Integrate database

.Rbuildignore

@@ -6,3 +6,27 @@
 ^pkgdown$
 ^\.github$
 ^vignettes/articles$
+^database$
+^database/.*$
+^.*[.]duckdb$
+^.*[.]parquet$
+^data/db_overlay$
+^data/raw$
+^dashboard$
+^dashboard/.*$
+^dashboard/dashboard_cache$
+^dashboard/_freeze$
+^python$
+^.*[.]RData$
+^.*[.]rdx$
+^\\.Rproj\\.user$
+^\\.all-contributorsrc$
+^\\.venv($|/)
+^venv($|/)
+^env($|/)
+^.env($|/)
+^__pycache__($|/)
+^python/__pycache__($|/)
+^\\.python-version$
+^.*[.]ipynb$
+^.*[.]ipynb_checkpoints($|/)

.Renviron

@@ -0,0 +1,2 @@
+RETICULATE_PYTHON=D:/GitHub/OSMdashboard/.venv/Scripts/python.exe
+DUCKDB_PATH=D:/GitHub/OSMdashboard/database/osm_changesets.duckdb

.gitignore

@@ -6,3 +6,23 @@
 .quarto
 docs
 inst/doc
+database/
+*.duckdb
+*.parquet
+data/db_overlay/
+data/raw/
+dashboard/dashboard_cache/
+dashboard/_freeze/
+dashboard/.quarto/
+dashboard/site_libs/
+.RData
+.Rhistory
+.Rproj.user/
+.venv/
+venv/
+env/
+.env/
+__pycache__/
+python/__pycache__/
+*.ipynb
+.ipynb_checkpoints/

DESCRIPTION

@@ -56,7 +56,12 @@ Imports:
     stringr,
     tidyr,
     tidyRSS,
-    WikipediR
+    WikipediR,
+    DBI,
+    duckdb,
+    readr,
+    sf,
+    reticulate
 Suggests: 
     knitr,
     rmarkdown,

NAMESPACE

@@ -2,6 +2,7 @@
 
 export(calc_stats_contributions_wiki)
 export(categorise_keys)
+export(connect_osm_db)
 export(create_dashboard)
 export(extract_and_combine_tags)
 export(get_changesets_details)

R/database.R

@@ -0,0 +1,33 @@
+#' Connect to the OSM DuckDB database
+#'
+#' @description Opens a DBI connection to the OSM Changesets DuckDB file.
+#' It first checks the environment variable `DATABASE_URL` for a URL of the form
+#' `duckdb:///path/to/osm_changesets.duckdb`. If not set, it falls back to
+#' the environment variable `DUCKDB_PATH` (default:
+#' `"database/osm_changesets.duckdb"`).
+#'
+#' @returns A DBI connection object (`DBI::DBIConnection`) to a DuckDB database.
+#'   You are responsible for closing it with `DBI::dbDisconnect(con)`.
+#' @export
+#'
+#' @examples
+#' \dontrun{
+#'   con <- connect_osm_db()
+#'   # ... use the connection ...
+#'   DBI::dbDisconnect(con)
+#' }
+
+#In the above example, dontrun is placed there as the database will not found on GitHub so these example
+#checks will not execute 
+connect_osm_db <- function() {
+  url <- Sys.getenv("DATABASE_URL", "")
+  if (nzchar(url) && startsWith(url, "duckdb:///")) {
+    # DATABASE_URL like "duckdb:///absolute/or/relative/path/to/file.duckdb"
+    dbfile <- sub("^duckdb:///", "", url)
+    return(DBI::dbConnect(duckdb::duckdb(), dbfile))
+  }
+
+  # Fallback: environment variable DUCKDB_PATH
+  duck_path <- Sys.getenv("DUCKDB_PATH", "database/osm_changesets.duckdb")
+  DBI::dbConnect(duckdb::duckdb(), duck_path)
+}

R/user_contributions.R

@@ -17,8 +17,6 @@ get_contributions_osm_users <- function(users) {
 
   for (user in users) {
 
-    print(user)
-
     # Sanitise url
     user_clean <- gsub(" ", "%20", user)
 

README.md

@@ -57,13 +57,25 @@ This package uses quarto, which is is already installed with RStudio and Positro
 
 ## Usage
 
+## Usage WITHOUT database
+
 1. Create the dashboard folder structure by calling `create_dashboard()`.
 2. Edit the `data/group_info.csv` and `data/group_users.csv` files to add your groups and users.
 3. Run `data_retrieval.R` to retrieve the data from OSM.
-4. Render `dashboard.qmd` to generate the dashboard.
+4. Render `dashboard.qmd` to generate the dashboard. Do this by running `quarto render dashboard.qmd`.
 
-Refer to the `vignette("dashboard-group-contributions")` for more details.
+## Usage WITH database
+
+In your main directory, the database must be called `osm_changesets.duckdb` and must be stored in a subdirectory named `database`.
 
+1. Create the dashboard folder structure by calling `create_dashboard()`.
+2. Edit the `data/group_info.csv` and `data/group_users.csv` files to add your groups and users.
+3. Run `data_retrieval.R` to retrieve the data from OSM.
+4. Run `quarto render dashboard.qmd -P use_db_overlay:true`.
+
+The `use_db_overlay:true` parameter ensures that all information within the database is loaded and used in place of information from APIs in the dashboard.
+
+Refer to the `vignette("dashboard-group-contributions")` for more details.
 
 ## Contributors