Add apr_parse_addr_port() for parsing the hostname:port portion of

URLs and similar strings.

Somebody with VC++ 6.0 needs to update the .dsp file(s) to include
inet_pton.c.  The BeOS Makefile.in wasn't updated because I think
BeOS uses the unix code now.


git-svn-id: http://svn.apache.org/repos/asf/apr/apr/trunk@60916 13f79535-47bb-0310-9956-ffa450edef68

6 files changed, 140 insertions, 0 deletions

diff --git a/CHANGES b/CHANGES
index 8afae6533..a27e309ff 100644
--- a/CHANGES
+++ b/CHANGES
@@ -1,4 +1,6 @@
 Changes with APR a9
+  *) Add apr_parse_addr_port() for parsing the hostname:port portion
+     of URLs and similar strings.  [Jeff Trawick]
 
   *) Add Win32 MMAP support [William Rowe]
 
diff --git a/aprlib.def b/aprlib.def
index 04efde19f..3ed3b2e1e 100644
--- a/aprlib.def
+++ b/aprlib.def
@@ -88,6 +88,7 @@ EXPORTS
         apr_get_port
 	apr_set_port
         apr_get_inaddr
+        apr_parse_addr_port
 
 ;
 ;
diff --git a/include/apr_network_io.h b/include/apr_network_io.h
index 6683f555d..219fbfbde 100644
--- a/include/apr_network_io.h
+++ b/include/apr_network_io.h
@@ -306,6 +306,39 @@ apr_status_t apr_getaddrinfo(apr_sockaddr_t **sa,
                              apr_pool_t *p);
 
 /**
+ * Parse hostname/IP address with scope id and port.
+ *
+ * Any of the following strings are accepted:
+ *   8080                  (just the port number)
+ *   www.apache.org        (just the hostname)
+ *   www.apache.org:8080   (hostname and port number)
+ *   [fe80::1]:80          (IPv6 numeric address string only)
+ *   [fe80::1%eth0]        (IPv6 numeric address string and scope id)
+ *
+ * Invalid strings:
+ *                         (empty string)
+ *   [abc]                 (not valid IPv6 numeric address string)
+ *   abc:65536             (invalid port number)
+ *
+ * @param addr The new buffer containing just the hostname.  On output, *addr 
+ *             will be NULL if no hostname/IP address was specfied.
+ * @param scope_id The new buffer containing just the scope id.  On output, 
+ *                 *scope_id will be NULL if no scope id was specified.
+ * @param port The port number.  On output, *port will be 0 if no port was 
+ *             specified.
+ * @param str The input string to be parsed.
+ * @param p The pool from which *addr and *scope_id are allocated.
+ * @tip If scope id shouldn't be allowed, check for scope_id != NULL in addition 
+ *      to checking the return code.  If addr/hostname should be required, check 
+ *      for addr == NULL in addition to checking the return code.
+ */
+apr_status_t apr_parse_addr_port(char **addr,
+                                 char **scope_id,
+                                 apr_port_t *port,
+                                 const char *str,
+                                 apr_pool_t *p);
+
+/**
  * Get name of the current machine
  * @param buf A buffer to store the hostname in.
  * @param len The maximum length of the hostname that can be stored in the
diff --git a/libapr.def b/libapr.def
index 04efde19f..3ed3b2e1e 100644
--- a/libapr.def
+++ b/libapr.def
@@ -88,6 +88,7 @@ EXPORTS
         apr_get_port
 	apr_set_port
         apr_get_inaddr
+        apr_parse_addr_port
 
 ;
 ;
diff --git a/network_io/os2/Makefile.in b/network_io/os2/Makefile.in
index f23522533..e5077ea8a 100644
--- a/network_io/os2/Makefile.in
+++ b/network_io/os2/Makefile.in
@@ -17,6 +17,7 @@ OBJS=poll.o \
 	sockopt.o \
 	sockaddr.o \
 	inet_ntop.o \
+	inet_pton.o \
 	os2calls.o
 
 .c.o:
diff --git a/network_io/unix/sa_common.c b/network_io/unix/sa_common.c
index f282df03f..3c0fa0c57 100644
--- a/network_io/unix/sa_common.c
+++ b/network_io/unix/sa_common.c
@@ -65,6 +65,10 @@
 
 #include "apr.h"
 
+#ifdef HAVE_STDLIB_H
+#include <stdlib.h>
+#endif
+
 apr_status_t apr_set_port(apr_sockaddr_t *sockaddr, apr_port_t port)
 {
     /* XXX IPv6: assumes sin_port and sin6_port at same offset */
@@ -178,6 +182,104 @@ apr_status_t apr_get_sockaddr(apr_sockaddr_t **sa, apr_interface_e which, apr_so
     return APR_SUCCESS;
 }
 
+apr_status_t apr_parse_addr_port(char **addr,
+                                 char **scope_id,
+                                 apr_port_t *port,
+                                 const char *str,
+                                 apr_pool_t *p)
+{
+    const char *ch, *lastchar;
+    int big_port;
+    apr_size_t addrlen;
+
+    *addr = NULL;         /* assume not specified */
+    *scope_id = NULL;     /* assume not specified */
+    *port = 0;            /* assume not specified */
+
+    /* First handle the optional port number.  That may be all that
+     * is specified in the string.
+     */
+    ch = lastchar = str + strlen(str) - 1;
+    while (ch >= str && apr_isdigit(*ch)) {
+        --ch;
+    }
+
+    if (ch < str) {       /* Entire string is the port. */
+        big_port = atoi(str);
+        if (big_port < 1 || big_port > 65535) {
+            return APR_EINVAL;
+        }
+        *port = big_port;
+        return APR_SUCCESS;
+    }
+
+    if (*ch == ':' && ch < lastchar) { /* host and port number specified */
+        if (ch == str) {               /* string starts with ':' -- bad */
+            return APR_EINVAL;
+        }
+        big_port = atoi(ch + 1);
+        if (big_port < 1 || big_port > 65535) {
+            return APR_EINVAL;
+        }
+        *port = big_port;
+        lastchar = ch - 1;
+    }
+
+    /* now handle the hostname */
+    addrlen = lastchar - str + 1;
+
+#if APR_HAVE_IPV6 /* XXX don't require this; would need to pass char[] for ipaddr and always define APR_INET6 */
+    if (*str == '[') {
+        const char *end_bracket = memchr(str, ']', addrlen);
+        struct in6_addr ipaddr;
+        const char *scope_delim;
+
+        if (!end_bracket || end_bracket != lastchar) {
+            *port = 0;
+            return APR_EINVAL;
+        }
+
+        /* handle scope id; this is the only context where it is allowed */
+        scope_delim = memchr(str, '%', addrlen);
+        if (scope_delim) {
+            if (scope_delim == end_bracket - 1) { /* '%' without scope identifier */
+                *port = 0;
+                return APR_EINVAL;
+            }
+            addrlen = scope_delim - str - 1;
+            *scope_id = apr_palloc(p, end_bracket - scope_delim);
+            memcpy(*scope_id, scope_delim + 1, end_bracket - scope_delim - 1);
+            (*scope_id)[end_bracket - scope_delim - 1] = '\0';
+        }
+        else {
+            addrlen = addrlen - 2; /* minus 2 for '[' and ']' */
+        }
+
+        *addr = apr_palloc(p, addrlen + 1);
+        memcpy(*addr,
+               str + 1,
+               addrlen);
+        (*addr)[addrlen] = '\0';
+        if (apr_inet_pton(AF_INET6, *addr, &ipaddr) != 1) {
+            *addr = NULL;
+            *scope_id = NULL;
+            *port = 0;
+            return APR_EINVAL;
+        }
+    }
+    else 
+#endif
+    {
+        /* XXX If '%' is not a valid char in a DNS name, we *could* check 
+         *     for bogus scope ids first.
+         */
+        *addr = apr_palloc(p, addrlen + 1);
+        memcpy(*addr, str, addrlen);
+        (*addr)[addrlen] = '\0';
+    }
+    return APR_SUCCESS;
+}
+
 apr_status_t apr_getaddrinfo(apr_sockaddr_t **sa, const char *hostname, 
                              apr_int32_t family, apr_port_t port,
                              apr_int32_t flags, apr_pool_t *p)