aboutsummaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2018-03-27Validate that units match between pipeline stepsKristian Evers
As suggested in [0], steps in a pipeline are now checked for compliance. If the right side units in step n differ from the left side units in step n+1 the pipeline can't be constructed and an error is raised. [0] https://lists.osgeo.org/pipermail/grass-dev/2018-March/088123.html
2018-03-26Merge pull request #902 from schwehr/gie-timeKristian Evers
gie.c: Initialize test_time other fields.
2018-03-26Initialize test_time other fields.Kurt Schwehr
Leaving x, y, z fields unset triggers an MSAN failure. fwd_prepare pj_fwd.c:40:9 pj_fwd4d pj_fwd.c:260:15 proj_trans proj_4D_api.c:166:20 test_time gie.c:1955:11
2018-03-26Tidy soci/proj4 referencesMike Toews
2018-03-26Restore PROJECT-NAME as PROJ4Mike Toews
2018-03-25Merge pull request #899 from schwehr/proj_etmerc-inf-zone-b76304619Kristian Evers
Catch over range lam0 in proj_etmerc.c
2018-03-25Catch over range lam0 in proj_etmerc.cKurt Schwehr
lam0 of inf caused a nan in the cast to int. Picked +/-1000 for lam0 as a guess for what constitutes reduculously large values. proj_etmerc.c:361:16: runtime error: nan is outside the range of representable values of type 'int' Found with autofuzz: UndefinedBehaviorSanitizer: float-cast-overflow
2018-03-26Rename CMake infrastructure to remove 4mwtoews
2018-03-25Merge pull request #900 from mwtoews/dot4Kristian Evers
Rename PACKAGE_NAME: PROJ.4 Projections -> PROJ
2018-03-25Add PACKAGE_URL for autoconf/CMakemwtoews
2018-03-25PACKAGE_NAME: PROJ.4 Projections -> PROJmwtoews
2018-03-25Merge pull request #898 from aaronpuchert/compile-flagsKristian Evers
Don't set -std=c89 in CMakeLists.txt
2018-03-25Merge pull request #893 from schwehr/horner-overflowKristian Evers
horner: Fail if the order is unreasonably large.
2018-03-24Don't set -std=c89 in CMakeLists.txtAaron Puchert
This allows us to take advantage of newer features when they are available. However, builds on Travis still use -std=c89 to ensure C89 compatibility. Locally this can be achieved with either * cmake -DCMAKE_C_STANDARD=90 .. * cmake -DCMAKE_C_FLAGS='-std=c89' .. * C_FLAGS='-std=c89' cmake .. We also reorder the warning flags: they are all part of the standard build now, but -Werror is only applied on Travis. Fixes #892.
2018-03-24Merge pull request #896 from cffk/assertNaNKristian Evers
geodtest: introduce assertNaN. geodesic: comment on non-use of pj_is_nan.
2018-03-24Merge pull request #895 from schwehr/use-pj-is-nanKristian Evers
Use pj_is_nan rather than x == x or x != x (#894)
2018-03-23Merge remote-tracking branch 'osgeo/master'Kristian Evers
2018-03-23Use the proper functions and variable names example programKristian Evers
2018-03-23Use pj_is_nan rather than x == x or x != x in pj_apply_vgridshift.c (#894)Kurt Schwehr
2018-03-23geodtest.c, checkNaN: print arg on failure.Charles Karney
2018-03-23assertEquals -> checkEquals, assertNaN -> checkNaN.Charles Karney
2018-03-23geodtest: introduce assertNaN. geodesic: comment on non-use of pj_is_nan.Charles Karney
2018-03-23horner: Fail if the order is unreasonably large.Kurt Schwehr
Overflow in horner_alloc with "2*(int)order" Found with autofuzz with UndefinedBehaviorSanitizer: signed-integer-overflow
2018-03-23Merge pull request #891 from cffk/isnanCharles Karney
If HAVE_C99_MATH, define pj_is_nan as isnan.
2018-03-23If HAVE_C99_MATH, define pj_is_nan as isnan.Charles Karney
Extend HAVE_C99_MATH checks in CMakeLists.txt and configure.ac to include test for C99 function isnan.
2018-03-23Undo mistaken push to master.Charles Karney
2018-03-23If HAVE_C99_MATH, then pj_is_nan is defined as isnan.Charles Karney
This required extending the HAVE_C99_MATH checks in CMakeLists.txt and configure.ac to include a test for the C99 function isnan.
2018-03-22Merge pull request #889 from mwtoews/typosEven Rouault
Correct typo: commiter -> committer
2018-03-23Correct typo: commiter -> committerMike Toews
2018-03-22Merge pull request #887 from schwehr/float-cast-overflowKristian Evers
Handle nan float cast overflow in PJ_robin.c and nad_intr.c
2018-03-22Handle nan float cast overflow in PJ_robin.c and nad_intr.cKurt Schwehr
Uses the new pj_is_nan to avoid x == x checks. Removes an assignment from an arg list
2018-03-22Add a simple implementation of pj_is_nan() to proj_internal.hKurt Schwehr
2018-03-22Merge pull request #886 from schwehr/log-levelKristian Evers
3 -> PJ_LOG_TRACE for pj_log level in pj_init.c
2018-03-22proj.h: update PROJ_VERSION_MAJOR.PROJ_VERSION_MINOR to 5.1Even Rouault
This will help avoiding issues for people tracking proj master and GDAL See https://lists.osgeo.org/pipermail/gdal-dev/2018-March/048285.html
2018-03-213 -> PJ_LOG_TRACE for pj_log level in pj_init.cKurt Schwehr
Bare literals are hard to follow and inconsisten with the rest of proj.
2018-03-21Update NEWS for 5.0.1Kristian Evers
2018-03-21Merge pull request #882 from rouault/fix_881Kristian Evers
Take into account +towgs84=0,0,0 in pipeline to still imply geodetic->cartesian->geodetic (fixes #881)
2018-03-21Take into account +towgs84=0,0,0 in pipeline to still imply ↵Even Rouault
geodetic->cartesian->geodetic (fixes #881)
2018-03-21Merge pull request #883 from schwehr/ctable-nullKristian Evers
Handle the case where nad_ctable2_init returns NULL.
2018-03-21Merge pull request #880 from mwtoews/rename-varKristian Evers
Avoid shadowed declaration errors with old gcc
2018-03-21Merge branch 'master' into ctable-nullKristian Evers
2018-03-21Merge pull request #878 from schwehr/logging-cleanupKristian Evers
Remove final \n from pj_log messages.
2018-03-20Handle the case where nad_ctable2_init returns NULL.Kurt Schwehr
Make CTABLE V2 handling the same as CTABLE. The pj_log call hit a NULL pointer during fuzzing. Found with autofuzz.
2018-03-21Avoid shadowed declaration errors with old gccMike Toews
2018-03-21Merge pull request #879 from mwtoews/tidyEven Rouault
Update .gitignore for newer binaries
2018-03-21Update .gitignore for newer binariesMike Toews
2018-03-21Add extra { } to make code clearerEven Rouault
2018-03-20Remove final \n from pj_log messages.Kurt Schwehr
There was a mix of pj_log calls with and without a trailing newline. pj_stderr_logger already adds a newline to the end of each msg, so pj_log calls don't need their own.
2018-03-20Merge pull request #877 from kbevers/fix-datum-param-in-pipelinesKristian Evers
Introduce late expansion of +datum parameters.
2018-03-20Merge pull request #876 from schwehr/b74944176-pj_gridinfo_initKristian Evers
Track the header size read in pj_gridinfo_init
generated by cgit v1.2.3 (git 2.25.1) at 2026-02-28 10:41:02 +0000