Solve Exercise 7-7: Match pattern in files

This one was tricky, but the solution was rather simple.

1 files changed, 128 insertions, 0 deletions

diff --git a/ch7/7-07_pattern-with-file.c b/ch7/7-07_pattern-with-file.c
new file mode 100644
index 0000000..46732c4
--- /dev/null
+++ b/ch7/7-07_pattern-with-file.c
@@ -0,0 +1,128 @@
+#include <stdio.h>
+#include <string.h>
+
+/* The C Programming Language: 2nd Edition
+ *
+ * Exercise 7-7: Modify the pattern finding program of Chapter 5 to take its
+ * input from a set of named files or, if no files are named as arguments, from
+ * the standard input. Should the file name be printed when a matching line is
+ * found?
+ *
+ * Notes: The program in question is on page 117 of the book. Note from various
+ * errata documents (found around the Web) that the line:
+ *
+ * while (c = *++argv[0])
+ *
+ * is not correct, since it's attempting to modify something that the standard
+ * makes no guarantee to be modifiable. So to monkey-patch it, I used the
+ * typical 'i is the iterator' variable.
+ *
+ * The biggest obstacle to this exercise was correcting the original code from
+ * page 117.
+ *
+ * A gotcha that I came across is figuring out how to make use of or save the
+ * pattern that the user will be searching for. I could have allocated a
+ * character array (string) and stored it directly, but it made more sense to
+ * create another character pointer and leverage the storage that argv was
+ * already making use of. So I ended up with two pointers to the same data,
+ * like hardlinks in a *nix filesystem.
+ *
+ * The filename that a pattern is found in *should* be present, because it's
+ * trivial to match text in multiple files. Generally when one is looking for
+ * text, they want to *edit* that text, and editing it is fruitless if you
+ * don't have the file name handy. That said, putting it behind an option is a
+ * great way to add extensibility to it.
+ */
+
+#define MAXLINE 1000
+
+int get_line(char *line, int max, FILE *f);
+
+int main(int argc, char *argv[]) {
+	char line[MAXLINE];
+	long lineno = 0;
+	int c, except = 0, number = 0, showfiles = 0, found = 0, i = 0;
+	char *pattern = NULL;
+	char *filename = NULL;
+	FILE *infile = NULL;
+	if (argc < 2) {
+		printf("pwf: please specify a pattern to search for\n");
+		return 1;
+	}
+	while (--argc > 0 && (*++argv)[0] == '-') {
+		i = 0;
+		while ((c = (*argv)[++i]) != '\0') {
+			switch (c) {
+				case 'x':
+					except = 1;
+					break;
+				case 'n':
+					number = 1;
+					break;
+				case 'f':
+					showfiles = 1;
+					break;
+				case 'h':
+					printf("Usage: pwf [-{xnfh}] PATTERN [file] [file] ...\n");
+					break;
+				default:
+					printf("pwf: illegal option %c\n", c);
+					argc = 0;
+					found = -1;
+					break;
+			}
+		}
+	}
+	/* The first argument _after_ options should be the pattern; without
+	 * a pattern we shouldn't be accepting anything at all.
+	 */
+	pattern = *argv;
+	/* This handles the edge-case of no file names given. It's somewhat of an
+	 * ugly hack since it fools argc, but it gets the job done. */
+	if (argc == 1) {
+		infile = stdin;
+		filename = "stdin";
+		argc++;
+	}
+	while (--argc > 0) {
+		/* Now we handle the case of a given filename */
+		if (infile != stdin) {
+			filename = *(++argv);
+			infile = fopen(filename, "r");
+			if (filename == NULL || infile == NULL) {
+				printf("File %s could not be opened; falling back to stdin...\n", filename);
+				infile = stdin;
+				filename = "stdin";
+			}
+		}
+		/* We have _some_ usable file to work with, now. Let's do the thing */
+		lineno = 0;
+		while (get_line(line, MAXLINE, infile) > 0) {
+			lineno++;
+			if ((strstr(line, pattern) != NULL) != except) {
+				if (showfiles) {
+					printf("%s:", filename);
+				}
+				if (number) {
+					printf("%ld:", lineno);
+				}
+				printf("%s", line);
+				found++;
+			}
+		}
+	}
+	return found;
+}
+
+int get_line(char *s, int lim, FILE *f) {
+	int c, i;
+	for (i = 0; i < lim - 1 && (c = getc(f)) != EOF && c != '\n'; ++i) {
+		*s++ = c;
+	}
+	if (c == '\n') {
+		*s++ = c;
+		i++;
+	}
+	*s = '\0';
+	return i;
+}