Getting first drop of generated EPSG database. #5
Conversation
It is based on epsg.properties from github.com/geotools/geotools/pull/2479. It is based on code generation template from github.com/fikin/wkt-crs-go.
|
this was quite some fun to pull the generation infra up and working. but now it is ok. there are few thinks we have to review:
how do you see it? |
|
That's a lot for me to look at, thanks for your effort! I will get back to you as soon as I have some more time.... |
|
Hi, sorry, I just got around to looking at the pull request. It looks to me like you can also add the coordinate systems to the wgs84.Repository (e.g. the EPSG() function). I also wonder if the current implementation as a hash map shouldn't be more efficient than a function with a switch case. |
|
Great you came around with this 😁
It was something I wondered about myself and we could generate such, if you
want?
In my current view use of package level map would force all functions to be
in-memory regardless of client use. Use of function with switch case would
allow compiler to drop unused functions if client bypasses that
main function.
How about increasing the supported identities? Would you happen to have any
ideas after looking at not generated yet codes?
…On Mon, 23 Oct 2023, 19:29 Malte Wrogemann, ***@***.***> wrote:
Hi, sorry, I just got around to looking at the pull request. It looks to
me like you can also add the coordinate systems to the wgs84.Repository
(e.g. the EPSG() function). I also wonder if the current implementation as
a hash map shouldn't be more efficient than a function with a switch case.
What is your opinion?
—
Reply to this email directly, view it on GitHub
<#5 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/ABOUTOH7KQWKP5ZFSBZ6CJTYA2LMXAVCNFSM6AAAAAA3ZYNEDSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTONZVGU3TCMZYGM>
.
You are receiving this because you authored the thread.Message ID:
***@***.***>
|
|
Okay, that makes sense. Then I could remove wgs84.Repository and we would just provide a function "EPSG(code int) wgs84.CoordinateReferenceSystem". |
|
Imho these would be nice changes.
…On Mon, 23 Oct 2023, 20:45 Malte Wrogemann, ***@***.***> wrote:
Okay, that makes sense. Then I could remove wgs84.Repository and we would
just provide a function "EPSG(code int) wgs84.CoordinateReferenceSystem".
I could clean up all the other issues with the package and provide a v2...
I think the Area Interface and Safe-functions (with errors) aren't really
needed and it would be better to make the projections and spheroid public.
What would you change?
—
Reply to this email directly, view it on GitHub
<#5 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/ABOUTOBIIJ2L4LZBOOOJJWLYA2UKRAVCNFSM6AAAAAA3ZYNEDSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTONZVG4YDKNBUGM>
.
You are receiving this because you authored the thread.Message ID:
***@***.***>
|
|
It would be nice if you could contribute to this. I am also open to having the CRS generated using your tool. However, I've had a little trouble using your module or the cli... |
|
I'll give it a try this week. |
It is based on epsg.properties from github.com/geotools/geotools/pull/2479.
It is based on code generation template from github.com/fikin/wkt-crs-go.
The generation currently recognizes only CRS with defined TOWGS84 entries.
And only those whose projection algorithm is supported by the lib.
All other definitions are listed at the end as comments.