Do not download current register from the URL on CRAN, in order to
avoid timeouts, which are suspected to cause the check failure on one of
the CRAN test systems of version 2.0.2
Avoid length one LHS argument for case_when in
R/use-rates.R, as this is deprecated in dplyr 1.2.0
version 2.0.2
Include the latest XML file published the registration authorities
instead of the file from December 2025, because in the current version
the information to correctly read the culture tree is included
again.
resolve_cultures(): The warning that was issued when
the function was used with the new format was removed, as everything
seems to be working fine again.
version 2.0.1 (only
published on github)
Get the substance type (i.e. ACTIVE_INGREDIENT, ADDITIVE_TO_DECLARE,
SAFENER, SYNERGIST) of ingredients also from XML files in the new
format. Thanks to Alexandre Gurba (FOEN) for reporting the problem.
version 2.0.0
Adapt to the new format of the XML file. The changes are explained
at
https://www.blv.admin.ch/blv/de/home/zulassung-pflanzenschutzmittel/zugelassene-pflanzenschutzmittel/pflanzenschutzmittelverzeichnis.html
Due to UUIDs being used for cross-referencing instead of integer
identifiers, the corresponding identifier columns in several tables in
the srppp_dm objects have changed their type from integer
to character.
The culture tree constructed from the new version of the XML has
changed, among other changes the cultures allg. Obstbau,
allg. Feldbau, allg. Gemüsebau and similar are
not defined any more as parent cultures of any of the cultures listed.
The functionality of resolve_cultures() and
alternative_products() is not yet adapted to this new
situation and will give a warning when working with a register read in
from the new format.
version 1.1.0 (CRAN
release: 2025-07-07)
Example code and vignettes: To satisfy CRAN policy, fall back to use
the register version distributed as test data with the package (see
below) in case downloading or reading from the standard URL fails,
instead of throwing an error. This makes the check process on CRAN
robust against server outage or a changed URL.
Include information on the completeness of the effect against each
target organism in the column type of the
pests table of the srppp_dm object. Possible
types are PEST_FULL_EFFECT,
PEST_PARTIAL_EFFECT and PEST_SIDE_EFFECT. In
the obligations table, include a column
varying_effect, that contains TRUE in case the
obligation text indicates that a full effect has not been demonstrated
in all cases.
R/resolve_cultures.R: Function to expand a dataframe
with culture names at the lowest level (leafs) of a culture tree
constructed by srppp_dm.
Add a culture tree as a Node object from the
data.tree package as attribute to srppp_dm
objects and show it in the main vignette.
Remove the vignette intended for JOSS, as it was rejected there
Remove duplicate ingredient entries for the same substance in the
same product (issue #6)
inst/testdata/: Add the zipped XML from 16 December
2024 for testing purposes and test the resulting srppp_dm
object for referential integrity
srppp_dm(): Add argument verbose with
default TRUE.
version 1.0.1 (CRAN
release: 2024-11-01)
Trim leading and trailing whitespace from descriptions that are read
in using the internal function get_descriptions(),
addressing an issue Elisabeth raised in srppphist
Remove html documentation built with pkgdown from the git
repository, the online documentation is built using the github
workflow
Address issues found by the package review on CRAN as follows:
Rename the package to make it clear that its purpose is to read in
data from the Swiss Register of Plant Protection Products
Wrap example code in to avoid CRAN notes for checks >5s. The code
is still tested in the testthat tests.
Document return values of all exported functions
version 1.0.0
Risk mitigation measures sw_drift_dist,
sw_runoff_dist, sw_runoff_points and
biotope_drift_dist are now stored as integers in the
obligations table.
In the table cultures, the primary keys of up to two
parent cultures are now included.
Fix the use rate calculations for liquid products for the case that
only a dosage in percent is given.
version 0.99.2
R/srppp-xml.R: Revise the structure of the srppp_dm
object, after verifying that the product sections of different products
with the same P-Number are identical, with the exception of the
permission holder. Therefore, all tables describing the products,
including the use definitions, are now tied to the pNbrs
table with the pNbr as the primary key, instead of the
products table which has the wNbr as a primary
key. Functions, example code, vignettes and tests were adapted
accordingly.
version 0.99.1
vignettes/srppp.rmd: Add an overview vignette which is displayed
with the link ‘Get started’ in the online documentation
version 0.3.4
Rename the package from ‘psmv’ to ‘srppp’
version 0.3.3
Format changes
Remove redundant information from the ingredients table by removing
all W-Numbers containing a dash. P-Numbers were added to the ingredient
table as well, so product compositions can more easily be obtained using
the ingredients table. Finally remove W-Numbers from the ingredients
table. As a consequence, products and ingredients must now be joined by
‘pNbr’, and the relationship is ‘many-to-many’, as a ‘pNbr’ can occur
more than once in the products table.
The grouping of the products table by P-Numbers was removed, as it
seemed not to be used anywhere and created spurious messages during
constraint checking.