* Rework java vl check in read/write. Handle old refs in h5dump
* Committing clang-format changes
* Java changes allow default VL reads
* Fix JNI utility for old refs
* HDFFV-11310 - implement vlen read/write for atomic types.
* format check fix
* Mostly format fixes
* More format issues
* Two format changes
* Use JNI names for sizeof
* format change
* fix size typo
* Change to older method to initialize list
* remove unused var
* format fix
* switch writeVL functions to use datatype instead of java type
* Add VL option to generic read/write check
* Correct function name typo
* Add JIRA issue
* Correct note to match change in code.
* HDFFV-11318 add VL references as byte arrays
* Add release note and format changes
* Another format update
* Fix unreleased allocations
* Fix format
* format correction
Co-authored-by: github-actions <41898282+github-actions[bot]@users.noreply.github.com>
* Fixed -Wreserved-id-macro warnings from header include guards
* Removed all __int64 and LL suffix stuff now that C99 is minimum requirement
* Rename `H5FD_CTL__` to `H5FD_CTL_` to fix -Wreserved-id-macro warnings
Double underscore is reserved in C++ and this public header should be C++ compatible.
* Never define __STDC_FORMAT_MACROS anymore
Defining it causes a -Wreserved-id-macro.
Happily, according to the C++11 standard:
"The macros defined by <stdint> are provided unconditionally. In particular, the symbols __STDC_LIMIT_MACROS and __STDC_CONSTANT_MACROS (mentioned in C99 footnotes 219, 220, and 222) play no role in C++."
https://cplusplus.github.io/LWG/issue984
So looks like it's not necessary to define it with reasonably new toolchains.
* Fixed some -Wunused-macros warnings, removed dead code
* Fixed all -Wdouble-promotion warnings in C++ files
* Fixed remaining -Wsuggest-destructor-override warnings
* Committing clang-format changes
Co-authored-by: github-actions <41898282+github-actions[bot]@users.noreply.github.com>
* Promote some GCC warnings back to errors and deal with build breakage.
I tested configuring and building with GCC 8.3.0 with
`.../configure --prefix=$HOME/wrk/install/manydsets --enable-build-mode=production --enable-symbols=-gdwarf-4 --enable-warnings-as-errors --enable-java --enable-hl --enable-fortran --enable-cxx --enable-mirror-vfd --enable-ros3-vfd --enable-direct-vfd`
and with the same options but `--enable-build-mode=debug`.
* Promote -Wrestrict to -Werror=restrict to help catch overlapping
arguments to strcpy, memcpy, et cetera, at compile time.
* NFCI: sort lines.
* Committing clang-format changes
* NFCI: remove whitespace at EOL.
* Re-use temporary variable `shared`.
* Remove merge oopsie.
* Mention that no op codes are understood per review comment.
* Change this back to the develop branch's way, since this change isn't
integral to the PR.
* Committing clang-format changes
* NFCI: reduce differences with `develop` branch in the hope that that
also kicks off CI.
Co-authored-by: github-actions <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: Dana Robinson <43805+derobins@users.noreply.github.com>
* Use the FAPL that was created earlier in the test (and delete an unused
variable). This allows 'make check-vfd' to pass with --enable-mirror-vfd.
* Check for testing directory before creating, to avoid warning from bash.
Clean out .libs directory before re-using it (after a failed test), to
remove any files generated by libtool.
* Committing clang-format changes
* Increment error count on failed file open and skip tests for VFDs that need
modified filenames.
* Skip the mirror VFD for 'make check-vfd' - the mirror VFD requires networking
configuration parameters and can't be provided for an automated test that
is configured with an environment variable.
Co-authored-by: github-actions <41898282+github-actions[bot]@users.noreply.github.com>
* Do not conditionally compile code that uses a pointer dereference
and assignment to copy a potentially unaligned variable to aligned
automatic storage, or vice versa. Instead, always use naked `memcpy(3)`s.
Disassembling the generated code reveals that the `memcpy(3)`s optimize
(`-O3`) to a single `mov` instruction for x86_64, which is not strict
about alignment.
This change reduces the size of code and scripts by 143 lines, eases
our way to cross-compilation, and avoids invoking undefined behavior.
* Committing clang-format changes
* Per discussion, use HD and add comments.
Co-authored-by: github-actions <41898282+github-actions[bot]@users.noreply.github.com>
* Updated plugins project required adjustments
* Updated plugins need to distinguish external libs
* Fix missed rename
* Use same name for windows as linux - avoid name clash
* H5Lexists docs: Removed reference to 1.8.16 since the change is the 1.8.x releases, HDFFV-11289
* H5Oget_info_by_name, name can be any object, not just a group
* Converted README.txt to README.md and updated files referring to README.txt to README.md.
* removed references to README.txt
* updated MANIFEST
Adds tool h5dwalk and configure options to enable building it.
Co-authored-by: Richard Warren <Richard.Warren@hdfgroup.org>
Co-authored-by: github-actions <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: Richard.Warren <richard.warren@jelly.ad.hdfgroup.org>
Co-authored-by: Larry Knox <lrknox@hdfgroup.org>
* Use stdout instead of file for configure check
* Make change requested by #1157
* Change fortran to use stderr for configure
* Correct typo
* remove obsolete file check
* Fortran statement fix
pointer or from pointer to integer (-Werror=int-conversion). Also,
make it an error if the library explicitly converts to pointer from an
integer of a different size (-Werror=int-to-pointer-cast).
* Enable usage of new-style references with dimension scale APIs
* Add API to check if an object ID represents a native connector object
* Modified code to use new function H5DSwith_new_ref to determine if new references should be used with
Dimension Scales. The new function return TRUE if non-native connector is used or if H5_DIMENSION_SCALES_WITH_NEW_REF
varible is define at configure time (--enable-dimension-scales-with-new-ref).
Tested on jelly.
ToDo: generate testing file on BE system and enable the test; add flag to CMake; test netCDF-4 with the new references.
* Adding new test files generated on BE system (hedgehog) created by 32 and 64-bit library.
test_ds chokes on test_ds_le_new_ref.h5 on BE system; test passes for test_ds_be_new_ref-32bit.h5
for the 32-bit library and fails for the 64-bit library, and vice versa. I am checking the files for further
investigation; but current implementation of the new references is not portable between LE and BE systems,
and 32 and 64-bit systems.
* Minor fixes for testing issues
* Update test_ds.c
Enabled broken test; tests pass now.
* Update RELEASE.txt
Documented new option to use new references with the HDF5 dimension scales APIs (H5DS*).
* Update MANIFEST for new 32-bit new-style references test file for H5DS APIs
* Update 'dimension scales w/ new-style refs' feature based on review
Co-authored-by: Elena <epourmal@hdfgroup.org>
* Move test utilities to utils/test folder
* Fix makefile assignment
* Add new dir
* add new folder
* Correct copied makefile
* Fix dir typo
* Add missing include dir
* Remove unnecessary lib link
* Correct dependent dirs
* Fix conditional checks
* Disable test if not built
* fix path to executable
* Use fixture for swmr_vfd check
* Add release note
* Correct shell tests and c++ flag warning
* Update autotools c++ warning
* Fix typo
* Avoid maintenance headaches: delete the `noerror-` variants of the
compiler flags files, since they essentially duplicate the `error-`
files modulo the replacement of `-Werror=` with `-W` and any changes in
comments. (I verified the duplication with a script.)
For autoconf, reinstate the use of the `demote_errors` shell function to
derive the `noerror-` content from the `error-` content. `demote_errors`
replaces `-Werror=` with `-W` when `WARNINGS_AS_ERRORS` is `no`.
Slightly reorder `configure.ac` so that the setting of
`WARNINGS_AS_ERRORS` takes effect before the `error-` files are sourced.
* Take a stab at updating the CMake files to match the changes I made to
the autoconf files to remove `noerror-` files. I'm not much of a CMake
user so these changes are quite rough.
Looks like the duplication can be reduced with the introduction of a new
macro.
* Delete `noerror-` files from the MANIFEST.
* Reduce duplication in the CMake files: perform the
HDF5_ENABLE_WARNINGS_AS_ERRORS test once in the ADD_H5_FLAGS macro.
* Add a release note.
* Implement support for loading of Virtual File Drivers as plugins
Fix plugin caching for VOL connector and VFD plugins
Fix plugin iteration to skip paths that can't be opened
* Enable dynamic loading of VFDs with HDF5_DRIVER environment variable
* Temporarily disable error reporting during H5F_open double file open
* Default to using HDstat in h5_get_file_size for unknown VFDs
* Use macros for some environment variables that HDF5 interprets
* Update "null" and "ctl testing" VFDs
