HTTP parser: fixed parsing of target after literal space character.

In theory, all space characters in request target must be encoded; however,
some clients may violate the specification.  For the sake of interoperability,
Unit supports unencoded space characters.

Previously, if there was a space character before the extension or arguments
parts, those parts weren't recognized.  Also, quoted symbols and complex
target weren't detected after a space character.

3 files changed, 63 insertions, 3 deletions

diff --git a/src/nxt_http_parse.c b/src/nxt_http_parse.c
index 4d806c61..5b009d96 100644
--- a/src/nxt_http_parse.c
+++ b/src/nxt_http_parse.c
@@ -164,6 +164,7 @@ nxt_http_parse_request_line(nxt_http_request_parse_t *rp, u_char **pos,
 {
     u_char                   *p, ch, *after_slash, *exten, *args;
     nxt_int_t                rc;
+    nxt_bool_t               rest;
     nxt_http_ver_t           ver;
     nxt_http_target_traps_e  trap;
 
@@ -256,6 +257,9 @@ nxt_http_parse_request_line(nxt_http_request_parse_t *rp, u_char **pos,
     after_slash = p + 1;
     exten = NULL;
     args = NULL;
+    rest = 0;
+
+continue_target:
 
     for ( ;; ) {
         p++;
@@ -312,6 +316,8 @@ nxt_http_parse_request_line(nxt_http_request_parse_t *rp, u_char **pos,
 
 rest_of_target:
 
+    rest = 1;
+
     for ( ;; ) {
         p++;
 
@@ -378,7 +384,12 @@ space_after_target:
         }
 
         rp->space_in_target = 1;
-        goto rest_of_target;
+
+        if (rest) {
+            goto rest_of_target;
+        }
+
+        goto continue_target;
     }
 
     /* " HTTP/1.1\r\n" or " HTTP/1.1\n" */
@@ -392,7 +403,12 @@ space_after_target:
         }
 
         rp->space_in_target = 1;
-        goto rest_of_target;
+
+        if (rest) {
+            goto rest_of_target;
+        }
+
+        goto continue_target;
     }
 
     nxt_memcpy(ver.str, &p[1], 8);
diff --git a/src/test/nxt_http_parse_test.c b/src/test/nxt_http_parse_test.c
index 6fbda25a..5498cb1f 100644
--- a/src/test/nxt_http_parse_test.c
+++ b/src/test/nxt_http_parse_test.c
@@ -206,13 +206,26 @@ static nxt_http_parse_test_case_t  nxt_http_test_cases[] = {
         }}
     },
     {
+        nxt_string("GET /na %20me.ext?args HTTP/1.0\r\n\r\n"),
+        NXT_DONE,
+        &nxt_http_parse_test_request_line,
+        { .request_line = {
+            nxt_string("GET"),
+            nxt_string("/na %20me.ext?args"),
+            nxt_string("ext"),
+            nxt_string("args"),
+            "HTTP/1.0",
+            0, 1, 1
+        }}
+    },
+    {
         nxt_string("GET / HTTP/1.0 HTTP/1.1\r\n\r\n"),
         NXT_DONE,
         &nxt_http_parse_test_request_line,
         { .request_line = {
             nxt_string("GET"),
             nxt_string("/ HTTP/1.0"),
-            nxt_null_string,
+            nxt_string("0"),
             nxt_null_string,
             "HTTP/1.1",
             0, 0, 1
diff --git a/test/test_python_application.py b/test/test_python_application.py
index 1f366570..5b6e2089 100644
--- a/test/test_python_application.py
+++ b/test/test_python_application.py
@@ -71,6 +71,37 @@ class TestPythonApplication(TestApplicationPython):
             'Query-String header',
         )
 
+    def test_python_application_query_string_space(self):
+        self.load('query_string')
+
+        resp = self.get(url='/ ?var1=val1&var2=val2')
+        self.assertEqual(
+            resp['headers']['Query-String'],
+            'var1=val1&var2=val2',
+            'Query-String space',
+        )
+
+        resp = self.get(url='/ %20?var1=val1&var2=val2')
+        self.assertEqual(
+            resp['headers']['Query-String'],
+            'var1=val1&var2=val2',
+            'Query-String space 2',
+        )
+
+        resp = self.get(url='/ %20 ?var1=val1&var2=val2')
+        self.assertEqual(
+            resp['headers']['Query-String'],
+            'var1=val1&var2=val2',
+            'Query-String space 3',
+        )
+
+        resp = self.get(url='/blah %20 blah? var1= val1 & var2=val2')
+        self.assertEqual(
+            resp['headers']['Query-String'],
+            ' var1= val1 & var2=val2',
+            'Query-String space 4',
+        )
+
     def test_python_application_query_string_empty(self):
         self.load('query_string')