Now variables are passed as props through vue so only one json update is
made by the full_table vue component

This is to solve the problem of fetching the monitorables in every
instance of a sub table. The full table will grab the data and
distribute it to the sub-tables

Now common functions and values that are needed for the tables are
imported.

Added vuejs and component js code into the FEDMonitorWeb code and then
turned DAQ, TTC, and Trigger Rate tables into vuejs components

The version of Vue is the lastest stable at time of commit. The trailing
whitespace had to be removed so it would pass the CI

The files were downloaded from:
https://unpkg.com/vue@3.2.37/dist/vue.global.js
https://github.com/FranckFreiburger/vue3-sfc-loader/releases/tag/v0.8.4

Miscellenaous bug-fixes from p5 deployment

See merge request !294

The DCS exporter was looking for monitorables into the JSON object root
instead than in each of the monitorable groups. Look in all groups until
there is a match instead.

While trying to deploy in production the layout tree from repository,
multiples inconsistencies arose. Fix them.

* Use in CMS the same convention between GE1/1 and GE2/1 for the display
  and wire names.

* Fix the GE2/1 YAML filenames. They should be based on the UID.

* Use the same name between the "parts" and the layout tree root
  directory. This is meant to simplify the deployment.

Recent updates of the DAQ at p5 show some glitches to reset the MGT,
usually just after the updates. Make the MGT reset more rugged and add
debugging information.

In theory, the run number is sent during the Start transition and
well-defined only during the Running state.

However, the current workflow imposes the GEMSupervisor to generate a
new local run number for every configuration during configuration. This
is required to provide the the calibration scans a pristine run number.

As a consequence, the GEM function manager must set the isLocalRunNumber
parameter of the GEMSupervisor prior to triggering the Configure state
transition.

Although the AnyState to Error transition was defined, the SetError
input and Error state were not included in the FSM description, leading to
a malformed state machine. Include all required states, inputs, and
state transtions.

During the JSON object key renaming of 22f39fb7, the FEDMonitor
JavaScript update script was forgotten. Consequently, the state value
was never updated in the FEDMonitor Expert Page tab, despite the state
being well defined. Fix the issue and get the current state from the
'fsm-state' key (instead of 'value').

Introduce local run numbers

Closes #221 and #217

See merge request !293

The current workflow imposes to generate a new local run number both at
'Configure' and 'Start'. The number generated at 'Configure' is
necessary for the calibration scans; the number generated at 'Start' is
necessary for the normal runs.

Allow different update times in the monitoring suite (AMCMonitor)

Closes #175

See merge request !158

Dylan Oliver Teague authored 3 years ago and Laurent Petre committed 2 years ago

Co-authored-by: Laurent Pétré <laurent.petre@cern.ch>

Misc toolchains updates (a.k.a. C++17 on the CTP7)

Closes #242 and #64

See merge request !287

Laurent Petre authored 2 years ago and Camilla Galloni committed 2 years ago

* {fmt} from 7.1.3 to 9.1.0

  For the new features and the improved documentation.

* libpqxx from 7.2.1 to 7.7.4

  For the new features and a ton of bugfixes. Not that it may be the
  last usable branch in cmsgemos before some time. Version 8.0.0 might
  require C++20.

* pybind11 from 2.6.2 to 2.7.1

  While more recent versions have been published (2.8.x, 2.9.x, and
  2.10.x), this is the last one that seems compatible with the CTP7
  target. Starting from version 2.8.x, any usage of gempy fails with an
  assertion within Python. [0] That version introduced changes in the
  thread-local storage handling which is clearly related to the
  assertion failure. Considering that starting from version 2.10.0,
  Python 2.7 is not supported by the pybind11 project anymore, chances
  of bugfixes are small.

  Anyway, update to benefit from nice new helpers and performance
  optimizations as well as bugfixes. We should really move on with the
  Python version!

  [0] Python-2.7.9/Python/thread.c:346: PyThread_set_key_value:
      Assertion `value != ((void *)0)' failed.

At the time of writing this commit, all external dependencies are now
updated to their last stable compatible version.

Laurent Petre authored 2 years ago and Camilla Galloni committed 2 years ago

libdim.so provided by the xDAQ packages does not have a SONAME. Mark it
as such in FindxDAQ.cmake for clean linking.

Laurent Petre authored 2 years ago and Camilla Galloni committed 2 years ago

The log4cplus dependency was accidentally built has a shared library.
It was actually setting the BUILD_SHARED_LIBS cache variable for the
whole CMake project which could have had an impact in the cmsgemos core
libraries.

In order to avoid such situations in the future, preset the global
variable BUILD_SHARED_LIBS to OFF. Would an adventurous developer play
with the option, it would be his responsibility.

Some binaries still require to be built as shared object. Those are
manually specified it on a library-per-library basis.

If conflicts between the FetchContent projects and the main project
continue to be seen, the ExternalProject option may be envisioned.

Laurent Petre authored 2 years ago and Camilla Galloni committed 2 years ago

This change aims at improving the compilatin speed. Notice that the
deployment is not affected due to the usage of static linking.

Laurent Petre authored 2 years ago and Camilla Galloni committed 2 years ago

The FetchContent calls have always been a weak point of the build
system. There are now many external dependencies so improveme this
weakness.

1. Perform all actions in Build*.cmake via a function. This prevent
   variables leaks throught the creation of a new scope.

2. Build on the targets we need, not all of them (i.e. add the
   EXCLUDE_FROM_ALL property on all external dependencies source
   directories).

Laurent Petre authored 2 years ago and Camilla Galloni committed 2 years ago

The register permissions (RO, WO, RW) stored in LMDB were corrupted and
were containing the register mode (single/block). The register
permission checks were also incorrect, leading to the bug being unoticed
and leaving the permissions not actually checked. Fix both bugs.

Found thanks to the warnings raised by Clang++ 14 in gemharware. Other
warnings have been fixed as well.

Laurent Petre authored 2 years ago and Camilla Galloni committed 2 years ago

Finally!

Please note that the CTP7 sysroot has been updated. An update of the
CTP7 libs on the back-end board is also required.

Care must be taken whenever introducing new C++ dependencies since those
have to be built against the recent libc++.

Laurent Petre authored 2 years ago and Camilla Galloni committed 2 years ago

Since the implementation of the templated RPC method calls in xHAL, the
Protobuf layer in gemrpc was essentially existing only for legacy
reasons. Facing the Protobuf cross-compilation complexity (the code
generator version must match the runtime library version) and the need
to support more targets, the decision was taken remove this useless
layer.

This commit replaces the key based messages by binary messages. Cereal,
already used for serialization of the RPC method arguments and return
values, is used to serialize defined message types.

Only the minimal amount of changes have been implemented in order to get
rid of the Protobuf dependency. Additional clean-up and changes may be
desirable in the future.

Laurent Petre authored 2 years ago and Camilla Galloni committed 2 years ago

Until now, Xerces-C++ was always assumed installed on the system (or
sysroot). This is an error prone assumption with the multiple supported
targets. Use the CMake facilities to find the package and fail as early
as possible if missing.