beln: Belgium Route Nationale / Gewestwegen

[si404]

N700, N701, N702 - fine

N712
 - N746. I'm not sure how the routes in Lommel go, but it's definitely not one point in the middle of the town not even at a junction! Both Google and OSM have concurrencies between N746 and N712 (but vary on which side of the roundabout at Sint-Pietersbadenkerk they are concurrent). It seems like the town centre is traffic calmed and routes stopped up for motor vehicles. N746 used to be signed into the town centre from the north, but the signs have been removed. There's also the N790 coming in from Lutlommel to add to the mix.

N713 - fine

N715 - FabStr -> N790?

N715b - HereStr -> N715a?

N716, N716a - fine

N717
 - N725_E -> N717b/N725?
 - N717c: either add route, or relabel point

N718 - fine

N719 - MuiStr -> BrugStr?

N721 - extends to R73?

N722 - Sch -> N722a?

N723
 - SteWeg -> N723a? if not then SteWeg_E
 - DorStr -> N730a?

N724, N725, N726, N729 - fine

N730
 - DorStr -> N730a?
 - add point for N723a/SteWeg

N739 - SintMarPla -> N739a?

N742, N743, N744, N745 - fine

N746 - see N712

N747 - GroStr -> GroStr_W

N748 - fine

N750, N751, N752, N753, N754, N754a, N755, N757, N758, N759, N760, N761, N762, N763, N765, N766, N768, N769, N770 - fine

N771 - GroStr -> WegOpi

N772 - PasMev -> PasMevLaan (for consistency with all the other points)

N773, N774, N775 - fine

N777 - OnzHerStr -> VinStr

N779, N784, N789 - fine

N790
 - continues west to Lommel?
 - add point at Leopoldlaan as N712a?

N793 - add point at Bocholterkiezel as N731?

[si404]

N800 - fine

N801 - N891 -> N891/N897?

N802, N803, N804, N805 - fine

N806 - extends west to N831?

N809 - N82/N875: is the one way ring R80? (also N82 and N875)

N810, N810Bou, N811, N812, N813, N814, N815, N816, N817, N818, N819, N820, N821, N822, N823, N825, N826, N827, N828, N829 - fine

N830 - fine

N831
 - Hin -> CheNan
 - PetBar -> N806?

N833 - CheMin & War -> N813_N & N813_S ?

N834, N835 - fine

N836 - add point for Contournement de Marche (N839?) as leads to N4 GSJ

N837, N838 - fine

N839 - continues south to N836?

N839Ize, N840, N840a, N841, N842, N843, N844, N845, N846, N847, N848, N849, N850, N851, N852, N853, N854, N855, N856, N857, N858, N859, N860 - fine

N861 - extends north?

N862, N864, N865, N866, N868 - fine

N870 - RueGau -> N852?

N871 - QuaOuy -> N871a?

N872 - do routes in LUX need sorting to match BEL/LUX? or vice versa? IIRC, N5E ends between the two half junctions, but should it extend to here as it does meet the N872.

N873 - fine

N874 - does it stay within shaping tolerance between Nef and N821

N875, N875a, N876, N877 - fine

N878 - Cou_E is off

N879 - Buz -> RueMon?

N880, N881, N882, N883 - fine

N884 - RueBab -> RueBab_E

N884a, N885 - fine

N886 - N88 -> N88_N

N887, N888, N889, N890, N891, N892, N893, N894, N895, N896, N897, N898, N899 - fine

[si404]

N902 -fine

N904 - add point for Nouvelle Route de Suarlee (leads to A15 exit 12)

N905, N907, N908, N909, N910, N911, N912, N913, N914 - fine

N915 - RueJeanLib -> RueAndLib

N917, N918, N919, N920, N921 - fine

N922 - N573 -> R53

N923, N924, N925, N928, N929, N930, N931, N931Pro - fine

N932
 - N92 -> N92/N947a
 - continues west to N978?

N933, N934, N935, N936, N937, N938, N939 - fine

N940, N941, N942, N942a, N943, N944, N945, N946, N947, N947a, N948, N949, N951, N952, N953, N954, N955, N957, N958, N959, N961, N963, N964 - fine

N966 - does the loop merit a bit more fancy shaping?

N967, N969 - fine

N971, N972, N973 - fine

N975
 - N576 -> R53 (tier 4 trumps tier 5 surely?)
 - is Rue de Walcourt (south of N574) N978?

N977 - RueGim -> RueVod

N978
 - continues north to Gerpinnes?
 - RueSaiDon is off and is it also N932?

N978a, N981, N982, N983, N983a, N984, N988, N989, N990, N991, N992, N998 - fine

[si404]

Don't forget to check datacheck and NMP, though everything currently is marked as false positives.

http://travelmapping.net/logs/nmpbyregion/bel.nmp
http://travelmapping.net/devel/datacheck.php?sys=null&rg=BEL

[michih]

N712
 - N746. I'm not sure how the routes in Lommel go

Same to N712 at the west. Sign into town was removed.
The large direction signs at the east was also removed.
And N746 at the south too - removed.

I've interrupted all routes at Lommel. L712 (East) already ends at N715. No evidence for extension nor where the it would end.

N715 - FabStr -> N790?
N790
 - continues west to Lommel?

no, nothing signed. I don't know where the end point should be.

N715b - HereStr -> N715a?

To the north as indicated on OSM or to the east as indicated on GM?
Keep as-is since there are also no signs on GSV.

N721 - extends to R73?

I don't think so. The extension of N721 and the existance of R73 are only indicated on OSM but no additional proof.

N790
 - add point at Leopoldlaan as N712a?

Only indicated on OSM but no idea where it should end.

[michih]

N801 - N891 -> N891/N897?

OSM is wrong. N897 is on a different route and THIS segment IS N891 according to 2019 GSV.

N809 - N82/N875: is the one way ring R80? (also N82 and N875)

Indicated on OSM but not anywhere else. Not on any list. Keep as-is.

N833 - CheMin & War -> N813_N & N813_S ?

OSM indicates it being N813 but there is nothing on GSV. N813 exists at another place and it's signed there according to 2019 GSV (it wasn't in 2010).

N861 - extends north?

Confirmed by 2019 GSV but where does it end? I've put the wp to the next junction in Thibessart.

N871 - QuaOuy -> N871a?

Yep, according to OSM. But GM indicates it being N87. From junction with N811 (unsigned on 2019 GSV) to the French border which is N871 on OSM and 2019 GSV. Don't wanna add it.

N872 - do routes in LUX need sorting to match BEL/LUX? or vice versa? IIRC, N5E ends between the two half junctions, but should it extend to here as it does meet the N872.

It's annoying, but... IMO leave as-is.
The closest comparison I can think of in my regions is ME113 & NH113B at the ME/NH border, though there's only one border point and super-short route segment there.

I'm open for other suggestions

N873 - fine

No, it extends to Aye according to 2019 GSV.

N874 - does it stay within shaping tolerance between Nef and N821

Sure! Why do you doubt?

[si404]

N874 - does it stay within shaping tolerance between Nef and N821

Sure! Why do you doubt?

It just looked off - exacerbated that the route is all to one side of the blue line. Checking it does get fairly close.

[michih]

N966 - does the loop merit a bit more fancy shaping?

I think it's pretty fine. I tried to avoid SA errors


I'll check data errors, NMPs and concurrencies tomorrow, and plan to activate the system after almost three years in preview. Thanks for the peer-review


Edit: https://github.com/TravelMapping/HighwayData/pull/3712

[michih]

Data errors, NMPs and concurrencies are fine, changes are committed, data check is running, French Fries are in the oven  

System will be active with the next site update 

[si404]

French Fries are in the oven  

I'm sure any Belgians reading this thread were appalled at this turn of phrase. 'French' and 'oven'. What horror is this! 

[michih]

Anyway, lunch was fine. And the pull request was submitted: https://github.com/TravelMapping/HighwayData/pull/3714

[yakra]

beln.csv:
beln;BEL;N712;Pel;Pelt;;bel.n712pel;
->
beln;BEL;N712;;Pel;Pelt;bel.n712pel;

[michih]

beln.csv:
beln;BEL;N712;Pel;Pelt;;bel.n712pel;
->
beln;BEL;N712;;Pel;Pelt;bel.n712pel;

How did you catch this? Is there a data error (data check output?) I didn't notice?
If not.... I think it should be a data error: abbrev entry exists but the city entry is missing.

[yakra]

How did you catch this? Is there a data error (data check output?) I didn't notice?

The C++ siteupdate program has a few extra error checks not in the Python version. I've been working on some changes to check all CSV data and data that goes into a DB field, in both versions. This should all be done... soon.

If not.... I think it should be a data error: abbrev entry exists but the city entry is missing.

This is a good idea. Thought about it but never implemented it for whatever reason. Adding to the work in progress.
Edit: Done. Tested with HighwayData @ 143811e9447d411958ff9e7660bbac96f8276e83

[yakra]

If not.... I think it should be a data error: abbrev entry exists but the city entry is missing.

This is a good idea. Thought about it but never implemented it for whatever reason. Adding to the work in progress.
Edit: Done. Tested with HighwayData @ 143811e9447d411958ff9e7660bbac96f8276e83

On second thought...
It's possible there could be some rare (but legitimate?) reasons we'd want to do this. Take KS171Pit. There used to be an ordinary KS171, but KDOT decommissioned it. It makes sense to remove the City field, as there's no other KS171 to differentiate from.
Could I & should I rename it KS171 & make KS171Pit an AltLabel in this case. Yeah. But the point is...
Imagine that the original plain KS171 became an AltRouteName for another route, and that it remained in-use in a .list file. KS171Pit would by necessity have to remain KS171Pit.

It makes sense to have these not be a showstopper that terminates the siteupdate program. They could go into datacheck, with the route flagged as per normal, just with no waypoint labels listed, maybe with the system name .csv in the "Info" column to provide a little clarity.
If we really wanna get fancy with it, we could check for the corresponding abbrev-less route in in-use AltLabels, and if found, not flag the error. I think this could even be a datacheck_always_error.