Have HDF5 write raise error if operator(s) requested #3951
Conversation
|
Thanks. I have to think about this. I am not sure if every user would want the code abort with an exception rather than quietly ignore the operator. On the other hand, as far as I know, you are the one and only user of the HDF5 engine. |
|
I would argue that silently not doing something that the docs explicitly saw will be done is worse that giving an error. |
rupertnash
force-pushed
the
hdf5-operator-error
branch
from
179e110
to
b5d3718
Compare
|
Hopefully made clang format happy now! |
We might think about a one-time rank-zero warning printout rather than an exception. (@anagainaru If we had a facility for this, we might also use it for warning about the overheads of reader-side derived variable computation.) |
|
Could there be an option to make warnings into errors (similar to |
Yes, I think this would be a good idea. I don't think we have the option now but we should look into this in the future. I am strongly in favor of giving at least some warning if you set a parameter that will be ignored. |
Taking this literally, we have long thought it might be useful to warn if you set an engine parameter that is never consumed (because you might have just misspelled it). But our disaggregated method of looking at parameters makes this hard. This case of ignoring an operator is at least easier to detect. But norbert's got a point too... |
rupertnash
force-pushed
the
hdf5-operator-error
branch
from
b5d3718
to
05be485
Compare
rupertnash
force-pushed
the
hdf5-operator-error
branch
from
05be485
to
8350f30
Compare
* master: Have HDF5 write raise error if operator(s) requested (ornladios#3951) fix for ASAN issue related to JoinedDimArray handling in BP5 deserializer (ornladios#3963) New operator MDR, for refactoring floating point arrays using MGARD's new MDR extension. (ornladios#3826) restricted http transport from windows builds. XMLConfigTest: Add RemoveIO test adios2::core::ADIOS: Initialize new IO objects with config file removed unsused variable Update readme for heat transfer example with new location and build instructions Ignore tests with defects for now Adapt libfabric dataplane of SST to Cray CXI provider (ornladios#3672) ci: fix path to lsan suppressions, fix broken gh status post Use adios2_mode_readRandomAccess in matlab open to make it work for BP5 (ornladios#3956) Add Global Array Capabilities and Limitations Add Section for Anatomy of an ADIOS Program Enable Shell-Check for gh-actions scripts Enable Shell-Check for circle CI scripts Enable Shell-Check for tau contract scripts Enable Shell-Check for scorpio contract scripts Enable Shell-Check for lammps contract scripts Delete VTK code in examples Fix MATLAB bindings for MacOS (ornladios#3950) Set the compiler for the Kokkos DataMan example to what is used to build Kokkos Fix the HIP architecture CMAKE variable (ornladios#3931) perfstubs 2023-11-27 (845d0702) (ornladios#3944) Revert "Only rank 0 should print the initialization message in perfstub" Formatting Formatting Revision Added buffered data receive in the client side. A socket version of HTTP connector. Proxy server host is hardwired to "localhost" and port to 9999 Remote bpls: bpls -E bp4 -T "Library=HTTP" /remote_path/myVector_cpp.bp -d bpInts
Discussion in #3942 suggests that support for operators in "secondary" (for want of a better term) engines is not planned. My suggestion that ADIOS should let the user know this won't happen is added in this PR.