Merge pull request #156 from isoverse/dev

update to version 1.3.0

DESCRIPTION

@@ -1,7 +1,7 @@
 Package: isoreader
 Title: Read Stable Isotope Data Files
 Description: Interface to the raw data file formats commonly encountered in scientific disciplines that make use of stable isotopes.
-Version: 1.2.8
+Version: 1.3.0
 Authors@R: 
     c(person(
         given = "Sebastian", family = "Kopf", 
@@ -41,15 +41,15 @@ Imports:
     lubridate (>= 1.7.9.2),
     readr (>= 1.4.0),
     progress (>= 1.2.2),
-    xml2 (>= 1.3.1),
-    rhdf5 (>= 2.0.0),
     UNF (>= 2.0.6)
 Suggests:
     devtools,
     testthat,
     feather (>= 0.3.5),
     readxl (>= 1.3.1),
     openxlsx (>= 4.1.5),
+    xml2 (>= 1.3.1),
+    rhdf5 (>= 2.0.0),
     knitr,
     rmarkdown,
     covr

Makefile

@@ -7,13 +7,13 @@ all: docu check
 docu:
 	Rscript -e "devtools::document(roclets=c('rd', 'collate', 'namespace'))"
 
+# test package functionality without all example files (= as if on CRAN)
 check:
 	R -q -e "devtools::check(env_vars = c())"
 
-# test package functionality without all example files
 auto_test:
-	R -q -e "rm(list = ls()); options("isoreader.run_file_tests" = FALSE); testthat::auto_test_package()"
+	R -q -e "rm(list = ls()); Sys.setenv(NOT_CRAN = \"false\"); testthat::auto_test_package()"
 
-# test all example files
+# test with all example files (= as if not on CRAN)
 auto_test_all:
 	R -q -e "rm(list = ls()); testthat::auto_test_package()"

NAMESPACE

@@ -224,10 +224,6 @@ importFrom(readr,parse_logical)
 importFrom(readr,parse_number)
 importFrom(readr,problems)
 importFrom(readr,stop_for_problems)
-importFrom(rhdf5,H5close)
-importFrom(rhdf5,h5ls)
-importFrom(rhdf5,h5read)
-importFrom(rhdf5,h5readAttributes)
 importFrom(rlang,"!!!")
 importFrom(rlang,"!!")
 importFrom(rlang,":=")
@@ -295,9 +291,3 @@ importFrom(vctrs,vec_ptype2.double)
 importFrom(vctrs,vec_ptype2.integer)
 importFrom(vctrs,vec_ptype_abbr)
 importFrom(vctrs,vec_ptype_full)
-importFrom(xml2,as_list)
-importFrom(xml2,read_xml)
-importFrom(xml2,xml_child)
-importFrom(xml2,xml_children)
-importFrom(xml2,xml_find_all)
-importFrom(xml2,xml_text)

R/isoread_flow_iarc.R

@@ -7,6 +7,24 @@ iso_read_flow_iarc <- function(ds, options = list()) {
   if(!iso_is_file(ds) || !is(ds, "continuous_flow")) 
     stop("data structure must be a 'continuous_flow' iso_file", call. = FALSE)
 
+  # check for availability of xml2
+  if (!requireNamespace("xml2", quietly = TRUE)) {
+    stop(
+      "'xml2' package is required to read .iarc, please run: install.packages('xml2')",
+      call. = FALSE
+    )
+    return(invisible(iso_files))
+  }
+
+  # check for availability of rhdf5
+  if (!requireNamespace("rhdf5", quietly = TRUE)) {
+    stop(
+      "'rhdf5' package is required to read .iarc, please run: install.packages('BiocManager'); BiocManager::install('rhdf5')",
+      call. = FALSE
+    )
+    return(invisible(iso_files))
+  }
+
   # unzipping iarc archive ====
   folder_name <- ds$file_info$file_path %>% basename() %>% { str_replace(., fixed(get_file_ext(.)), "") }
   folder_path <- file.path(tempdir(), folder_name)
@@ -170,18 +188,18 @@ process_iarc_sample_data <- function(iso_file, task, gas_configs, folder_path) {
 # will also add H3 factor if part of the gas configuration
 # @param iso_file
 read_irms_data_file <- function(iso_file, filepath, gas_config, run_time.s, data_units = "nA", data_scaling = 1e-9) {
-  if (!"DataSet" %in% h5ls(filepath)$name)
+  if (!"DataSet" %in% rhdf5::h5ls(filepath)$name)
     stop("expected DataSet attribute not present in HDF5 data file", call. = FALSE)
 
   # attributes (NOTE: not sure what to do with the $Tuning information (usually not filled))
-  dataset_attributes <- h5readAttributes(filepath, "DataSet")
+  dataset_attributes <- rhdf5::h5readAttributes(filepath, "DataSet")
   if (!dataset_attributes$Species %in% names(gas_config$species))
     stop("gas configuration for species ", dataset_attributes$Species, " not specified", call. = FALSE)
   config <- gas_config$species[[dataset_attributes$Species]]
 
   # read irms data and determine which beams are used
-  irms_data <- h5read(filepath, "DataSet") %>% dplyr::as_tibble()
-  H5close() # garbage collect
+  irms_data <- rhdf5::h5read(filepath, "DataSet") %>% dplyr::as_tibble()
+  rhdf5::H5close() # garbage collect
 
   if (!"Scan" %in% names(irms_data)) 
     stop("Scan column missing from data file ", basename(filepath), call. = FALSE)

R/package.R

@@ -14,8 +14,6 @@
 #' @importFrom stringr str_c str_detect str_to_title str_replace str_replace_all str_replace_na str_match str_match_all str_interp str_subset str_extract fixed
 #' @importFrom methods is
 #' @importFrom utils unzip head tail modifyList packageVersion
-#' @importFrom xml2 xml_find_all xml_child xml_text read_xml xml_children as_list
-#' @importFrom rhdf5 h5ls h5read h5readAttributes H5close
 #' @importFrom UNF unf
 NULL
 

R/utils_xml_files.R

@@ -4,7 +4,7 @@
 map_xml_children <- function(nodes, select = NULL) {
   nodes %>% 
     map_df(function(node) {
-      as_list(node) %>% 
+      xml2::as_list(node) %>% 
         # if select is specific, only take the children specific
       { if(is.null(select)) . else .[select[select %in% names(.)]] } %>% 
         # map all as text ignoring everything that does not have exactly 1 value
@@ -17,22 +17,22 @@ map_xml_children <- function(nodes, select = NULL) {
 # retrieve Identifier/Value pairs from 'container' type children of current node
 xml_fetch_container_value <- function(xml, ids, container = "PersistedPropertyBagProperty") {
   sapply(ids, function(id) {
-    xml %>% xml_find_all(str_c(".//", container, "[Identifier[.='", id, "']]")) %>% 
-      xml_child("Value") %>% xml_text() %>% list()
+    xml %>% xml2::xml_find_all(str_c(".//", container, "[Identifier[.='", id, "']]")) %>% 
+      xml2::xml_child("Value") %>% xml2::xml_text() %>% list()
   })
 }
 
 # XML iarc xml file processing ========
 
 # process iarc info xml file
 process_iarc_info_xml <- function(filepath) {
-  info_xml <- read_xml(filepath, encoding = "UTF-8")
-  info_version <- info_xml %>% xml_child("Version") %>% xml_text()
+  info_xml <- xml2::read_xml(filepath, encoding = "UTF-8")
+  info_version <- info_xml %>% xml2::xml_child("Version") %>% xml2::xml_text()
 
   # retrieve processing lists information
   processing_lists <- 
-    info_xml %>% xml_child("ProcessingLists") %>% 
-    xml_children() %>% 
+    info_xml %>% xml2::xml_child("ProcessingLists") %>% 
+    xml2::xml_children() %>% 
     map_xml_children() 
 
   # version safety check
@@ -69,12 +69,12 @@ process_iarc_methods_xml <- function(filepaths) {
   method_params <- 
     filepaths %>% 
     lapply(function(methods_file) {
-      method_xml <- read_xml(methods_file, encoding = "UTF-8")
+      method_xml <- xml2::read_xml(methods_file, encoding = "UTF-8")
       # id
-      method_id <- method_xml %>% xml_child("Id") %>% xml_text()
+      method_id <- method_xml %>% xml2::xml_child("Id") %>% xml2::xml_text()
       # method parameters
       method_xml %>% 
-        xml_find_all(".//SerialisedFlowParameter") %>% 
+        xml2::xml_find_all(".//SerialisedFlowParameter") %>% 
         map_xml_children() %>% 
         mutate(MethodId = method_id,
                MethodFile = basename(methods_file))
@@ -101,20 +101,20 @@ process_iarc_tasks_xml <- function(filepaths, method_parameters) {
 
   process_iarc_task_xml <- function(task_file) {
     # read file
-    task_xml <- read_xml(task_file, encoding = "UTF-8")  
+    task_xml <- xml2::read_xml(task_file, encoding = "UTF-8")  
 
     # retrieve general task info
     task_info <- 
       c("GlobalIdentifier", "Name", "Id", 
         "AcquisitionStartDate", "AcquisitionEndDate", # not sure these are useful
         "CompletionState", "MethodId", "ProcessingListTypeIdentifier") %>% 
-      sapply(function(child) task_xml %>% xml_child(child) %>% xml_text() %>% list())
+      sapply(function(child) task_xml %>% xml2::xml_child(child) %>% xml2::xml_text() %>% list())
 
     # retrieve task values based on methods information (if there is any)
     if (nrow(method_parameters) > 0) {
       task_values <-
         task_xml %>% 
-        xml_find_all(".//SerialisableTaskValue") %>% 
+        xml2::xml_find_all(".//SerialisableTaskValue") %>% 
         map_xml_children() %>% 
         # link with parameters defined in methods
         mutate(
@@ -131,7 +131,7 @@ process_iarc_tasks_xml <- function(filepaths, method_parameters) {
     # retrieve task data (where the real information is recorded)
     task_data <- 
       task_xml %>% 
-      xml_find_all(".//SerialisableDataSet") %>% 
+      xml2::xml_find_all(".//SerialisableDataSet") %>% 
       map_xml_children(
         select = c("Id", "AcquireDataStatus", "AcquireStartDate", "AcquireEndDate", "TypeIdentifier")) %>% 
       mutate(
@@ -188,8 +188,8 @@ process_iarc_processing_xml <- function(processing_list_id, filepath) {
   Label <- NumeratorBeamChannel <- numerator_mass <- DenominatorBeamChannel <- denominator_mass <- NULL
 
   # read file
-  xml <- read_xml(filepath, encoding = "UTF-8")
-  global_id <- xml %>% xml_child("DefinitionUniqueIdentifier") %>% xml_text()
+  xml <- xml2::read_xml(filepath, encoding = "UTF-8")
+  global_id <- xml %>% xml2::xml_child("DefinitionUniqueIdentifier") %>% xml2::xml_text()
 
   # safety check
   if (global_id != processing_list_id) {
@@ -201,16 +201,16 @@ process_iarc_processing_xml <- function(processing_list_id, filepath) {
   # find the species
   xml_find_species <- function(node) {
     # potentially useful(?): DetectionBeamChannel
-    node %>% xml_child("SerialisedPropertyBagProperties") %>%
+    node %>% xml2::xml_child("SerialisedPropertyBagProperties") %>%
       xml_fetch_container_value("Species") %>% { .$Species }
   }
 
   # find the channel masses from the beam ratio definitions
   xml_find_channel_masses <- function(node) {
     # find the beam ratio definitions
     ratio_defs <-
-      node %>% xml_child("SerialisedChildPropertyBags") %>% 
-      xml_find_all(".//SerialisablePropertyBag[Identifier[.='{42D28191-A6E9-4B7B-8C3D-0F0037624F7D}']]") %>%  
+      node %>% xml2::xml_child("SerialisedChildPropertyBags") %>% 
+      xml2::xml_find_all(".//SerialisablePropertyBag[Identifier[.='{42D28191-A6E9-4B7B-8C3D-0F0037624F7D}']]") %>%  
       map(xml_fetch_container_value, c("NumeratorBeamChannel", "DenominatorBeamChannel", "Label")) %>%
       bind_rows() 
     if (nrow(ratio_defs) == 0) return (tibble(channel = character(), mass = character()))  
@@ -237,7 +237,7 @@ process_iarc_processing_xml <- function(processing_list_id, filepath) {
   # find the H3 factor
   xml_find_H3_factor <- function(node) {
     H3_factor <- 
-      node %>% xml_child("SerialisedPropertyBagProperties") %>%
+      node %>% xml2::xml_child("SerialisedPropertyBagProperties") %>%
       xml_fetch_container_value(c("ApplyH3CorrectionFactor", "H3CorrectionFactor"))
     if (!is.na(H3_factor$ApplyH3CorrectionFactor) && H3_factor$ApplyH3CorrectionFactor == "True") 
       return(as.numeric(H3_factor$H3CorrectionFactor))
@@ -246,7 +246,7 @@ process_iarc_processing_xml <- function(processing_list_id, filepath) {
 
   # process channel configurations
   species_config <- xml %>% 
-    xml_find_all("//SerialisablePropertyBag[Identifier[.='10DC1602-5ED4-4D62-BAB0-2693E3FBC3AF']]") %>% 
+    xml2::xml_find_all("//SerialisablePropertyBag[Identifier[.='10DC1602-5ED4-4D62-BAB0-2693E3FBC3AF']]") %>% 
     sapply(function(node) {
       species <- xml_find_species(node)
       if (is.null(species) || is.na(species)) # no species definition found

README.Rmd

@@ -37,15 +37,23 @@ You can install the latest release of isoreader from [CRAN](https://cran.r-proje
 install.packages("isoreader")
 ```
 
-To install the current development version directly from GitHub, please use the devtools package:
+Some isoreader features including Excel and feather export depend on optional packages that are not required for the core functionality of isoreader. To use this functionality, please install the following packages manually if not already installed (isoreader will throw an informative warning if they are needed but missing):
+
+```{r, optional-installation, eval = FALSE}
+# optional extensions
+install.packages(c("feather", "openxlsx", "xml2", "BiocManager"))
+BiocManager::install("rhdf5")
+```
+
+To install the current development version of isoreader directly from GitHub, please use the devtools package:
 
 ```{r gh-installation, eval = FALSE}
 # installs the development tools package if not yet installed
 if(!requireNamespace("devtools", quietly = TRUE)) install.packages("devtools")
 devtools::install_github("isoverse/isoreader")
 ```
 
-Troubleshooting note: depending on your workspace and operating system, you may have to re-start your R session or manually install some dependencies. For example, the [Bioconductor](https://www.bioconductor.org/) package manager may need manual installation: `install.packages("BiocManager")`; also, the `digest` package sometimes causes trouble - re-install with `remove.packages("digest"); install.packages("digest")`.
+Troubleshooting note: depending on your workspace and operating system, you may have to re-start your R session or manually install some dependencies. For example, the `digest` package sometimes causes trouble - re-install with `remove.packages("digest"); install.packages("digest")`.
 
 ## Show me some code
 

README.md

@@ -53,8 +53,20 @@ You can install the latest release of isoreader from
 install.packages("isoreader")
 ```
 
-To install the current development version directly from GitHub, please
-use the devtools package:
+Some isoreader features including Excel and feather export depend on
+optional packages that are not required for the core functionality of
+isoreader. To use this functionality, please install the following
+packages manually if not already installed (isoreader will throw an
+informative warning if they are needed but missing):
+
+``` r
+# optional extensions
+install.packages(c("feather", "openxlsx", "xml2", "BiocManager"))
+BiocManager::install("rhdf5")
+```
+
+To install the current development version of isoreader directly from
+GitHub, please use the devtools package:
 
 ``` r
 # installs the development tools package if not yet installed
@@ -64,11 +76,9 @@ devtools::install_github("isoverse/isoreader")
 
 Troubleshooting note: depending on your workspace and operating system,
 you may have to re-start your R session or manually install some
-dependencies. For example, the
-[Bioconductor](https://www.bioconductor.org/) package manager may need
-manual installation: `install.packages("BiocManager")`; also, the
-`digest` package sometimes causes trouble - re-install with
-`remove.packages("digest"); install.packages("digest")`.
+dependencies. For example, the `digest` package sometimes causes trouble
+- re-install with `remove.packages("digest");
+install.packages("digest")`.
 
 ## Show me some code
 
@@ -89,7 +99,7 @@ iso_files <- iso_read_scan(data_folder)
 #> Info: reading file 'full_scan_example.scn' with '.scn' reader...
 #> Info: reading file 'peak_shape_scan_example.scn' with '.scn' reader...
 #> Info: reading file 'time_scan_example.scn' with '.scn' reader...
-#> Info: finished reading 4 files in 0.94 secs
+#> Info: finished reading 4 files in 1.00 secs
 
 iso_files
 #> Data from 4 scan iso files: 

tests/testthat/test-continuous-flow.R

@@ -18,17 +18,11 @@ test_that("test that parameter checks are performed", {
 })
 
 
-test_that("test that dxf files can be read", {
+test_that("test that continous flow files can be read", {
 
   # skip on CRAN to reduce checktime to below 10 minutes
   skip_on_cran()
 
-  # check if tests are enabled
-  run_file_tests <- getOption("isoreader.run_file_tests")
-  if (!is.null(run_file_tests) && identical(run_file_tests, FALSE)) {
-    skip("Currently not testing all continuous flow data files.")
-  }
-
   # test specific files
   iso_turn_reader_caching_off()
 
@@ -61,14 +55,22 @@ test_that("test that dxf files can be read", {
   expect_equal(nrow(iso_get_vendor_data_table(cf)), 6)
   expect_equal(ncol(iso_get_vendor_data_table(cf)), 61)
 
+  # skip if optional dependencies are not installed
+  skip_if_not_installed("xml2")
+  skip_if_not_installed("rhdf5")
   expect_true(file.exists(file <- iso_get_reader_example("continuous_flow_example.iarc")))
   expect_is(iarc <- iso_read_continuous_flow(file), "iso_file_list")
   expect_equal(nrow(problems(iarc)), 0)
 
+})
+
+test_that("test that additional continous flow files can be read", {
+
   # additional test files (skip on CRAN because test files not includes due to tarball size limits) =====
   skip_on_cran()
   test_folder <- file.path("test_data") # test_folder <- file.path("tests", "testthat", "test_data") # direct
-
+  iso_turn_reader_caching_off()
+
   # testing wrapper
   check_continuous_flow_test_file <- function(file) {
     file_path <- get_isoreader_test_file(file, local_folder = test_folder)