Contributing
Adding images
Anyone can contribute images to this repository by submitting a pull request at the github repo.
Step 1: adding images to images
branch
The following steps should be followed to add an image/imageset to the repository:
-
Check the images for copyright or license issues before adding images.
-
Do
git checkout images
in your local repository folder.Pkg.dir("TestImages")
gives the location of the repo. -
Add the image locally to the
images/
folder on your machine. -
If you have metadata, a more common name, a url for the source or an author, you can add them to the
images/metadata.toml
. -
Do
git add --all
to stage the changes for a commit. -
git commit -m "Adds <filename> to the repository from link <link>"
-
git push <fork> images
-
Now submit a pull request to the
images
branch. Once accepted, the image or imageset will be stored athttps://github.com/JuliaImages/TestImages.jl/blob/images/images/<filename>
.
Now that the images are added to the repository, maintainers will then create a tag(not release) v*-artifacts (e.g., v1.5.0-artifacts
) for images
branch and push it to the repo. After that, github action CI defined in images
branch will be triggered, build and release an tarball artifact for this commit.
Step 2: update master
branch
The next step is to modify the source files of the TestImages.jl
package to make them available for download, i.e., add it to remotefiles
list.
-
Do
git checkout master
to return to themaster
branch. -
Modify
src/TestImages.jl
: Add an entry with the filename to the remotefiles dictionary. -
Modify
Artifacts.toml
: updates[images]
section with the newly released artifacts. -
Do
git add --all
to add the changed source files. -
git commit -m "Adds <filename> to package"
-
git push <fork> master
-
Now submit a pull request to the
master
branch. Once accepted, the image or imageset will be available for download to users ofTestImages.jl
.
See PR#123 for an example.
Others
Feel free to open issues or pull-requests. If you have any questions, you can ask in Discourse or #image-processing
channel in Slack.