From 0f0bfe3b0d038924160cf0091e8d1396405561d0 Mon Sep 17 00:00:00 2001
From: tan <tanmaykm@gmail.com>
Date: Mon, 27 Oct 2025 10:16:53 +0530
Subject: [PATCH 1/3] feat: support minio credentials for datasets

Handle minio credentials for dataset operations.

fixes: https://github.com/JuliaComputing/JuliaHub/issues/20486
---
 src/datasets.jl | 28 +++++++++++++++++++++++-----
 1 file changed, 23 insertions(+), 5 deletions(-)

diff --git a/src/datasets.jl b/src/datasets.jl
index 8a42010a9..d0ca2f0b5 100644
--- a/src/datasets.jl
+++ b/src/datasets.jl
@@ -801,7 +801,7 @@ end
 function _upload_dataset(upload_config, local_path; progress::Bool)
     type = upload_config["upload_type"]
     vendor = upload_config["vendor"]
-    if type != "S3" || vendor != "aws"
+    if type != "S3" || !(vendor in ("aws", "minio"))
         throw(JuliaHubError("Unknown upload type ($type) or vendor ($vendor)"))
     end
     mktemp() do rclone_conf_path, rclone_conf_io
@@ -868,19 +868,29 @@ function _write_rclone_config(
     access_key_id::AbstractString,
     secret_access_key::AbstractString,
     session_token::AbstractString,
+    provider::AbstractString="AWS",
+    endpoint::AbstractString="",
 )
+    if lowercase(provider) == "aws"
+        provider = "AWS"
+    elseif lowercase(provider) == "minio"
+        provider = "Minio"
+    else
+        throw(JuliaHubError("Unknown storage backend $(provider)"))
+    end
+
     write(
         io,
         """
 [juliahub_remote]
 type = s3
-provider = AWS
+provider = $provider
 env_auth = false
 access_key_id = $access_key_id
 secret_access_key = $secret_access_key
 session_token = $session_token
 region = $region
-endpoint =
+endpoint = $endpoint
 location_constraint = $region
 acl = private
 server_side_encryption =
@@ -894,7 +904,11 @@ function _write_rclone_config(io::IO, upload_config::Dict)
     access_key_id = upload_config["credentials"]["access_key_id"]
     secret_access_key = upload_config["credentials"]["secret_access_key"]
     session_token = upload_config["credentials"]["session_token"]
-    _write_rclone_config(io; region, access_key_id, secret_access_key, session_token)
+    provider = upload_config["vendor"]
+    endpoint = get(upload_config["credentials"], "endpoint_url", "")
+    _write_rclone_config(
+        io; region, access_key_id, secret_access_key, session_token, provider, endpoint
+    )
 end
 
 function _get_dataset_credentials(auth::Authentication, dataset::Dataset)
@@ -988,9 +1002,11 @@ function download_dataset(
         throw(InvalidRequestError("Dataset '$(dataset.name)' does not have version 'v$version'"))
 
     credentials = Mocking.@mock _get_dataset_credentials(auth, dataset)
-    credentials["vendor"] == "aws" ||
+    provider = credentials["vendor"]
+    provider in ("aws", "minio") ||
         throw(JuliaHubError("Unknown 'vendor': $(credentials["vendor"])"))
     credentials = credentials["credentials"]
+    endpoint = get(credentials, "endpoint_url", "")
 
     bucket = dataset._storage.bucket
     prefix = dataset._storage.prefix
@@ -1019,6 +1035,8 @@ function download_dataset(
                 access_key_id=credentials["access_key_id"],
                 secret_access_key=credentials["secret_access_key"],
                 session_token=credentials["session_token"],
+                provider=provider,
+                endpoint=endpoint,
             )
             close(rclone_conf_io)
 

From 702a89cd2609f9ad7495ac34d57ad6d4d47ae9d0 Mon Sep 17 00:00:00 2001
From: tan <tanmaykm@gmail.com>
Date: Fri, 31 Oct 2025 08:11:31 +0530
Subject: [PATCH 2/3] update changelog

---
 CHANGELOG.md | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index 0fea68941..3a0b58def 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -4,6 +4,10 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
 
 ## Unreleased
 
+### Added
+
+* Support for JuliaHub instances using MinIO as datasets storage backend.
+
 ### Fixed
 
 * The `.message` property _should_ always be present in the log messages, but there are a few versions out there where it is sometimes omitted due to a backend bug. We default to an empty string in those cases. ([#111])

From bda3a91b281a65c3a1e5c4e97cb533ba1ced73a3 Mon Sep 17 00:00:00 2001
From: Morten Piibeleht <morten.piibeleht@juliahub.com>
Date: Mon, 3 Nov 2025 20:03:51 +1300
Subject: [PATCH 3/3] Document MinIO backend requirements for datasets

Add information about MinIO backend requirements for JuliaHub datasets.
---
 docs/src/reference/datasets.md | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/docs/src/reference/datasets.md b/docs/src/reference/datasets.md
index 7dc55a414..c01c15277 100644
--- a/docs/src/reference/datasets.md
+++ b/docs/src/reference/datasets.md
@@ -38,6 +38,10 @@ A JuliaHub dataset can have zero or more versions.
 A newly created dataset _usually_ has at least one version, but it may have zero versions if, for example, the upload did not finish.
 The versions are indexed with a linear list of integers starting from `1`.
 
+## MinIO backend
+
+JuliaHub instances with the MinIO backend for data storage require at least JuliaHub.jl v0.1.6 for dataset uploads and downloads.
+
 ## Reference
 
 ```@docs