Cyclestreets::journey failing for routes of one segment in length about cyclestreets-r

[Robinlovelace]

Good find. I suspected there would be more adjustments. Idea: provide the complete vector (e.g. 44,45,45,46,46,45,45) of elevations and distances as an optional text column so we can go back. Another option: set output_list and raw_output as true to get the complete output from CS (disadvantage: looses OD attribute data).

[joeytalbot]

I think we need to prevent smooth_with_cutoffs() from attempting to smooth any route with less than 3 segments

[joeytalbot]

output_list and raw_output don't help. For ten routes including the problem one:

> system.time({routes_lsoa_1 = route(l = od_lsoas_short_list_10[[82]], route_fun = cyclestreets::journey, cl = cl, smooth_gradient = TRUE)})
  |++++++++++++++++++++++++++++++++++++++++++++++++++| 100% elapsed=02s  
Most common output is sf
These routes failed: 8
The first of which was:
<simpleError in stats::filter(x, rep(1/n, n), sides = 2): 'filter' is longer than time series>
   user  system elapsed 
  0.053   0.004   1.993 
> system.time({routes_lsoa_1 = route(l = od_lsoas_short_list_10[[82]], route_fun = cyclestreets::journey, cl = cl, smooth_gradient = TRUE, output_list = TRUE, raw_output = TRUE)})
  |++++++++++++++++++++++++++++++++++++++++++++++++++| 100% elapsed=00s  
Most common output is simpleError
These routes failed: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10
The first of which was:
<simpleError in FUN(ldf[i, 1:2], ldf[i, 3:4], ...): unused arguments (output_list = TRUE, raw_output = TRUE)>
Returning list
   user  system elapsed 
  0.019   0.004   0.112 

[Robinlovelace]

I think we need to prevent smooth_with_cutoffs() from attempting to smooth any route with less than 3 segments

Sounds like a viable solution.

[Robinlovelace]

Duplicated by #22