(not suited for this release Dave)
Changes committed to git://git.pcp.io/nathans/pcp.git master
Nathan Scott (3):
Revert "configure & python pmapi wrapper changes: structs timeval and
timespec"
python api: reduce earlier timeval fix to just problem case
qa: add a check for required perl threading module in qa/1031
config.mingw | 1
configure | 196 ---
configure.ac | 184 ---
qa/1031 | 3
src/include/pcp/config.h.in | 5
src/python/pcp/.gitignore | 1
src/python/pcp/pmapi.py | 23
src/python/pcp/pmapi.py.in | 2370 --------------------------------------------
src/python/pmapi.c | 5
9 files changed, 75 insertions(+), 2713 deletions(-)
commit 345b3b7084990ac977819eb3f1cf482b013fccb4
Author: Nathan Scott <nathans@xxxxxxxxxx>
Date: Fri Mar 18 10:37:44 2016 +1100
qa: add a check for required perl threading module in qa/1031
commit 638d82bfd3cf7f70533bcc4ac6a210703d7c6f9e
Author: Nathan Scott <nathans@xxxxxxxxxx>
Date: Fri Mar 18 10:16:39 2016 +1100
python api: reduce earlier timeval fix to just problem case
This is a simplified version of Kens earlier struct timeval
fix, which deals just with the problematic suseconds_t case.
This is handled more like the way differences in long/int/
pointer are handled now (in terms of configure - not python
specific), and more like the way the pmUnits structure is
conditionally setup (in terms of pmapi.py).
While researching alternative approaches, I came across the
base ctypes support for int32/uint32/int64/uint64 which we
should be using in pmAtomValue instead of guessing - so that
trivial type rectification is made here as well.
commit 4fe1f097250d4b092f559ac7083b378972a87ecd
Author: Nathan Scott <nathans@xxxxxxxxxx>
Date: Fri Mar 18 08:48:48 2016 +1100
Revert "configure & python pmapi wrapper changes: structs timeval and
timespec"
This reverts commit 24de23a2ee39dc94f66699e8323642afaf1dbe78.
|