2014-12-03 00:04:00 +00:00
|
|
|
from .typedefs cimport hash_t, flags_t, id_t, len_t, tag_t, attr_t
|
2014-09-10 18:41:37 +00:00
|
|
|
|
2014-10-23 13:59:17 +00:00
|
|
|
from .utf8string cimport StringStore
|
2014-10-09 08:53:30 +00:00
|
|
|
|
|
|
|
|
|
2014-12-03 00:04:00 +00:00
|
|
|
# Reserve 64 values for flag features
|
|
|
|
|
cpdef enum attr_id_t:
|
|
|
|
|
FLAG0
|
|
|
|
|
FLAG1
|
|
|
|
|
FLAG2
|
|
|
|
|
FLAG3
|
|
|
|
|
FLAG4
|
|
|
|
|
FLAG5
|
|
|
|
|
FLAG6
|
|
|
|
|
FLAG7
|
|
|
|
|
FLAG8
|
|
|
|
|
FLAG9
|
|
|
|
|
FLAG10
|
|
|
|
|
FLAG11
|
|
|
|
|
FLAG12
|
|
|
|
|
FLAG13
|
|
|
|
|
FLAG14
|
|
|
|
|
FLAG15
|
|
|
|
|
FLAG16
|
|
|
|
|
FLAG17
|
|
|
|
|
FLAG18
|
|
|
|
|
FLAG19
|
|
|
|
|
FLAG20
|
|
|
|
|
FLAG21
|
|
|
|
|
FLAG22
|
|
|
|
|
FLAG23
|
|
|
|
|
FLAG24
|
|
|
|
|
FLAG25
|
|
|
|
|
FLAG26
|
|
|
|
|
FLAG27
|
|
|
|
|
FLAG28
|
|
|
|
|
FLAG29
|
|
|
|
|
FLAG30
|
|
|
|
|
FLAG31
|
|
|
|
|
FLAG32
|
|
|
|
|
FLAG33
|
|
|
|
|
FLAG34
|
|
|
|
|
FLAG35
|
|
|
|
|
FLAG36
|
|
|
|
|
FLAG37
|
|
|
|
|
FLAG38
|
|
|
|
|
FLAG39
|
|
|
|
|
FLAG40
|
|
|
|
|
FLAG41
|
|
|
|
|
FLAG42
|
|
|
|
|
FLAG43
|
|
|
|
|
FLAG44
|
|
|
|
|
FLAG45
|
|
|
|
|
FLAG46
|
|
|
|
|
FLAG47
|
|
|
|
|
FLAG48
|
|
|
|
|
FLAG49
|
|
|
|
|
FLAG50
|
|
|
|
|
FLAG51
|
|
|
|
|
FLAG52
|
|
|
|
|
FLAG53
|
|
|
|
|
FLAG54
|
|
|
|
|
FLAG55
|
|
|
|
|
FLAG56
|
|
|
|
|
FLAG57
|
|
|
|
|
FLAG58
|
|
|
|
|
FLAG59
|
|
|
|
|
FLAG60
|
|
|
|
|
FLAG61
|
|
|
|
|
FLAG62
|
|
|
|
|
FLAG63
|
2014-10-09 08:53:30 +00:00
|
|
|
|
2014-12-03 00:04:00 +00:00
|
|
|
ID
|
|
|
|
|
SIC
|
|
|
|
|
NORM
|
|
|
|
|
SHAPE
|
|
|
|
|
ASCIIED
|
|
|
|
|
PREFIX
|
|
|
|
|
SUFFIX
|
2014-11-02 02:19:05 +00:00
|
|
|
|
2014-12-03 00:04:00 +00:00
|
|
|
LENGTH
|
|
|
|
|
CLUSTER
|
|
|
|
|
POS_TYPE
|
|
|
|
|
SENSE_TYPE
|
2014-11-02 13:13:51 +00:00
|
|
|
|
2014-10-09 08:53:30 +00:00
|
|
|
|
2014-10-23 13:59:17 +00:00
|
|
|
cdef struct Lexeme:
|
2014-12-03 00:04:00 +00:00
|
|
|
flags_t flags
|
2014-10-29 12:19:38 +00:00
|
|
|
|
2014-12-03 00:04:00 +00:00
|
|
|
attr_t id
|
|
|
|
|
attr_t sic
|
|
|
|
|
attr_t norm
|
|
|
|
|
attr_t shape
|
|
|
|
|
attr_t asciied
|
|
|
|
|
attr_t prefix
|
|
|
|
|
attr_t suffix
|
|
|
|
|
|
|
|
|
|
attr_t length
|
|
|
|
|
attr_t cluster
|
|
|
|
|
attr_t pos_type
|
|
|
|
|
attr_t sense_type
|
2014-10-10 08:17:22 +00:00
|
|
|
|
2014-10-23 13:59:17 +00:00
|
|
|
float prob
|
2014-12-03 00:04:00 +00:00
|
|
|
float upper_pc
|
|
|
|
|
float title_pc
|
2014-10-30 04:42:15 +00:00
|
|
|
|
2014-10-22 14:57:59 +00:00
|
|
|
|
2014-10-23 13:59:17 +00:00
|
|
|
cdef Lexeme EMPTY_LEXEME
|
2014-09-10 18:41:37 +00:00
|
|
|
|
2014-12-03 00:04:00 +00:00
|
|
|
|
|
|
|
|
cpdef Lexeme init(id_t i, unicode string, hash_t hashed, StringStore store,
|
|
|
|
|
dict props) except *
|
2014-10-29 12:19:38 +00:00
|
|
|
|
2014-10-09 03:10:46 +00:00
|
|
|
|
2014-12-03 04:44:25 +00:00
|
|
|
cdef inline bint check_flag(const Lexeme* lexeme, attr_id_t flag_id) nogil:
|
2014-10-23 13:59:17 +00:00
|
|
|
return lexeme.flags & (1 << flag_id)
|
2014-12-03 00:04:00 +00:00
|
|
|
|
|
|
|
|
|
2014-12-03 04:44:25 +00:00
|
|
|
cdef attr_t get_attr(const Lexeme* lex, attr_id_t attr_id)
|
