clone depth does what? Why do I care about this setting?

[NPxComplete]

This question is in reference to Atlassian Documentation: Configure bitbucket-pipelines.yml

Ask your question here...

[Philip Hodder]

Just to extend on what Johannes said: Bitbucket Pipelines uses a "shallow clone" by default to clone your repository into the build environment. If you have a large repository, a shallow clone will decrease the amount of time it takes for you repository to be cloned into the build environment. Typically you will only need to build/test recent commits, so there is no need to get the older ones.

Some deployment tools require the entire repository history in order to work. So the full clone feature exists to support these tools. There were also some issues with old tags not being in the clone (as they were more than 50 commits behind HEAD). So using a full clone (or a larger clone depth) means you can see all the tags in your repository. 

If you don't hit any issues at the moment in regards to cloning, it's likely you can stick with the default. 

Thanks

Phil

[screig]

Hi

I would like to change the depth number on cloning in my pipeline jobs. How can i do that?

Thanks

Sean

[Philip Hodder]

Hi Sean,

Try this:

clone:
  depth: 5 # or some other number of commits. You can also configure 'full' if you need the entire commit history.

pipelines:
   ... # standard pipelines stuff from here

 Look here for more details: https://confluence.atlassian.com/bitbucket/configure-bitbucket-pipelines-yml-792298910.html?_ga=2.243827725.922211975.1534723907-825437565.1515570924#Configurebitbucket-pipelines.yml-ci_cloneclone

[screig]

Great, thanks!

[Ihor Sviziev]

Hi @Philip Hodder,

Is it possible to define "depth" only for specific steps? I have one step that requires depth: full, but all another steps doesn't. How can I  do that?

[Shuhrat Dehkanov]

@Ihor Sviziev

Adding `git fetch --unshallow` to that specific step's script block worked for me. E.g.,:

tags:
  release/*:
    - step:
        script:
          - git fetch --unshallow
          # moar commands

Cc: @Philip Hodder 

[Philip Hodder]

Whoops. I missed the question.  @Ihor Sviziev not at the moment. I suggest you open a feature request here for us to track: https://bitbucket.org/site/master/issues/new

[Ihor Sviziev]

@Philip Hodder issue created. https://bitbucket.org/site/master/issues/18006/define-clone-depth-only-for-specific

Thank you!

[Mateusz Przybyłek]

You can use it directly in step

    - step: &dummy
        name: Dummy step
        clone:
          depth: 2
        script:
          - echo "Hello, I am pointless"

[Marc Wüest]

If I shallow-clone (e.g. depth 1) and later realize I need a full clone, would I then start over or can I 'complete' the shallow clone to a full clone?

[Johannes Kilian]

"Clone depth" is a feature of git to reduce server load: Instead of cloning the complete repository (as usually done with git), using clone depth just clones the last clone-depth-number revisions of your repository.

In literature this is also called "shallow clone"

For example See here: Git Beyond the Basics: Using shallow clones

[Johannes Kilian]

As I don't use either bitbucket-cloud nor bitbucket-pipelines, I cannot say what clone depth is used for in this context