Merge branch 'main' into auto-merge-options

Author: davidorme

.Rprofile

@@ -19,5 +19,4 @@ jobs:
           with:
             r-version: 'release'
             use-public-rspm: true
-        # - uses: r-lib/actions/setup-renv@v2
         - uses: pre-commit/[email protected]

.github/workflows/ci.yaml

@@ -1,5 +1,5 @@
 linters: linters_with_defaults(
     cyclocomp_linter(complexity_limit = 10L),
-    indentation_linter(indent = 4L),
+    indentation_linter(indent = 2L),
     line_length_linter(length = 88L)
   )

.lintr

@@ -6,7 +6,6 @@ repos:
     rev: v5.0.0
     hooks:
     - id: trailing-whitespace
-      exclude: renv/activate.R
     - id: end-of-file-fixer
     - id: check-yaml
     - id: check-added-large-files
@@ -32,7 +31,7 @@ repos:
     rev: dcd2a60
     hooks:
     - id: style-files
-      args: [--style_pkg=styler, --indent_by=4, --style_fun=tidyverse_style]
+      args: [--style_pkg=styler, --style_fun=tidyverse_style]
     - id: lintr
     - id: parsable-R
     - id: no-browser-statement

.pre-commit-config.yaml

@@ -2,10 +2,6 @@
     "r.alwaysUseActiveTerminal": true,
     "r.bracketedPaste": true,
     "r.rpath.mac": "/opt/homebrew/bin/R",
-    "r.useRenvLibPath": true,
     "r.workspaceViewer.clearPrompt": false,
     "r.rterm.mac": "radian",
-    "r.libPaths": [
-        "/Users/dorme/Research/Virtual_Rainforest/ve_data_science/renv/library/macos/R-4.4/aarch64-apple-darwin23.6.0 "
-    ]
 }

.vscode/settings.json

@@ -21,9 +21,9 @@ As a brief overview, this involves setting up the following tools:
   tools.
 * **Poetry**. This is a Python package manager that we use to manage a shared set of
   Python packages used across the project.
-* **R**. We will be using R extensively for analysis and data visualisation.
-* **renv**. This is an R package manager that, like `poetry`, is being used to manage
-  the shared set of R packages used across the project.
+* **R**. We will be using R extensively for analysis and data visualisation. At the
+  moment, we are managing package use and versioning with a simple list of packages.
+  This is currently very light touch and we may use something stricter in the future.
 * **pre-commit**. This is a QA tool - we have a set of configured checks that run
   whenever you try and `git commit` some changes to the repo. If the changes fail any of
   the checks, then you will have to fix them and try again. Sometimes, the QA checks can

README.md

@@ -0,0 +1,95 @@
+#' ---
+#' title: Script metadata checker
+#'
+#' description: |
+#'     This file provides the function `check_metadata` that can be used to check that
+#'     a given file contains the expected metadata and that it can read correctly.
+#'
+#'     We aren't sure if we are going to support R markdown rather than Jupyter
+#'     notebooks but we'll start with providing some support.
+#'
+#' VE_module: None
+#'
+#' author:
+#'   - name: David Orme
+#'
+#' status: wip
+#'
+#' package_dependencies:
+#'     - rmarkdown
+#'     - yaml
+#'     - tools
+#'
+#' usage_notes: |
+#'     From within R use:
+#'
+#'     source('templates/metadata_check.R')
+#'     yaml <- check_metadata('path/to/file.R')
+#' ---
+
+library(rmarkdown)
+library(yaml)
+library(tools)
+
+check_metadata <- function(file_path) {
+  #' Check VE data science script and notebook metadata
+  #'
+  #' Checks that a script file or notebook contains expected metadata. It raises an
+  #' error if the file is missing metadata, if the metadata is badly formatted or if
+  #' it contains unexpected values.
+  #'
+  #' @param file_path The path to the file to be checked
+  #'
+  #' @return The function returns the YAML metadata if it can be read cleanly.
+  #'
+  #'
+
+  if (!file.exists(file_path)) {
+    simpleError("File path not found.")
+  }
+
+  # Extract the file extension
+  file_type <- tolower(file_ext(file_path))
+
+  # For R Markdown files, can just use the standard YAML parser
+  if (file_type == "rmd") {
+    yaml <- rmarkdown::yaml_front_matter(file_path)
+  } else if (file_type == "r") {
+    # Load the script as text
+    content <- readLines("templates/R_script_template.r")
+
+    # Locate the YAML document blocks
+    document_markers <- grep("#' ---", content)
+    n_doc_markers <- length(document_markers)
+
+    # Check there are two...
+    if (n_doc_markers != 2) {
+      simpleError(sprintf("Found %i not 2 YAML metadata markers.", n_doc_markers))
+    }
+
+    # And that the first one is at the start of the file
+    if (document_markers[1] != 1) {
+      simpleError(
+        sprintf("First YAML metadata markers is not at the file start.")
+      )
+    }
+
+    # Extract the block and strip the comments
+    yaml_block <- content[document_markers[1]:document_markers[2]]
+    yaml_block <- sub("^#' ?", "", yaml_block)
+
+    yaml <- try(yaml.load(yaml_block), silent = TRUE)
+    if (inherits(yaml, "try-error")) {
+      simpleError(
+        paste0(
+          "Could not load YAML metadata. The YAML loader repoorts:\n",
+          yaml[1]
+        )
+      )
+    }
+  }
+
+  # TODO - this could be extended to validate the contents.
+
+  return(yaml)
+}

maintenance/metadata_check.R

@@ -0,0 +1,11 @@
+# This file is used to record R packages used by the project.
+#
+# We did look at using `renv` to manage this automatically but it added a layer of
+# complexity and unexpected issues and workflow problems and so we have dropped it for
+# now.
+
+# This is required for R support in VSCode
+library(languageserver)
+
+# This is required to use R with Jupyter
+library(IRkernel)