Merge pull request #10 from jrdnbradford/tax-update-action

Add GHA to check taxonomy

.github/workflows/taxonomy-check.yaml

@@ -0,0 +1,89 @@
+name: taxonomy-check
+
+on:
+  schedule:
+    - cron: '0 13 * * 1' # Weekly on Mondays at 9AM EST
+
+env:
+  ISSUE_TITLE: 'eBird taxonomy different than expected'
+  ISSUE_BODY: |
+    This issue has been opened by the GitHub Action workflow `taxonomy-check`.
+
+    The `tax` dataset is different than data retrieved using `rebird::ebirdtaxonomy()`. Potential causes of this could include:
+      * an update to the eBird taxonomy
+      * issues with eBird's API
+      * different versions of R/packages resulting in data being loaded, retrieved, compared, and/or saved differently
+
+    Check if there has been an update to eBird's taxonomy data. If not, investigate the other potential causes.
+
+jobs:
+  taxonomy-check-job:
+    runs-on: ubuntu-latest
+    env:
+      GITHUB_PAT: ${{ secrets.GITHUB_TOKEN }}
+      R_KEEP_PKG_SOURCE: yes
+    outputs:
+      result: ${{ steps.check-taxonomy.outputs.result }}
+    steps:
+      - name: checkout-step
+        uses: actions/checkout@v4
+
+      - name: setup-r-step
+        uses: r-lib/actions/setup-r@v2
+        with:
+          use-public-rspm: true
+          r-version: 'release'
+
+      - name: setup-r-dependencies-step
+        uses: r-lib/actions/setup-r-dependencies@v2
+        with:
+          extra-packages: local::.
+
+      - name: check-taxonomy-step
+        id: check-taxonomy
+        run: Rscript data-raw/tax_check.R
+
+  check-and-create-issue-job:
+    needs: taxonomy-check-job
+    if: needs.taxonomy-check-job.outputs.result == 'TRUE'
+    runs-on: ubuntu-latest
+    steps:
+      - name: check-issue-step
+        id: check-issue
+        uses: actions/github-script@v7
+        with:
+          github-token: ${{ secrets.GITHUB_TOKEN }}
+          script: |
+            const issueTitle = process.env.ISSUE_TITLE;
+            const { owner, repo } = context.repo;
+
+            const issues = await github.rest.issues.listForRepo({
+              owner,
+              repo,
+              state: 'open',
+            });
+
+            const existingIssue = issues.data.find(issue => issue.title === issueTitle);
+
+            if (existingIssue) {
+              core.setOutput('exists', 'true');
+            } else {
+              core.setOutput('exists', 'false');
+            }
+
+      - name: create-issue-step
+        if: ${{ steps.check-issue.outputs.exists == 'false' }}
+        uses: actions/github-script@v7
+        with:
+          github-token: ${{ secrets.GITHUB_TOKEN }}
+          script: |
+            const issueTitle = process.env.ISSUE_TITLE;
+            const issueBody = process.env.ISSUE_BODY;
+            const { owner, repo } = context.repo;
+
+            await github.rest.issues.create({
+              owner,
+              repo,
+              title: issueTitle,
+              body: issueBody,
+            });

R/ebirdtaxonomy.R

@@ -67,5 +67,8 @@ ebirdtaxonomy <- function(cat=NULL, locale=NULL, species = NULL, key = NULL, ...
     key <- ""
   }
 
-  ebird_GET(paste0(ebase(), 'ref/taxonomy/ebird'), args, key = key, ...)
+  tax <- ebird_GET(paste0(ebase(), 'ref/taxonomy/ebird'), args, key = key, ...)
+  tax$comNameCodes <- sort_comma_separated(tax$comNameCodes)
+  tax$sciNameCodes <- sort_comma_separated(tax$sciNameCodes)
+  tax
 }

R/zzz.R

@@ -88,3 +88,15 @@ ebird_GET <- function(url, args, key = NULL, ...){
   }
 }
 
+sort_comma_separated <- function(x) {
+  sapply(strsplit(as.character(x), ","), function(x) {
+    x <- x[x != ""]
+    if (length(x) == 0) {
+      return(NA_character_)
+    } else if (length(x) == 1) {
+      return(x)
+    } else {
+      return(paste(sort(x), collapse = ","))
+    }
+  })
+}

README.Rmd

@@ -20,6 +20,7 @@ knitr::opts_chunk$set(
 [![Codecov test coverage](https://codecov.io/gh/ropensci/rebird/branch/master/graph/badge.svg)](https://app.codecov.io/gh/ropensci/rebird?branch=master)
 [![rstudio mirror downloads](https://cranlogs.r-pkg.org/badges/rebird)](https://github.com/r-hub/cranlogs.app)
 [![cran version](https://www.r-pkg.org/badges/version/rebird)](https://cran.r-project.org/package=rebird/)
+[![taxonomy-check](https://github.com/ropensci/rebird/actions/workflows/taxonomy-check.yaml/badge.svg)](https://github.com/ropensci/rebird/actions/workflows/taxonomy-check.yaml)
 <!-- badges: end -->
 
 `rebird` is a package to interface with the eBird webservices.

README.md

@@ -15,6 +15,8 @@ coverage](https://codecov.io/gh/ropensci/rebird/branch/master/graph/badge.svg)](
 downloads](https://cranlogs.r-pkg.org/badges/rebird)](https://github.com/r-hub/cranlogs.app)
 [![cran
 version](https://www.r-pkg.org/badges/version/rebird)](https://cran.r-project.org/package=rebird/)
+[![taxonomy-check](https://github.com/ropensci/rebird/actions/workflows/taxonomy-check.yaml/badge.svg)](https://github.com/ropensci/rebird/actions/workflows/taxonomy-check.yaml)
+
 <!-- badges: end -->
 
 `rebird` is a package to interface with the eBird webservices.

data-raw/tax.R

@@ -1,7 +1,7 @@
 # eBird taxonomy to look up species codes
 # 
-# This creates the internal taxonomy data frame used by rebird, which
-# is stored ind 'R/sysdata.rda'. We update the package in CRAN when
+# This creates the taxonomy data frame used by rebird, which
+# is stored in 'data/tax.rda'. We update the package in CRAN when
 # the eBird taxonomy update is completed, but in the event you need
 # to update the taxonomy yourself you can do so by running the code below.
 

data-raw/tax_check.R

@@ -0,0 +1,21 @@
+#!/usr/bin/env Rscript
+compare_dataframes <- function(df1, df2) {
+  df1_sorted <- df1[do.call(order, df1),]
+  df2_sorted <- df2[do.call(order, df2),]
+  identical(df1_sorted, df2_sorted)
+}
+
+print("Loading current taxonomy...")
+temp_env <- new.env()
+data("tax", envir = temp_env)
+old_tax <- temp_env$tax
+paste("Current rebird taxonomy has", ncol(old_tax), "columns and", nrow(old_tax), "rows")
+
+print("Retrieving taxonomy from eBird...")
+new_tax <- rebird::ebirdtaxonomy()
+paste("The latest taxonomy from eBird has", ncol(new_tax), "columns and", nrow(new_tax), "rows")
+
+needs_update <- !(compare_dataframes(old_tax, new_tax))
+print(paste("Needs update?:", needs_update))
+if (needs_update) print("The next job to create an issue should run.")
+cat(paste0("result=", needs_update), file = Sys.getenv("GITHUB_OUTPUT"), append = TRUE)