You are viewing a plain text version of this content. The canonical link for it is here.
Posted to cvs@httpd.apache.org by fa...@apache.org on 2001/01/18 04:34:26 UTC
cvs commit: apache-1.3/src/main http_protocol.c
fanf 01/01/17 19:34:25
Modified: src CHANGES
src/main http_protocol.c
Log:
Rewrite byterange handling so that:
* the syntax is parsed properly
* unsatisfiable requests cause a 416 error
* the API is completely preserved
PR: 6973
Submitted by: fanf, wrowe
Reviewed by: jim, stoddard, trawick, (dirkx, martin)
Revision Changes Path
1.1613 +4 -0 apache-1.3/src/CHANGES
Index: CHANGES
===================================================================
RCS file: /home/cvs/apache-1.3/src/CHANGES,v
retrieving revision 1.1612
retrieving revision 1.1613
diff -u -u -r1.1612 -r1.1613
--- CHANGES 2001/01/04 04:12:41 1.1612
+++ CHANGES 2001/01/18 03:34:25 1.1613
@@ -1,5 +1,9 @@
Changes with Apache 1.3.15
+ *) Fix Range header processing to properly parse the syntax specified
+ in RFC 2616, and properly handle unsatisfiable requests by returning
+ a 416 error. PR#6973 [Tony Finch, William Rowe]
+
*) Remove some human-readable fluff from the machine-readable mod_status
display. PR#7025 [Youichirou Koga <y-...@apache.or.jp>]
1.297 +202 -139 apache-1.3/src/main/http_protocol.c
Index: http_protocol.c
===================================================================
RCS file: /home/cvs/apache-1.3/src/main/http_protocol.c,v
retrieving revision 1.296
retrieving revision 1.297
diff -u -u -r1.296 -r1.297
--- http_protocol.c 2001/01/15 17:05:01 1.296
+++ http_protocol.c 2001/01/18 03:34:25 1.297
@@ -141,48 +141,139 @@
return type;
}
-static int parse_byterange(char *range, long clength, long *start, long *end)
+enum byterange_token {
+ BYTERANGE_OK,
+ BYTERANGE_EMPTY,
+ BYTERANGE_BADSYNTAX,
+ BYTERANGE_UNSATISFIABLE
+};
+
+static enum byterange_token
+ parse_byterange(request_rec *r, long *start, long *end)
{
- char *dash = strchr(range, '-');
+ /* parsing first, semantics later */
- if (!dash)
- return 0;
+ while (ap_isspace(*r->range))
+ ++r->range;
- if ((dash == range)) {
- /* In the form "-5" */
- *start = clength - atol(dash + 1);
- *end = clength - 1;
+ /* check for an empty range, which is OK */
+ if (*r->range == '\0') {
+ return BYTERANGE_EMPTY;
}
- else {
- *dash = '\0';
- dash++;
- *start = atol(range);
- if (*dash)
- *end = atol(dash);
- else /* "5-" */
- *end = clength - 1;
+ else if (*r->range == ',') {
+ ++r->range;
+ return BYTERANGE_EMPTY;
}
+
+ if (ap_isdigit(*r->range))
+ *start = strtol(r->range, (char **)&r->range, 10);
+ else
+ *start = -1;
+
+ while (ap_isspace(*r->range))
+ ++r->range;
+
+ if (*r->range != '-')
+ return BYTERANGE_BADSYNTAX;
+ ++r->range;
- if (*start < 0)
- *start = 0;
+ while (ap_isspace(*r->range))
+ ++r->range;
- if (*end >= clength)
- *end = clength - 1;
+ if (ap_isdigit(*r->range))
+ *end = strtol(r->range, (char **)&r->range, 10);
+ else
+ *end = -1;
- if (*start > *end)
- return 0;
+ while (ap_isspace(*r->range))
+ ++r->range;
- return (*start > 0 || *end < clength - 1);
+ /* check the end of the range */
+ if (*r->range == ',') {
+ ++r->range;
+ }
+ else if (*r->range != '\0') {
+ return BYTERANGE_BADSYNTAX;
+ }
+
+ /* parsing done; now check the numbers */
+
+ if (*start < 0) { /* suffix-byte-range-spec */
+ if (*end < 0) /* no numbers */
+ return BYTERANGE_BADSYNTAX;
+ *start = r->clength - *end;
+ if (*start < 0)
+ *start = 0;
+ *end = r->clength - 1;
+ }
+ else {
+ if (*end >= 0 && *start > *end) /* out-of-order range */
+ return BYTERANGE_BADSYNTAX;
+ if (*end < 0 || *end >= r->clength)
+ *end = r->clength - 1;
+ }
+ /* RFC 2616 is somewhat unclear about what we should do if the end
+ * is missing and the start is after the clength. The robustness
+ * principle says we should accept it as an unsatisfiable range.
+ * We accept suffix-byte-range-specs like -0 for the same reason.
+ */
+ if (*start >= r->clength)
+ return BYTERANGE_UNSATISFIABLE;
+
+ return BYTERANGE_OK;
}
-static int internal_byterange(int, long *, request_rec *, const char **, long *,
- long *);
+/* If this function is called with output=1, it will spit out the
+ * correct headers for a byterange chunk. If output=0 it will not
+ * output anything but just return the number of bytes it would have
+ * output. If start or end are less than 0 then it will do a byterange
+ * chunk trailer instead of a header.
+ */
+static int byterange_boundary(request_rec *r, long start, long end, int output)
+{
+ int length = 0;
+#ifdef CHARSET_EBCDIC
+ /* determine current setting of conversion flag,
+ * set to ON (protocol strings MUST be converted)
+ * and reset to original setting before returning
+ */
+ PUSH_EBCDIC_OUTPUTCONVERSION_STATE(r->connection->client, 1);
+#endif /*CHARSET_EBCDIC*/
+
+ if (start < 0 || end < 0) {
+ if (output)
+ ap_rvputs(r, CRLF "--", r->boundary, "--" CRLF, NULL);
+ else
+ length = 4 + strlen(r->boundary) + 4;
+ }
+ else {
+ const char *ct = make_content_type(r, r->content_type);
+ char ts[MAX_STRING_LEN];
+
+ ap_snprintf(ts, sizeof(ts), "%ld-%ld/%ld", start, end, r->clength);
+ if (output)
+ ap_rvputs(r, CRLF "--", r->boundary, CRLF "Content-type: ",
+ ct, CRLF "Content-range: bytes ", ts, CRLF CRLF,
+ NULL);
+ else
+ length = 4 + strlen(r->boundary) + 16
+ + strlen(ct) + 23 + strlen(ts) + 4;
+ }
+
+#ifdef CHARSET_EBCDIC
+ POP_EBCDIC_OUTPUTCONVERSION_STATE(r->connection->client);
+#endif /*CHARSET_EBCDIC*/
+
+ return length;
+}
+
API_EXPORT(int) ap_set_byterange(request_rec *r)
{
const char *range, *if_range, *match;
- long range_start, range_end;
-
+ long length, start, end, one_start, one_end;
+ int ranges, empty;
+
if (!r->clength || r->assbackwards)
return 0;
@@ -202,6 +293,7 @@
if (!range || strncasecmp(range, "bytes=", 6)) {
return 0;
}
+ range += 6;
/* Check the If-Range header for Etag or Date.
* Note that this check will return false (as required) if either
@@ -218,129 +310,100 @@
return 0;
}
- if (!strchr(range, ',')) {
- /* A single range */
- if (!parse_byterange(ap_pstrdup(r->pool, range + 6), r->clength,
- &range_start, &range_end))
- return 0;
-
- r->byterange = 1;
+ /*
+ * Parse the byteranges, counting how many of them there are and
+ * the total number of bytes we will send to the client. This is a
+ * dummy run for the while(ap_each_byterange()) loop that the
+ * caller will perform if we return 1.
+ */
+ r->range = range;
+ r->boundary = ap_psprintf(r->pool, "%lx%lx",
+ r->request_time, (long) getpid());
+ length = 0;
+ ranges = 0;
+ empty = 1;
+ do {
+ switch (parse_byterange(r, &start, &end)) {
+ case BYTERANGE_UNSATISFIABLE:
+ empty = 0;
+ break;
+ default:
+ /* be more defensive here? */
+ case BYTERANGE_BADSYNTAX:
+ r->boundary = NULL;
+ r->range = NULL;
+ return 0;
+ case BYTERANGE_EMPTY:
+ break;
+ case BYTERANGE_OK:
+ ++ranges;
+ length += byterange_boundary(r, start, end, 0)
+ + end - start + 1;
+ /* save in case of unsatisfiable ranges */
+ one_start = start;
+ one_end = end;
+ break;
+ }
+ } while (*r->range != '\0');
+ if (ranges == 0) {
+ /* no ranges or only unsatisfiable ranges */
+ if (empty || if_range) {
+ r->boundary = NULL;
+ r->range = NULL;
+ return 0;
+ }
+ else {
+ ap_table_setn(r->headers_out, "Content-Range",
+ ap_psprintf(r->pool, "bytes */%ld", r->clength));
+ r->boundary = NULL;
+ r->range = range;
+ r->header_only = 1;
+ r->status = HTTP_RANGE_NOT_SATISFIABLE;
+ return 1;
+ }
+ }
+ else if (ranges == 1) {
+ /* simple handling of a single range -- no boundaries */
ap_table_setn(r->headers_out, "Content-Range",
ap_psprintf(r->pool, "bytes %ld-%ld/%ld",
- range_start, range_end, r->clength));
- ap_table_setn(r->headers_out, "Content-Length",
- ap_psprintf(r->pool, "%ld", range_end - range_start + 1));
+ one_start, one_end, r->clength));
+ ap_table_setn(r->headers_out, "Content-Length",
+ ap_psprintf(r->pool, "%ld", one_end - one_start + 1));
+ r->boundary = NULL;
+ r->byterange = 1;
+ r->range = range;
+ r->status = PARTIAL_CONTENT;
+ return 1;
}
else {
- /* a multiple range */
- const char *r_range = ap_pstrdup(r->pool, range + 6);
- long tlength = 0;
- int ret;
-
- r->byterange = 2;
- r->boundary = ap_psprintf(r->pool, "%lx%lx",
- r->request_time, (long) getpid());
- do {
- /* Loop while we have another range spec to process */
- ret = internal_byterange(0, &tlength, r, &r_range, NULL, NULL);
- } while (ret == 1);
- /* If an error occured processing one of the range specs, we
- * must fail */
- if (ret < 0) {
- r->byterange = 0;
- return 0;
- }
- ap_table_setn(r->headers_out, "Content-Length",
- ap_psprintf(r->pool, "%ld", tlength));
+ /* multiple ranges */
+ length += byterange_boundary(r, -1, -1, 0);
+ ap_table_setn(r->headers_out, "Content-Length",
+ ap_psprintf(r->pool, "%ld", length));
+ r->byterange = 2;
+ r->range = range;
+ r->status = PARTIAL_CONTENT;
+ return 1;
}
-
- r->status = PARTIAL_CONTENT;
- r->range = range + 6;
-
- return 1;
}
API_EXPORT(int) ap_each_byterange(request_rec *r, long *offset, long *length)
{
- return internal_byterange(1, NULL, r, &r->range, offset, length);
-}
-
-/* If this function is called with realreq=1, it will spit out
- * the correct headers for a byterange chunk, and set offset and
- * length to the positions they should be.
- *
- * If it is called with realreq=0, it will add to tlength the length
- * it *would* have used with realreq=1.
- *
- * Either case will return 1 if it should be called again, 0 when done,
- * or -1 if an error occurs AND realreq=0.
- */
-static int internal_byterange(int realreq, long *tlength, request_rec *r,
- const char **r_range, long *offset, long *length)
-{
- long range_start, range_end;
- char *range;
-#ifdef CHARSET_EBCDIC
- /* determine current setting of conversion flag,
- * set to ON (protocol strings MUST be converted)
- * and reset to original setting before returning
- */
- PUSH_EBCDIC_OUTPUTCONVERSION_STATE(r->connection->client, 1);
-#endif /*CHARSET_EBCDIC*/
-
- if (!**r_range) {
- if (r->byterange > 1) {
- if (realreq)
- ap_rvputs(r, CRLF "--", r->boundary, "--" CRLF, NULL);
- else
- *tlength += 4 + strlen(r->boundary) + 4;
- }
-#ifdef CHARSET_EBCDIC
- POP_EBCDIC_OUTPUTCONVERSION_STATE(r->connection->client);
-#endif /*CHARSET_EBCDIC*/
- return 0;
- }
-
- range = ap_getword(r->pool, r_range, ',');
- if (!parse_byterange(range, r->clength, &range_start, &range_end)) {
-#ifdef CHARSET_EBCDIC
- POP_EBCDIC_OUTPUTCONVERSION_STATE(r->connection->client);
-#endif /*CHARSET_EBCDIC*/
- if (!realreq)
- /* Return error on invalid syntax */
- return -1;
- else
- /* Should never get here */
- return 0;
- }
+ long start, end;
- if (r->byterange > 1) {
- const char *ct = make_content_type(r, r->content_type);
- char ts[MAX_STRING_LEN];
-
- ap_snprintf(ts, sizeof(ts), "%ld-%ld/%ld", range_start, range_end,
- r->clength);
- if (realreq)
- ap_rvputs(r, CRLF "--", r->boundary, CRLF "Content-type: ",
- ct, CRLF "Content-range: bytes ", ts, CRLF CRLF,
- NULL);
- else
- *tlength += 4 + strlen(r->boundary) + 16 + strlen(ct) + 23 +
- strlen(ts) + 4;
- }
-
- if (realreq) {
- *offset = range_start;
- *length = range_end - range_start + 1;
- }
- else {
- *tlength += range_end - range_start + 1;
- }
-#ifdef CHARSET_EBCDIC
- POP_EBCDIC_OUTPUTCONVERSION_STATE(r->connection->client);
-#endif /*CHARSET_EBCDIC*/
- return 1;
+ do {
+ if (parse_byterange(r, &start, &end) == BYTERANGE_OK) {
+ if (r->byterange > 1)
+ byterange_boundary(r, start, end, 1);
+ *offset = start;
+ *length = end - start + 1;
+ return 1;
+ }
+ } while (*r->range != '\0');
+ if (r->byterange > 1)
+ byterange_boundary(r, -1, -1, 1);
+ return 0;
}
API_EXPORT(int) ap_set_content_length(request_rec *r, long clength)