r14953@tombo: nickm | 2007-11-25 15:56:40 -0500

Replace evbuffer_readline with a more powerful evbuffer_readln that can handle more EOL styles, and that can give useful results when there are NUL characters inside the returned values. Includes regression tests.


svn:r550

buffer.c

@@ -211,45 +211,91 @@ evbuffer_remove(struct evbuffer *buf, void *data, size_t datlen)
  * Reads a line terminated by either '\r\n', '\n\r' or '\r' or '\n'.
  * The returned buffer needs to be freed by the called.
  */
-
 char *
 evbuffer_readline(struct evbuffer *buffer)
+{
+	return evbuffer_readln(buffer, NULL, EVBUFFER_EOL_ANY);
+}
+
+char *
+evbuffer_readln(struct evbuffer *buffer, size_t *n_read_out,
+		enum evbuffer_eol_style eol_style)
 {
 	u_char *data = EVBUFFER_DATA(buffer);
+	u_char *start_of_eol, *end_of_eol;
 	size_t len = EVBUFFER_LENGTH(buffer);
 	char *line;
-	unsigned int i;
+	unsigned int i, n_to_copy, n_to_drain;
 
-	for (i = 0; i < len; i++) {
-		if (data[i] == '\r' || data[i] == '\n')
-			break;
+	/* depending on eol_style, set start_of_eol to the first character
+	 * in the newline, and end_of_eol to one after the last character. */
+	switch (eol_style) {
+	case EVBUFFER_EOL_ANY:
+		for (i = 0; i < len; i++) {
+			if (data[i] == '\r' || data[i] == '\n')
+				break;
+		}
+		if (i == len)
+			return (NULL);
+		start_of_eol = data+i;
+		++i;
+		for ( ; i < len; i++) {
+			if (data[i] != '\r' && data[i] != '\n')
+				break;
+		}
+		end_of_eol = data+i;
+		break;
+	case EVBUFFER_EOL_CRLF:
+		end_of_eol = memchr(data, '\n', len);
+		if (!end_of_eol)
+			return (NULL);
+		if (end_of_eol > data && *(end_of_eol-1) == '\r')
+			start_of_eol = end_of_eol - 1;
+		else
+			start_of_eol = end_of_eol;
+		end_of_eol++; /*point to one after the LF. */
+		break;
+	case EVBUFFER_EOL_CRLF_STRICT: {
+		u_char *cp = data;
+		while ((cp = memchr(cp, '\r', len-(cp-data)))) {
+			if (cp < data+len-1 && *(cp+1) == '\n')
+				break;
+			if (++cp >= data+len) {
+				cp = NULL;
+				break;
+			}
+		}
+		if (!cp)
+			return (NULL);
+		start_of_eol = cp;
+		end_of_eol = cp+2;
+		break;
+	}
+	case EVBUFFER_EOL_LF:
+		start_of_eol = memchr(data, '\n', len);
+		if (!start_of_eol)
+			return (NULL);
+		end_of_eol = start_of_eol + 1;
+		break;
+	default:
+		return (NULL);
 	}
 
-	if (i == len)
-		return (NULL);
+	n_to_copy = start_of_eol - data;
+	n_to_drain = end_of_eol - data;
 
-	if ((line = event_malloc(i + 1)) == NULL) {
+	if ((line = event_malloc(n_to_copy+1)) == NULL) {
 		fprintf(stderr, "%s: out of memory\n", __func__);
-		evbuffer_drain(buffer, i);
+		evbuffer_drain(buffer, n_to_drain);
 		return (NULL);
 	}
 
-	memcpy(line, data, i);
-	line[i] = '\0';
+	memcpy(line, data, n_to_copy);
+	line[n_to_copy] = '\0';
 
-	/*
-	 * Some protocols terminate a line with '\r\n', so check for
-	 * that, too.
-	 */
-	if ( i < len - 1 ) {
-		char fch = data[i], sch = data[i+1];
-
-		/* Drain one more character if needed */
-		if ( (sch == '\r' || sch == '\n') && sch != fch )
-			i += 1;
-	}
-
-	evbuffer_drain(buffer, i + 1);
+	evbuffer_drain(buffer, n_to_drain);
+	if (n_read_out)
+		*n_read_out = (size_t)n_to_copy;
 
 	return (line);
 }

event.h

@@ -952,18 +952,40 @@ int evbuffer_add(struct evbuffer *, const void *, size_t);
  */
 int evbuffer_remove(struct evbuffer *, void *, size_t);
 
+/** Used to tell evbuffer_readln what kind of line-ending to look for.
+ */
+enum evbuffer_eol_style {
+	/** Any sequence of CR and LF characters is acceptable as an EOL. */
+	EVBUFFER_EOL_ANY,
+	/** An EOL is an LF, optionally preceded by a CR.  This style is
+	 * most useful for implementing text-based internet protocols. */
+	EVBUFFER_EOL_CRLF,
+	/** An EOL is a CR followed by an LF. */
+	EVBUFFER_EOL_CRLF_STRICT,
+	/** An EOL is a LF. */
+	EVBUFFER_EOL_LF
+};
 
 /**
  * Read a single line from an event buffer.
  *
- * Reads a line terminated by either '\r\n', '\n\r' or '\r' or '\n'.
- * The returned buffer needs to be freed by the caller.
+ * Reads a line terminated by an EOL as determined by the evbuffer_eol_style
+ * argument.  Returns a newly allocated nul-terminated string; the caller must
+ * free the returned value.  The EOL is not included in the returned string.
  *
  * @param buffer the evbuffer to read from
+ * @param n_read_out if non-NULL, points to a size_t that is set to the
+ *       number of characters in the returned string.  This is useful for
+ *       strings that can contain NUL characters.
+ * @param eol_style the style of line-ending to use.
  * @return pointer to a single line, or NULL if an error occurred
  */
-char *evbuffer_readline(struct evbuffer *);
+char *evbuffer_readln(struct evbuffer *, size_t *, enum evbuffer_eol_style);
 
+/**
+   Obsolete alias for evbuffer_readln(buffer, NULL, EOL_STYLE_ANY).
+ **/
+char *evbuffer_readline(struct evbuffer *);
 
 /**
   Move data from one evbuffer into another evbuffer.

http.c

@@ -685,7 +685,7 @@ evhttp_handle_chunked_read(struct evhttp_request *req, struct evbuffer *buf)
 	while ((len = EVBUFFER_LENGTH(buf)) > 0) {
 		if (req->ntoread < 0) {
 			/* Read chunk size */
-			char *p = evbuffer_readline(buf);
+			char *p = evbuffer_readln(buf, NULL, EVBUFFER_EOL_CRLF);
 			char *endp;
 			int error;
 			if (p == NULL)
@@ -1239,7 +1239,7 @@ evhttp_parse_lines(struct evhttp_request *req, struct evbuffer* buffer)
 	int done = 0;
 
 	struct evkeyvalq* headers = req->input_headers;
-	while ((line = evbuffer_readline(buffer)) != NULL) {
+	while ((line = evbuffer_readln(buffer, NULL, EVBUFFER_EOL_CRLF)) != NULL) {
 		char *skey, *svalue;
 
 		if (*line == '\0') { /* Last header - Done */

test/regress.c

@@ -808,6 +808,115 @@ test_evbuffer(void) {
 	cleanup_test();
 }
 
+void
+test_evbuffer_readln(void)
+{
+	struct evbuffer *evb = evbuffer_new();
+	const char *s;
+	char *cp = NULL;
+	size_t sz;
+	setup_test("Testing evbuffer_readln(): ");
+
+	/* Test EOL_ANY. */
+	s = "complex silly newline\r\n\n\r\n\n\rmore\0\n";
+	evbuffer_add(evb, s, strlen(s)+2);
+	cp = evbuffer_readln(evb, &sz, EVBUFFER_EOL_ANY);
+	if (!cp || sz != strlen(cp) || strcmp(cp, "complex silly newline"))
+		goto done;
+	free(cp);
+	cp = evbuffer_readln(evb, &sz, EVBUFFER_EOL_ANY);
+	if (!cp || sz != 5 || memcmp(cp, "more\0\0", 6))
+		goto done;
+	if (EVBUFFER_LENGTH(evb) != 0)
+		goto done;
+	s = "\nno newline";
+	evbuffer_add(evb, s, strlen(s));
+	free(cp);
+	cp = evbuffer_readln(evb, &sz, EVBUFFER_EOL_ANY);
+	if (!cp || sz || strcmp(cp, ""))
+		goto done;
+	free(cp);
+	cp = evbuffer_readln(evb, &sz, EVBUFFER_EOL_ANY);
+	if (cp)
+		goto done;
+	evbuffer_drain(evb, EVBUFFER_LENGTH(evb));
+	if (EVBUFFER_LENGTH(evb) != 0)
+		goto done;
+
+	/* Test EOL_CRLF */
+	s = "Line with\rin the middle\nLine with good crlf\r\n\nfinal\n";
+	evbuffer_add(evb, s, strlen(s));
+	cp = evbuffer_readln(evb, &sz, EVBUFFER_EOL_CRLF);
+	if (!cp || sz != strlen(cp) || strcmp(cp, "Line with\rin the middle"))
+		goto done;
+
+	free(cp);
+	cp = evbuffer_readln(evb, &sz, EVBUFFER_EOL_CRLF);
+	if (!cp || sz != strlen(cp) || strcmp(cp, "Line with good crlf"))
+		goto done;
+	free(cp);
+	cp = evbuffer_readln(evb, &sz, EVBUFFER_EOL_CRLF);
+	if (!cp || sz != strlen(cp) || strcmp(cp, ""))
+		goto done;
+	free(cp);
+	cp = evbuffer_readln(evb, &sz, EVBUFFER_EOL_CRLF);
+	if (!cp || sz != strlen(cp) || strcmp(cp, "final"))
+		goto done;
+	s = "x";
+	evbuffer_add(evb, s, 1);
+	free(cp);
+	cp = evbuffer_readln(evb, &sz, EVBUFFER_EOL_CRLF);
+	if (cp)
+		goto done;
+
+	/* Test CRLF_STRICT */
+	s = " and a bad crlf\nand a good one\r\n\r\nMore\r";
+	evbuffer_add(evb, s, strlen(s));
+	cp = evbuffer_readln(evb, &sz, EVBUFFER_EOL_CRLF_STRICT);
+	if (!cp || sz != strlen(cp) ||
+	    strcmp(cp, "x and a bad crlf\nand a good one"))
+		goto done;
+	free(cp);
+	cp = evbuffer_readln(evb, &sz, EVBUFFER_EOL_CRLF_STRICT);
+	if (!cp || sz != strlen(cp) || strcmp(cp, ""))
+		goto done;
+	free(cp);
+	cp = evbuffer_readln(evb, &sz, EVBUFFER_EOL_CRLF_STRICT);
+	if (cp)
+		goto done;
+	evbuffer_add(evb, "\n", 1);
+	cp = evbuffer_readln(evb, &sz, EVBUFFER_EOL_CRLF_STRICT);
+	if (!cp || sz != strlen(cp) || strcmp(cp, "More"))
+		goto done;
+	if (EVBUFFER_LENGTH(evb) != 0)
+		goto done;
+
+	/* Test LF */
+	s = "An\rand a nl\n\nText";
+	evbuffer_add(evb, s, strlen(s));
+	cp = evbuffer_readln(evb, &sz, EVBUFFER_EOL_LF);
+	if (!cp || sz != strlen(cp) || strcmp(cp, "An\rand a nl"))
+		goto done;
+	free(cp);
+	cp = evbuffer_readln(evb, &sz, EVBUFFER_EOL_LF);
+	if (!cp || sz != strlen(cp) || strcmp(cp, ""))
+		goto done;
+	free(cp);
+	cp = evbuffer_readln(evb, &sz, EVBUFFER_EOL_LF);
+	if (cp)
+		goto done;
+	evbuffer_add(evb, "\n", 1);
+	cp = evbuffer_readln(evb, &sz, EVBUFFER_EOL_LF);
+	if (!cp || sz != strlen(cp) || strcmp(cp, "Text"))
+		goto done;
+
+	test_ok = 1;
+ done:
+	evbuffer_free(evb);
+	if (cp) free(cp);
+	cleanup_test();
+}
+
 void
 test_evbuffer_find(void)
 {
@@ -1263,6 +1372,7 @@ main (int argc, char **argv)
 	test_priorities(3);
 
 	test_evbuffer();
+	test_evbuffer_readln();
 	test_evbuffer_find();
 	
 	test_bufferevent();