This docker image is for serving brat via a full web server using Apache. As the official website said:
For security reasons, we strongly recommend serving brat via a full web server such as Apache in production environments.
- Not create volume like the author do (especially for Mac/OS X), instead use a folder in the Host
**Reason: **if usedocker volume create
to create a volume, you have to usedocker volume inspect
to show the local storage in your PC, like the following:
[
{
"CreatedAt": "2019-11-14T03:24:19Z",
"Driver": "local",
"Labels": {},
"Mountpoint": "/var/lib/docker/volumes/brat_data/_data",
"Name": "brat_data",
"Options": {},
"Scope": "local"
}
]
you think the data is stored in /var/lib/docker/volumes/brat_data/_data
? NO!
reference to https://stackoverflow.com/questions/38532483/where-is-var-lib-docker-on-mac-os-x,
this folder just resides on the Docker host VM (the Alpine Linux ran by xhyve). You can run
screen ~/Library/Containers/com.docker.docker/Data/vms/0/tty
then, press Enter
to connect the Mac VM, and you can reach the data in /var/lib/docker/volumes/brat_data/_data
Now, you can use copy
to copy the data to this folder.
Some commands you may need during the process:
- disconnect that session but leave it open in background
Ctrl-a d
- list that session that's still running in background
screen -ls
- reconnect to that session (don't open a new one, that won't work and 2nd tty will give you garbled screen)
screen -r
- kill this session (window) and exit
Ctrl-a k
NOTE:
If you want to use a local folder, and you are using docker in OS X
, you should
configure shared paths from Docker -> Preferences... -> File Sharing
, add the full
path of the folder you want to mount, and when you run the image, you should set
-v $FULL_PATH:VOLUME_IN_CONTAINER:consistent
, see it in Run
docker: Error response from daemon: Mounts denied:
The paths /brat_cfg and /brat_data
are not shared from OS X and are not known to Docker.
You can configure shared paths from Docker -> Preferences... -> File Sharing.
- Not use
brat_cfg
for user configuration, because we only use one user, which can be set when run the image, changebrat_crf
to store project configuration.
- modify
Dockerfile
, addline 36
, commentline 51
- modify
brat_install_wrapper.sh
, don't runuser_patch.py
I am no longer doing anything with brat and am not maintaining this at all.
This is a docker container deploying an instance of brat.
There are two ways to create the mounted folder:
- You will need two volumes to pass annotation data and user configuration to the container. Start by creating a named volume for each of them like this:
docker volume create --name brat_data
docker volume create --name brat_cfg
- You can create two folder pass annotation data and user configuration to the container.
mkdir brat_data
mkdir brat_cfg
The folder brat_data
should be linked to your annotation data, the folder brat_cfg
should be linked to your annotation project configurations.
and the brat-cfg
should contain a file called users.json
.
To add multiple users to the server use users.json
to list your users and their passwords like so:
{
"user1": "password",
"user2": "password",
...
}
The data in these directories will persist even after stopping or removing the container.
You can then start another brat container as above and you should see the same data.
Note that if you are using docker < 1.9.0
named volumes are not available and
you'll have to use a data-only container and --volumes-from
instead.
You can also add data and edit the users from within the container. To add some data directly inside the container do something like:
$ docker run --name=brat-tmp -it -v brat_data:/bratdata cassj/brat /bin/bash
$ cd /bratdata
$ wget http://my.url/some-dataset.tgz
$ tar -xvzf some-dataset.tgz
$ exit
$ docker rm brat-tmp
Or, if you have data on the host machine, you can just copy the data into there from your host.
Build a brat image named brat:v0.1
docker build -t brat:v0.1 .
To run the container you need to specify a username, password and email address for BRAT as environment variables
when you start the container. This user will have editor permissions.
According to Preparation, there are two ways to create mounted
folder, so there are also two ways to run the image:
- run:
docker run --name=brat -d -p 80:80 -v brat_data:/bratdata -v brat_cfg:/bratcfg -e BRAT_USERNAME=brat -e BRAT_PASSWORD=brat -e [email protected] brat:v0.1
- Firstly, configure the shared paths from
Docker -> Preferences... -> File Sharing
, add the full path of the folder you want to mount, then run:
docker run --name=brat -d -p 80:80 -v $FULL_PATH_OF_brat_data:/bratdata -v $FULL_PATH_OF_brat_cfg:/bratcfg -e BRAT_USERNAME=brat -e BRAT_PASSWORD=brat -e [email protected] brat:v0.1