Skip to content

Implement AnnexB RegExp.prototype.compile() #579

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Aug 24, 2015
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
167 changes: 167 additions & 0 deletions jerry-core/ecma/builtin-objects/ecma-builtin-regexp-prototype.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -45,6 +45,173 @@
* @{
*/

#ifndef CONFIG_ECMA_COMPACT_PROFILE_DISABLE_ANNEXB_BUILTIN

/**
* The RegExp.prototype object's 'compile' routine
*
* See also:
* ECMA-262 v5, B.2.5.1
*
* @return completion value
* Returned value must be freed with ecma_free_completion_value.
*/
static ecma_completion_value_t
ecma_builtin_regexp_prototype_compile (ecma_value_t this_arg, /**< this argument */
ecma_value_t pattern_arg, /**< pattern or RegExp object */
ecma_value_t flags_arg) /**< flags */
{
ecma_completion_value_t ret_value = ecma_make_empty_completion_value ();
ecma_object_t *this_obj_p = ecma_get_object_from_value (this_arg);

if (ecma_object_get_class_name (this_obj_p) != LIT_MAGIC_STRING_REGEXP_UL)
{
/* Compile can only be called on RegExp objects. */
ret_value = ecma_make_throw_obj_completion_value (ecma_new_standard_error (ECMA_ERROR_TYPE));
}
else
{
ecma_string_t *pattern_string_p = NULL;
uint8_t flags = 0;

if (ecma_is_value_object (pattern_arg)
&& ecma_object_get_class_name (ecma_get_object_from_value (pattern_arg)) == LIT_MAGIC_STRING_REGEXP_UL)
{
if (!ecma_is_value_undefined (flags_arg))
{
ret_value = ecma_raise_type_error ("Invalid argument of RegExp compile.");
}
else
{
/* Compile from existing RegExp pbject. */
ecma_object_t *target_p = ecma_get_object_from_value (pattern_arg);

/* Get source. */
ecma_string_t *magic_string_p = ecma_get_magic_string (LIT_MAGIC_STRING_SOURCE);
ecma_property_t *prop_p = ecma_get_named_data_property (target_p, magic_string_p);
pattern_string_p = ecma_get_string_from_value (ecma_get_named_data_property_value (prop_p));
ecma_deref_ecma_string (magic_string_p);

/* Get flags. */
magic_string_p = ecma_get_magic_string (LIT_MAGIC_STRING_GLOBAL);
prop_p = ecma_get_named_data_property (target_p, magic_string_p);

if (ecma_is_value_true (ecma_get_named_data_property_value (prop_p)))
{
flags |= RE_FLAG_GLOBAL;
}

ecma_deref_ecma_string (magic_string_p);
magic_string_p = ecma_get_magic_string (LIT_MAGIC_STRING_IGNORECASE_UL);
prop_p = ecma_get_named_data_property (target_p, magic_string_p);

if (ecma_is_value_true (ecma_get_named_data_property_value (prop_p)))
{
flags |= RE_FLAG_IGNORE_CASE;
}

ecma_deref_ecma_string (magic_string_p);
magic_string_p = ecma_get_magic_string (LIT_MAGIC_STRING_MULTILINE);
prop_p = ecma_get_named_data_property (target_p, magic_string_p);

if (ecma_is_value_true (ecma_get_named_data_property_value (prop_p)))
{
flags |= RE_FLAG_MULTILINE;
}

ecma_deref_ecma_string (magic_string_p);

/* Get bytecode property. */
ecma_property_t *bc_prop_p = ecma_get_internal_property (this_obj_p,
ECMA_INTERNAL_PROPERTY_REGEXP_BYTECODE);
re_bytecode_t *old_bc_p = ECMA_GET_NON_NULL_POINTER (re_bytecode_t,
bc_prop_p->u.internal_property.value);

FIXME ("We currently have to re-compile the bytecode, because we can't copy it without knowing its length.")
re_bytecode_t *new_bc_p = NULL;
ecma_completion_value_t bc_comp = re_compile_bytecode (&new_bc_p, pattern_string_p, flags);
/* Should always succeed, since we're compiling from a source that has been compiled previously. */
JERRY_ASSERT (ecma_is_completion_value_empty (bc_comp));

mem_heap_free_block (old_bc_p);
ECMA_SET_POINTER (bc_prop_p->u.internal_property.value, new_bc_p);

re_initialize_props (this_obj_p, pattern_string_p, flags);

ret_value = ecma_make_normal_completion_value (ecma_make_simple_value (ECMA_SIMPLE_VALUE_UNDEFINED));
}
}
else
{
/* Get source string. */
if (!ecma_is_value_undefined (pattern_arg))
{
ECMA_TRY_CATCH (regexp_str_value,
ecma_op_to_string (pattern_arg),
ret_value);

if (ecma_string_get_length (ecma_get_string_from_value (regexp_str_value)) == 0)
{
pattern_string_p = ecma_get_magic_string (LIT_MAGIC_STRING_EMPTY_NON_CAPTURE_GROUP);
}
else
{
pattern_string_p = ecma_copy_or_ref_ecma_string (ecma_get_string_from_value (regexp_str_value));
}

ECMA_FINALIZE (regexp_str_value);
}
else
{
pattern_string_p = ecma_get_magic_string (LIT_MAGIC_STRING_EMPTY_NON_CAPTURE_GROUP);
}

/* Parse flags. */
if (ecma_is_completion_value_empty (ret_value) && !ecma_is_value_undefined (flags_arg))
{
ECMA_TRY_CATCH (flags_str_value,
ecma_op_to_string (flags_arg),
ret_value);

ECMA_TRY_CATCH (flags_dummy,
re_parse_regexp_flags (ecma_get_string_from_value (flags_str_value), &flags),
ret_value);
ECMA_FINALIZE (flags_dummy);
ECMA_FINALIZE (flags_str_value);
}

if (ecma_is_completion_value_empty (ret_value))
{
ecma_property_t *bc_prop_p = ecma_get_internal_property (this_obj_p,
ECMA_INTERNAL_PROPERTY_REGEXP_BYTECODE);
re_bytecode_t *old_bc_p = ECMA_GET_NON_NULL_POINTER (re_bytecode_t,
bc_prop_p->u.internal_property.value);

/* Try to compile bytecode from new source. */
re_bytecode_t *new_bc_p = NULL;
ECMA_TRY_CATCH (bc_dummy, re_compile_bytecode (&new_bc_p, pattern_string_p, flags), ret_value);

/* Replace old bytecode with new one. */
mem_heap_free_block (old_bc_p);
ECMA_SET_POINTER (bc_prop_p->u.internal_property.value, new_bc_p);
re_initialize_props (this_obj_p, pattern_string_p, flags);
ret_value = ecma_make_normal_completion_value (ecma_make_simple_value (ECMA_SIMPLE_VALUE_UNDEFINED));

ECMA_FINALIZE (bc_dummy);
}

if (pattern_string_p != NULL)
{
ecma_deref_ecma_string (pattern_string_p);
}
}
}

return ret_value;
} /* ecma_builtin_regexp_prototype_compile */

#endif /* !CONFIG_ECMA_COMPACT_PROFILE_DISABLE_ANNEXB_BUILTIN */

/**
* The RegExp.prototype object's 'exec' routine
*
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -86,6 +86,9 @@ NUMBER_VALUE (LIT_MAGIC_STRING_LASTINDEX_UL,
ECMA_PROPERTY_NOT_ENUMERABLE,
ECMA_PROPERTY_NOT_CONFIGURABLE)

#ifndef CONFIG_ECMA_COMPACT_PROFILE_DISABLE_ANNEXB_BUILTIN
ROUTINE (LIT_MAGIC_STRING_COMPILE, ecma_builtin_regexp_prototype_compile, 2, 1)
#endif
ROUTINE (LIT_MAGIC_STRING_EXEC, ecma_builtin_regexp_prototype_exec, 1, 1)
ROUTINE (LIT_MAGIC_STRING_TEST, ecma_builtin_regexp_prototype_test, 1, 1)
ROUTINE (LIT_MAGIC_STRING_TO_STRING_UL, ecma_builtin_regexp_prototype_to_string, 0, 0)
Expand Down
154 changes: 102 additions & 52 deletions jerry-core/ecma/operations/ecma-regexp-object.cpp
Original file line number Diff line number Diff line change
Expand Up @@ -62,7 +62,7 @@
* @return completion value
* Returned value must be freed with ecma_free_completion_value
*/
static ecma_completion_value_t
ecma_completion_value_t
re_parse_regexp_flags (ecma_string_t *flags_str_p, /**< Input string with flags */
uint8_t *flags_p) /**< Output: parsed flag bits */
{
Expand Down Expand Up @@ -119,6 +119,101 @@ re_parse_regexp_flags (ecma_string_t *flags_str_p, /**< Input string with flags
return ret_value;
} /* re_parse_regexp_flags */

/*
* Initializes the source, global, ignoreCase, multiline, and lastIndex properties of RegExp instance.
*/
void
re_initialize_props (ecma_object_t *re_obj_p, /**< RegExp obejct */
ecma_string_t *source_p, /**< source string */
uint8_t flags) /**< flags */
{
/* Set source property. ECMA-262 v5, 15.10.7.1 */
ecma_string_t *magic_string_p = ecma_get_magic_string (LIT_MAGIC_STRING_SOURCE);
ecma_property_t *prop_p = ecma_find_named_property (re_obj_p, magic_string_p);

if (prop_p == NULL)
{
prop_p = ecma_create_named_data_property (re_obj_p,
magic_string_p,
false, false, false);
}

ecma_deref_ecma_string (magic_string_p);
JERRY_ASSERT (prop_p->type == ECMA_PROPERTY_NAMEDDATA);
ecma_named_data_property_assign_value (re_obj_p,
prop_p,
ecma_make_string_value (source_p));

ecma_simple_value_t prop_value;

/* Set global property. ECMA-262 v5, 15.10.7.2 */
magic_string_p = ecma_get_magic_string (LIT_MAGIC_STRING_GLOBAL);
prop_p = ecma_find_named_property (re_obj_p, magic_string_p);

if (prop_p == NULL)
{
prop_p = ecma_create_named_data_property (re_obj_p,
magic_string_p,
false, false, false);
}

ecma_deref_ecma_string (magic_string_p);
prop_value = flags & RE_FLAG_GLOBAL ? ECMA_SIMPLE_VALUE_TRUE : ECMA_SIMPLE_VALUE_FALSE;
JERRY_ASSERT (prop_p->type == ECMA_PROPERTY_NAMEDDATA);
ecma_set_named_data_property_value (prop_p, ecma_make_simple_value (prop_value));

/* Set ignoreCase property. ECMA-262 v5, 15.10.7.3 */
magic_string_p = ecma_get_magic_string (LIT_MAGIC_STRING_IGNORECASE_UL);
prop_p = ecma_find_named_property (re_obj_p, magic_string_p);

if (prop_p == NULL)
{
prop_p = ecma_create_named_data_property (re_obj_p,
magic_string_p,
false, false, false);
}

ecma_deref_ecma_string (magic_string_p);
prop_value = flags & RE_FLAG_IGNORE_CASE ? ECMA_SIMPLE_VALUE_TRUE : ECMA_SIMPLE_VALUE_FALSE;
JERRY_ASSERT (prop_p->type == ECMA_PROPERTY_NAMEDDATA);
ecma_set_named_data_property_value (prop_p, ecma_make_simple_value (prop_value));

/* Set multiline property. ECMA-262 v5, 15.10.7.4 */
magic_string_p = ecma_get_magic_string (LIT_MAGIC_STRING_MULTILINE);
prop_p = ecma_find_named_property (re_obj_p, magic_string_p);

if (prop_p == NULL)
{
prop_p = ecma_create_named_data_property (re_obj_p,
magic_string_p,
false, false, false);
}

ecma_deref_ecma_string (magic_string_p);
prop_value = flags & RE_FLAG_MULTILINE ? ECMA_SIMPLE_VALUE_TRUE : ECMA_SIMPLE_VALUE_FALSE;
JERRY_ASSERT (prop_p->type == ECMA_PROPERTY_NAMEDDATA);
ecma_set_named_data_property_value (prop_p, ecma_make_simple_value (prop_value));

/* Set lastIndex property. ECMA-262 v5, 15.10.7.5 */
magic_string_p = ecma_get_magic_string (LIT_MAGIC_STRING_LASTINDEX_UL);
prop_p = ecma_find_named_property (re_obj_p, magic_string_p);

if (prop_p == NULL)
{
prop_p = ecma_create_named_data_property (re_obj_p,
magic_string_p,
true, false, false);
}

ecma_deref_ecma_string (magic_string_p);

ecma_number_t *lastindex_num_p = ecma_alloc_number ();
*lastindex_num_p = ECMA_NUMBER_ZERO;
JERRY_ASSERT (prop_p->type == ECMA_PROPERTY_NAMEDDATA);
ecma_named_data_property_assign_value (re_obj_p, prop_p, ecma_make_number_value (lastindex_num_p));
ecma_dealloc_number (lastindex_num_p);
} /* re_initialize_props */

/**
* RegExp object creation operation.
*
Expand Down Expand Up @@ -155,64 +250,19 @@ ecma_op_create_regexp_object (ecma_string_t *pattern_p, /**< input pattern */
ecma_property_t *class_prop_p = ecma_create_internal_property (obj_p, ECMA_INTERNAL_PROPERTY_CLASS);
class_prop_p->u.internal_property.value = LIT_MAGIC_STRING_REGEXP_UL;

/* Set source property. ECMA-262 v5, 15.10.7.1 */
ecma_string_t *magic_string_p = ecma_get_magic_string (LIT_MAGIC_STRING_SOURCE);
ecma_property_t *source_prop_p = ecma_create_named_data_property (obj_p,
magic_string_p,
false, false, false);
ecma_deref_ecma_string (magic_string_p);
ecma_set_named_data_property_value (source_prop_p,
ecma_make_string_value (ecma_copy_or_ref_ecma_string (pattern_p)));

ecma_simple_value_t prop_value;

/* Set global property. ECMA-262 v5, 15.10.7.2*/
magic_string_p = ecma_get_magic_string (LIT_MAGIC_STRING_GLOBAL);
ecma_property_t *global_prop_p = ecma_create_named_data_property (obj_p,
magic_string_p,
false, false, false);
ecma_deref_ecma_string (magic_string_p);
prop_value = flags & RE_FLAG_GLOBAL ? ECMA_SIMPLE_VALUE_TRUE : ECMA_SIMPLE_VALUE_FALSE;
ecma_set_named_data_property_value (global_prop_p, ecma_make_simple_value (prop_value));

/* Set ignoreCase property. ECMA-262 v5, 15.10.7.3*/
magic_string_p = ecma_get_magic_string (LIT_MAGIC_STRING_IGNORECASE_UL);
ecma_property_t *ignorecase_prop_p = ecma_create_named_data_property (obj_p,
magic_string_p,
false, false, false);
ecma_deref_ecma_string (magic_string_p);
prop_value = flags & RE_FLAG_IGNORE_CASE ? ECMA_SIMPLE_VALUE_TRUE : ECMA_SIMPLE_VALUE_FALSE;
ecma_set_named_data_property_value (ignorecase_prop_p, ecma_make_simple_value (prop_value));


/* Set multiline property. ECMA-262 v5, 15.10.7.4*/
magic_string_p = ecma_get_magic_string (LIT_MAGIC_STRING_MULTILINE);
ecma_property_t *multiline_prop_p = ecma_create_named_data_property (obj_p,
magic_string_p,
false, false, false);
ecma_deref_ecma_string (magic_string_p);
prop_value = flags & RE_FLAG_MULTILINE ? ECMA_SIMPLE_VALUE_TRUE : ECMA_SIMPLE_VALUE_FALSE;
ecma_set_named_data_property_value (multiline_prop_p, ecma_make_simple_value (prop_value));

/* Set lastIndex property. ECMA-262 v5, 15.10.7.5*/
magic_string_p = ecma_get_magic_string (LIT_MAGIC_STRING_LASTINDEX_UL);
ecma_property_t *lastindex_prop_p = ecma_create_named_data_property (obj_p,
magic_string_p,
true, false, false);
ecma_deref_ecma_string (magic_string_p);

ecma_number_t *lastindex_num_p = ecma_alloc_number ();
*lastindex_num_p = ECMA_NUMBER_ZERO;
ecma_named_data_property_assign_value (obj_p, lastindex_prop_p, ecma_make_number_value (lastindex_num_p));
ecma_dealloc_number (lastindex_num_p);
re_initialize_props (obj_p, pattern_p, flags);

/* Set bytecode internal property. */
ecma_property_t *bytecode_prop_p;
bytecode_prop_p = ecma_create_internal_property (obj_p, ECMA_INTERNAL_PROPERTY_REGEXP_BYTECODE);

/* Compile bytecode. */
ECMA_TRY_CATCH (empty, re_compile_bytecode (bytecode_prop_p, pattern_p, flags), ret_value);
re_bytecode_t *bc_p = NULL;
ECMA_TRY_CATCH (empty, re_compile_bytecode (&bc_p, pattern_p, flags), ret_value);

ECMA_SET_POINTER (bytecode_prop_p->u.internal_property.value, bc_p);
ret_value = ecma_make_normal_completion_value (ecma_make_object_value (obj_p));

ECMA_FINALIZE (empty);

if (ecma_is_completion_value_throw (ret_value))
Expand Down
8 changes: 8 additions & 0 deletions jerry-core/ecma/operations/ecma-regexp-object.h
Original file line number Diff line number Diff line change
Expand Up @@ -65,6 +65,14 @@ re_set_result_array_properties (ecma_object_t *array_obj_p,
uint32_t num_of_elements,
int32_t index);

extern ecma_completion_value_t
re_parse_regexp_flags (ecma_string_t *flags_str_p, uint8_t *flags_p);

extern void
re_initialize_props (ecma_object_t *re_obj_p,
ecma_string_t *source_p,
uint8_t flags);

/**
* @}
* @}
Expand Down
1 change: 1 addition & 0 deletions jerry-core/lit/lit-magic-strings.inc.h
Original file line number Diff line number Diff line change
Expand Up @@ -211,6 +211,7 @@ LIT_MAGIC_STRING_DEF (LIT_MAGIC_STRING_INVALID_DATE_UL, "Invalid Date")
LIT_MAGIC_STRING_DEF (LIT_MAGIC_STRING_APPLY, "apply")
LIT_MAGIC_STRING_DEF (LIT_MAGIC_STRING_CALL, "call")
LIT_MAGIC_STRING_DEF (LIT_MAGIC_STRING_BIND, "bind")
LIT_MAGIC_STRING_DEF (LIT_MAGIC_STRING_COMPILE, "compile")
LIT_MAGIC_STRING_DEF (LIT_MAGIC_STRING_EXEC, "exec")
LIT_MAGIC_STRING_DEF (LIT_MAGIC_STRING_TEST, "test")
LIT_MAGIC_STRING_DEF (LIT_MAGIC_STRING_NAME, "name")
Expand Down
Loading