Represent byte, char and tree counts as 32 bit numbers

The parser spends the majority of its time allocating and freeing trees and stack nodes.
Also, the memory footprint of the AST is a significant concern when using tree-sitter
with large files. This library is already unlikely to work very well with source files
larger than 4GB, so representing rows, columns, byte lengths and child indices as
unsigned 32 bit integers seems like the right choice.
This commit is contained in:
Max Brunsfeld 2016-11-14 12:15:24 -08:00
parent 11e767bd81
commit 535879a2bd
25 changed files with 268 additions and 263 deletions

View file

@ -19,7 +19,7 @@ enum {
symbol9, symbol10
};
Length operator*(const Length &length, size_t factor) {
Length operator*(const Length &length, uint32_t factor) {
return {length.bytes * factor, length.chars * factor, {0, length.extent.column * factor}};
}
@ -54,7 +54,7 @@ vector<StackEntry> get_stack_entries(Stack *stack, StackVersion version) {
ts_stack_iterate(
stack,
version,
[](void *payload, TSStateId state, TreeArray *trees, size_t tree_count, bool is_done, bool is_pending) -> StackIterateAction {
[](void *payload, TSStateId state, TreeArray *trees, uint32_t tree_count, bool is_done, bool is_pending) -> StackIterateAction {
auto entries = static_cast<vector<StackEntry> *>(payload);
StackEntry entry = {state, tree_count};
if (find(entries->begin(), entries->end(), entry) == entries->end())