Preparation for PR reprojection 'a la volee' #2422
Conversation
ftoromanoff
force-pushed
the
prepAlaVolee
branch
9 times, most recently
from
eefe617
to
add66c8
Compare
…er' and add 'foot'
…Back, errCallBack)
ftoromanoff
force-pushed
the
prepAlaVolee
branch
from
add66c8
to
3616127
Compare
| @@ -61,7 +64,7 @@ function toUnit(crs) { | |||
| case 'EPSG:4326' : return UNIT.DEGREE; | |||
| case 'EPSG:4978' : return UNIT.METER; | |||
| default: { | |||
| const p = proj4.defs(formatToEPSG(crs)); | |||
| const p = proj4.defs(crs.startsWith('TMS') ? formatToEPSG(crs) : crs); | |||
There was a problem hiding this comment.
Are we sure that there is no part of the code which depends on the assumption that a crs starts either by EPSG: or TMS:? Maybe this conversion is not needed anymore with #2412.
| } else if (projunit === 'foot') { | ||
| return UNIT.FOOT; |
There was a problem hiding this comment.
I'm all for splitting support of foot unit and support of non-EPSG/TMS crs in two different commits.
| if (this.onPointsCreated) { | ||
| this.onPointsCreated(layer, pts); | ||
| } | ||
|
|
There was a problem hiding this comment.
This is not documented, nor used (except for one unit test). Shall we mark this as breaking change?
| if (proj4.defs('unknown').type === 'COMPD_CS') { | ||
| console.warn('CopcSource: compound coordinate system is not yet supported.'); | ||
| metadata.wkt = metadata.wkt.slice(metadata.wkt.search('PROJCS'), metadata.wkt.search(',VERT_CS')); | ||
| proj4.defs('unknown', metadata.wkt); | ||
| } | ||
|
|
||
| const projCS = proj4.defs('unknown'); | ||
| if (projCS.AUTHORITY) { | ||
| const authority = Object.keys(projCS.AUTHORITY)[0]; | ||
| this.crs = `${authority}:${projCS.AUTHORITY[authority]}`; | ||
| if (!proj4.defs(this.crs)) { | ||
| proj4.defs(this.crs, proj4.defs('unknown')); | ||
| } | ||
| } else { | ||
| this.crs = projCS.name || 'EPSG:4326'; | ||
| } |
There was a problem hiding this comment.
Seems we are patching a missing feature of proj4.js, I'm for proposing a PR directly upstream instead of patching it in iTowns...
| if (metadata.srs) { | ||
| if (metadata.srs.authority && metadata.srs.horizontal) { | ||
| this.crs = `${metadata.srs.authority}:${metadata.srs.horizontal}`; | ||
| if (!proj4.defs(this.crs)) { | ||
| proj4.defs(this.crs, metadata.srs.wkt); | ||
| } | ||
| } else if (metadata.srs.wkt) { | ||
| proj4.defs('unknown', metadata.srs.wkt); | ||
| this.crs = proj4.defs('unknown').name; | ||
| proj4.defs(this.crs, proj4.defs('unknown')); |
There was a problem hiding this comment.
Same here, the wkt parser breaks the type invariant that what type returned by proj4.defs(crs) (i.e. the ProjectionDefinition type) shall return a defined title property. We could use this property to uniquely define a projection.
I suggest that we could contribute to proj4 such fix (using srs.authority) to define the title property.
No description provided.