metadataTools

metadataTools is an R package designed to simplify the process of documenting datasets with structured metadata. It provides tools to:

• Annotate data frames with variable-level descriptions and units\
• Generate human-readable README files for data repositories (e.g., DataverseNO)\
• Document and validate naming conventions for datasets

Features

📝 Add Metadata to Columns

• add_desc() – Adds a "description" attribute to columns. If the description contains square brackets (e.g., "Length [cm]"), the content inside the brackets is stored as a "unit" attribute.
• add_unit() – Adds a "unit" attribute to columns directly.
• Use attr(df, "description") to describe the dataset as a whole.

📄 Generate README Files

• create_readme() produces a structured README.txt file with:
  • Dataset title, DOI, contact info
  • Dataset description and methodology
  • File- and variable-level metadata
• set_metadata_defaults() lets you store default metadata values (e.g., name, institution, email, ORCID) used by create_readme() to streamline README generation.
• Compatible with the DataverseNO metadata schema.

📊 Create Metadata Tables

• create_metadata_table() builds a formatted flextable summarizing variable names, types, units, and descriptions.

✅ Validate Projects' Naming Conventions

• validate_naming_convention() checks whether a dataset contains the mandatory variables and only valid entries as defined in a project's naming convention (e.g., EcoForest_naming). It also adds standardized variable descriptions to the dataset.
• view_naming_convention() prints a human-readable overview of the naming convention.

🔧 Installation

You can install the development version of metadataTools directly from GitHub using the remotes package:

# Install remotes if not already installed
install.packages("remotes")

# Install metadataTools from GitHub
remotes::install_github("asplundj/metadataTools")