This new configuration parameter causes paths matching "<dir>/index.html" to be stored as "<dir>/" remotely. This simplifies the cloud configuration needed for some use cases, such as CloudFront distributions with S3 bucket origins. Before this change, users must configure their S3 buckets as public websites (which is incompatible with certain authentication / authorization schemes), or users must add a CloudFormation function to add index.html to the end of incoming requests. After this change, users can simply use an ordinary CloudFront distribution (no additional code) with an ordinary S3 bucket origin (and not an S3 website). This adds tests to ensure that functionality like matchers is unaffected by this change. I have also tested that the functionality works as expected when deploying to a real S3 / CloudFront website. Closes #12607
10 KiB
title | description | categories | keywords | menu | weight | toc | |||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Hugo Deploy | Upload your site to GCS, S3, or Azure |
|
|
|
20 | true |
You can use the "hugo deploy" command to upload your site directly to a Google Cloud Storage (GCS) bucket, an AWS S3 bucket, and/or an Azure Storage container.
Assumptions
- You have completed the Quick Start or have a Hugo website you are ready to deploy and share with the world.
- You have an account with the service provider (Google Cloud, AWS, or Azure) that you want to deploy to.
- You have authenticated.
- Google Cloud: Install the CLI and run
gcloud auth login
. - AWS: Install the CLI and run
aws configure
. - Azure: Install the CLI and run
az login
. - NOTE: Each service supports alternatives for authentication, including using environment variables. See here for more details.
- Google Cloud: Install the CLI and run
- You have created a bucket to deploy to. If you want your site to be
public, be sure to configure the bucket to be publicly readable as a static website.
- Google Cloud: create a bucket and host a static website
- Amazon S3: create a bucket and host a static website
- Microsoft Azure: create a storage container and host a static website
Configuring your first deployment
In the configuration file for your site, add a [deployment]
section
and a [[deployment.targets]]
subsection. The only required parameters are
the name and URL:
[deployment]
[[deployment.targets]]
# An arbitrary name for this target.
name = "production"
# URL specifies the Go Cloud Development Kit URL to deploy to. Examples:
URL = "<FILL ME IN>"
# Google Cloud Storage -- see https://gocloud.dev/howto/blob/#gcs
#URL = "gs://<Bucket Name>"
# Amazon Web Services S3; see https://gocloud.dev/howto/blob/#s3
# For S3-compatible endpoints, see https://gocloud.dev/howto/blob/#s3-compatible
#URL = "s3://<Bucket Name>?region=<AWS region>"
# Microsoft Azure Blob Storage; see https://gocloud.dev/howto/blob/#azure
#URL = "azblob://$web"
Deploy
To deploy to a target:
hugo deploy [--target=<target name>]
The deploy process recursively walks through your local publish directory
(public
by default) and syncs it to the destination bucket, to ensure
that the local and remote contents match.
If you don't specify a target, Hugo will deploy to the first target in your configuration.
See hugo help deploy
or the deploy command-line documentation for more command-line options.
How the file list works
The first thing hugo deploy
does is create file lists for local and remote by
traversing the local publish directory and remote bucket.
For both local and remote, the file list includes and excludes files according to the deployment target's configuration --
- If the configuration specifies an
include
pattern, all files are skipped by default except those matching the pattern. - If the configuration specifies an
exclude
pattern, files matching the pattern are skipped.
{{% note %}}
When creating the local file list, a few additional skips apply: first, Hugo always
skips files named .DS_Store
.
Second, Hugo always skips local hidden directories
(directories with names starting with a period, e.g. .git
) and does not
traverse into them, except for the special hidden directory named
.well-known
, which is
traversed if it exists.
{{% /note %}}
How the local and remote file lists are compared
In the second step, Hugo compares the two file lists to figure out what changes actually need to be made on the remote. File names are compared first; if the local and remote files both exist then the sizes and md5sums are compared. Any difference means that the file will be (re-)uploaded.
Specifying the --force
flag will ensure all files are re-uploaded even
if Hugo cannot detect any differences between local and remote.
Files are deleted from the remote bucket if they are not present in the local file list.
{{% note %}} If a remote file is excluded from the file list generation using the exclude/include configs, then the comparison step will not know to delete the file -- so it will remain on the remote even if it isn't present locally. {{% /note %}}
If the --confirm
or --dryRun
flags are given, Hugo displays
what differences it has found and either pauses or stops here.
How synchronization works
Hugo applies the list of changes to the remote storage bucket. Missing and/or changed files are uploaded, and files missing locally but present remotely are deleted. As files are uploaded, their headers are also configured on the remote according to the matchers configuration.
{{% note %}}
As a safety measure to help prevent accidents, if there are more than 256 files
to delete, Hugo won't delete any files from the remote. Use the --maxDeletes
command line flag to override this.
{{% /note %}}
Advanced configuration
Here's a full example deployment configuration:
[deployment]
# By default, files are uploaded in an arbitrary order.
# If you specify an `order` list, files that match regular expressions
# in this list will be uploaded first, in the specified order.
order = [".jpg$", ".gif$"]
[[deployment.targets]]
# Define one or more targets, e.g., staging and production.
# Each target gets its own [[deployment.targets]] section.
# An arbitrary name for this target.
name = "mydeployment"
# The Go Cloud Development Kit URL to deploy to. Examples:
URL = "<FILL ME IN>"
# GCS; see https://gocloud.dev/howto/blob/#gcs
#URL = "gs://<Bucket Name>"
# S3; see https://gocloud.dev/howto/blob/#s3
# For S3-compatible endpoints, see https://gocloud.dev/howto/blob/#s3-compatible
#URL = "s3://<Bucket Name>?region=<AWS region>"
# Azure Blob Storage; see https://gocloud.dev/howto/blob/#azure
#URL = "azblob://$web"
# You can use a "prefix=" query parameter to target a subfolder of the bucket:
#URL = "gs://<Bucket Name>?prefix=a/subfolder/"
# If you are using a CloudFront CDN, deploy will invalidate the cache as needed.
#cloudFrontDistributionID = "<FILL ME IN>"
# Include or exclude specific files when deploying to this target:
# If exclude is non-empty, and a local or remote file's path matches it, that file is not synced.
# If include is non-empty, and a local or remote file's path does not match it, that file is not synced.
# Note: local files that don't pass the include/exclude filters are not uploaded to remote,
# and remote files that don't pass the include/exclude filters are not deleted.
#
# The pattern syntax is documented here: https://godoc.org/github.com/gobwas/glob#Glob
# Patterns should be written with forward slashes as separator.
#
#include = "**.html" # would only include files with ".html" suffix
#exclude = "**.{jpg, png}" # would exclude files with ".jpg" or ".png" suffix
# Map any file named "<dir>/index.html" to the remote file "<dir>/". This does
# not affect the root "index.html" file, and it does not affect matchers below.
# This works when deploying to key-value cloud storage systems, such as Amazon
# S3 (general purpose buckets, not directory buckets), Google Cloud Storage, and
# Azure Blob Storage. This makes it so the canonical URL will match the object
# key in cloud storage, except for the root index.html file.
#
#stripIndexHTML = true
#######################
[[deployment.matchers]]
# Matchers enable special caching, content type and compression behavior for
# specified file types. You can include any number of matcher blocks; the first one
# matching a given file pattern will be used.
# See https://golang.org/pkg/regexp/syntax/ for pattern syntax.
# Pattern searching is stopped on first match.
# This is not affected by stripIndexHTML, above.
pattern = "<FILL ME IN>"
# If true, Hugo will gzip the file before uploading it to the bucket.
# With many storage services, this will save on storage and bandwidth costs
# for uncompressed file types.
#gzip = false
# If true, Hugo always re-uploads this file even if size and md5 match.
# This is useful if Hugo isn't reliably able to determine whether to re-upload
# the file on its own.
#force = false
# Content-type header to configure for this file when served.
# By default this can be determined from the file extension.
#contentType = ""
# Cache-control header to configure for this file when served.
# The default is the empty string.
#cacheControl = ""
# Content-encoding header to configure for this file when served.
# By default, if gzip is True, this will be filled with "gzip".
#contentEncoding = ""
# Samples:
[[deployment.matchers]]
# Cache static assets for 1 year.
pattern = "^.+\\.(js|css|svg|ttf)$"
cacheControl = "max-age=31536000, no-transform, public"
gzip = true
[[deployment.matchers]]
pattern = "^.+\\.(png|jpg)$"
cacheControl = "max-age=31536000, no-transform, public"
gzip = false
[[deployment.matchers]]
# Set custom content type for /sitemap.xml
pattern = "^sitemap\\.xml$"
contentType = "application/xml"
gzip = true
[[deployment.matchers]]
pattern = "^.+\\.(html|xml|json)$"
gzip = true