Skip to content

Latest commit

 

History

History
848 lines (594 loc) · 35.5 KB

develop.md

File metadata and controls

848 lines (594 loc) · 35.5 KB

Develop ParaView with Git

This page documents how to develop ParaView using GitLab and Git. See the README for more information.

Git is an extremely powerful version control tool that supports many different "workflows" for individual development and collaboration. Here we document procedures used by the ParaView development community. In the interest of simplicity and brevity we do not provide an explanation of why we use this approach.

Initial Setup

Before you begin, perform initial setup:

  1. Register GitLab Access to create an account and select a user name.

  2. Fork ParaView into your user's namespace on GitLab.

  3. Follow the download instructions to create a local clone of the main ParaView repository.

    $ git clone --recursive https://gitlab.kitware.com/paraview/paraview.git ParaView
    

    The main repository will be configured as your origin remote.

  4. Run the developer setup script to prepare your ParaView work tree and create Git command aliases used below:

    $ ./Utilities/SetupForDevelopment.sh
    

    This will prompt for your GitLab username and configure a remote called gitlab to refer to your fork. It will also setup a data directory for you. No need to do anything else.

Quick Start Guide

This is a quick start guide so that you can start contributing to ParaView easily. To understand the process more deeply, you can jump to the workflow section.

Developement

Create a local branch for your changes:

git checkout -b your_branch

Make the needed changes in ParaView and use git locally to create logically separated commits. There is no strict requirements regarding git commit messages syntax but a good rule of thumb to follow is: General domain: reason for change, General domain being a class, a module , a specific system like build or CI.

git commit -m "General domain: Short yet informative reason for the change"

Build ParaView following the guide and fix any build warnings or issues that arise and seems related to your changes.

Bringing in VTK changes

If you are working on a change that is across both VTK and ParaView, you can work directly on the VTK that is inside your repository of ParaView. Just follow the steps in the VTK develop guide starting at the usage of SetupForDevelopment.sh script.

You can then develop in the VTK directory as if it was any VTK repository, where you can make changes, commit and push to your VTK fork.

If you want to test the integration of your VTK changes in ParaView continuous integration, commit and push your changes to your VTK fork, then commit the VTK submodule update with your changes in a separate commit in your ParaView branch. Your VTK commit will be found without issue.

Once your VTK changes are merged into VTK master, you can cleanup your history and commit the submodule update cleanly like this (this will remove local non commited changes and also rebase your branch on the last master):

cd VTK
git fetch origin
git checkout master
git pull
cd ../
git fetch origin
git rebase -i origin/master # delete any VTK submodule update commit
git submodule update
git bump VTK master

Testing

Every change and new features needs to be tested. In ParaView, there are mainly two types of tests. Python tests and XML tests. While both types of tests are as valid to add, XML tests should be preferred for standard feature tests when possible as they exercise the ParaView user interface as well as core features.

XML Tests

First, as XML tests are impacted by ParaView's user settings, it is recommanded to always run paraview in dry run (option --dr).

To add a XML test, use the Tools -> Record Test menu in ParaView. After specifying name for the XML test, a new window named Recording User Input will pop up. This one will record every action you will perfom in ParaView to test his feature.

It is also recommanded to lock the view size when we want to create a baseline, choose Tools -> Lock View Size Custom... and set it to a 400x400 window as it works well.

On top of recording actions, the most important feature in this window is the check mark button which allows you to:

  • record the value of a property in the Properties Panel.
  • save an intermediary baseline. (in that case, use TESTS_WITH_INLINE_COMPARES)
  • check any value in the spreadsheet view.

Advanced features are:

  • Continuous Flush: writes events to file as soon as they are recorded. Especially useful when user environment isn't stable.
  • Record Interaction Timings: record each pause between interactions as a pause event.

It is also possible to stop the test recording by using the pause button and add comments in the XML test.

Finally, you can save the XML test by using the Stop recording button.

You may need to edit this file slightly to make it work as expected, do not hesitate to look at other files.

Add the file to Client/ParaView/Testing/XML folder and list it in one of the test.XXX.cmake file. Those files are conditionally used, typically depending on a build option. Look at Client/ParaView/Testing/XML/CMakeLists.txt to find the appropriate one. By default, tests should be added in test.common.cmake.

There are different categories of tests depending on what you want to test :

  • DATA_WITH_BASELINES: for general use case, in most of the case the safest bet is this category.
  • TESTS_WITH_INLINE_COMPARES: for when you need to do multiple image comparison.
  • TESTS_WITHOUT_BASELINES: in some context, a valid XML test can be made without baseline.

You can then configure ParaView and run your test from the build directory and check that they pass:

cmake . && cmake --build .
ctest -VV -R yourTest

Python Tests

To add a Python test, write a pvpython script testing the feature, add it in Client/ParaView/Testing/Python folder and list it in Client/ParaView/Testing/Python/CMakeLists.txt. There are different categories of tests, but the safest bet is probably the paraview_add_test_python category.

cmake . && cmake --build .
ctest -VV -R yourTest

Upload

Push your changes to the gitlab fork that you created in the initial setup stage:

git push gitlab

Data

If your test uses new data or baselines, you will need to add it to your fork. For data, add the file names to the list in the corresponding Clients/ParaView/Testing/Python/CMakeLists.txt or Clients/ParaView/Testing/XML/test.common.cmake and drop the files in Testing/Data/. For baselines, just drop the file in Clients/ParaView/Testing/Data/Baseline/ and run the following commands from your build directory:

cmake . && cmake --build .

This will transform your files into .sha512 files. Check your test is passing by running from your build directory:

ctest -VV -R yourTest

If it passes, add these .sha512 files and commit them, then push with:

git gitlab-push

Create a Merge Request

Once you are happy with the state of your development on your fork, the next step is to create a merge request back into the main ParaView repository.

Open in a browser, select your branch in the list and create a Merge Request against master.

In the description, write an informative explanation of your added features or bugfix. If there is an associated issue, link it with the #number in the description.

Tag some ParaView maintainers in the description to ensure someone will see it, eg: @cory.quammen, @ben.boeckel or @mwestphal.

Robot Checks

Once the MR is created, our gitlab robot will check multiple things and make automated suggestions. Please read them and try to follow the instructions. The two standard suggestions are related to formatting errors and adding markdown changelog.

To fix the formatting, just add a comment containing:

Do: reformat

Then, once the robot has fixed the formatting, fetch the changes locally (this will remove any local changes to your branch)

git fetch gitlab
git reset --hard gitlab/your_branch

To fix the changelog warning, create, add, commit and push a markdown (.md) file in Documentation/release/dev folder. In this file, write a small markdown paragraph describing the development. See other .md files in this folder for examples. It may look like this:

## Development title

A new feature that does this and that has been introduced.
This specific issue has been fixed in this particular way.

Suggestions and best practices on writing the changelog can be found in the Documentation/release/dev/0-sample-topic.md file. This is an optional step but recommended to do for any new feature and user facing issues.

Reviews

ParaView maintainers and developers will review your MR by leaving comments on it. Try to follow their instructions and be patient. It can take a while to get a MR into mergeable form. This is a mandatory step, and it is absolutely normal to get change requests.

Review comments can be resolved, please resolve a comment once you've taken it into account and pushed related changes or once you've reached an agreement with the commenter that nothing should be changed.

Once a reviewer is happy with your changes, they will add a +X comment. You need at least one +2 or higher to consider merging the MR. Two +1s do not equal a +2. If a reviewer leave a -1 comment, please discuss with them to understand what is the issue and how it could be fixed.

Once you have pushed new changes, please tag reviewers again so that they can take a look. If you do not tag reviewers, they may not know to revisit your changes. Do not hesitate to tag them and ask for help.

Continuous Integration

Before merging a MR, the ParaView continuous integration (CI) needs to run and be green. For CI to be functional, please read and follow this guide. It was written for VTK but is as valid for ParaView.

To run the CI:

  • Click on the Pipelines Tab
  • Click on the last pipeline status badge
  • Press the Play all manual arrows on top of the Build and Test stages

Do not hesitate to tag a ParaView developer for help if needed.

You then need to wait for CI to run, it can take a while, up to a full day.

A successful CI should be fully green. If that is so, then your MR is ready !

If not, you need to analyze the issues and fix them. Recover the failure information this way:

Click on the pipelines tab, then on the last status badge, then on the cdash-commit job. It will take you to the related CDash report where you will find all information.

Everything in the CDash report should be green except the NotRun and Time column. Take a look into each issue and fix them locally. If there are issues in the pipeline but nothing is visible in the CDash, please ask a maintainer for help to figure out if anything should be done. You can always try to rerun the failed job by clicking on the arrow of the job in the pipeline.

Once you have fixed some issues locally, commit and push them to gitlab, run the CI again and tag reviewers again for follow-up reviews.

Merging

Once the MR has green CI and you have at least one +2, you can ask for a merge. Before that please make sure that:

  • Your commit history is logical (or squashed into a single commit) and cleaned up with good commit messages
  • You are rebased on a fairly recent version of master

If that is not the case, please rebase on master using the following commands:

git fetch origin
git rebase -i origin/master
git push gitlab -f

The interactive rebase will let you squash commits, reorganize commits and edit commit messages.

After the force push, make sure to run CI again.

Once all is done, tag a ParaView developer so that they can perform the merge command.

Congratulations ! You just contributed to ParaView !

Workflow

ParaView development uses a branchy workflow based on topic branches. Our collaboration workflow consists of three main steps:

  1. Local Development:

  2. Code Review (requires GitLab Access):

  3. Integrate Changes:

Update

  1. Update your local master branch:

    $ git checkout master
    $ git pullall
    
  2. Optionally push master to your fork in GitLab:

    $ git push gitlab master
    

    to keep it in sync. The git gitlab-push script used to Share a Topic below will also do this.

Create a Topic

All new work must be committed on topic branches. Name topics like you might name functions: concise but precise. A reader should have a general idea of the feature or fix to be developed given just the branch name. Additionally, it is preferred to have an issue associated with every topic. The issue can document the bug or feature to be developed. In such cases, being your topic name with the issue number.

  1. To start a new topic branch:

    $ git fetch origin
    

    If there is an issue associated with the topic, assign the issue to yourself using the "Assignee" field, and add the workflow:active-development label to it.

  2. For new development, start the topic from origin/master:

    $ git checkout -b my-topic origin/master
    

    For release branch fixes, start the topic from origin/release, and by convention use a topic name starting in release-:

    $ git checkout -b release-my-topic origin/release
    

    If subdmodules may have changed, the run:

    $ git submodule update
    
  3. Edit files and create commits (repeat as needed):

    $ edit file1 file2 file3
    $ git add file1 file2 file3
    $ git commit
    

    Caveats:

    • To add data follow these vtk instructions.
    • To add icons, Kitware's graphic designer may be able to help create an SVG icon.

    Commit messages must contain a brief description as the first line and a more detailed description of what the commit contains. If the commit contains a new feature, the detailed message must describe the new feature and why it is needed. If the commit contains a bug fix, the detailed message must describe the bug behavior, its underlying cause, and the approach to fix it. If the bug is described in the bug tracker, the commit message must contain a reference to the bug number.

  4. Update a submodule if needed

    If you need to update a submodule (eg: VTK) in order to access a specific bugfix or features, first make sure that the needed developments have been merged into the main branch of the submodule. You can then use the command:

     $ git bump my-submodule my-hash-or-branch
    

    my-submodule being the submodule folder (eg: VTK), my-hash-or-branch being either a hash or a branch provided by any of your remote or your local repository, typically, origin/master.

    This will add a new commit which update the submodule and prefill the commit message with information about the different commits in the submodule. Make sure to still add some information about the reason for the bump.

    Please note you can run CI on a submodule commit in another remote, see [continuous integration] for more info.

  5. Add some tests

    Every changes and new features need to be tested in ParaView. Depending on what you implement, you can perform an image comparison with a baseline of expected result or checking a property at anytime.

    They are mainly 2 types of test in ParaView, python and XML testing. Generally we use a XML test, for more details regarding this topic, it's highly recommanded to check this section.

  6. Add release notes

    Notable changes should create a new file in Documentation/release/dev/ named my-topic.md (replace my-topic with the name of your topic). This is not necessary for branches which are "trivial" such as fixing typos, updating test baselines, or are developer-oriented.

Share a Topic

When a topic is ready for review and possible inclusion, share it by pushing to a fork of your repository in GitLab. Be sure you have registered and signed in for GitLab Access and created your fork by visiting the main ParaView GitLab repository page and using the "Fork" button in the upper right.

  1. Checkout the topic if it is not your current branch:

    $ git checkout my-topic
    
  2. Check what commits will be pushed to your fork in GitLab:

    $ git prepush
    
  3. Push commits in your topic branch to your fork in GitLab:

    $ git gitlab-push
    

    Notes:

    • If you are revising a previously pushed topic and have rewritten the topic history, add -f or --force to overwrite the destination.
    • If the topic adds data see this note.
    • The gitlab-push script also pushes the master branch to your fork in GitLab to keep it in sync with the upstream master.

    The output will include a link to the topic branch in your fork in GitLab and a link to a page for creating a Merge Request.

Create a Merge Request

(If you already created a merge request for a given topic and have reached this step after revising it, skip to the next step.)

Visit your fork in GitLab, browse to the "Merge Requests" link on the left, and use the "New Merge Request" button in the upper right to reach the URL printed at the end of the previous step. It should be of the form:

https://gitlab.kitware.com/<username>/paraview/-/merge_requests/new

Follow these steps:

  1. In the "Source branch" box select the <username>/paraview repository and the my-topic branch.

  2. In the "Target branch" box select the paraview/paraview repository and the master branch. It should be the default.

    If your change is a fix for the release branch, you should still select the master branch as the target because the change needs to end up there too.

  3. Use the "Compare branches" button to proceed to the next page and fill out the merge request creation form.

  4. In the "Title" field provide a one-line summary of the entire topic. This will become the title of the Merge Request.

    Example Merge Request Title:

    Wrapping: Add Java 1.x support
    
  5. In the "Description" field provide a high-level description of the change the topic makes and any relevant information about how to try it.

    • Use @username syntax to draw attention of specific developers. This syntax may be used anywhere outside literal text and code blocks. Or, wait until the next step and add comments to draw attention of developers.
    • If your change is a fix for the release branch, indicate this so that a maintainer knows it should be merged to release.
    • Optionally use a fenced code block with type message to specify text to be included in the generated merge commit message when the topic is merged.

    Example Merge Request Description:

    This branch requires Java 1.x which is not generally available yet.
    Get Java 1.x from ... in order to try these changes.
    
    ```message
    Add support for Java 1.x to the wrapping infrastructure.
    ```
    
    Cc: @user1 @user2
    
  6. The "Assign to", "Milestone", and "Labels" fields may be left blank.

  7. Use the "Submit merge request" button to create the merge request and visit its page.

Guidelines

For guidelines with respect to licensing, naming conventions, code formatting, documentation and further, please check guidelines

Review a Merge Request

Add comments mentioning specific developers using @username syntax to draw their attention and have the topic reviewed. After typing @ and some text, GitLab will offer completions for developers whose real names or user names match.

Comments use GitLab Flavored Markdown for formatting. See GitLab documentation on Special GitLab References to add links to things like merge requests and commits in other repositories.

When a merge request is ready for review, developers can use the triage:ready-for-review to indicate the same to the reviewers. If reviewers deems that it needs more work, they can add the triage:needswork label. This can be repeated as many times as needed adding/removing labels as appropriate.

If a merge request is waiting on dashboards, use the triage:pending-dashboards label.

Human Reviews

Reviewers may add comments providing feedback or to acknowledge their approval.

If a reviewer raises an issue via a thread, the contributor may resolve the thread if the solution is indisputable. In case of cited lines within the issue, the corresponding changes of the commit prior to a comment are shown. So it is better to first push the change and then comment/resolve. When code is cited, the last line is usually the line of interest, the above lines just provide the context. Providing a comment just saying 'done' is unnecessary as GitLab will indicate that the relevant code has changed.

It shall be a common goal for reviewers and contributors to limit the amount of generated emails. Therefore reviewers are encouraged to comment systematic issues (e.g. missing this->) only once but to indicate the general application. Contributors do not need to comment on trivial fixes (e.g. typos in comments) but may simply solved the threads after fixing.

To raise the attention of individuals, e.g. a reviewer for another round of review, this person can be pinged by addressing via @ in a comment.

Special Comments

Lines of specific forms will be extracted during merging and included as trailing lines of the generated merge commit message:

The leading line of a comment may optionally be exactly one of the following votes followed by nothing but whitespace before the end of the line:

  • -1 or :-1: indicates "the change is not ready for integration".
  • +1 or :+1: indicates "I like the change". This adds an Acked-by: trailer to the merge commit message.
  • +2 indicates "the change is ready for integration". This adds a Reviewed-by: trailer to the merge commit message.
  • +3 indicates "I have tested the change and verified it works". This adds a Tested-by: trailer to the merge commit message.

The middle lines of a comment may be free-form GitLab Flavored Markdown.

Zero or more trailing lines of a comment may each contain exactly one of the following votes followed by nothing but whitespace before the end of the line:

  • Rejected-by: me means "The change is not ready for integration."
  • Acked-by: me means "I like the change but defer to others."
  • Reviewed-by: me means "The change is ready for integration."
  • Tested-by: me means "I have tested the change and verified it works."

Each me reference may instead be an @username reference or a full Real Name <user@domain> reference to credit someone else for performing the review. References to me and @username will automatically be transformed into a real name and email address according to your GitLab account profile.

Fetching Changes

One may fetch the changes associated with a merge request by using the git fetch command line shown at the top of the Merge Request page. It is of the form:

$ git fetch https://gitlab.kitware.com/$username/paraview.git $branch

This updates the local FETCH_HEAD to refer to the branch.

There are a few options for checking out the changes in a work tree:

  • One may checkout the branch:

    $ git checkout FETCH_HEAD -b $branch
    

    or checkout the commit without creating a local branch:

    $ git checkout FETCH_HEAD
    
  • Or, one may cherry-pick the commits to minimize rebuild time:

    $ git cherry-pick ..FETCH_HEAD
    

Robot Reviews

The "Kitware Robot" automatically performs basic checks including clang-format on the commits and adds a comment acknowledging or rejecting the topic. This will be repeated automatically whenever the topic is pushed to your fork again.

Automatic formatting can be triggered by adding a comment with a single line.

Do: reformat

A re-check may be explicitly requested by

Do: check

A topic cannot be merged until the automatic review succeeds. It is necessary to correct the specific commit via rebase.

Continuous Integration

ParaView uses GitLab CI to test merge requests, configured by the top-level .gitlab-ci.yml file. Results may be seen both on the merge request's pipeline page and on the ParaView CDash Page. Filtered CDash results showing just the pipeline's jobs can be reached by selecting the cdash-commit job in the External stage of the pipeline. Note that due to GitLab changes, the External stage may be in a separate pipeline for the same commit.

Lint build jobs run automatically after every push. Actual CI jobs require a manual trigger to run:

  • Merge request authors may visit their merge request's pipeline and click the "Play" button on one or more jobs manually. If the merge request has the "Allow commits from members who can merge to the target branch" check box enabled, ParaView maintainers may use the "Play" button too. When in doubt, it's a good idea to run a few jobs as smoke tests to catch early build/test failures before a full CI run that would tie up useful resources. Note that, as detailed below, a full CI run is necessary before the request can be merged.

  • When working simultaneously on ParaView and a submodule, e.g., VTK, it may be useful to run the CI before merging changes in the submodule in question. This is perfectly supported, just push your change on your remote for the submodule and update the submodule manually in a temporary commit, then run the CI. Make sure to remove this commit before the merge and use git bump when performing the actual submodule update.

  • ParaView GitLab Project Developers may trigger CI on a merge request by adding a comment with a command among the comment trailing lines:

      Do: test
    

    @kwrobot will add an award emoji to the comment to indicate that it was processed and also trigger all manual jobs in the merge request's pipeline.

    The Do: test command accepts the following arguments:

    • --named <regex>, -n <regex>: Trigger jobs matching <regex> anywhere in their name. Job names may be seen on the merge request's Pipelines tab.

    • --stage <stage>, -s <stage>: Only affect jobs in a given stage. Stage names may be seen on the merge request's Pipelines tab. Note that the names are determined by what is in the .gitlab-ci.yml file and may be capitalized in the web page, so lowercasing the webpage's display name for stages may be required.

    • --action <action>, -a <action>: The action to perform on the jobs. Possible actions:

      • manual (the default): Start jobs awaiting manual interaction.
      • unsuccessful: Start or restart jobs which have not completed successfully.
      • failed: Restart jobs which have completed, but without success.
      • completed: Restart all completed jobs.

If the merge request topic branch is updated by a push, a new manual trigger using one of the above methods is needed to start CI again. Currently running jobs will generally be canceled automatically.

Before the merge, all the jobs must be run and reviewed, see below.

If you have any question about the CI process, do not hesitate to ask a CI maintainer:

  • ben.boeckel
  • mwestphal

Reading CI Results

Reading CI results is a very important part of the merge request process and is the responsibility of the author of the merge request, although reviewers can usually help. There are two locations to read the results, GitLab CI and CDash. Both should be checked and considered clean before merging.

To read GitLab CI result, click on the Pipelines tab then on the last pipeline. It is expected to be fully green. If there is a yellow warning job, please consult CDash. If there is a red failed job, click on it to see the reason for the failure. It should clearly appears on the bottom of the log. Possible failures are:

  • Timeouts: please rerun the job and report to CI maintainers
  • Memory related errors: please rerun the job and report to CI maintainers
  • Testing errors: please consult CDash for more information, usually an issue in your code
  • Non disclosed error: please consult CDash, usually a build error in your code

To read the CDash results, on the job page, click on the "cdash-commit" external job which will open the commit-specific CDash page. Once it is open, make sure to show "All Build" at the bottom left of the page. CDash results display error, warnings, and test failures for all the jobs. It is expected to be green except for the "NoRun" and "Test Timings" categories, which can be ignored.

  • Configure warnings: there must not be any; to fix before the merge
  • Configure errors: there must not be any; to fix before the merge
  • Build warnings: there must not be any; to fix before the merge. If unrelated to your code, rerun the job and report to CI maintainers.
  • Build Errors: there must not be any; to fix before the merge. If unrelated to your code, report to CI maintainers.
  • NotRun test: ignore; these tests have self-diagnosed that they are not relevant on the testing machine.
  • Testing failure: there should not be any, ideally, to fix before the merge. If unrelated to your code, check the test history to see if it is a flaky test and report to CI maintainers.
  • Testing success: if your MR creates or modifies tests, please check that your test are listed there.
  • Test timings errors: can be ignored, but if it is all red, you may want to report it to CI maintainers.

To check the history of a failing test, on the test page, click on the "Summary" link to see a summary of the test for the day, then click on the date controls on the top of the page to go back in time. If the test fails on other MRs or on master, this is probably a flaky test, currently in the process of being fixed or excluded. A flaky test can be ignored.

As a reminder, here is our current policy regarding CI results. All the jobs must be run before merging. Configure warnings and errors are not acceptable to merge and must be fixed. Build warning and errors are not acceptable to merge and must be fixed. Testing failure should be fixed before merging but can be accepted if a flaky test has been clearly identified.

Revise a Topic

If a topic is approved during GitLab review, skip to the next step. Otherwise, revise the topic and push it back to GitLab for another review as follows:

  1. Checkout the topic if it is not your current branch:

    $ git checkout my-topic
    
  2. To revise the 3rd commit back on the topic:

    $ git rebase -i HEAD~3
    

    (Substitute the correct number of commits back, as low as 1.) Follow Git's interactive instructions.

  3. Return to the above step to share the revised topic.

Merge a Topic

Once review has concluded that the MR topic is ready for integration (at least one +2), authorized developers may add a comment with a single trailing line:

Do: merge

to ask that the change be merged into the upstream repository. By convention, do not request a merge if any -1 or Rejected-by: review comments have not been resolved and superseded by at least +1 or Acked-by: review comments from the same user.

Developers are encouraged to merge their own merge requests on review. However, please do not merge unless you are available to address any dashboard issues that may arise. Developers who repeatedly ignore dashboard issues following their merges may loose developer privileges to the repository temporarily (or permanently)!

Merge Success

If the merge succeeds the topic will appear in the upstream repository master branch and the Merge Request will be closed automatically. Any issues associated with the Merge Request will generally get closed automatically. If not, the developer merging the changes should close such issues and add a workflow:customer-review tag to the issue(s) addressed by the change. Reporters and testers can then review the fix. Try to add enough information to the Issue or the Merge Request to indicate how to test the functionality if not obvious from the original Issue.

Merge Failure

If the merge fails (likely due to a conflict), a comment will be added describing the failure. In the case of a conflict, fetch the latest upstream history and rebase on it:

$ git fetch origin
$ git rebase origin/master

(If you are fixing a bug in the latest release then substitute origin/release for origin/master.)

Return to the above step to share the revised topic.

Delete a Topic

After a topic has been merged upstream the Merge Request will be closed. Now you may delete your copies of the branch.

  1. In the GitLab Merge Request page a "Remove Source Branch" button will appear. Use it to delete the my-topic branch from your fork in GitLab.

  2. In your work tree checkout and update the master branch:

    $ git checkout master
    $ git pull
    
  3. Delete the local topic branch:

    $ git branch -d my-topic
    

    The branch -d command works only when the topic branch has been correctly merged. Use -D instead of -d to force the deletion of an unmerged topic branch (warning - you could lose commits).

Contribute VTK Changes

If you have any VTK changes, then you are required to get your changes incorporated into VTK using VTK's development workflow. Once your VTK topic has been approved and merged into VTK, then:

  1. Create a ParaView topic if you haven't already.

  2. Add your VTK topic head (or the latest VTK origin/master which includes your VTK topic head).

     $ cd VTK
     $ git checkout master
     $ cd ..
     $ git add VTK
     $ git commit
    
  3. Follow the merge process documented earlier.