We need dedicated pipeline for release branch and release tags. We also need documentation for whole release process.

We need to add unit tests co code related with pipeline objects.

As in the title. This is for keeping up with best Go practices

We should also build binaries for macOS in CI pipelines.

This command needs to highlight how local and Spinnaker resources differ.

Is your feature request related to a problem? Please describe.
We need to increase 5 seconds timeout in Floodgate config so we need to convert it from static to parametrical. If you have too many applications and pipelines, default timeout duration may cause api errors.

Describe the solution you'd like
Add Timeout parameter in config struct
type Config struct { Endpoint string json:"endpoint"Timeout int64 json:"timeout"Insecure bool json:"insecure"Auth auth.Configjson:"auth"Libraries []string json:"libraries"Resources []string json:"resources" }

Change Timeout from static to parametric in http.Client
var gateHTTPClient = &http.Client{ Timeout: time.Duration(floodgateConfig.Timeout) * time.Second, Transport: &http.Transport{ TLSClientConfig: &tls.Config{InsecureSkipVerify: floodgateConfig.Insecure}, }, Jar: cookieJar, }

I want to make this change a new branch and create a pull request then.

ATM floodgate needs connection to spinnaker to render json files by floodgate render command.

Steps to reproduce the behaviour:

1. Make sure you don't have configured connection to spinnaker in your floodgate config.
2. Execute floodgate render command on exaples directory of this repo.
3. Nothing is generated and no error is given.

Expected behaviour
Json files should be generated or there should be error informing that spinnaker connection is required to generate json files.

Currently supported Spinnaker versions are

• 1.22.x
• 1.21.x
• 1.20.x

Is your feature request related to a problem? Please describe.
There is no Pull requests template.

Describe the solution you'd like
Add PULL_REQUEST_TEMPLATE.md.

Describe alternatives you've considered
No.

Additional context
Add any other context or screenshots about the feature request here.

Basic flow implementation for all resources:

(start) -> read local -> isChanged? -> update remote -> isChanged? -> (stop)

Sometimes there is situation you are testing changes on separate instance of spinnaker, or you would like to hydrate only one template / diff or sync just one pipeline.

I think it would be useful to be able to set such options like --endpoint or if at the end of command are some dirs/files just hydrate/diff/sync/render those files.
But generally speaking it's about possibility to override config options in cli command.

Currently there is to few information about how to start using floodgate for example using binary only.

Floodgate should support Spinnaker project management.

floodgate sync --dry-run should show changes in resources, but shouldn't propagate them to Spinnaker

Implement a command, ie. floodgate download --project project_name --output-dir output_dir. Floodgate would download all resources related to provided project and save them in appropriate subdirectories in provided output directory, e.g. project_name/applications, project_name/pipelines etc.

Also, it would be nice if Floodgate could take care of unwanted fields (e.g. update timestamp) while downloading and saving the resources.

https://app.circleci.com/pipelines/github/codilime/floodgate/784/workflows/4f9815e0-20aa-42d8-88fb-425479fbd1dd/jobs/3486

#!/bin/bash -eo pipefail
echo "Print version using version flag"
/floodgate/bin/floodgate --version
echo "Print version using version command"
/floodgate/bin/floodgate version
echo "Comare changes - clean Spinnaker"
/floodgate/bin/floodgate -q --config ~/floodgate.yaml compare && exit 1 || echo "Found changes"
echo "Apply local resources"
/floodgate/bin/floodgate -q --config ~/floodgate.yaml sync
echo "Compare changes - synced resources"
/floodgate/bin/floodgate -q --config ~/floodgate.yaml compare

Print version using version flag
version 0.3.0-pull/126
Print version using version command
Version:		0.3.0-pull/126
Gate API version:	1.21.x
Go version:		1.14.4
Git commit:		9099505e23301e5e55cdf1dfe00f645ea0d0f806
Built:			2020-10-05_20:25:28
Comare changes - clean Spinnaker
FATA[0007] unexpected end of JSON input                 
Found changes
Apply local resources
Compare changes - synced resources

CircleCI received exit code 0

When there is such error Floodgate should exit with different code than 1 because 1 is used when there is difference between local and remote state.

Floodgate should be able to authenticate using OAuth2 authentication.
This method is required if Spinnaker is configured with Okta or GitHub.

Floodgate should be able to authenticate using X.509 authentication as it is one of standard authentication mechanism used in Spinnaker.

Implement tagging in pipeline and pipeline-templates.
This will allow us to test multiple versions of pipelines on one Spinnaker instance.
https://www.spinnaker.io/guides/user/tagging/

It would be nice to have an option to trigger pipeline for example using webhook and be able to track status execution using Floodgate.

Currently I need to use curl to trigger webhook, and then with combination of spin and curl command I'm able to track execution status.
I believe that Floodgate could do this on its own.

It could look like this:

$ floodgate execute deploy-nging-webhook -p parameters.json
triggering 'deploy-nging-webhook'
waiting for execution id................
execution id is 12345463DCAF
execution status: RUNNING.....................
pipeline finished with status: FAILED
$ echo $?
1

Currently we have implemented diff using jsondiff. It works but is far from perfect. We should find better library or implement it by ourselves.

It would be nice to create and publish docker image for Floodgate.
Docker image should be published in some public repository like docker hub or Quay.io

We should be able to build Floodgate using Gate API for different Spinnaker versions.

With this feature Floodgate will be able to apply independent parts in the same time.
This should reduce total time required for applying configuration in big deployments.

--quiet or -q flag should squelch non-essential output

floodgate hydrate command should allow to create pipelines based on pipeline templates and pipeline configurations.

floodgate inspect should fetch and show state of the resources from Spinnaker

Currently messages about loading files are set to level INFO, but they should probably be on DEBUG and should contain full path to the loaded file (maybe also the "detected type" - JSON/YAML/JSONNET)

Applications are dumped into logs but they shouldn't be there.

Describe the bug
When we synth our resources, Floodgate makes changes to our existing resources but it never creates our new resources. When we investigate, we should make change on spinnaker resource directory as like that:

if api != nil { if err := a.LoadRemoteState(api); err != nil { err := a.SaveLocalState(api) return err } }

If loadRemoteState has an error (does not exist error) local state should be saved on remote. This is quick solution. If you want to resolve this issue on me. I can do clean solution.

We need to create some kind of CLI interface based on https://github.com/spf13/cobra

Interface should have at least:

• dry-run option
• print diff option
• print raw and hydrated pipeline
• save to spinnaker option
• print states from spinnaker

floodgate render should allow to render and store JSON objects rendered from Jsonnet files from directories provided in configuration file

Gate API definitions can be different for different Spinnaker versions. We need to generate automatically swagger code in CI according to pinned Spinnaker version.

As in the topic...

We should be able to build Floodgate with all supported Spinnaker gateapi versions. Usually it is current version and two previous versions.

Is your feature request related to a problem? Please describe.
I use macOS but the artifacts in GitHub Releases are not executable.

Describe the solution you'd like
Release a .tar.gz that contains the binary of the release by using goreleaser.

Describe alternatives you've considered
Release the .tar.gz manually.

Additional context
Add any other context or screenshots about the feature request here.

Parser should return proper objects.