py/parse: Optimise away parse node that's just parenthesis around expr.
Before this patch, (x+y)*z would be parsed to a tree that contained a
redundant identity parse node corresponding to the parenthesis. With
this patch such nodes are optimised away, which reduces memory
requirements for expressions with parenthesis, and simplifies the
compiler because it doesn't need to handle this identity case.
A parenthesis parse node is still needed for tuples.
diff --git a/py/compile.c b/py/compile.c
index f349dca..8a30518 100644
--- a/py/compile.c
+++ b/py/compile.c
@@ -299,14 +299,12 @@
if (jump_if == false) {
EMIT_ARG(jump, label);
}
- } else if (MP_PARSE_NODE_IS_STRUCT_KIND(pns->nodes[0], PN_testlist_comp)) {
+ } else {
+ assert(MP_PARSE_NODE_IS_STRUCT_KIND(pns->nodes[0], PN_testlist_comp));
// non-empty tuple, acts as true for the condition
if (jump_if == true) {
EMIT_ARG(jump, label);
}
- } else {
- // parenthesis around 1 item, is just that item
- c_if_cond(comp, pns->nodes[0], jump_if, label);
}
return;
}
@@ -420,7 +418,6 @@
// assigns top of stack to pn
STATIC void c_assign(compiler_t *comp, mp_parse_node_t pn, assign_kind_t assign_kind) {
- tail_recursion:
assert(!MP_PARSE_NODE_IS_NULL(pn));
if (MP_PARSE_NODE_IS_LEAF(pn)) {
if (MP_PARSE_NODE_IS_ID(pn)) {
@@ -462,16 +459,13 @@
if (MP_PARSE_NODE_IS_NULL(pns->nodes[0])) {
// empty tuple
goto cannot_assign;
- } else if (MP_PARSE_NODE_IS_STRUCT_KIND(pns->nodes[0], PN_testlist_comp)) {
+ } else {
+ assert(MP_PARSE_NODE_IS_STRUCT_KIND(pns->nodes[0], PN_testlist_comp));
if (assign_kind != ASSIGN_STORE) {
goto bad_aug;
}
pns = (mp_parse_node_struct_t*)pns->nodes[0];
goto testlist_comp;
- } else {
- // parenthesis around 1 item, is just that item
- pn = pns->nodes[0];
- goto tail_recursion;
}
break;
@@ -885,7 +879,10 @@
}
} else if (MP_PARSE_NODE_IS_STRUCT_KIND(pn, PN_atom_paren)) {
pn = ((mp_parse_node_struct_t*)pn)->nodes[0];
- if (MP_PARSE_NODE_IS_STRUCT_KIND(pn, PN_testlist_comp)) {
+ if (MP_PARSE_NODE_IS_NULL(pn)) {
+ goto cannot_delete;
+ } else {
+ assert(MP_PARSE_NODE_IS_STRUCT_KIND(pn, PN_testlist_comp));
mp_parse_node_struct_t *pns = (mp_parse_node_struct_t*)pn;
// TODO perhaps factorise testlist_comp code with other uses of PN_testlist_comp
@@ -915,12 +912,9 @@
c_del_stmt(comp, pns->nodes[0]);
c_del_stmt(comp, pns->nodes[1]);
}
- } else {
- // tuple with 1 element
- c_del_stmt(comp, pn);
}
} else {
- // TODO is there anything else to implement?
+ // some arbitrary statment that we can't delete (eg del 1)
goto cannot_delete;
}
@@ -2185,7 +2179,8 @@
if (MP_PARSE_NODE_IS_NULL(pns->nodes[0])) {
// an empty tuple
c_tuple(comp, MP_PARSE_NODE_NULL, NULL);
- } else if (MP_PARSE_NODE_IS_STRUCT_KIND(pns->nodes[0], PN_testlist_comp)) {
+ } else {
+ assert(MP_PARSE_NODE_IS_STRUCT_KIND(pns->nodes[0], PN_testlist_comp));
pns = (mp_parse_node_struct_t*)pns->nodes[0];
assert(!MP_PARSE_NODE_IS_NULL(pns->nodes[1]));
if (MP_PARSE_NODE_IS_STRUCT(pns->nodes[1])) {
@@ -2209,9 +2204,6 @@
tuple_with_2_items:
c_tuple(comp, MP_PARSE_NODE_NULL, pns);
}
- } else {
- // parenthesis around a single item, is just that item
- compile_node(comp, pns->nodes[0]);
}
}
diff --git a/py/parse.c b/py/parse.c
index 1741757..6de0a0f 100644
--- a/py/parse.c
+++ b/py/parse.c
@@ -647,6 +647,20 @@
#endif
STATIC void push_result_rule(parser_t *parser, size_t src_line, const rule_t *rule, size_t num_args) {
+ // optimise away parenthesis around an expression if possible
+ if (rule->rule_id == RULE_atom_paren) {
+ // there should be just 1 arg for this rule
+ mp_parse_node_t pn = peek_result(parser, 0);
+ if (MP_PARSE_NODE_IS_NULL(pn)) {
+ // need to keep parenthesis for ()
+ } else if (MP_PARSE_NODE_IS_STRUCT_KIND(pn, RULE_testlist_comp)) {
+ // need to keep parenthesis for (a, b, ...)
+ } else {
+ // parenthesis around a single expression, so it's just the expression
+ return;
+ }
+ }
+
#if MICROPY_COMP_CONST_FOLDING
if (fold_constants(parser, rule, num_args)) {
// we folded this rule so return straight away
@@ -864,8 +878,6 @@
// if a rule has the RULE_ACT_ALLOW_IDENT bit set then this
// rule should not be emitted if it has only 1 argument
- // NOTE: can't set this flag for atom_paren because we need it
- // to distinguish, for example, [a,b] from [(a,b)]
if (rule->act & RULE_ACT_ALLOW_IDENT) {
emit_rule = false;
}