New HB showroute implementation

[yakra]

Screenshot of a DISCONNECTED_ROUTE error:

[Jim]

Marker popups now have links to intersecting/connected routes that should be the same as what we have in the production HB code.

[Jim]

I made a few minor fixes and enhancements here and there this afternoon, and then added some copy-to-clipboard functionality.

So like with the new Mapview a little while ago, I would like some help with testing, to make sure the new showroute on tmtest is still doing everything the old HB version of the same did for you, as well as get any additional comments/suggestions/problem reports about the new functionality.

I'm hopeful that the .list Entry Toolbox functionality (which I am going to leave with only segment-based selections, at least until after the initial version goes live) will be really helpful for new users, and even for experienced users who have been baffled over the years by beltways.

I'd like to merge this into the master branch and install it on the main site before I move on to other things, so if our regulars could take a look sooner rather than later, that would be appreciated.

[michih]

It happens to me quite often that the "Processing Route Stats Data..." messages doesn't disappear. I jumped from small routes of about 10..20 wps to other small routes in Norway by using the intersecting routes link.

When I click on the intersecting routes link, the map is first loaded for Siena and then moved to the correct location. It doesn't happen with the old HB version. Loading times for jumping between intersecting routes are smaller with the old version.

When clicking on a wp on the wp list table, the map was centered with the old version but isn't with the new one.

[Jim]

It happens to me quite often that the "Processing Route Stats Data..." messages doesn't disappear. I jumped from small routes of about 10..20 wps to other small routes in Norway by using the intersecting routes link.

There's a timing issue there that I will resolve, hopefully today.

When I click on the intersecting routes link, the map is first loaded for Siena and then moved to the correct location. It doesn't happen with the old HB version. Loading times for jumping between intersecting routes are smaller with the old version.

For any showroute load, it shows the default map location until the data is loaded, then fits the map bounds to the data.  This is because the DB reads for almost everything have moved from PHP on page load to PHP in AJAX calls, so the initial page shows up (with the default map) while data is loading.  I'd expect overall loading times to be similar, since it's pretty much the same data, but I'm surprised that the new one is slower in the end given that we're no longer transferring and executing a giant chunk of JS code but instead getting only the data.  Maybe the conversion to JSON and back takes more time than I think.

When clicking on a wp on the wp list table, the map was centered with the old version but isn't with the new one.

Do we want it to center?  I can do that if that's the preference.

[michih]

When clicking on a wp on the wp list table, the map was centered with the old version but isn't with the new one.

Do we want it to center?  I can do that if that's the preference.

I think so!

[Jim]

It happens to me quite often that the "Processing Route Stats Data..." messages doesn't disappear. I jumped from small routes of about 10..20 wps to other small routes in Norway by using the intersecting routes link.

There's a timing issue there that I will resolve, hopefully today.

I think I've fixed this, please let me know if you encounter a message that doesn't go away.

[yakra]

Do we want it to center?  I can do that if that's the preference.

I think so!

Seconded.

[michih]

It happens to me quite often that the "Processing Route Stats Data..." messages doesn't disappear. I jumped from small routes of about 10..20 wps to other small routes in Norway by using the intersecting routes link.

There's a timing issue there that I will resolve, hopefully today.

I think I've fixed this, please let me know if you encounter a message that doesn't go away.

I think it happened in about 20..30% of the cases. I made about 30 tests now and it never happened again!

About centering at wp: It happend to me now that the route was not on the screen at all. Many cases where it was on the margin(?) but once the route was only on the visible screen when I zoomed out three times.

[Jim]

Centering on all marker popups is now in place.  I did not do the same for segment popups, since I think all the bouncing around could be distracting for users trying to mark segments for .list entry generation.

Anything else missing or need fixing before this goes live on the main site?

[Jim]

This is live on the main site.  Please report problems here or in a GitHub Issue.

[rickmastfan67]

This is live on the main site.  Please report problems here or in a GitHub Issue.

Links need to be updated on the 'updates' page so we can 'bypass' the "Please continue with the new showroute page for this route" message and go to the new page right off the bat.
https://travelmapping.net/devel/updates.php

[Jim]

Thanks.  Just fixed this and the datacheck.php links.

[Jim]

Would you prefer the "View Connected Route" link on chopped route pages in the new showroute.php to behave as it does now, where you are taken to an overview of the entire connected route, or would you prefer to have it link to the connected route with the current position and zoom level?

The current behavior is nice, because the link is static, and the functionality already works.  It's easy to copy the link without following it, as there's no JS involved in it.

The other behavior would be nice for long routes, where you might be interested in seeing visible parts in neighboring regions without having to zoom your way back down.  Plus, if you really do want the overview, Zoom to Fit is a single additional click to get it.  It would also make this link consistent in behavior to the Connected/Intersecting Routes links in the popups.

[michih]

The other behavior

This.