fmt 12.1.0 calls std::isalpha in locale_ref's constructor only to check
that it is found via ADL, and discards the result. MSVC's standard
library marks the function [[nodiscard]], so compiling fmt's format.cc
gives C4834. Disable that warning for the fetched fmt target only.
The decode tests build uint64_t input words with 1UL << bit. MSVC's
unsigned long is 32 bits, so with a runtime bit index that is a 32-bit
shift widened afterwards, which it reports as C4334 and which would
silently drop any bit above 31. Use 1ULL for the two shifts whose
amount is not a compile-time constant.
read_header rewinds past the header with seek(-header_size), but
header_size is a size_t, so the minus wrapped around and only became
-16 again in the conversion to ssize_t. MSVC flags the unsigned
negation as C4146. Cast to ssize_t before negating; the offset passed
to seek is the same.
Thirteen catch clauses named the exception without using it. gcc and
clang accept that silently, but MSVC reports each one as C4101. Drop
the name; the handlers are unchanged.
The pybind11 bindings wrapped their definitions in a GCC diagnostic
pragma that ignored -Wunused-parameter. MSVC does not understand the
pragma, so the same parameters showed up there as 17 C4100 warnings.
Mark the parameters [[maybe_unused]], which every compiler honours,
and drop the pragma from all ten headers; four of them had no unused
parameter left.
Collection of fixes to make aare build on Windows. Requires the previous
PRs for compilation to succeed.
- Replaced `or` keyword with ||
- updated cmake file
- fix for 32 bit long in fseek
closes #395
closes #396
logger.hpp pulled in <sys/time.h> and used gettimeofday and localtime_r,
which do not exist on MSVC. Build the timestamp from std::chrono instead
and use localtime_s on MSVC. The output format is unchanged: <colour>-
HH:MM:SS.mmm LEVEL: message<reset>. The unused __SHORT_FORM_OF_FILE__
helper now also strips backslash separators. Add logger tests that pin
the timestamp and line format.
closes #394
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
On Windows std::filesystem::path holds wchar_t, so assigning a path
component to a std::string or passing path::c_str() to fopen does not
compile. Use path::string() at those sites, and drop a redundant path
round trip in NumpyFile.
closes #399
- MSVC has no ssize_t, which made every translation unit fail on
Windows.
- Define aare::ssize_t in defs.hpp: the platform ssize_t on POSIX (so
nothing changes there) and std::ptrdiff_t on MSVC.
- Code inside the namespace picks it up unchanged; the few files that
use ssize_t outside the namespace get a using-declaration. FilePtr.hpp
now includes defs.hpp since it had no aare include.
Closing #393
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
## Summary
Makes the minimizer behind `FitModel` selectable and adds two built-in,
dependency-free solvers next to Minuit2's Migrad and Fumili.
`Minimizer.Migrad` stays the default, so existing fits behave as before.
- **`Minimizer` enum** (C++ `FitModel::SetMinimizer`, Python `minimizer`
constructor argument and property): `Migrad`, `Fumili`,
`LevenbergMarquardt`, `VarPro`.
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
Co-authored-by: Khalil Ferjaoui <khalilferjaoui@yahoo.fr>
- For set_nSigma only call update_pedestal if we have accumulated enough
samples otherwise just write it
- For update threshold check if pd is ready and if not throw
closes #390
- Introduce a standard way to bind numpy arrays: py::array_t<T> and
.noconvert() on the argument
- Take py::array_t by value since it's just a small wrapper
- Update bindings
- Removed faulty move (pointed out by: #227)
- Fixed license for the queue
- Runtime check for capacity=>2
Thank you @kferjaoui for finding the issue. In the end I found it better
to remove the operators rather than modifying the queue
Cleaned up the helper functions by:
- Added template for make_view (and new make_const_view) instead of the
copy paste implementation
- checks c_style and dimensions inside make_view
- test binding to be able to probe the checks in make_view
Call sites can be reviewed in a second PR since usage of const or
strictness of conversion depends...
We have no code in aare to deal with RawFiles that have missing packets
if they are not padded to full size or frames with packet loss dropped.
Added:
- check that we don't try to read files that are not supported
- python bindings for frame discard policy
at least we get the check in:
closes #171
---------
Co-authored-by: AliceMazzoleni99 <alice.mazzoleni@psi.ch>
- error upon warning when building documentation (warnings often
indicate malrendered documentation)
- fixed some warnings
Co-authored-by: Erik Fröjdh <erik.frojdh@psi.ch>
- seek to the edge of a file + read_into would select the wrong file
- Visible for example in f.read_frame(index) in python
Added fix, release notes and a small test.
closes #371
Co-authored-by: AliceMazzoleni99 <alice.mazzoleni@psi.ch>
- renamed pedestal.hpp to bind_Pedestal.hpp for python bindings
- variance is now always calculated in double and made private
- cost correctness in a few places
- removed push_fast and explicit updates of mean from Pedestal. If
performance is needed use FastPedestal
- Pedestal no longer caches std
Improved API for ClusterFile
- read_frame now returns a std::optional<ClusterVector> to differentiate
between an empty frame and end of file
- f.frames() produces an iterator over frames
- f.chunks() produces an iterator reading chunks of clusters
- python string repr for Cluster and Eta
- Context manager for RawMasterFile
- slice operator for ROI
---------
Co-authored-by: Erik Fröjdh <erik.frojdh@psi.ch>
- Updated error message to mention out of range and which file it
applies to
- If there are different number of frames in the subfiles, take the
minimum and print a warning
Closes: #363
Closes: #362
- fixed test in ClusterFinderMT (use minimum one pedestal to accumulate
dummy pedestal)
- added minimum_pedestal_samples as additional constructor argument in
ClusterFinder and ClusterFinderMT
- bugfix: race conditions for empty queues in collect() - check directly
when needed
- Default `ClusterVector` move operations simplifying the code and
fixing a bug.
- Fixed inconsistent frame number type (uint64/int32)
- Validate Python masks as one-dimensional, C-contiguous Boolean arrays,
handle empty masks safely, and reserve filtered storage based on the
selected cluster count.
- Align the C++ and Python API documentation with the implementation,
including concise `hitmap` and reduction documentation and a correctly
rendered constructor example.
- handles file reading of disabled udp ports
- adds members disabled_udp_ports in master file as an optional
- adds member diasbled_udp_port_types in master file as an optional
- Treats enabled udp ports as ROIs
- merges/combines consecutive ROis into one
### To discuss:
- right now can only handle either ROI or disabled udp ports (disabled
udp ports has higher precedence)
- Should Frames in File be removed from Master File as value is nonsense
for disabled udp ports
---------
Co-authored-by: Erik Fröjdh <erik.frojdh@psi.ch>
Optimized pedestal tracking with FastPedestal and performance improvements to the cluster finder.
- ~2x faster hitting 14k FPS on benchmark dataset on 16 core threadripper pro
**Changes**
- No bounds check on interior of frame. (only within half of a cluster)
(~15% improvement)
- Cache threshold (std*n_rms) Biggest improvement comes from not
computing std for each access
- Cache pedestal subtracted frame
- Switched to FastPedestal which assumes we reached steady state
**Things tried but rejected:**
- Always store cluster values, commit on val==max (~15% drop in frame
rate)
**Options**
- using 16 bit clusters and 16 bit pedestal. (slows down the single
threaded case but allow us to reach 12k with multi threading)
**Other notes on performance**
- Passing in memory frames from python and doing nothing: 0.8us/frame
- Passing + pedestal subtraction: 28us/frame
- Passing + pedestal + cluster finding (no store): 652us/frame
- Passing + pedestal + cluster finding: 886 us/frame
Minor API change:
- NDView<T> implicitly converts to NDView<const T> (as std::span in
C++20)
- expand4to8bit and expand24to32bit now takes a view over const uint8_t
to show that the input is not changes
closes #291
Before v8.0 we should always try to read analog samples. This bug
resulted in old Mönch03 data not being reordered.
1. Added test for master file parsing
2. Fixed code
closing #348
- Reading files from nfs shares tops out at ~1GB/s for single threaded
reads.
- MultiThreadedFileReader uses our File wrapper to read a generic file
in parallel
- Placed in aare::experimental to show that it's not production ready
Tomas Celko
