ios - Does MKOverlayPathView need drawMapRect? -

i'm having inconsistencies modifying breadcrumb example, have crumbpathview subclassed mkoverlaypathview (like it's supposed to) rather subclassed mkoverlayview.

trouble is, docs limited in stating difference in how these 2 should implemented. subclass of mkoverlaypathview it's advised use:

- createpath - applystrokepropertiestocontext:atzoomscale: - strokepath:incontext: 

but in place of drawmaprect, or in addition to? doesn't seem point if it's in addition to, because both used similar implementations. using instead of drawmaprect, leaves line choppy , broken.

struggling find real world examples of subclassing mkoverlaypathview too...is there point?

update - modified code drawmaprect, should work:

- (void)createpath {     crumbpath *crumbs = (crumbpath *)(self.overlay);      cgmutablepathref newpath = [self createpathforpoints:crumbs.points                                     pointcount:crumbs.pointcount];      if (newpath != nil) {         cgpathaddpath(newpath, null, self.path);                 [self setpath:newpath];     }      cgpathrelease(newpath); }   - (void)applystrokepropertiestocontext:(cgcontextref)context atzoomscale:(mkzoomscale)zoomscale {     cgcontextsetstrokecolorwithcolor(context, [[uicolor greencolor] cgcolor]);      cgfloat linewidth = mkroadwidthatzoomscale(zoomscale);     cgcontextsetlinewidth(context, linewidth);      cgcontextsetlinejoin(context, kcglinejoinround);     cgcontextsetlinecap(context, kcglinecapround); }   - (void)strokepath:(cgpathref)path incontext:(cgcontextref)context {     cgcontextaddpath(context, path);     cgcontextstrokepath(context);      [self setpath:path]; } 

this draws initial line, fails continue line...it doesn't add path. i've confirmed applystrokepropertiestocontext , strokepath getting called, upon every new location.

here's screenshot of broken line results (it draws createpath, not after that):

here's screenshot of "choppy" path happens when drawmaprect included createpath:

without having seen more of code i'm guessing, here goes.

i suspect path being broken segments, a->b, c->d, e->f rather path points a,b,c,d, e , f. sure of we'd need see happening self.overlay , whether being reset @ point.

in strokepath set self.path 1 being stroked. doubt idea since stroking happen @ time viewforannotations.

as choppiness may side effect or poor bounds calculation on apple's part. if ends near boundary of tile apple uses cover map prompt map draw 1 line within. stroke width extends neighbouring tile hasn't been draw. i'm guessing again test out moving point north of w in "queen st w" fraction south, or increasing stroke width , see if cut off line stays in same place geographically.