Cleanup; add a missing test.

mwparserfromhell/parser/tokenizer.c

@@ -2783,12 +2783,6 @@ Tokenizer_handle_table_cell(Tokenizer* self, const char *markup,
                           line_context))
             return -1;
         padding = Tokenizer_parse_as_table_style(self, '|', 0);
-        if (BAD_ROUTE) {
-            trash = Tokenizer_pop(self);
-            Py_XDECREF(trash);
-            self->head = reset;
-            return 0;
-        }
         if (!padding)
             return -1;
         style = Tokenizer_pop(self);

mwparserfromhell/parser/tokenizer.py

@@ -1325,7 +1325,7 @@ class Tokenizer(object):
             elif this in ("\n", ":") and self._context & contexts.DL_TERM:
                 self._handle_dl_term()
                 if this == "\n":
-                    # kill potential table contexts
+                    # Kill potential table contexts
                     self._context &= ~contexts.TABLE_CELL_LINE_CONTEXTS
             # Start of table parsing
             elif this == "{" and next == "|" and (self._read(-1) in ("\n", self.START) or

tests/tokenizer/tags_wikimarkup.mwtest

@@ -447,6 +447,13 @@ output: [TagOpenOpen(wiki_markup=":"), Text(text="dd"), TagCloseSelfclose(), Tag
 
 ---
 
+name:   dt_dd_mix4
+label:  another example of correct dt/dd usage, with a trigger for a specific parse route
+input:  ";foo]:bar"
+output: [TagOpenOpen(wiki_markup=";"), Text(text="dt"), TagCloseSelfclose(), Text(text="foo]"), TagOpenOpen(wiki_markup=":"), Text(text="dd"), TagCloseSelfclose(), Text(text="bar")]
+
+---
+
 name:   ul_ol_dt_dd_mix
 label:  an assortment of uls, ols, dds, and dts
 input:  ";:#*foo\n:#*;foo\n#*;:foo\n*;:#foo"