This project is mirrored from https://github.com/haskell/text. Pull mirroring updated .
  1. 21 May, 2017 6 commits
  2. 03 May, 2017 1 commit
  3. 07 Dec, 2016 2 commits
  4. 06 Dec, 2016 6 commits
  5. 02 Dec, 2016 2 commits
  6. 11 Oct, 2016 1 commit
  7. 18 Aug, 2016 1 commit
  8. 17 Aug, 2016 1 commit
  9. 16 Aug, 2016 5 commits
  10. 02 Jul, 2016 1 commit
  11. 22 Jun, 2016 1 commit
  12. 17 May, 2016 1 commit
  13. 14 May, 2016 1 commit
    • harendra's avatar
      Fix benchmarking · a3247022
      harendra authored
      france.html is not available in the test data. Commenting out the corresponding
      benchmark for now so that benchmarking does not fail.
      a3247022
  14. 06 May, 2016 1 commit
    • harendra's avatar
      Optimize unstream loop · 09843692
      harendra authored
      Separate the infrequent code from bulk processing loop so that the common case
      loop is smaller and more likely to use processor pipelining more efficiently.
      
      With this change the ddump-simpl output of the unsream loop was reduced by
      around 10 times and the performance of just stream followed by unstream got
      improved by 3 times. The relevant macro benchmark results on my machine are as
      follows:
      
      Benchmark                       Before After     Change
      -------------------------------------------------------
      Pure/intersperse/Text+ascii     1.552 1.025 (ms) 34%
      Pure/cons/Text+ascii            830.0 564.7 (us) 32%
      Pure/map/Text+ascii             824.2 604.5 (us) 27%
      
      Pure/filter/Text+ascii          342.5 320.2 (us) 6.5%
      Pure/toLower/Text+ascii         3.447 3.253 (ms) 6%
      Pure/zipWith/Text+ascii         1.252 1.207 (ms) 3.6%
      Pure/replicate char/Text+ascii  697.7 682.4 (us) 2%
      09843692
  15. 05 May, 2016 1 commit
  16. 30 Mar, 2016 2 commits
  17. 29 Mar, 2016 1 commit
  18. 17 Mar, 2016 5 commits
  19. 13 Feb, 2016 1 commit