The hardware and bandwidth for this mirror is donated by dogado GmbH, the Webhosting and Full Service-Cloud Provider. Check out our Wordpress Tutorial.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]dogado.de.
Several changes have been made to reduce the package size from over 20 MB to around 5 MB:
BioCro.Rmd) that simply
redirects readers to the documentation websiteAddressed a missing-field-initializers warning from
the compiler by explicitly setting iterations to 0 in the
output from rue_leaf_photosynthesis
Addressed a mistake in
thermal_time_and_frost_senescence.h where the leaf death
rate due to frost had been unintentionally set to 0 in all conditions.
This mistake was caught by a compiler that reported a “ignoring return
value of function declared with ‘nodiscard’ attribute” warning.
Changed the minimum version of macOS checked by the R-CMD-check from 3.6.0 to 4.2.0.
Another bug was corrected in
src/module_library/c3photoC.cpp: The photorespiration value
Rp is now calculated using the value of Ci
from the current loop iteration (rather than the previous loop
iteration).
Modified some testing-related functions so that warnings due to
mismatched framework versions do not trigger test failures: the
tryCatch call in test_module now only catches
errors (not warnings) when evaluating the module, and
test_plant_model (in
crop_model_testing_helper_functions.R) now uses
expect_no_error instead of
expect_silent.
Changed the energy_par_content constant to 0.219. This is a conversion rate from photon flux density (in micromoles per square meter per second) to energy flux density (in joules per square meter per second or watts per square meter) for photosynthetically active radiation (PAR). It equals 1/4.57, 4.57 being a commonly used constant to convert PAR in W m^-2 to micromole m^-2 s^-1. Users should take care to ensure that if processing of radiation data is required to prepare it for use with BioCro, the same conversion factor is used. See more details in Plant Growth Chamber Handbook. CHAPTER 1 – RADIATION– John C. Sager and J. Craig McFarlane. Table 2, Pg 3 (https://www.controlledenvironments.org/wp-content/uploads/sites/6/2017/06/Ch01.pdf)
The C++ framework has been updated to v1.1.3. Since the framework
is included as a git submodule, it will be necessary to use the
--recurse-submodule flag when using git pull,
git checkout, or git switch to update a local
copy of the BioCro repository, or to move to or from this
branch.
Replaced the inc/boost directory with a submodule
pointing to the new biocro/boost repository.
The (unexported) lightME function has been removed
from the R package, since its functionality can be reproduced using the
BioCro:solar_position_michalsky and
BioCro:shortwave_atmospheric_scattering modules.
All instances of fabs or unqualified
abs have been replaced by std::abs. The use of
unqualified abs in
src/module_library/c3photoC.cpp had been causing test
failures when running BioCro on Windows using R version 3.6.0.
This version adds a description of the BioCro git branching model
to contribution_guidelines.Rmd and clarifies the process of
updating NEWS.md.
The R-CMD-check workflow has been changed in the following ways:
When the check workflow is run manually, there are two new input options:
The user can now choose whether or not to run R CMD check with the –as-cran option. Formerly, this was always used.
The user can choose whether and when to throw an error on R CMD check failures. Formerly, an error was thrown whenever the R CMD check failure was either “warning” or “error”.
Output that was formerly shown only on manual runs when the “debug” checkbox was selected is now always shown. The “debug” option has been changed to “dry-run”, which results in the debug output being shown but the actual check, and those set-up steps needed only to carry out the check, are skipped.
The debug output steps are grouped together when possible, and the output is shown earlier on in the workflow.
The R-CMD-check workflow has been modified to work around a problem with testing R version 3.6.0 on Windows. And for all platforms, we now specify the tested minimum R version as 3.6.0 rather than simply 3.6 in order to ensure that we are actually testing the minimum required R version specified in the DESCRIPTION file, rather than some later 3.6.x version such as version 3.6.3.
Modified the R-CMD-check workflow so that the manual is not checked when the workflow runs automatically. This has also been made the default when the workflow is run manually.
GitHub workflows and actions in the repository have been updated to use the latest versions of all GitHub and 3rd-party actions.
Updates related to changing the GitHub hosting organization from “ebimodeling” to “biocro”:
Most references to the ebimodeling GitHub organization have been removed; references to ebimodeling/biocro have been updated to point to biocro/biocro instead. Most of these occurred in various places in the documentation. Most links to the online documentation have been replaced with links to https://biocro.github.io, with (in some cases) instructions on how to navigate to the particular section of the documentation of interest. This decreases dependence on the precise layout of the online documentation. Some other changes and clarifications to the documentation have been made as well.
Addressed some format-security compiler warnings
related to calling Rf_error and Rprintf
without a format specifier; a format specifier of "%s"
should always be used when printing the value of a string
variable.
biocro/BioCro-documentation repository, and it changes the
triggers of the workflow so that automatic publication happens when a
new release is published. Additionally, a symlink is created to link the
URL https://biocro.github.io/BioCro-documentation/latest/pkgdown/ to
https://biocro.github.io/BioCro-documentation/This version introduces the concept of distinct module libraries,
allowing users to develop modules in private and to create collections
of related modules. There is an associated syntax change, where modules
must now be specified using fully-qualified names that include
a module library name and the local name of a module within that
library; for example, the module that was previously known as
thermal_time_linear must now be referred to as
BioCro:thermal_time_linear.
Any R package representing a BioCro module library must now have
four non-exported functions related to accessing its modules:
get_all_modules, get_all_quantities,
module_creators, and framework_version. When a
fully-qualified module name such as
library_name:local_module_name is passed to a function such
as module_info, an internal call to
library_name:::module_creators(module_name) will be made to
retrieve a pointer to a module. Hence, library_name must be
the same as the module library package name. This is a required part of
a method for passing C objects from a module library to the core BioCro
framework via R; the full details are not discussed here.
The code in the src directory has been reorganized
to reflect the division between framework code, module code, and
“R-to-C” code discussed in the manuscript; now,
src/framework contains the core C++ code,
src/module_library contains the module code, and the
“R-to-C” code can be found directly in src.
The C++ framework code in src/framework has been
moved to a separate repository and licensed under the GNU LGPL; it is
included in the BioCro R package repository as a Git submodule. This
allows us to use a permissive license for the BioCro R package while
still protecting the code that assembles and solves models. Associated
with this change, the BioCro R package is now licensed under the MIT
license. See LICENSE.note for details.
The soil_type_selector module has been removed and
replaced with a data object called soil_parameters. For
crop models that previously set the soil_type_indicator to
6 to choose the soil parameter values, this parameter has
been replaced with the values from
soil_parameters$clay_loam.
Crop model definitions are now stored as single lists rather than
as multiple objects; for example, soybean_parameters and
soybean_initial_values are now stored as two elements of
the soybean list: soybean$parameters and
soybean$initial_values.
The function returned by partial_run_biocro can now
properly respond to vectors and lists of named elements.
Module testing functions have been added to the package namespace.
Many small improvements have been made to the documentation and
the module code in src/module_library; these changes are
too numerous to list here.
The elements of the arg_names input to
partial_run_biocro can now be in any order; previously,
they were required to be supplied in the same order as the appear in the
other inputs to partial_run_biocro, for example, the names
of any initial values were required to come before the names of any
parameters. If the arguments were supplied in the wrong order, then the
inputs to the function returned by partial_run_biocro would
be interpreted in the wrong order.
A bug that sometimes caused the last time point of a simulation
to contain NaN for all quantities has been fixed; this was related to an
out-of-bounds error when accessing vector elements in the C++ function
dynamical_system::update_drivers.
These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.
Health stats visible at Monitor.