optimize named capture assignment

* compile.c (compile_named_capture_assign): optimize named capture
  assignments, by replacing repeating global variable accesses
  with `dup`, and by returning the matched result instead of
  re-getting it from the MatchData.

* parse.y (reg_named_capture_assign_gen): build just assignment
  nodes for the optimization.

ex. `/(?<x>.)/ =~ "bar"`

- old
  ```
  0000 putstring        "bar"
  0002 opt_regexpmatch1 /(?<x>.)/
  0004 pop
  0005 getglobal        $~
  0007 branchunless     25
  0009 getglobal        $~
  0011 putobject        :x
  0013 opt_aref         <callinfo!mid:[], argc:1, ARGS_SIMPLE>
  0016 setlocal_OP__WC__0 2
  0018 getglobal        $~
  0020 putobject_OP_INT2FIX_O_0_C_
  0021 opt_send_without_block <callinfo!mid:begin, argc:1, ARGS_SIMPLE>
  0024 leave
  0025 putobject        nil
  0027 setlocal_OP__WC__0 2
  0029 putobject        nil
  0031 leave
  ```

- new
  ```
  0000 putstring        "bar"
  0002 opt_regexpmatch1 /(?<x>.)/
  0004 getglobal        $~
  0006 dup
  0007 branchunless     14
  0009 putobject        :x
  0011 opt_aref         <callinfo!mid:[], argc:1, ARGS_SIMPLE>
  0014 setlocal_OP__WC__0 2
  0016 leave
  ```

git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@54100 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
This commit is contained in:
nobu 2016-03-14 07:53:39 +00:00
Родитель a4e6f7d707
Коммит ea5e885a95
4 изменённых файлов: 95 добавлений и 45 удалений

Просмотреть файл

@ -1,3 +1,13 @@
Mon Mar 14 16:53:37 2016 Nobuyoshi Nakada <nobu@ruby-lang.org>
* compile.c (compile_named_capture_assign): optimize named capture
assignments, by replacing repeating global variable accesses
with `dup`, and by returning the matched result instead of
re-getting it from the MatchData.
* parse.y (reg_named_capture_assign_gen): build just assignment
nodes for the optimization.
Mon Mar 14 16:02:59 2016 Nobuyoshi Nakada <nobu@ruby-lang.org>
* file.c (ruby_is_fd_loadable): now return -1 if loadable but

Просмотреть файл

@ -3648,6 +3648,65 @@ build_postexe_iseq(rb_iseq_t *iseq, LINK_ANCHOR *ret, NODE *body)
return Qnil;
}
static void
compile_named_capture_assign(rb_iseq_t *iseq, LINK_ANCHOR *ret, NODE *node)
{
NODE *vars;
LINK_ELEMENT *last;
int line = nd_line(node);
LABEL *fail_label = NEW_LABEL(line), *end_label = NEW_LABEL(line);
#if !(defined(NAMED_CAPTURE_BY_SVAR) && NAMED_CAPTURE_BY_SVAR-0)
ADD_INSN1(ret, line, getglobal, ((VALUE)rb_global_entry(idBACKREF) | 1));
#else
ADD_INSN2(ret, line, getspecial, INT2FIX(1) /* '~' */, INT2FIX(0));
#endif
ADD_INSN(ret, line, dup);
ADD_INSNL(ret, line, branchunless, fail_label);
for (vars = node; vars; vars = vars->nd_next) {
INSN *cap;
if (vars->nd_next) {
ADD_INSN(ret, line, dup);
}
last = ret->last;
COMPILE_POPED(ret, "capture", vars->nd_head);
last = last->next; /* putobject :var */
cap = new_insn_send(iseq, line, idAREF, INT2FIX(1),
NULL, INT2FIX(0), NULL);
INSERT_ELEM_PREV(last->next, (LINK_ELEMENT *)cap);
#if !defined(NAMED_CAPTURE_SINGLE_OPT) || NAMED_CAPTURE_SINGLE_OPT-0
if (!vars->nd_next && vars == node) {
/* only one name */
DECL_ANCHOR(nom);
INIT_ANCHOR(nom);
ADD_INSNL(nom, line, jump, end_label);
ADD_LABEL(nom, fail_label);
# if 0 /* $~ must be MatchData or nil */
ADD_INSN(nom, line, pop);
ADD_INSN(nom, line, putnil);
# endif
ADD_LABEL(nom, end_label);
(nom->last->next = cap->link.next)->prev = nom->last;
(cap->link.next = nom->anchor.next)->prev = &cap->link;
return;
}
#endif
}
ADD_INSNL(ret, line, jump, end_label);
ADD_LABEL(ret, fail_label);
ADD_INSN(ret, line, pop);
for (vars = node; vars; vars = vars->nd_next) {
last = ret->last;
COMPILE_POPED(ret, "capture", vars->nd_head);
last = last->next; /* putobject :var */
((INSN*)last)->insn_id = BIN(putnil);
((INSN*)last)->operand_size = 0;
}
ADD_LABEL(ret, end_label);
}
/**
compile each node
@ -5351,6 +5410,10 @@ iseq_compile_each(rb_iseq_t *iseq, LINK_ANCHOR *ret, NODE * node, int poped)
ADD_SEND(ret, line, idEqTilde, INT2FIX(1));
}
if (node->nd_args) {
compile_named_capture_assign(iseq, ret, node->nd_args);
}
if (poped) {
ADD_INSN(ret, line, pop);
}

6
node.c
Просмотреть файл

@ -583,8 +583,12 @@ dump_node(VALUE buf, VALUE indent, int comment, NODE *node)
ANN("format: [nd_recv] =~ [nd_value]");
ANN("example: /foo/ =~ 'foo'");
F_NODE(nd_recv, "regexp (receiver)");
LAST_NODE;
if (!node->nd_args) LAST_NODE;
F_NODE(nd_value, "string (argument)");
if (node->nd_args) {
LAST_NODE;
F_NODE(nd_args, "named captures");
}
break;
case NODE_MATCH3:

61
parse.y
Просмотреть файл

@ -502,8 +502,8 @@ static void reg_fragment_setenc_gen(struct parser_params*, VALUE, int);
#define reg_fragment_setenc(str,options) reg_fragment_setenc_gen(parser, (str), (options))
static int reg_fragment_check_gen(struct parser_params*, VALUE, int);
#define reg_fragment_check(str,options) reg_fragment_check_gen(parser, (str), (options))
static NODE *reg_named_capture_assign_gen(struct parser_params* parser, VALUE regexp, NODE *match);
#define reg_named_capture_assign(regexp,match) reg_named_capture_assign_gen(parser,(regexp),(match))
static NODE *reg_named_capture_assign_gen(struct parser_params* parser, VALUE regexp);
#define reg_named_capture_assign(regexp) reg_named_capture_assign_gen(parser,(regexp))
static void parser_heredoc_dedent(struct parser_params*,NODE*);
# define heredoc_dedent(str) parser_heredoc_dedent(parser, (str))
@ -2327,9 +2327,12 @@ arg : lhs '=' arg
{
/*%%%*/
$$ = match_op($1, $3);
if (nd_type($1) == NODE_LIT && RB_TYPE_P($1->nd_lit, T_REGEXP)) {
$$ = reg_named_capture_assign($1->nd_lit, $$);
}
if (nd_type($1) == NODE_LIT) {
VALUE lit = $1->nd_lit;
if (RB_TYPE_P(lit, T_REGEXP)) {
$$->nd_args = reg_named_capture_assign(lit);
}
}
/*%
$$ = dispatch3(binary, $1, ID2SYM(idEqTilde), $3);
%*/
@ -10600,8 +10603,6 @@ typedef struct {
struct parser_params* parser;
rb_encoding *enc;
NODE *succ_block;
NODE *fail_block;
int num;
} reg_named_capture_assign_t;
static int
@ -10614,13 +10615,7 @@ reg_named_capture_assign_iter(const OnigUChar *name, const OnigUChar *name_end,
long len = name_end - name;
const char *s = (const char *)name;
ID var;
arg->num++;
if (arg->succ_block == 0) {
arg->succ_block = NEW_BEGIN(0);
arg->fail_block = NEW_BEGIN(0);
}
NODE *node, *succ;
if (!len || (*name != '_' && ISASCII(*name) && !rb_enc_islower(*name, enc)) ||
(len < MAX_WORD_LENGTH && rb_reserved_word(s, (int)len)) ||
@ -10632,48 +10627,26 @@ reg_named_capture_assign_iter(const OnigUChar *name, const OnigUChar *name_end,
rb_warning1("named capture conflicts a local variable - %"PRIsWARN,
rb_id2str(var));
}
arg->succ_block = block_append(arg->succ_block,
newline_node(node_assign(assignable(var,0),
NEW_CALL(
gettable(idBACKREF),
idAREF,
NEW_LIST(NEW_LIT(ID2SYM(var))))
)));
arg->fail_block = block_append(arg->fail_block,
newline_node(node_assign(assignable(var,0), NEW_LIT(Qnil))));
node = newline_node(node_assign(assignable(var, 0), NEW_LIT(ID2SYM(var))));
succ = arg->succ_block;
if (!succ) succ = NEW_BEGIN(0);
succ = block_append(succ, node);
arg->succ_block = succ;
return ST_CONTINUE;
}
static NODE *
reg_named_capture_assign_gen(struct parser_params* parser, VALUE regexp, NODE *match)
reg_named_capture_assign_gen(struct parser_params* parser, VALUE regexp)
{
reg_named_capture_assign_t arg;
arg.parser = parser;
arg.enc = rb_enc_get(regexp);
arg.succ_block = 0;
arg.fail_block = 0;
arg.num = 0;
onig_foreach_name(RREGEXP_PTR(regexp), reg_named_capture_assign_iter, &arg);
if (arg.num == 0)
return match;
return
block_append(
newline_node(match),
NEW_IF(gettable(idBACKREF),
block_append(
newline_node(arg.succ_block),
newline_node(
NEW_CALL(
gettable(idBACKREF),
rb_intern("begin"),
NEW_LIST(NEW_LIT(INT2FIX(0)))))),
block_append(
newline_node(arg.fail_block),
newline_node(
NEW_LIT(Qnil)))));
if (!arg.succ_block) return 0;
return arg.succ_block->nd_next;
}
static VALUE