Switch to compact-SRID WKB format #664
Conversation
| @@ -202,7 +202,6 @@ else if(isStrict) | |||
|
|
|||
| // determine if SRIDs are present | |||
| hasSRID = (typeInt & 0x20000000) != 0; | |||
| int SRID = 0; | |||
There was a problem hiding this comment.
Yes, definitely problematic. This is a bug in the GEOS implementation IMO.
This should just use the SRID of the top-most geometry (or 0 if not set). JTS does not support mixed SRIDs in collections.
There was a problem hiding this comment.
Shouldn't we throw a ParseException in that case instead of using the SRID of top-most geometry?
There was a problem hiding this comment.
It's nice to allow reading as many WKB inputs as possible. Even if they are invalid, they can only be fixed if they are able to be read.
To preserve the most information, the SRIDs should always be read, and then applied to any missing SRIDs in child geometries. This will require passing the most recent SRID down the call stack as a parameter.
There was a problem hiding this comment.
Added a commit, PTAL.
Signed-off-by: Halil İbrahim Şener <[email protected]>
hisener
force-pushed
the
hsener/gh-663
branch
from
86095d7 to
f1261ac
Compare
Signed-off-by: Halil İbrahim Şener <[email protected]>
hisener
force-pushed
the
hsener/gh-663
branch
from
e22f1af to
b542462
Compare
|
Can we include this change in the next release, and is there a planned date for it? |
|
Yes, I will commit this. There should be a new release by the middle of 2021, I think. Is that soon enough for you? |
dr-jts
changed the title
|
Is it going to be part of the next major release? I was hoping for a minor in the next couple of weeks. 🙈 |
|
We might be able to get out a minor release this quarter. A minor release takes just as much work as a major release. |
|
Ack! This quarter would be nice. Thanks! 🙏 |
Fixes #663.
I have ported GEOS WKBWriter and changed
WKBReaderto use the parent's SRID in sub geometries.Added tests to verify that
WKBWriterwrites compact WKB formatWKBReadercan read from both format types