kopia lustrzana https://github.com/micropython/micropython
re: Add support for start- and endpos.
Pattern objects have two additional parameters for the ::search and ::match methods to define the starting and ending position of the subject within the string to be searched. This allows for searching a sub-string without creating a slice. However, one caveat of using the start-pos rather than a slice is that the start anchor (`^`) remains anchored to the beginning of the text.pull/14179/head
rodzic
9d27183bde
commit
6f3c6d7de9
|
@ -195,10 +195,11 @@ static void re_print(const mp_print_t *print, mp_obj_t self_in, mp_print_kind_t
|
|||
}
|
||||
|
||||
static mp_obj_t re_exec(bool is_anchored, uint n_args, const mp_obj_t *args) {
|
||||
(void)n_args;
|
||||
mp_obj_re_t *self;
|
||||
bool was_compiled = false;
|
||||
if (mp_obj_is_type(args[0], (mp_obj_type_t *)&re_type)) {
|
||||
self = MP_OBJ_TO_PTR(args[0]);
|
||||
was_compiled = true;
|
||||
} else {
|
||||
self = MP_OBJ_TO_PTR(mod_re_compile(1, args));
|
||||
}
|
||||
|
@ -206,6 +207,27 @@ static mp_obj_t re_exec(bool is_anchored, uint n_args, const mp_obj_t *args) {
|
|||
size_t len;
|
||||
subj.begin_line = subj.begin = mp_obj_str_get_data(args[1], &len);
|
||||
subj.end = subj.begin + len;
|
||||
|
||||
if (was_compiled && n_args > 2) {
|
||||
// Arg #2 is starting-pos
|
||||
size_t startpos = mp_obj_get_int(args[2]);
|
||||
if (startpos >= len) {
|
||||
startpos = len;
|
||||
}
|
||||
subj.begin += startpos;
|
||||
if (n_args > 3) {
|
||||
// Arg #3 is ending-pos
|
||||
size_t endpos = mp_obj_get_int(args[3]);
|
||||
if (endpos > len) {
|
||||
endpos = len;
|
||||
}
|
||||
else if (endpos < startpos) {
|
||||
endpos = startpos;
|
||||
}
|
||||
subj.end = subj.begin_line + endpos;
|
||||
}
|
||||
}
|
||||
|
||||
int caps_num = (self->re.sub + 1) * 2;
|
||||
mp_obj_match_t *match = m_new_obj_var(mp_obj_match_t, caps, char *, caps_num);
|
||||
// cast is a workaround for a bug in msvc: it treats const char** as a const pointer instead of a pointer to pointer to const char
|
||||
|
|
|
@ -0,0 +1,59 @@
|
|||
# test start and end pos specification
|
||||
|
||||
try:
|
||||
import re
|
||||
except ImportError:
|
||||
print("SKIP")
|
||||
raise SystemExit
|
||||
|
||||
|
||||
def print_groups(match):
|
||||
print("----")
|
||||
try:
|
||||
if match is not None:
|
||||
i = 0
|
||||
while True:
|
||||
print(match.group(i))
|
||||
i += 1
|
||||
except IndexError:
|
||||
pass
|
||||
|
||||
p = re.compile(r"o")
|
||||
m = p.match("dog")
|
||||
print_groups(m)
|
||||
|
||||
m = p.match("dog", 1)
|
||||
print_groups(m)
|
||||
|
||||
m = p.match("dog", 2)
|
||||
print_groups(m)
|
||||
|
||||
m = p.match("dog", 3) # Past end of input
|
||||
print_groups(m)
|
||||
|
||||
m = p.match("dog", 0, 1)
|
||||
print_groups(m)
|
||||
|
||||
# Caret only matches the actual beginning
|
||||
p = re.compile(r"^o")
|
||||
m = p.match("dog", 1)
|
||||
print_groups(m)
|
||||
|
||||
# End at begging means searching empty string
|
||||
p = re.compile(r"o")
|
||||
m = p.match("dog", 1, 1)
|
||||
print_groups(m)
|
||||
|
||||
# Search also works
|
||||
print('--search')
|
||||
|
||||
p = re.compile(r"o")
|
||||
m = p.search('dog')
|
||||
print_groups(m)
|
||||
|
||||
m = p.search('dog', 1)
|
||||
print_groups(m)
|
||||
|
||||
m = p.search('dog', 2)
|
||||
print_groups(m)
|
||||
|
|
@ -0,0 +1,14 @@
|
|||
----
|
||||
----
|
||||
o
|
||||
----
|
||||
----
|
||||
----
|
||||
----
|
||||
----
|
||||
--search
|
||||
----
|
||||
o
|
||||
----
|
||||
o
|
||||
----
|
Ładowanie…
Reference in New Issue