[transliteration] Using revisit strings instead of keeping a backtrack count so we don't have to later map logical characters to the actual string, removing any duplicate keys in the table builder so that if any rules happen to overlap within a step, the first will take precedence

This commit is contained in:
Al
2015-05-29 16:53:28 -04:00
parent a9d5b91ac0
commit a278cfd12c
5 changed files with 6801 additions and 6763 deletions

View File

@@ -49,14 +49,14 @@ VECTOR_INIT(group_capture_array, group_capture_t)
typedef struct transliteration_replacement {
uint32_t string_index;
int32_t move;
uint32_t revisit_index;
size_t num_groups;
group_capture_array *groups;
} transliteration_replacement_t;
transliteration_replacement_t *transliteration_replacement_new(
uint32_t string_index,
int32_t move,
uint32_t string_index,
uint32_t revisit_index,
group_capture_array *groups
);
@@ -74,6 +74,7 @@ typedef struct transliteration_table {
transliteration_replacement_array *replacements;
cstring_array *replacement_strings;
cstring_array *revisit_strings;
} transliteration_table_t;
#define NAMESPACE_SEPARATOR_CHAR "|"