Store a mapping to ensure no two symbols map to the same metadata

This commit is contained in:
Max Brunsfeld 2019-12-05 17:21:46 -08:00
parent d6c7b243a7
commit 56c620c005
7 changed files with 128 additions and 33 deletions

View file

@ -3,8 +3,28 @@
#include "./error_costs.h"
#include <string.h>
void ts_language_table_entry(const TSLanguage *self, TSStateId state,
TSSymbol symbol, TableEntry *result) {
uint32_t ts_language_symbol_count(const TSLanguage *self) {
return self->symbol_count + self->alias_count;
}
uint32_t ts_language_version(const TSLanguage *self) {
return self->version;
}
uint32_t ts_language_field_count(const TSLanguage *self) {
if (self->version >= TREE_SITTER_LANGUAGE_VERSION_WITH_FIELDS) {
return self->field_count;
} else {
return 0;
}
}
void ts_language_table_entry(
const TSLanguage *self,
TSStateId state,
TSSymbol symbol,
TableEntry *result
) {
if (symbol == ts_builtin_sym_error || symbol == ts_builtin_sym_error_repeat) {
result->action_count = 0;
result->is_reusable = false;
@ -19,31 +39,41 @@ void ts_language_table_entry(const TSLanguage *self, TSStateId state,
}
}
uint32_t ts_language_symbol_count(const TSLanguage *language) {
return language->symbol_count + language->alias_count;
}
uint32_t ts_language_version(const TSLanguage *language) {
return language->version;
}
TSSymbolMetadata ts_language_symbol_metadata(const TSLanguage *language, TSSymbol symbol) {
TSSymbolMetadata ts_language_symbol_metadata(
const TSLanguage *self,
TSSymbol symbol
) {
if (symbol == ts_builtin_sym_error) {
return (TSSymbolMetadata){.visible = true, .named = true};
} else if (symbol == ts_builtin_sym_error_repeat) {
return (TSSymbolMetadata){.visible = false, .named = false};
} else {
return language->symbol_metadata[symbol];
return self->symbol_metadata[symbol];
}
}
const char *ts_language_symbol_name(const TSLanguage *language, TSSymbol symbol) {
TSSymbol ts_language_public_symbol(
const TSLanguage *self,
TSSymbol symbol
) {
if (symbol == ts_builtin_sym_error) return symbol;
if (self->version >= TREE_SITTER_LANGUAGE_VERSION_WITH_SYMBOL_DEDUPING) {
return self->public_symbol_map[symbol];
} else {
return symbol;
}
}
const char *ts_language_symbol_name(
const TSLanguage *self,
TSSymbol symbol
) {
if (symbol == ts_builtin_sym_error) {
return "ERROR";
} else if (symbol == ts_builtin_sym_error_repeat) {
return "_ERROR";
} else {
return language->symbol_names[symbol];
return self->symbol_names[symbol];
}
}
@ -59,13 +89,22 @@ TSSymbol ts_language_symbol_for_name(
TSSymbolMetadata metadata = ts_language_symbol_metadata(self, i);
if (!metadata.visible || metadata.named != is_named) continue;
const char *symbol_name = self->symbol_names[i];
if (!strncmp(symbol_name, string, length) && !symbol_name[length]) return i;
if (!strncmp(symbol_name, string, length) && !symbol_name[length]) {
if (self->version >= TREE_SITTER_LANGUAGE_VERSION_WITH_SYMBOL_DEDUPING) {
return self->public_symbol_map[i];
} else {
return i;
}
}
}
return 0;
}
TSSymbolType ts_language_symbol_type(const TSLanguage *language, TSSymbol symbol) {
TSSymbolMetadata metadata = ts_language_symbol_metadata(language, symbol);
TSSymbolType ts_language_symbol_type(
const TSLanguage *self,
TSSymbol symbol
) {
TSSymbolMetadata metadata = ts_language_symbol_metadata(self, symbol);
if (metadata.named) {
return TSSymbolTypeRegular;
} else if (metadata.visible) {
@ -75,15 +114,10 @@ TSSymbolType ts_language_symbol_type(const TSLanguage *language, TSSymbol symbol
}
}
uint32_t ts_language_field_count(const TSLanguage *self) {
if (self->version >= TREE_SITTER_LANGUAGE_VERSION_WITH_FIELDS) {
return self->field_count;
} else {
return 0;
}
}
const char *ts_language_field_name_for_id(const TSLanguage *self, TSFieldId id) {
const char *ts_language_field_name_for_id(
const TSLanguage *self,
TSFieldId id
) {
uint32_t count = ts_language_field_count(self);
if (count) {
return self->field_names[id];

View file

@ -10,6 +10,7 @@ extern "C" {
#define ts_builtin_sym_error_repeat (ts_builtin_sym_error - 1)
#define TREE_SITTER_LANGUAGE_VERSION_WITH_FIELDS 10
#define TREE_SITTER_LANGUAGE_VERSION_WITH_SYMBOL_DEDUPING 11
#define TREE_SITTER_LANGUAGE_VERSION_WITH_SMALL_STATES 11
typedef struct {
@ -22,6 +23,8 @@ void ts_language_table_entry(const TSLanguage *, TSStateId, TSSymbol, TableEntry
TSSymbolMetadata ts_language_symbol_metadata(const TSLanguage *, TSSymbol);
TSSymbol ts_language_public_symbol(const TSLanguage *, TSSymbol);
static inline bool ts_language_is_symbol_external(const TSLanguage *self, TSSymbol symbol) {
return 0 < symbol && symbol < self->external_token_count + 1;
}

View file

@ -415,13 +415,15 @@ TSPoint ts_node_end_point(TSNode self) {
}
TSSymbol ts_node_symbol(TSNode self) {
return ts_node__alias(&self)
? ts_node__alias(&self)
: ts_subtree_symbol(ts_node__subtree(self));
TSSymbol symbol = ts_node__alias(&self);
if (!symbol) symbol = ts_subtree_symbol(ts_node__subtree(self));
return ts_language_public_symbol(self.tree->language, symbol);
}
const char *ts_node_type(TSNode self) {
return ts_language_symbol_name(self.tree->language, ts_node_symbol(self));
TSSymbol symbol = ts_node__alias(&self);
if (!symbol) symbol = ts_subtree_symbol(ts_node__subtree(self));
return ts_language_symbol_name(self.tree->language, symbol);
}
char *ts_node_string(TSNode self) {

View file

@ -742,8 +742,8 @@ TSQuery *ts_query_new(
) {
// Work around the fact that multiple symbols can currently be
// associated with the same name, due to "simple aliases".
// In the next language ABI version, this map should be contained
// within the language itself.
// In the next language ABI version, this map will be contained
// in the language's `public_symbol_map` field.
uint32_t symbol_count = ts_language_symbol_count(language);
TSSymbol *symbol_map = ts_malloc(sizeof(TSSymbol) * symbol_count);
for (unsigned i = 0; i < symbol_count; i++) {