Hi,
FYI, sqlite 3.41.0 has just been released yesterday, and I've noticed it
causes a major performance regression with PROJ which I've reported to
https://sqlite.org/forum/forumpost/bcc4375032
Hopefully there will be a fix on SQLite side in a bugfix release.
Otherwise we might have to workaround the issue on PROJ side.
Even
Le 22/02/2023 à 15:03, Kristian Evers a écrit :
All,
We are closing in on March 1st which is the release date of both PROJ 9.2.0 and
PROJ-data 1.13.
In preparation for the final releases I have uploaded release candidates for
both packages.
Download the archives here:
https://download.osgeo.org/proj/proj-9.2.0RC1.tar.gz
https://download.osgeo.org/proj/proj-9.2.0RC1.zip
https://download.osgeo.org/proj/proj-data-1.13RC1.tar.gz
https://download.osgeo.org/proj/proj-data-1.13RC1.zip
The PROJ release includes a number of improvements, new features and bug fixes.
See the release notes below.
The PROJ-data release adds the following transformations grids:
- American NADCON5 grids
- An updated version of no_kv_arcgp-2006-sk.tif
- Slovenian geoid model SLO-VRP2016/Koper
Please test the release candidates and report any issues you may find with
them. Either here
on the mailing list or on GitHub.
/Kristian
————————————————————————————————————
9.2.0 Release Notes
-------------------
Updates
--------
+ Database
o EPSG updated to EPSG 10.082 (#3609)
o ESRI objects updated to v3.1.0 (#3575)
o Added Slovenian geoid model SLO-VRP2016/Koper (#3610)
+ Projections
o Added generic Space Oblique Mercator (#3338)
o Inverse eck4: Add checks for projection domain (#3625)
+ Transformations
o Added support for NADCON5 grids and transformation method (#3510)
o Added option ONLY_BEST=YES (#3535)
When enabled only use the best possible transformation between two CRS's
can be
instantiated. If the best possible transfomations can't be instantiated
it will
fail instead of opting for the next best transformation.
With ONLY_BEST=YES a transformation will fail if a needed grid is
unavailable.
The option can be toggled in proj.ini and with --only-best in cs2cs.
o Implement 'Similarity transformation' (EPSG:9621) and import related
tranformations from EPSG (#3578)
o Optimise transformations between GDA94, GDA2020 and WGS84 (#3621)
+ Performance improvements
o Pass PJ_COORD by reference for fwd4d and inv4d callbacks (#3503)
o Improved of inverse meridian length calculations (#3516)
o Avoid very poor performance on a geographic CRS whose datum name is the
alias
of an official one (#3624)
+ Well-Known Text improvements
o Support WKT:2019 COORDINATEMETADATA[] construct (#2711)
o implement alterCSLinearUnit for CompoundCRS (#3325)
o Implement alterCSLinearUnit for BoundCRS (#3327)
o Added PJ_TYPE_DERIVED_PROJECTED_CRS (#3488)
o Added support for ISO19111 CoordinateMetadata class (#3562)
o Update WKT import to support WKT2 corrigendum (#3573)
o WKT import: Make axis direction and range meaning enumeration case
insensitive as mandated by spec (#3577)
o Implement AffineCS and handle it in WKT (#3579)
o WKT1 parser: Support inf as inverse flattening value (#3626)
+ PROJJSON
o Version number changed to 0.6
o Added an additional optional source_crs property in
abridged_transformation (#3454)
o Added support for ISO19111 CoordinateMetadata class (#3562)
o Implement AffineCS and handle it in PROJJSON (#3579)
+ C/C++ API
o C API: add proj_get_domain_count(), proj_get_area_of_use_ex() and
proj_get_scope_ex() (#3588)
DEPRECATIONS
The following functions have been deprecated and will be removed in a
future version.
o C API function proj_rtodms. Use proj_rtodms2 instead (#3431)
o C experimental API function proj_create_conversion_tunisia_mapping_grid().
Use proj_create_conversion_tunisia_mining_grid() instead (#3559)
o C++ API function createTunisiaMapingGrid(). Use createTunisiaMiningGrid()
instead (#3559)
o Added explicit NaN handling to proj_trans and gie (#3603)
+ CMake
o Added uninstall target (#3362)
o Adopted use of TIFF:TIFF and CURL::libcurl targets (#3374)
+ Security improvements
o Replaced uses of sprintf with the safer snprintf (#3421, #3431)
+ Various other improvements
o Implemented platform independent mutex for better results on lesser used
platforms (#3425)
o Improved name and remarks in descriptions of some vertical transformations
(#3334)
o Implemented stricter matching by name to avoid false positives (#3487)
o Improved numerical accuracy of the meridian length calculations and its
inverse (#3516)
o Added PROJ_IGNORE_CELESTIAL_BODY=YES environment variable that allows
better matching of
non-earth shaped ellipsoids to other celestial bodies (#3569)
Bug fixes
----------
o rtodms(): fix potential buffer overflow not dealt with (#3441)
o NetworkFile::open(): fix memory leak in unlikely error code path (#3443)
o When importing LOCAL_CS["foo"], generates a non-empty name for the datum
(#3491)
o Avoid unit mix-up in CRS::alterCSLinearUnit for DerivedProjectedCRS (#3499)
o WKT to PROJ conversions: Map 'Lambert Azimuthal Equal Area (Spherical)' to
'+proj=laea +R_A' (#3526)
o ProjectedCRS::identify(): Relax trust in id/authority in definition to
identify (#3532)
o Fix instantion of http://www.opengis.net/def/crs/IAU/2015/xxx URNs (#3569)
o Fix proj_normalize_for_visualization() on a transformation with
FORCE_OVER=YES (#3570)
o createBoundCRSToWGS84IfPossible(): Do not report +towgs84 for a CRS with
multiple extents,
such as EPSG:6316 (#3587)
o pj_get_suggested_operation(): handle longitudes outside of [-180,180] for
coordinate
operation selection (#3595)
o WKT1 parser: only emit warning when encountering invalid UNKNOWN WKT1 axis
direction (#3618)
_______________________________________________
PROJ mailing list
[email protected]
https://lists.osgeo.org/mailman/listinfo/proj
--
http://www.spatialys.com
My software is free, but my time generally not.
_______________________________________________
PROJ mailing list
[email protected]
https://lists.osgeo.org/mailman/listinfo/proj
