stop_times.txt::shape_dist_traveled and shapes.txt::shape_dist_traveled

[Frumin, Michael]

Brian,

 

Currently our GTFS has neither stop_times.txt::shape_dist_traveled nor shapes.txt::shape_dist_traveled so OBA calculates both itself, apparently correctly in every case I’ve looked at.

 

I’m considering trying to add these fields, but I want to know how much risk this will create around how OBA digests and uses our GTFS.  Does OBA even use these fields *if* they are there, or does it always re-calculate them for itself?

 

My concern is that if we messed it up at first, we might be breaking something that’s not currently broken.

 

Thanks,

Mike

[Brian Ferris]

I'm pretty sure OBA ignores both shape_dist_traveled fields, even when
specified, because there is no guarantee from the spec about what
units will be used for these values (we use meters internally) and we
generally trust ourselves to compute these values just as accurately
if not more so than the GTFS providers. I will check the code later
tonight to confirm.

Brian

> --
> You received this message because you are subscribed to the Google Groups
> "onebusaway-developers" group.
> To post to this group, send email to onebusaway...@googlegroups.com.
> To unsubscribe from this group, send email to
> onebusaway-devel...@googlegroups.com.
> For more options, visit this group at
> http://groups.google.com/group/onebusaway-developers?hl=en.

[Frumin, Michael]

Brian, thanks.

Should I take your response on the transit-developers list to indicate that in fact OBA ignores these entirely, or at least sufficiently so as to not affect anything we are doing?

Thanks,
Mike

[Brian Ferris]

Indeed. I checked the code. We're not using these values at all on import.