Sort out axis=neu for EPSG:4326

[Don-vip]

We define EPSG:4326 with axis=neu since r8584 to get WMTS working:

# WGS 84
<4326> +proj=lonlat +datum=WGS84 +axis=neu +bounds=-180,-90,180,90  <>

Yet we have currently the Saxony latest aerial imagery WMTS in Maps/Germany which ​defines its EPSG:4326 tile matrix as follows:

<TileMatrixSet>
  <ows:Identifier>grid_4326</ows:Identifier>
  <ows:SupportedCRS>EPSG:4326</ows:SupportedCRS>
  <TileMatrix>
    <ows:Identifier>00</ows:Identifier>
    <ScaleDenominator>17471320.7509</ScaleDenominator>
    <TopLeftCorner>11.7 51.7</TopLeftCorner>

We cannot use this matrix with JOSM as it assumes the coordinates being inverted, resulting in negative tile coordinates. This makes the imagery integration test fail (see #16073).

Who's wrong? Saxony's WMTS server or us?

[stoecker]

Is there an 1.3/1.1 difference similar to WMS for WMTS?

[anonymous]

Replying to stoecker:

Is there an 1.3/1.1 difference similar to WMS for WMTS?

No, there is only one version of WMTS specification.

I've cross-checked in QGIS and there - projection 4326 also fails (using default settings).

But in QGis there is an option: "Ignore axis orientation (WMS 1.3/WMTS) and with this checked - this layer works in 4326.

So it looks like "well known bug" in WMTS implementation in some servers

[Don-vip]

OK, I've sent them an e-mail and will ignore the error until we get a response.

[Don-vip]

First answer: my request has been received and was forwarded to the support team for processing.

[Don-vip]

I completely forgot this, but I received today this answer:

Hello,

we adjusted the configuration and corrected the alignment. Thanks for the note.

Greetings

GDI Servicezentrum
_
STAATSBETRIEB GEOBASISINFORMATION UND VERMESSUNG SACHSEN (GeoSN)
GDI Servicezentrum
Olbrichtplatz 3 | 01099 Dresden | Postanschrift: Postfach 10 02 44 | 01072 Dresden
servicedesk@… | www.landesvermessung.sachsen.de

I removed the ignore rule accordingly.

[Don-vip]

Error is still present.