Changes between Version 4 and Version 5 of Ticket #12427, comment 10


Ignore:
Timestamp:
2016-01-24T15:12:40+01:00 (10 years ago)
Author:
cmuelle8

Legend:

Unmodified
Added
Removed
Modified

  • Ticket #12427, comment 10

    v4 v5  
    1212I've already written in the ticket description that I do not intend to use curvatureDistance() as a general replacement, but for high zoom situations only (in which case LatLon.getCenter() /is/ guaranteed to deliver a point on the shortest path, at least for [https://en.wikipedia.org/wiki/Mercator_projection merc projections] (if the prime meridian is not greenwich, we're just talking about an offset fix)).
    1313
    14 To use it generally (low zoom), the shortest path problem could IMO be fixed by simply using the antipodal point to {{{m}}} if {{{this.distance(other) > 180}}} degrees.
     14EDIT1: To use it generally (low zoom), the shortest path problem could IMO be fixed by simply using the antipodal point to {{{m}}} if {{{this.distance(other) > 180}}} degrees.
     15
     16EDIT2: It is true, that {{{a.getCenter(b)}}} will not produce a point at exactly half of true distance, maybe this is what you're referring to, but please explain why it should not lie ''exactly'' on that line!? For the algorithm to work correctly we ''solely'' need it to be on that line, it does ''not'' need to hit the middle of the line.