如何在AWS的RStudio中保存数据到S3并同步生成的网站文件

Great questions! Let's break this down step by step for your AWS RStudio environment, where your files live in ~/Home/Myproject.

First, you'll need to set up your R environment to interact with AWS S3. The aws.s3 package is the most straightforward tool for this.

1. Install and load the package #

install.packages("aws.s3")
library(aws.s3)

2. Configure AWS Credentials #

• Best practice (if using AWS compute like EC2): Attach an IAM role to your RStudio instance with permissions for s3:PutObject, s3:ListBucket, and any other S3 actions you need. The aws.s3 package will automatically pick up these credentials—no manual key entry needed.
• Manual credentials (if not using an IAM role): Set your AWS keys as environment variables in R (never hardcode them in scripts!):

  Sys.setenv(AWS_ACCESS_KEY_ID = "your-access-key-here",
             AWS_SECRET_ACCESS_KEY = "your-secret-key-here",
             AWS_REGION = "your-region") # e.g., "us-east-1"
  

Option A: Save R objects directly to S3 #

You can skip writing to your local filesystem and save R objects (like data frames, models) straight to S3:

• For RDS files (ideal for preserving R object structure):

  # Save a data frame 'my_dataset' to S3
  s3write_rds(my_dataset, 
              bucket = "your-target-s3-bucket", 
              object = "path/inside/bucket/my_data.rds")
  

• For CSV files (good for sharing with non-R users):

  s3write_csv(my_dataset, 
              bucket = "your-target-s3-bucket", 
              object = "path/inside/bucket/my_data.csv")
  

Option B: Upload existing local files to S3 #

If you already have a file saved in ~/Home/Myproject (like a CSV or Excel file), use put_object() to upload it:

put_object(file = "~/Home/Myproject/my_local_file.csv",
           bucket = "your-target-s3-bucket",
           object = "path/inside/bucket/my_uploaded_file.csv")

Step 1: Find your S3 website bucket path #

• Head to the AWS S3 Console, locate the bucket you're using to host your website. The bucket name (e.g., my-personal-website-bucket) is your base "path". If you're using a subdirectory inside the bucket (like docs), your full target path is s3://my-personal-website-bucket/docs.
• Double-check that your bucket is configured for static website hosting (you can set this up in the S3 Console under Properties > Static website hosting).

Step 2: Render your Rmarkdown site #

First, make sure you're in your project directory, then run the render command. By default, this will generate all site files in a _site folder:

# Navigate to your project root
setwd("~/Home/Myproject")

# Render the site from your .Rmd files
rmarkdown::render_site()

Step 3: Sync the generated site to S3 #

The sync_folder() function is perfect here—it only uploads files that have changed since your last sync, saving time:

# Sync the local '_site' folder to your S3 website bucket
sync_folder(from = "~/Home/Myproject/_site",
            to = "s3://your-website-bucket/path/to/site",
            region = "your-region",
            acl = "public-read") # Add this if your site needs public access

Key Notes: #

• Ensure your IAM role/user has permissions for s3:PutObject, s3:DeleteObject, and s3:ListBucket on the website bucket to enable syncing.
• The acl = "public-read" argument makes your site files accessible to the public—only use this if your website is meant to be public!

内容的提问来源于stack exchange，提问作者Sebastian Zeki