Website screenshot microservice written in Golang that caches screenshots on disk or S3 storage
Compile and run with:
go build github.com/mkalus/goggler/cmd/goggler && GOGGLER_DEBUG=1 ./goggler
Point your Browser to the URL: http://localhost:8080/?url=https%3A%2F%2Fduckduckgo.com%2F&wait=2000
Note: Goggler uses ChromeDP and needs Chrome or Chromium to work. This means it has to be able to access and start Chrome or Chromium somehow (Chrome in path variable or the like).
Valid URI parameters:
url
URL to take screenshot from (required)width
Width of image and viewport (might be scaled up or down, see below, default: 1920)height
Height of image and viewport (might be scaled up or down, see below, default: 1024)scale
Scale of image and viewport (final width and height are multiplied by this, default: 0.2)quality
PNG image quality (default: 90)wait
Time in ms to wait until screenshot is taken (to wait for rendering to finish, default: 2000 = 2 secs)timeout
Timeout in ms after which request is cancelled (default: 60000 = 60 secs)maxage
Maximum age of cache file in seconds (!) before it gets reloaded (set to 0 to never renew files, default: 2592000 = 30 days)force
Force update (set to1
or any other non-empty value)
Defaults may be changed by setting environmental variables when running goggle.
Changes defaults and sets some other elements:
GOGGLER_WIDTH
Set default image widthGOGGLER_HEIGHT
Set default image heightGOGGLER_SCALE
Set default image scaleGOGGLER_QUALITY
Set default image qualityGOGGLER_WAIT
Set default wait timeGOGGLER_TIMEOUT
Set default timeout timeGOGGLER_MAXAGE
Set default max age timeGOGGLER_LISTEN
Set default listen address (default::8080
)GOGGLER_DEBUG
Enable debugging logGOGGLER_CACHE
Type of cache (local
ors3
, defaultlocal
)GOGGLER_CACHE_CLEANUP_INTERVAL
Interval in seconds at which cleanup service is run to clean stale data (maxage has to be greater than 0, also: set to 0 to never clean up old files, s3 needs full days to work properly, default: 2592000 = 30 days)GOGGLER_CACHE_LOCAL_PATH
Path to local cache (default: OS specific temp dir like/tmp/goggler
)GOGGLER_CACHE_S3_URL
Endpoint for S3 storage (e.g.s3.amazonaws.com
)GOGGLER_CACHE_S3_BUCKETNAME
Bucket nameGOGGLER_CACHE_S3_ACCESSKEY
Access Key ID for S3 storageGOGGLER_CACHE_S3_SECRETKEY
Secret Access Key ID for S3 storageGOGGLER_CACHE_S3_REGION
S3 region (might be skipped in Amazon, e.g.us-east-1
)GOGGLER_CACHE_S3_SKIPSSL
Set to any value to skip secure SSL/TLS connectionGOGGLER_CACHE_S3_CREATEBUCKET
Set to any value to create bucket if it does not exist
Examples:
# Basic test
GOGGLER_DEBUG=1 GOGGLER_LISTEN=127.0.0.1:9090 ./goggler
# Local storage
GOGGLER_CACHE_LOCAL_PATH=~/mydata ./goggler
# S3 storage
GOGGLER_DEBUG=1 GOGGLER_CACHE=s3 GOGGLER_CACHE_S3_BUCKETNAME=mytestbucket \
GOGGLER_CACHE_S3_ACCESSKEY=_KEY_ GOGGLER_CACHE_S3_SECRETKEY=_KEY_ ./goggler
# Minio storage
GOGGLER_DEBUG=1 GOGGLER_CACHE=s3 GOGGLER_CACHE_S3_URL=127.0.0.1:9000 \
GOGGLER_CACHE_S3_BUCKETNAME=test GOGGLER_CACHE_S3_ACCESSKEY=minioadmin \
GOGGLER_CACHE_S3_SECRETKEY=minioadmin GOGGLER_CACHE_S3_SKIPSSL=1 \
GOGGLER_CACHE_S3_CREATEBUCKET=1 ./goggler
There is a Docker container of goggler including a headless version of Chromium. Try it using:
docker run --rm -p8080:8080 --init ronix/goggler
Important: The --init
option is needed to get rid of zombie processes that will spawn if you run the container.
Full example with persistent volume:
docker run -d -p8080:8080 -v /tmp/goggler:/tmp/goggler \
--name goggler --init ronix/goggler
Full example with S3 storage:
docker run -d -p8080:8080 --name goggler -e "GOGGLER_CACHE=s3" \
-e "GOGGLER_CACHE_S3_BUCKETNAME=mytestbucket" \
-e "GOGGLER_CACHE_S3_ACCESSKEY=_KEY_" \
-e "GOGGLER_CACHE_S3_SECRETKEY=_KEY_" \
--init ronix/goggler
Building Docker image from this source:
docker build -t goggler .