Page 164 of 237

Re: [Script] WME Validator 0.5.8 (BETA) / 31.01.2014

PostPosted: Sun Feb 02, 2014 6:57 am
by pumrum
I didn't see anything specific in the road naming guide for the US, but the following are being highlighted by #94 "lowercase street name (except country-specific words)":

el
la
de

These show up in some states (CA/AZ/NM) and elsewhere as lowercase on street signs. Should they remain lowercase in WME?

if not, let's add them to the exception list for US
if so, I will correct these in WME to be UPPER whenever I find them

Re: [Script] WME Validator 0.5.8 (BETA) / 31.01.2014

PostPosted: Sun Feb 02, 2014 3:58 pm
by pumrum
dbraughlr wrote:I don't see why they should be exceptions. All the lowercase ones in California are wrong and should be capitalized.


I stand corrected. I was doing edits in the SJC area and I could swear I had seen road signs that matched the lower case name in WME, but I have been unable to locate. I was likely mistaken. For now it seems like lower case names would be the exception, and 'la' and 'el' should not be added to the exception list. I think the summary from the other forum thread was "match the case of the road sign"

thanks!

Re: [Script] WME Validator 0.6.3 (BETA) / 12.02.2014

PostPosted: Mon Feb 17, 2014 9:11 pm
by pumrum
Feature Suggestion:

Highlight railroad segments that form junctions with non-railroad segments.

According to the Wiki:
DO NOT create a junction between the driven road and the railroad.

Source: https://www.waze.com/wiki/Railroad#Railroad

I see this frequently in the northeast US where there are a lot of regional railroads that go through towns and for one reason or another someone formed a junction between a railroad and a driveable segment.

Re: junctions with drivable and non-drivable

PostPosted: Mon Feb 17, 2014 10:21 pm
by pumrum
dbraughlr wrote:A Railroad segment must not be at a junction with another type.
A Runway/taxiway segment must not be at a junction with another type.
There must not be a junction with any drivable type and any non-drivable type.
A ramp must be connected at both ends and the junction at one end must have a freeway/highway/ramp segment.


Good point, perhaps instead my feature request should be:

Highlight Non-drivable segments that form junctions with segments of a different type.

This way if a railroad formed a junction with anything but a railroad, it would get highlighted. If a runway formed an intersection with a walking path, it would get highlighted. etc etc

Re: [Script] WME Validator 0.8.3 (BETA) / 14.03.2014

PostPosted: Mon Mar 17, 2014 10:44 pm
by pumrum
[post edited to reflect actual update]
I saw in one of the recent updates:

- UPD 'Non-drivable connected to drivable': excluded short dead-end segments and segments with HNs


It seems that it's ignoring non-drivable segments all together. I have found intersections between streets and runways, railroads, and walking trails - none are being flagged by WMV 0.8.3

Example:

https://www.waze.com/editor/?zoom=7&lat ... 78&env=usa

Re: [Script] WME Validator 0.9.0 (BETA) / 20.03.2014

PostPosted: Fri Mar 21, 2014 5:18 am
by pumrum
Suggestion:

Condition: Street name or alternate street name ends in 'Exd' or any case variant
Text: Street name ends in 'Exd'. Change to 'Ext' to support proper TTS instructions
Wiki Source: https://wiki.waze.com/wiki/Abbreviations_and_acronyms#Standard_suffix_abbreviations
Alert level: Note

I see a lot of segments where the suffix 'Exd' is incorrectly used to abbreviate 'Extension', when it should be 'Ext'

Re: [Script] WME Validator 0.9.0 (BETA) / 20.03.2014

PostPosted: Fri Mar 21, 2014 1:35 pm
by pumrum
berestovskyy wrote:In the latest 0.9.0 you can use a custom check (Settings->custom):
Template: ${street}
RegExp: / exd$/i


very cool! I hadn't had a chance to fiddle around with the custom filters yet, but that should work for me in the interim.

berestovskyy wrote:If we create a dedicated check, I suggest we check at least few incorrect abbreviations, not just Exd.


Agreed. Here are a couple suffixes (abbreviations) that come to mind for the USA. Perhaps other people have suggestions?

Format: [incorrect] ==> [correct]
Cntr ==> Ctr
Exd ==> Ext
La ==> Ln
Plc ==> Pl
Rvr == Riv
Tnpk ==> Tpk
Wy ==> Way


I did a quick peruse through the available checks, and didn't see anything for non-abbreviated street suffixes in the US, such as:

Road ==> Rd
Street ==> St

Would there be value in adding these for the US servers (I see some for Pl, etc)

Re: [Script] WME Validator 0.9.0 (BETA) / 20.03.2014

PostPosted: Fri Mar 21, 2014 1:39 pm
by pumrum
berestovskyy wrote:- Removed 'Toolbox/WMECH: Report highlighted segments' (now it is on by default)


I liked having this option. I spend a lot of time with highlights enabled and don't always want the validator button lighting up or populating reports. If it's not too much trouble to put it back in... even if not on the front page(s), that would be great! :)

Re: [Script] WME Validator 0.9.4 (BETA) / 30.03.2014

PostPosted: Mon Mar 31, 2014 8:30 pm
by pumrum
berestovskyy wrote:1. No road continuation (for Primary Streets and up)

yes!
berestovskyy wrote:2. No right* turn (for any drivable road)

yes! you might want to go wider than +/-5*, but I haven't given that much thought so there may be reason not to
berestovskyy wrote:3. Geometry points are too close (for any drivable road)

yes!

Re: [Script] WME Validator 0.9.6 (BETA) / 08.04.2014

PostPosted: Tue Apr 08, 2014 12:33 pm
by pumrum
olestas wrote:Hiding problems in locked segments is bad. I report those to corresponding level editors in my area. You could add option that hides them by default after installation of script.. but it should be possible to change it.


Agreed. I find that the top level editors are busy fixing major infrastructure issues... It's up to us lower level editors to scour the maps looking for everything they don't have time to search for. I frequently find issues with rank 5/6 segments as the validator gets better updates. I post them to the unlock forum, and the CMs fix it when they have time. Good system.

The rank lock system is an invaluable tool to prevent damage to the critical infrastructure, and ignoring locked segments makes the incorrect assumption that the last editor was perfect in their changes :)