Get a sneak peek at whats next for Permanent Hazards on our April 7th Office Hours!
Discussion for the unofficial, community-developed addons, extensions and scripts built for the Waze Map Editor.

The official index of these tools is the Community Plugins, Extensions and Tools wiki page.
Post by dbraughlr
iainhouse wrote:... the scripts can only perform actions that can already be performed by editors.
I believe that some scripts to perform functions which are not possible with the WME user interface but are possible with the Waze object model. The functions for manipulating roundabouts are such functions. This amazing ability to modify a property which should be read-only could be a pair of ordinary bugs.
dbraughlr
Posts: 569
Has thanked: 164 times
Been thanked: 98 times
Send a message

Post by dbraughlr
The statement "This roundabout is correct." should not be displayed when there is an unsaved change to the roundabout which could cause it to become incorrect.

After redoing a roundabout' but before it is saved, it should not be reported as still incorrect.

Redoing a roundabout should set all turns including disallowing U-turns. I'm seeing soft turns.
dbraughlr
Posts: 569
Has thanked: 164 times
Been thanked: 98 times
Send a message

Post by dbraughlr
This segment has no City and/or Street name.
You must fix that or it won't be displayed on the client.
Anywhere such a pop-up message occurs, it needs to be tagged in a manner that identifies the script which generated it.

As for substance of the message, the first sentence is inaccurate. The second sentence is not true and propagates a myth that a segment with nothing set for City and/or Street name is not displayed in the client.

It is true that freshly paved roads do not appear in the client. But the test here is not for new paving, but for missing values - which is not the same condition that suppresses a segment from the client.
If you can detect that a segment is new paving, that would be terrific; but a subset of the cases where name properties have not been set.

Please reduce the message to a single, factual statement and include a signature (e.g., WME Toolbox).
dbraughlr
Posts: 569
Has thanked: 164 times
Been thanked: 98 times
Send a message
Last edited by dbraughlr on Thu Apr 03, 2014 6:16 pm, edited 1 time in total.

Post by dbraughlr
This roundabout is correct.
If you want, you can change it into a standard round road.
Only do that if you're sure it's not a roundabout!
Proposed change to condition: A dead-end roundabout is never correct. All roundabouts mush have at least three consecutively-numbered one-way segments in the direction of driving.

Proposed wording:
WME Toolbox:
This roundabout is correct.
If it is not actually a roundabout, change it into a standard round road.


WME Toolbox:
This roundabout is incorrect.
If it is not actually a roundabout, change it into a standard round road; otherwise, redo this roundabout.


WME Toolbox:
This roundabout is incorrect and does not join at least 3 segments.
If it is not actually a roundabout, change it into a standard round road; or delete this roundabout and reconnect the roads.


Clicking the underlined portion should invoke the correct tool and close the pop-up.
dbraughlr
Posts: 569
Has thanked: 164 times
Been thanked: 98 times
Send a message
Last edited by dbraughlr on Thu Apr 03, 2014 7:32 pm, edited 1 time in total.

Post by dbraughlr
doctorkb wrote: Personally, I prefer the small popups that don't have a "signature".
Because there are scripts with overlapping functionality, a message is necessary.
I am not the only person who has ever blamed the wrong script for something.
At the time I posted, I was not 100% certain the the source was Toolbox.
Actually, the test seems to be for the "thin red line" (aka freshly paved / uninitialised) segments. Please cite your source if you believe a freshly drawn road in WME (without a city/street or "no city / no street" being set) will display on the client.
I am the source. I cite myself. I refer to the NA server. You can easily test this. Draw a freeway square over the block where you live but do not set any name properties.
dbraughlr
Posts: 569
Has thanked: 164 times
Been thanked: 98 times
Send a message

Post by dbraughlr
doctorkb wrote:I understood all "thin red line" roads would not appear -- perhaps this is a new development
Not new.
doctorkb wrote:the key to knowing which script to "blame" for something lies in disabling scripts until your issue goes away.
That is an unnecessary effort. I have seen posts about a problem with one script on the thread for another script. It is so much simpler to tag messages.
doctorkb wrote: The scripts aren't "production" features -- they're items that have been added on by the community to support the community. If something is going haywire, you need to start disabling scripts until that stops -- not just hoping that a popup box tells you who is popping it up. Ultimately, the issue could be caused by an interaction between two scripts... so a "signed" popup doesn't necessarily even fix your concern.
Exactly because they are add-ons is why the messages need to be tagged (a small icon in the lower right corner for ltr languages or a watermark would be fine and wouldn't increase the size of the pop-up box).

A signed pop-up fixes my concern. Your method won't identify the problem script any better than mine does.
dbraughlr
Posts: 569
Has thanked: 164 times
Been thanked: 98 times
Send a message
Last edited by dbraughlr on Fri Apr 04, 2014 9:05 am, edited 1 time in total.

Post by dbraughlr
CBenson wrote:I've seen routing issue, like routing the wrong way on roundabouts with only two roads that connect to them. I do think they are a problem.
Timbones wrote:I disagree with this. A roundabout with 2 segments is a valid roundabout, and segments do not need to be consecutively-numbered. I know of many instances of both, and they function just fine.
I'm just repeating what I thought I heard. I am happy to see the facts brought forth.
There are a lot of misconceptions that have taken on a life of their own.

As for the consecutive numbering, I understand that all roundabouts are meticulously done in the UK. For the USA, we have this:
WME Toolbox: Roundabout which may cause issues (#1)
Junction IDs of the roundabout segments are not consecutive.
Is this explanation of error in need of additional qualifications?
dbraughlr
Posts: 569
Has thanked: 164 times
Been thanked: 98 times
Send a message

Post by dbraughlr
doctorkb wrote:There are probably instances where this causes a problem, and others where it doesn't.
If it might ever be a problem, why risk it when Toolbox makes it so easy to fix?
If it isn't ever a problem why have a warning at all? :?
dbraughlr
Posts: 569
Has thanked: 164 times
Been thanked: 98 times
Send a message

Post by dbraughlr
arbaot wrote:If it was mandatory to do something this process would have become automatized
Although I suspect that most of what you wrote is wrong or relies on an extremely strict interpretation of mandatory (like if it won't make the database server crash, then it's not mandatory), the idea that if there were a real problem fixing it would be automated is clearly untrue.

Regardless, for this specific alert, how precisely do you decide whether to do something about it?
dbraughlr
Posts: 569
Has thanked: 164 times
Been thanked: 98 times
Send a message

Post by dbraughlr
Timbones wrote:
dbraughlr wrote:All roundabouts mush have at least three consecutively-numbered one-way segments in the direction of driving.
I disagree with this. A roundabout with 2 segments is a valid roundabout, …
A roundabout with 2 segments is flagged for "Same connection segments".
If there is never a problem, then Toolbox really shouldn't flag it.
dbraughlr
Posts: 569
Has thanked: 164 times
Been thanked: 98 times
Send a message