proxy: fix subtle parser bug for some cmds1.6.16

in the request parser the new endcap token was always the end of the
line, but it should be the start of the next token. For some use cases
the scanning stops early, or if we have too many tokens it would make
the final token look like the rest of the line.

1 files changed, 14 insertions, 2 deletions

diff --git a/proxy_request.c b/proxy_request.c
index 3face1f..497e57c 100644
--- a/proxy_request.c
+++ b/proxy_request.c
@@ -23,29 +23,41 @@ static int _process_tokenize(mcp_parser_t *pr, const size_t max) {
     while (s != end) {
         switch (state) {
             case 0:
+                // scanning for first non-space to find a token.
                 if (*s != ' ') {
                     pr->tokens[curtoken] = s - pr->request;
                     if (++curtoken == max) {
-                        goto endloop;
+                        s++;
+                        state = 2;
+                        break;
                     }
                     state = 1;
                 }
                 s++;
                 break;
             case 1:
+                // advance over a token
                 if (*s != ' ') {
                     s++;
                 } else {
                     state = 0;
                 }
                 break;
+            case 2:
+                // hit max tokens before end of the line.
+                // keep advancing so we can place endcap token.
+                if (*s == ' ') {
+                    goto endloop;
+                }
+                s++;
+                break;
         }
     }
 endloop:
 
     // endcap token so we can quickly find the length of any token by looking
     // at the next one.
-    pr->tokens[curtoken] = len;
+    pr->tokens[curtoken] = s - pr->request;
     pr->ntokens = curtoken;
     P_DEBUG("%s: cur_tokens: %d\n", __func__, curtoken);