Skip to content
gerhardol edited this page Mar 15, 2015 · 2 revisions

Currently implementing

  • ever continuing improvements

Maybe

  • Use Trail activity category to store activities used as Trails. This would also be a simple way to get Trail Categories
  • Use "Unique Routes - Common Stretches" when resyncing differences (Resync Diff at Trail Points) time/distance differences.

Less likely

  • Summary line: subtrails
  • Stretches of sub trails in the result (to get other "start points")
  • Resize multi graphs
  • Custom algorithms for smoothing (in addition to ST averaging).
  • Thread calculations, speed up handling for multiple cores. Will not make much difference though after other optimizations.
  • Insert results - compare to others. The intention is that it should be easy to paste from result services. Check a few result services, but goteborgsvarvet.com uses "Start (0 km) 00:00:00 00:00 13:33:47" and stockholmmarathon.se "5K 0.26.18 26.18 05.16 11.41 4990".
    • Parse textbox
    • Ignore lines where first character is # completely, truncate lines after #
    • Make sure no of non commentlines match trail points (or a multiple, add several at once)
    • Split fields using tab, semicolon, comma, space, trim space (in this order).
    • Regexp for time: (((\d{1,2})[.:]){1,2}(\d{1,2})(,.?)). Locale dependent?
    • Guess field number from data.
    • If there is no data for a point, do not set it. This means that if first and last is set, the speed will be constant and if 2 and 3 only is set, the speed is even between 2 and 3 only, no other changes.
    • Optional:
      • Configure name, field to parse? in # comment, using keywords?
    • Some easy way to export results as activities (can be done from Edit Trail now).
    • Also parse DateTime and lap times?

Not so likely

  • Export/Import trails in files. Probably as GPX routes