Fixed bug #61038; "Z" and better behavior for unpack()

Added new "Z" argument to pack/unpack, now allowing "a" to return data without stripping, and "A" strips all trailing white space, while "Z" will strip everything after the first null.
14 years ago · 4968fa644b
5 changed files with 104 additions and 10 deletions
--- a/4
+++ b/4
@ -9,9 +9,11 @@ PHP                                                                        NEWS
  . Support constant array/string dereferencing. (Laruence)

 - Core:
+  . Fixed bug #61681 (Malformed grammar). (Nikita Popov, Etienne, Laruence).
+  . Fixed bug #61038 (unpack("a5", "str\0\0") does not work as expected).
+    (srgoogleguy, Gustavo)
  . Implemented FR #60738 (Allow 'set_error_handler' to handle NULL).
    (Laruence, Nikita Popov)
-  . Fixed bug #61681 (Malformed grammar). (Nikita Popov, Etienne, Laruence).

 - cURL:
  . Added support for CURLOPT_FTP_RESPONSE_TIMEOUT, CURLOPT_APPEND, 
--- a/6
+++ b/6
@ -46,6 +46,12 @@ PHP X.Y UPGRADE NOTES
 4. Changed Functions
 ========================================

+- pack()/unpack() had the following changes, which bring it more in line
+  with Perl's behavior:
+  - Implemented format character "Z": NUL-padded string
+  - "a" now does not remove trailing NUL characters on unpack() anymore
+  - "A" will now strip all trailing ASCII whitespace on unpack() (it used to
+    remove only trailing spaces.

 ========================================
 5. New Functions
--- a/ext/standard/pack.c
+++ b/ext/standard/pack.c
@ -99,7 +99,7 @@ static void php_pack(zval **val, int size, int *map, char *output)
 /* }}} */

 /* pack() idea stolen from Perl (implemented formats behave the same as there)
- * Implemented formats are A, a, h, H, c, C, s, S, i, I, l, L, n, N, f, d, x, X, @.
+ * Implemented formats are Z, A, a, h, H, c, C, s, S, i, I, l, L, n, N, f, d, x, X, @.
 */
 /* {{{ proto string pack(string format, mixed arg1 [, mixed arg2 [, mixed ...]])
   Takes one or more arguments and packs them into a binary string according to the format argument */
@ -170,6 +170,7 @@ PHP_FUNCTION(pack)
 			/* Always uses one arg */
 			case 'a': 
 			case 'A': 
+			case 'Z': 
 			case 'h': 
 			case 'H':
 				if (currentarg >= num_args) {
@ -250,6 +251,7 @@ PHP_FUNCTION(pack)

 			case 'a': 
 			case 'A':
+			case 'Z':
 			case 'c': 
 			case 'C':
 			case 'x':
@ -315,7 +317,8 @@ PHP_FUNCTION(pack)
 		switch ((int) code) {
 			case 'a': 
 			case 'A': 
-				memset(&output[outputpos], (code == 'a') ? '\0' : ' ', arg);
+			case 'Z': 
+				memset(&output[outputpos], (code == 'a' || code == 'Z') ? '\0' : ' ', arg);
 				val = argv[currentarg++];
 				if (Z_ISREF_PP(val)) {
 					SEPARATE_ZVAL(val);
@ -511,7 +514,7 @@ static long php_unpack(char *data, int size, int issigned, int *map)
 * chars1, chars2, and ints.
 * Numeric pack types will return numbers, a and A will return strings,
 * f and d will return doubles.
- * Implemented formats are A, a, h, H, c, C, s, S, i, I, l, L, n, N, f, d, x, X, @.
+ * Implemented formats are Z, A, a, h, H, c, C, s, S, i, I, l, L, n, N, f, d, x, X, @.
 */
 /* {{{ proto array unpack(string format, string input)
   Unpack binary string into named array elements according to format argument */
@ -586,6 +589,7 @@ PHP_FUNCTION(unpack)

 			case 'a': 
 			case 'A':
+			case 'Z':
 				size = arg;
 				arg = 1;
 				break;
@ -662,9 +666,24 @@ PHP_FUNCTION(unpack)

 			if ((inputpos + size) <= inputlen) {
 				switch ((int) type) {
-					case 'a': 
+					case 'a': {
+						/* a will not strip any trailing whitespace or null padding */
+						char pad = ' ';
+						int len = inputlen - inputpos;	/* Remaining string */
+
+						/* If size was given take minimum of len and size */
+						if ((size >= 0) && (len > size)) {
+							len = size;
+						}
+
+						size = len;
+
+						add_assoc_stringl(return_value, n, &input[inputpos], len, 1);
+						break;
+					}
 					case 'A': {
-						char pad = (type == 'a') ? '\0' : ' ';
+						/* A will strip any trailing whitespace */
+						char padn = '\0'; char pads = ' '; char padt = '\t'; char padc = '\r'; char padl = '\n';
 						int len = inputlen - inputpos;	/* Remaining string */

 						/* If size was given take minimum of len and size */
@ -674,15 +693,45 @@ PHP_FUNCTION(unpack)

 						size = len;

-						/* Remove padding chars from unpacked data */
+						/* Remove trailing white space and nulls chars from unpacked data */
 						while (--len >= 0) {
-							if (input[inputpos + len] != pad)
+							if (input[inputpos + len] != padn
+								&& input[inputpos + len] != pads
+								&& input[inputpos + len] != padt
+								&& input[inputpos + len] != padc
+								&& input[inputpos + len] != padl
+							)
 								break;
 						}

 						add_assoc_stringl(return_value, n, &input[inputpos], len + 1, 1);
 						break;
 					}
+					/* New option added for Z to remain in-line with the Perl implementation */
+					case 'Z': {
+						/* Z will strip everything after the first null character */
+						char pad = '\0';
+						int len = inputlen - inputpos;	/* Remaining string */
+
+						/* If size was given take minimum of len and size */
+						if ((size >= 0) && (len > size)) {
+							len = size;
+						}
+
+						size = len;
+
+						/* Remove everything after the first null */
+						int s = 0;
+						while (s++ <= len) {
+							if (input[inputpos + s] == pad)
+								break;
+						}
+						len = s;
+
+						add_assoc_stringl(return_value, n, &input[inputpos], len, 1);
+						break;
+					}
+
 					
 					case 'h': 
 					case 'H': {
--- a/ext/standard/tests/strings/bug61038.phpt
+++ b/ext/standard/tests/strings/bug61038.phpt
@ -0,0 +1,37 @@
+--TEST--
+BugFix #61038
+--FILE--
+<?php
+	var_dump(unpack("Z4", pack("Z4", "foo")));
+	var_dump(unpack("a4", pack("a4", "foo")));
+	var_dump(unpack("A4", pack("A4", "foo")));
+	var_dump(unpack("a9", pack("a*", "foo\x00bar\x00 ")));
+	var_dump(unpack("A9", pack("a*", "foo\x00bar\x00 ")));
+	var_dump(unpack("Z9", pack("a*", "foo\x00bar\x00 ")));
+?>
+--EXPECTF--
+array(1) {
+  [1]=>
+  string(3) "foo"
+}
+array(1) {
+  [1]=>
+  string(4) "foo%c"
+}
+array(1) {
+  [1]=>
+  string(3) "foo"
+}
+array(1) {
+  [1]=>
+  string(9) "foo%cbar%c "
+}
+array(1) {
+  [1]=>
+  string(7) "foo%cbar"
+}
+array(1) {
+  [1]=>
+  string(3) "foo"
+}
+
--- a/ext/standard/tests/strings/unpack_error.phpt
+++ b/ext/standard/tests/strings/unpack_error.phpt
@ -19,7 +19,7 @@ var_dump(unpack("I", pack("I", 65534), $extra_arg));

 echo "\n-- Testing unpack() function with invalid format character --\n";
 $extra_arg = 10;
-var_dump(unpack("Z", pack("I", 65534)));
+var_dump(unpack("G", pack("I", 65534)));
 ?>
 ===DONE===
 --EXPECTF--
@ -37,6 +37,6 @@ NULL

 -- Testing unpack() function with invalid format character --

-Warning: unpack(): Invalid format type Z in %s on line %d
+Warning: unpack(): Invalid format type G in %s on line %d
 bool(false)
 ===DONE===