Add R package for users

BiocHubsIngestR/.Rbuildignore

@@ -0,0 +1,5 @@
+^.*\.Rproj$
+^\.Rproj\.user$
+^_pkgdown\.yml$
+^docs$
+^pkgdown$

BiocHubsIngestR/.gitignore

@@ -0,0 +1,5 @@
+.Rproj.user
+.Rhistory
+.RData
+.Ruserdata
+docs

BiocHubsIngestR/DESCRIPTION

@@ -0,0 +1,20 @@
+Package: BiocHubsIngestR
+Title: Upload Data to Bioconductor Hubs Ingest Endpoints
+Version: 0.99.0
+Authors@R: 
+    person("Alexandru", "Mahmoud", email = "[email protected]", role = c("aut", "cre"))
+Description: Facilitates uploading data to temporary S3 endpoints created by the Bioconductor Hubs Ingest stack.
+    Provides simple functions to configure credentials and upload local data directories.
+License: Artistic-2.0
+Encoding: UTF-8
+Roxygen: list(markdown = TRUE)
+RoxygenNote: 7.3.2
+Imports:
+    aws.s3
+Suggests:
+    testthat (>= 3.0.0),
+    knitr,
+    rmarkdown
+Config/testthat/edition: 3
+biocViews: Infrastructure
+VignetteBuilder: knitr

BiocHubsIngestR/NAMESPACE

@@ -0,0 +1,4 @@
+# Generated by roxygen2: do not edit by hand
+
+export(auth)
+export(upload)

BiocHubsIngestR/R/credentials.R

@@ -0,0 +1,30 @@
+#' Set AWS credentials for Bioconductor Hubs Ingest endpoint
+#'
+#' @param username Character string of the username provided by administrator
+#' @param password Character string of the password/key provided by administrator
+#' @param endpoint Optional custom endpoint URL. If NULL, constructs from username
+#'
+#' @return Invisible NULL, sets environment variables as side effect
+#' @export
+#'
+#' @examples
+#' \dontrun{
+#' BiocHubsIngestR::auth("myusername", "mypassword")
+#' }
+auth <- function(username, password, endpoint = NULL) {
+    if (!is.character(username) || !is.character(password))
+        stop("Username and password must be character strings")
+
+    if (is.null(endpoint)) {
+        endpoint <- sprintf("https://%s.hubsingest.bioconductor.org", username)
+    }
+
+    Sys.setenv(
+        AWS_ACCESS_KEY_ID = username,
+        AWS_SECRET_ACCESS_KEY = password,
+        AWS_DEFAULT_REGION = "",
+        AWS_S3_ENDPOINT = endpoint
+    )
+
+    invisible(NULL)
+}

BiocHubsIngestR/R/upload.R

@@ -0,0 +1,44 @@
+#' Upload local directory to Bioconductor Hubs Ingest S3 endpoint
+#'
+#' @param path Character string path to local directory or file to upload
+#' @param bucket Character string of bucket name, defaults to "userdata"
+#'
+#' @return Invisible list of uploaded files
+#' @export
+#'
+#' @examples
+#' \dontrun{
+#' BiocHubsIngestR::upload("path/to/data")
+#' }
+upload <- function(path, bucket = "userdata") {
+    if (!file.exists(path))
+        stop("Path does not exist: ", path)
+
+    if (!aws.s3::bucket_exists(bucket)) {
+        message("Creating bucket: ", bucket)
+        aws.s3::put_bucket(bucket)
+    }
+
+    if (file.info(path)$isdir) {
+        files <- list.files(path, recursive = TRUE, full.names = TRUE)
+    } else {
+        files <- path
+    }
+
+    uploaded <- lapply(files, function(f) {
+        rel_path <- if(file.info(path)$isdir) {
+            sub(paste0("^", path, "/?"), "", f)
+        } else {
+            basename(f)
+        }
+        message("Uploading: ", rel_path)
+        aws.s3::put_object(
+            file = f,
+            object = rel_path,
+            bucket = bucket
+        )
+        return(rel_path)
+    })
+
+    invisible(uploaded)
+}

BiocHubsIngestR/tests/testthat.R

@@ -0,0 +1,4 @@
+library(testthat)
+library(BiocHubsIngestR)
+
+test_check("BiocHubsIngestR")

BiocHubsIngestR/tests/testthat/test-credentials.R

@@ -0,0 +1,22 @@
+test_that("auth sets environment variables correctly", {
+    old_env <- Sys.getenv(c("AWS_ACCESS_KEY_ID", "AWS_SECRET_ACCESS_KEY", 
+                           "AWS_DEFAULT_REGION", "AWS_S3_ENDPOINT"))
+    on.exit(do.call(Sys.setenv, as.list(old_env)))
+
+    username <- "testuser"
+    password <- "testpass"
+    auth(username, password)
+
+    expect_equal(Sys.getenv("AWS_ACCESS_KEY_ID"), username)
+    expect_equal(Sys.getenv("AWS_SECRET_ACCESS_KEY"), password)
+    expect_equal(Sys.getenv("AWS_DEFAULT_REGION"), "")
+    expect_equal(
+        Sys.getenv("AWS_S3_ENDPOINT"),
+        "https://testuser.hubsingest.bioconductor.org"
+    )
+})
+
+test_that("auth validates inputs", {
+    expect_error(auth(123, "pass"))
+    expect_error(auth("user", 123))
+})

BiocHubsIngestR/vignettes/BiocHubsIngestR.Rmd

@@ -0,0 +1,54 @@
+---
+title: "Using BiocHubsIngestR"
+author: "Alexandru Mahmoud"
+date: "`r Sys.Date()`"
+output: rmarkdown::html_vignette
+vignette: >
+  %\VignetteIndexEntry{Using BiocHubsIngestR}
+  %\VignetteEngine{knitr::rmarkdown}
+  %\VignetteEncoding{UTF-8}
+---
+
+```{r setup, include=FALSE}
+knitr::opts_chunk$set(echo = TRUE)
+```
+
+## Introduction
+
+The `BiocHubsIngestR` package provides functionality for uploading data to temporary S3 endpoints created by the Bioconductor Hubs Ingest stack. Before proceeding with data upload, you must:
+
+1. Have an S3 ingestion endpoint generated by a Bioconductor admin
+2. Receive access credentials (username and password) from the admin
+
+## Authentication
+
+First, set up your credentials using the `auth()` function:
+
+```{r eval=FALSE}
+library(BiocHubsIngestR)
+
+# Replace with your actual credentials provided by the admin
+auth(username = "your_username", password = "your_password")
+```
+
+## Uploading Data
+
+Once authenticated, you can upload either individual files or entire directories:
+
+```{r eval=FALSE}
+# Upload a single file
+upload("path/to/your/file.txt")
+
+# Upload an entire directory
+upload("path/to/your/data/directory")
+```
+
+The `upload()` function will create a bucket if it doesn't exist and handle the transfer of files to the S3 endpoint. Progress messages will be displayed during the upload process.
+
+## Important Notes
+
+1. After completing your data upload, notify the Bioconductor admin you are coordinating with.
+2. Once you inform the admin that the upload is complete, the endpoint will be locked for processing and no further uploads will be possible.
+3. Make sure all your data is uploaded completely before notifying the admin.
+
+For questions or issues, please contact the Bioconductor team or open an issue on the package's GitHub repository.