1. 01 Apr, 2021 1 commit
  2. 21 Mar, 2021 1 commit
    • David Faure's avatar
      Add support for non-standard "whitsun" = easter +49 days · 019eef28
      David Faure authored
      In 9948 entries with errors, whitsun appeared 20 times.
      Not a lot, but still worth automating.
      It probably appears a lot more in UK/US, actually, for now I'm
      only looking at FR+DE+IT so it's even surprising that it came up.
      019eef28
  3. 13 Mar, 2021 1 commit
  4. 07 Mar, 2021 1 commit
  5. 21 Feb, 2021 2 commits
  6. 20 Feb, 2021 1 commit
    • David Faure's avatar
      Add abbreviated French days · 4f54362b
      David Faure authored
      This adds
      Ma, Me, Je, Ve
      (but not Mar because of March)
      
      AFAICS they don't conflict with other languages,
      and Me+Ve also work for italian.
      4f54362b
  7. 14 Feb, 2021 1 commit
  8. 13 Feb, 2021 2 commits
  9. 24 Jan, 2021 1 commit
  10. 03 Jan, 2021 3 commits
  11. 01 Jan, 2021 3 commits
  12. 30 Dec, 2020 3 commits
  13. 18 Dec, 2020 3 commits
  14. 17 Dec, 2020 2 commits
  15. 16 Dec, 2020 1 commit
  16. 14 Dec, 2020 2 commits
  17. 13 Dec, 2020 3 commits
    • Volker Krause's avatar
      Accept localized day names in more languages · 73a2ff23
      Volker Krause authored
      Also, skip a few filler words and accept states in other languages. This
      covers another ~500 expressions in the full OSM corpus.
      73a2ff23
    • Volker Krause's avatar
      Support alternative range separators · fdfc39dd
      Volker Krause authored
      That is words like "to" (in various languages). That makes about 1.8k
      more expressions from the full OSM opening hours set parse.
      fdfc39dd
    • Volker Krause's avatar
      Move time parsing from lexer to parser · abee559b
      Volker Krause authored
      This allows us to handle e.g. spaces inside times since we have enough
      context to distinguish this from other places where a time-like token
      sequence can occur. This also adds support for a few Unicode separators
      found in use.
      
      The am/pm handling is a bit messy so we retain the ability to distinguish
      "10a" and "10 a 12" (with 'a' being an ASCII-fied version of 'à', as a
      French alternative range separator).
      
      However, I failed to do that with the LALR(1) parsing mode, as the correct
      interpretation of especially the colon and number tokens depends on more
      look-ahead. Switching to GLR mode helps with that. That however is a very
      heavy tool due to its (in theory) exponential complexity. We seem to have
      sufficiently few different parsing paths to check and typically very short
      inputs so that the cost impact on the entire 493k OSM opening hours corpus
      is only about 900ms, which is a little less than doubling the previous
      parsing cost.
      
      With this another ~850 expressions in the full corpus get accepted now.
      abee559b
  18. 08 Dec, 2020 2 commits
    • Volker Krause's avatar
      Parse times in am/pm format · 6869dff9
      Volker Krause authored
      Again not technically valid, but covers another 600 or so entries in the
      full OSM opening hours corpus.
      6869dff9
    • David Faure's avatar
      Remember whether off or closed was used · 28aecd96
      David Faure authored
      No public API changed, Interval::State still only knows Closed.
      The distinction is internal, using a separate State enum in the parser,
      and a flag in the rule. A simple bool could have been enough, I made
      room for other possible flags but I doubt it'll be useful. At least
      it's good for readability...
      28aecd96
  19. 07 Dec, 2020 2 commits
  20. 06 Dec, 2020 2 commits
  21. 05 Dec, 2020 1 commit
  22. 04 Dec, 2020 2 commits