1 files changed, 187 insertions, 0 deletions
diff --git a/a8cat.c b/a8cat.c
new file mode 100644
index 0000000..752e96f
--- /dev/null
+++ b/a8cat.c
@@ -0,0 +1,187 @@
+#include <stdio.h>
+#include <string.h>
+#include <stdlib.h>
+#include <unistd.h>
+#include <locale.h>
+#include <wchar.h>
+#include <errno.h>
+
+#include "atables.h"
+#include "wtable.h"
+
+const char **table = ata2utf;
+
+const char *inverse_on  = "\x1b[7m";
+const char *inverse_off = "\x1b[0m";
+
+int underline = 0, reverse = 0, textmode = 0, ics = 0;
+
+void print_help(void) {
+	printf("Usage: a8cat [-i] [-u] [file ...]\n");
+}
+
+FILE *open_input(const char *file) {
+	FILE *input;
+
+	if(file[0] == '-' && file[1] == 0) {
+		if(freopen(NULL, "rb", stdin)) {
+			input = stdin;
+		} else {
+			perror("(standard input)");
+			return NULL;
+		}
+	} else if(!(input = fopen(file, "rb"))) {
+		perror(file);
+		return NULL;
+	}
+
+	return input;
+}
+
+int handle_escape_seq(int inv, FILE *input) {
+	int count, c;
+	char buf[5] = { 0x1b, 0, 0, 0, 0 };
+
+	for(count = 1; count < 4; count++) {
+		c = fgetwc(input);
+		if(c == WEOF) break;
+		buf[count] = c; /* FIXME: might be a wide char! */
+	}
+
+	if(strcmp(inverse_on, buf) == 0) {
+		return 0x80;
+	} else if(strcmp(inverse_off, buf) == 0) {
+		return 0;
+	} else {
+		fputs(buf, stdout);
+		return inv;
+	}
+}
+
+int a8revcat(const char *file) {
+	FILE *input;
+	int c, d, inv = 0;
+
+	if( !(input = open_input(file)) )
+		return 1;
+
+	setlocale(LC_CTYPE, "en_US.UTF-8");
+	while( (c = fgetwc(input)) != WEOF ) {
+		if(c == 0x1b) {
+			inv = handle_escape_seq(inv, input);
+		} else if(c == '\n') {
+			putchar(0x9b);
+		} else if(c < 0x80) {
+			putchar(c | inv);
+		} else {
+			d = wchar2atascii(c, ics);
+			if(d == -1) {
+				fprintf(stderr, "warning: unrecognized Unicode character %04x\n", c);
+			} else {
+				putchar(d | inv);
+			}
+		}
+	}
+
+	return 0;
+}
+
+/* XXX: hard-coded ANSI/vt100 escape sequences. would be
+   better but more complex to use terminfo to support any ol'
+   terminal... */
+void inverse(int onoff) {
+	fputs((onoff ? inverse_on : inverse_off ), stdout);
+}
+
+int a8cat(const char *file) {
+	FILE *input;
+	int c, inv = 0;
+
+	if( !(input = open_input(file)) )
+		return 1;
+
+	while( (c = fgetc(input)) != EOF ) {
+		if(c == 0x9b) {
+			putchar('\n');
+			continue;
+		}
+
+		if(textmode) {
+			switch(c) {
+				case 0x09: /* Atari TAB is same as ASCII */
+					putchar('\t');
+					continue;
+				case 0xfd: /* bell */
+					putchar('\a');
+					continue;
+				case 0x7e: /* backspace */
+					putchar('\b');
+					continue;
+				default: break;
+			}
+		}
+
+		if(!underline) {
+			/* strings of inverse chars only get one "inverse on" ANSI
+			   sequence, and one "inverse off" afterwards. */
+			if(c & 0x80) {
+				if(!inv) {
+					inv = 1;
+					inverse(1);
+				}
+			} else {
+				if(inv) {
+					inv = 0;
+					inverse(0);
+				}
+			}
+		}
+
+		fputs(table[c & 0x7f], stdout);
+
+		if(underline && (c & 0x80)) {
+			putchar('\b');
+			putchar('_');
+		}
+	}
+
+	/* gotta turn off inverse, so if there's another file after this one,
+	   it doesn't start out being printed in inverse. */
+	if(inv && !underline) inverse(0);
+
+	fclose(input);
+	return 0;
+}
+
+int main(int argc, char **argv) {
+	int opt, result = 0;
+
+	while( (opt = getopt(argc, argv, "ihurt")) != -1) {
+		switch(opt) {
+			case 'i': table = ics2utf; ics = 1; break;
+			case 'h': print_help(); exit(0); break;
+			case 'u': underline = 1; break;
+			case 'r': reverse = 1; break;
+			case 't': textmode = 1; break;
+			default: print_help(); exit(1); break;
+		}
+	}
+
+	if(reverse) {
+		if(underline || textmode) {
+			fprintf(stderr, "-t and -u options don't make sense with -r.\n");
+			exit(1);
+		}
+	}
+
+	if(optind >= argc) {
+		result = (reverse ? a8revcat("-") : a8cat("-"));
+	} else {
+		while(optind < argc) {
+			result += (reverse ? a8revcat(argv[optind]) : a8cat(argv[optind]));
+			optind++;
+		}
+	}
+
+	exit(result);
+}