diff options
| author | KOBAYASHI Shuji <[email protected]> | 2020-11-24 16:30:52 +0900 |
|---|---|---|
| committer | KOBAYASHI Shuji <[email protected]> | 2020-11-24 16:30:52 +0900 |
| commit | db00fb238be221f58414d5b5763321706e262f24 (patch) | |
| tree | 2860aa510340d14a35f80fab9e5ce3dd294ec78f /Rakefile | |
| parent | 36e3c4404af102411b6949e672aa6b73768945d5 (diff) | |
| download | mruby-db00fb238be221f58414d5b5763321706e262f24.tar.gz mruby-db00fb238be221f58414d5b5763321706e262f24.zip | |
Optimize `presym_find`
Chang to compare string length first.
### Benchmark
#### Code
* https://github.com/shuujii/mruby-presym_find-benchmark
#### Result
```console
Previous: 10.240772M i/s (25M times in 2.441222s)
New: 16.412985M i/s (25M times in 1.523184s)
```
Diffstat (limited to 'Rakefile')
| -rw-r--r-- | Rakefile | 2 |
1 files changed, 1 insertions, 1 deletions
@@ -192,7 +192,7 @@ file presym_file => cfiles+rbfiles+psfiles+[__FILE__] do src.scan(/[ \(\[\{]:'([^']+)'/) ] end - symbols = (symbols+csymbols+rbsymbols+op_table.keys).flatten.compact.uniq.sort.grep_v(/#/).map{|x| x.gsub("\n", '\n')} + symbols = (symbols+csymbols+rbsymbols+op_table.keys).flatten.compact.uniq.grep_v(/#/).map{|x| x.gsub("\n", '\n')}.sort_by!{|x| [x.bytesize, x]} presyms = File.readlines(presym_file) rescue [] presyms.each{|x| x.chomp!} if presyms != symbols |