiles], "w+")) == NULL) {
 		fprintf(stderr, "Unable to open temporary file '%s' for writing.\n", tmpNames[tmpFiles]);
 		fclose(fp);
 		return;
+	}
 	if(verbose) {
 		fprintf(stdout, "Writing to temporary files.\r");
+	}
 	linesWritten = 0;
 	lines = 0;
 	while(run && fscanf(fp, "%c", &c) == 1) {
 #if defined ___AmigaOS___
 		// Check if CTRL+C was pressed and abort the program.
 		if(SetSignal(0L, SIGBREAKF_CTRL_C) & SIGBREAKF_CTRL_C) {
 			run = FALSE;
+		}
 #endif
 		switch(c) {
 			case '\n':
 				// Increment the total written lines.
 				++linesWritten;
 				if(verbose) {
 					fprintf(stdout, "Writing to temporary files: %d%%.\r", (int)(((float)linesWritten / total) * 100.0));
+				}
 				// Write the newline character back.
 				if(fprintf(tp, "%c", c) != 1) {
 					fprintf(stderr, "Unable to write to temporary file '%s'.\n", tmpNames[tmpFiles]);
 					fclose(tp);
 					fclose(fp);
 					return;
+				}
 				// Switch to the next temporary file.
 				if(++lines >= tmpLines) {
 					// If there are no temporary files left then run till the end.
 					if(tmpFiles - 1 < 0) {
 						break;
+					}
 					// Close the previous temporary file and write to the next temporary file.
 					fclose(tp);
 					if((tp = fopen(tmpNames[--tmpFiles], "w+")) == NULL) {
 						fprintf(stderr, "Unable to open temporary file '%s' for writing.\n", tmpNames[tmpFiles]);
 						fclose(tp);
 						fclose(fp);
+					}
 					lines = 0;
 					break;
+				}
 				break;
 			default:
 				if(fprintf(tp, "%c", c) != 1) {
 					fprintf(stderr, "Unable to write to temporary file '%s'.\n", tmpNames[tmpFiles]);
 					fclose(tp);
 					fclose(fp);
 					return;
+				}
 				break;
+		}
+	}
 	fprintf(stdout, "\n");
 	fclose(tp);
 	fclose(fp);
+}
 /*
+	*
 	* Skips a line in a database file "fp".
 	*/
 void SkipDatabaseLine(FILE *fp) {
 	char c;
 	while(fscanf(fp, "%c", &c) == 1) {
 		if(c == '\n') {
 			break;
+		}
+	}
 	return;
+}
 /*
+	*
 	* Reads a line from the database file "fp".
 	*/
 char *ReadDatabaseLine(FILE *fp) {
 	char c;
 	char *line;
 	int line_size;
 	int i;
 	line_size = LINE_BUF;
 	line = malloc(line_size * sizeof(char));
 	i = 0;
 	while(run && fscanf(fp, "%c", &c) == 1) {
 #if defined ___AmigaOS___
 		// Check if CTRL+C was pressed and abort the program.
 		if(SetSignal(0L, SIGBREAKF_CTRL_C) & SIGBREAKF_CTRL_C) {
 			run = FALSE;
+		}
 #endif
 		switch(c) {
 			case '\n':
 				// Rewind the file by the number of read characters.
 				fseek(fp, -(i + 1), SEEK_CUR);
 				return line;
 			default:
 				if(strlen(line) == line_size) {
 					line_size = line_size * 1.5;
 					line = realloc(line, line_size * sizeof(char));
+				}
 				//line = realloc(line, (chars + 1 + 1) * sizeof(char));
 				line[i] = c;
 				line[i + 1] = '\0';
 				break;
+		}
 		++i;
+	}
 	return NULL;
+}
 /*
+	*
 	* Merges temporary files "tmpNames" into a database "dbFile".
 	*/
 void MergeDatabase(char *dbFile, char **tmpNames, int files, int lines) {
 	FILE *fp;
 	FILE **tp;
 	int i;
 	char *tmp;
 	char *tmpMin;
 	int idxMin;
 	int count;
 	if((fp = fopen(dbFile, "w+")) == NULL) {
 		fprintf(stderr, "Unable to open gather database for writing.\n");
 		return;
+	}
 	// Allocate as many file pointers as temporary files.
 	tp = malloc(files * sizeof(FILE *));
 	// Open all temporary files for reading.
 	for(i = 0; i < files; ++i) {
 		if((tp[i] = fopen(tmpNames[i], "r")) == NULL) {
 			fprintf(stderr, "Unable to open temporary file '%s' for reading.\n", tmpNames[i]);
 			// Close all temporary files.
 			--i;
 			while(i >= 0) {
 				fclose(tp[i]);
+			}
 			return;
+		}
+	}
 	if(verbose) {
 		fprintf(stdout, "Merging all database lines in temporary files.\r");
+	}
 	count = lines;
 	idxMin = 0;
 	while(run && --count > -1) {
 #if defined ___AmigaOS___
 		// Check if CTRL+C was pressed and abort the program.
 		if(SetSignal(0L, SIGBREAKF_CTRL_C) & SIGBREAKF_CTRL_C) {
 			run = FALSE;
+		}
 #endif
 		// Find the smallest line in all temporary files.
 		if(verbose) {
 			fprintf(stdout, "Merging all database lines in temporary files: %d%%.\r", 100 - (int)(((float)count / lines) * 100.0));
+		}
 		tmpMin = NULL;
 		for(i = 0; i < files; ++i) {
 			tmp = ReadDatabaseLine(tp[i]);
 			if(tmp == NULL) {
 				continue;
+			}
-			if(tmpMin == NULL || strcmp(tmp, tmpMin) < 0) {
+			if(tmpMin == NULL || strncmp(tmp, tmpMin, strlen(tmp)) < 0) {
 				if(tmpMin != NULL) {
 					// Free previous instance.
 					free(tmpMin);
+				}
 				tmpMin = malloc((strlen(tmp) + 1) * sizeof(char));
 				sprintf(tmpMin, "%s", tmp);
 				// Remember the index of the file where the smallest entry has been found.
 				idxMin = i;
 				free(tmp);
 				continue;
+			}
 			free(tmp);
+		}
 		// Forward the file where the smallest line was found.
 		SkipDatabaseLine(tp[idxMin]);
 		// Write the smallest line.
 		if(tmpMin != NULL) {
 			fprintf(fp, "%s\n", tmpMin);
 			free(tmpMin);
+		}
+	}
 	// Write out any remaining contents from the temporary files.
 	for(i = 0; i < files; ++i) {
 		tmp = ReadDatabaseLine(tp[i]);
 		if(tmp == NULL) {
 			continue;
+		}
 		fprintf(fp, "%s\n", tmp);
+	}
 	// Close and delete all temporary files.
 	for(i = 0; i < files; ++i) {
 		fclose(tp[i]);
 		// Delete temporary file.
 		remove(tmpNames[i]);
+	}
 	if(verbose) {
 		fprintf(stdout, "\n");
+	}
 	fclose(fp);
+}
 /*
+	*
 	* Indexes a "path" by creating a database "dbFile".
 	*/
 void Gather(char *dbFile, char *path) {
 	stringStack *stack = stringStackCreate(1);
 	stats *stats = malloc(sizeof(stats));
 	char **tmpNames;
 	int dbSize, dbLines, tmpFiles, tmpLines;
 	int i;
  	// Initialize metrics.
  	stats->dirs = 0;
  	stats->files = 0;
 	// Push the first path onto the stack.
 	stringStackPush(stack, path);
 	// Generate the database file.
 	UpdateDatabase(dbFile, stack, stats);
 	// Get the database metrics.
 	dbSize = GetDatabaseSize(dbFile);
 	dbLines = CountDatabaseLines(dbFile);
 	// Compute the amount of temporary files needed.
 	tmpFiles = dbSize / MAX_MEM;
    // In case no temporary files are required,
 	// just sort the database and terminate.
 	if(tmpFiles <= 1) {
 		SortDatabase(dbFile);
 		return;
+	}
 	tmpLines = dbLines / tmpFiles;
 	// Create temporary files.
 	if((tmpNames = CreateTempFiles(tmpFiles)) == NULL) {
 		fprintf(stderr, "Unable to create temporary files.\n");
 		return;
+	}
 	// Write "tmpLines" to temporary files in "tmpFiles" from "dbFile".
 	WriteTempFiles(dbFile, tmpNames, tmpFiles, tmpLines, dbLines);
 	// Sort the temporary files.
 	for(i = 0; i < tmpFiles; ++i) {
 		SortDatabase(tmpNames[i]);
+	}
 	MergeDatabase(dbFile, tmpNames, tmpFiles, dbLines);
+}
 void usage(char *name) {
 	fprintf(stdout, "Hunt & Gather - %s, a file index generating tool.        \n", name);
    fprintf(stdout, "                                                         \n");
 	fprintf(stdout, "SYNTAX: %s [-q] DATABASE                                 \n", name);
 	fprintf(stdout, "                                                         \n");
 	fprintf(stdout, "    -q             Do not print out any messages.        \n");
    fprintf(stdout, "                                                         \n");
 	fprintf(stdout, "DATABASE is a path to where the indexed results will be  \n");
 	fprintf(stdout, "stored for searching with the Hunt tool.                 \n");
 	fprintf(stdout, "                                                         \n");
 	fprintf(stdout, "(c) 2021 Wizardry and Steamworks, MIT.                   \n");
+}
 /*
+	*
 	* Main entry point.
 	*/
 int main(int argc, char **argv) {
 	int option;
 	char *dbFile;
 	struct stat dirStat;
 	// Bind handler to SIGINT.
  	signal(SIGINT, SignalHandler);
 	dbFile = DEFAULT_DATABASE_FILE;
 	while((option = getopt(argc, argv, "hqd:")) != -1) {
 		switch(option) {
 			case 'd':
 				dbFile = optarg;
 				break;
 			case 'q':
 				verbose = FALSE;
 				break;
 			case 'h':
 				usage(argv[0]);
 				return 0;
 			case '?':
 				fprintf(stderr, "Invalid option %ct.\n", optopt);
 				return 1;
+		}
+	}
 	if(optind >= argc) {
 		usage(argv[0]);
 		return 1;
+	}
    stat(argv[optind], &dirStat);
 	if(!S_ISDIR(dirStat.st_mode)) {
 		fprintf(stderr, "Path '%s' is not a directory.\n", argv[optind]);
 		return 1;
+	}
 	if(verbose) {
    	fprintf(stdout, "Gathering to database file: %s\n", dbFile);
+	}
  	// Gather.
  	Gather(dbFile, argv[optind]);
    return 0;
+}