Fix destructuring alternation

Attempting to use the existing FORK_OPT opcode resulted in difficulty
knowing when to pop an error message off the stack and when not to. This
commit makes DESTRUCTURE_ALT a real opcode that is identical to
FORK_OPT, except for never pushing the error message onto the stack when
continuing from an error backtrack.

Some small changes were necessary to the DUP/POP behavior surrounding
destructuring to accomodate this.
This commit is contained in:
William Langford committed 2018-08-17 23:15:48 -04:00
1 parent 3dc5f4e948
commit 0673dee1b3
4 files changed
+122 -21

No files matched your search

+9 -3
View File
@@ -799,21 +799,27 @@ jv jq_next(jq_state *jq) {
}
case FORK_OPT:
case DESTRUCTURE_ALT:
case FORK: {
stack_save(jq, pc - 1, stack_get_pos(jq));
pc++; // skip offset this time
break;
}
case ON_BACKTRACK(FORK_OPT): {
case ON_BACKTRACK(FORK_OPT):
case ON_BACKTRACK(DESTRUCTURE_ALT): {
if (jv_is_valid(jq->error)) {
// `try EXP ...` backtracked here (no value, `empty`), so we backtrack more
jv_free(stack_pop(jq));
goto do_backtrack;
}
// `try EXP ...` exception caught in EXP
jv_free(stack_pop(jq)); // free the input
stack_push(jq, jv_invalid_get_msg(jq->error)); // push the error's message
// DESTRUCTURE_ALT doesn't want the error message on the stack,
// as we would just want to throw it away anyway.
if (opcode != ON_BACKTRACK(DESTRUCTURE_ALT)) {
jv_free(stack_pop(jq)); // free the input
stack_push(jq, jv_invalid_get_msg(jq->error)); // push the error's message
}
jq->error = jv_null();
uint16_t offset = *pc++;
pc += offset;