* Add ans implementation
* Remove todo comment
* Rename test function
Co-authored-by: Linda Siebert <39100394+LindaSieb@users.noreply.github.com>
* Better wording
Co-authored-by: Linda Siebert <39100394+LindaSieb@users.noreply.github.com>
* Add reading of response body function
* Use http pkg ReadResponseBody
* Check read error
* Better test case description
* Fix formatting
* Create own package for read response body
* Omit empty nested resource struct
* Separate Resource struct from Event struct
* Merge and unmarshall instead of only unmarshalling
* Improve status code error message
* Remove unchangeable event fields
* Separate event parts
* Change log level setter function
* Restructure ans send test
* Revert exporting readResponseBody function
Instead the code is duplicated in the xsuaa and ans package
* Add check correct ans setup request
* Add set options function for mocking
* Review fixes
* Correct function name
* Use strict unmarshalling
* Validate event
* Move functions
* Add documentation comments
* improve test
* Validate event
* Add logrus hook for ans
* Set defaults on new hook creation
* Fix log level on error
* Don't alter entry log level
* Set severity fatal on 'fatal error' log message
* Ensure that log entries don't affect each other
* Remove unnecessary correlationID
* Use file path instead of event template string
* Improve warning messages
* Add empty log message check
* Allow configuration from file and string
* Add sourceEventId to tags
* Change resourceType to Pipeline
* Use structured config approach
* Use new log level set function
* Check correct setup and return error
* Mock http requests
* Only send log level warning or higher
* Use new function name
* One-liner ifs
* Improve test name
* Fix tests
* Prevent double firing
* Reduce Fire test size
* Add error message to test
* Reduce newANSHook test size
* Further check error
* Rename to defaultEvent in hook struct
* Reduce ifs further
* Fix set error category test
The ansHook Fire test cannot run in parallel, as it would affect the
other tests that use the error category.
* Change function name to SetServiceKey
* Validate event
* Rename to eventTemplate in hook struct
* Move copy to event.go
* Fix function mix
* Remove unnecessary cleanup
* Remove parallel test
The translation fails now and again when parallel is on.
* Remove prefix test
* Remove unused copyEvent function
* Fix ifs
* Add docu comment
* Register ans hook from pkg
* register hook and setup event template seperately
* Exclusively read eventTemplate from environment
* setupEventTemplate tests
* adjust hook levels test
* sync tests- wlill still fail
* migrate TestANSHook_registerANSHook test
* fixes
* Add ans send event step
* Fix tests
* Add groovy wrapper
* Add groovy wrapper test
* Fix function names
* Reduce ifs
* Fix docu
* We always set the timestamp
* Validate event
* Test unknown field in json
* Make test list test
* Set all event fields as separate parameters
* Generate and fix code
* Review fixes
* Format test file
* Format go code
* Fix common steps tests
* Print event to console if verbose
Co-authored-by: Linda Siebert <39100394+LindaSieb@users.noreply.github.com>
Co-authored-by: Roland Stengel <r.stengel@sap.com>
* Add ans implementation
* Remove todo comment
* Rename test function
Co-authored-by: Linda Siebert <39100394+LindaSieb@users.noreply.github.com>
* Better wording
Co-authored-by: Linda Siebert <39100394+LindaSieb@users.noreply.github.com>
* Add reading of response body function
* Use http pkg ReadResponseBody
* Check read error
* Better test case description
* Fix formatting
* Create own package for read response body
* Omit empty nested resource struct
* Separate Resource struct from Event struct
* Merge and unmarshall instead of only unmarshalling
* Improve status code error message
* Remove unchangeable event fields
* Separate event parts
* Change log level setter function
* Restructure ans send test
* Revert exporting readResponseBody function
Instead the code is duplicated in the xsuaa and ans package
* Add check correct ans setup request
* Add set options function for mocking
* Review fixes
* Correct function name
* Use strict unmarshalling
* Validate event
* Move functions
* Add documentation comments
* improve test
* Validate event
* Add logrus hook for ans
* Set defaults on new hook creation
* Fix log level on error
* Don't alter entry log level
* Set severity fatal on 'fatal error' log message
* Ensure that log entries don't affect each other
* Remove unnecessary correlationID
* Use file path instead of event template string
* Improve warning messages
* Add empty log message check
* Allow configuration from file and string
* Add sourceEventId to tags
* Change resourceType to Pipeline
* Use structured config approach
* Use new log level set function
* Check correct setup and return error
* Mock http requests
* Only send log level warning or higher
* Use new function name
* One-liner ifs
* Improve test name
* Fix tests
* Prevent double firing
* Reduce Fire test size
* Add error message to test
* Reduce newANSHook test size
* Further check error
* Rename to defaultEvent in hook struct
* Reduce ifs further
* Fix set error category test
The ansHook Fire test cannot run in parallel, as it would affect the
other tests that use the error category.
* Change function name to SetServiceKey
* Validate event
* Rename to eventTemplate in hook struct
* Move copy to event.go
* Fix function mix
* Remove unnecessary cleanup
* Remove parallel test
The translation fails now and again when parallel is on.
* Remove prefix test
* Remove unused copyEvent function
* Fix ifs
* Add docu comment
* Register ans hook from pkg
* register hook and setup event template seperately
* Exclusively read eventTemplate from environment
* setupEventTemplate tests
* adjust hook levels test
* sync tests- wlill still fail
* migrate TestANSHook_registerANSHook test
* fixes
* review - cleanup, reuse poke
* Apply suggestions from code review
* Change subject
* Review fixes
* Set stepName 'n/a' if not available
* Fix fire tests
Co-authored-by: Linda Siebert <39100394+LindaSieb@users.noreply.github.com>
Co-authored-by: Roland Stengel <r.stengel@sap.com>
* Add ans implementation
* Remove todo comment
* Rename test function
Co-authored-by: Linda Siebert <39100394+LindaSieb@users.noreply.github.com>
* Better wording
Co-authored-by: Linda Siebert <39100394+LindaSieb@users.noreply.github.com>
* Add reading of response body function
* Use http pkg ReadResponseBody
* Check read error
* Better test case description
* Fix formatting
* Create own package for read response body
* Omit empty nested resource struct
* Separate Resource struct from Event struct
* Merge and unmarshall instead of only unmarshalling
* Improve status code error message
* Remove unchangeable event fields
* Separate event parts
* Change log level setter function
* Restructure ans send test
* Revert exporting readResponseBody function
Instead the code is duplicated in the xsuaa and ans package
* Add check correct ans setup request
* Add set options function for mocking
* Review fixes
* Correct function name
* Use strict unmarshalling
* Validate event
* Move functions
* Add documentation comments
* improve test
Co-authored-by: Linda Siebert <39100394+LindaSieb@users.noreply.github.com>
Co-authored-by: Roland Stengel <r.stengel@sap.com>
* Add bearer token retrieval function
Retrieving a bearer token from the xsuaa service on BTP is always the
same. With these functions one can retrieve a bearer token and set it
to the given header as 'Authorization'.
* CodeClimate fixes
* Refactor test
* Add basic auth to token retrieve request
Co-authored-by: Thorsten Duda <thorsten.duda@sap.com>
* Make credential id name more descriptive
* Improve serviceKey description
* Change descriptions to add word service
Co-authored-by: Linda Siebert <linda.siebert@sap.com>
* Introduce docker config.json parameter for kubectl
* Use docker config.json parameter for helm
* Export definition of kube secret parameters
* fix username password existence check
* choose more fitting name for secret spec
* Adopt review suggestions
Co-authored-by: Roland Stengel <r.stengel@sap.com>
* Skip checkout flag
Defaults to null==false
We need the option to provide an scmInfo object by the user
* Check skipCheckout is Boolean
* Require scmInfo map in case checkout is skipped
* Move pull image parameters to general section
* Allow no nesting override
* Test dockerPullImage can be set to true
As the default is false, we need to make sure it can still be set to true.
First a bug fix is addressed in which the pull policy could not be configured to false by configuring the general configuration. It could neither be configured via dockerExecute or dockerExecuteOnKubernetes, even though this parameter is docker specific. Only by configuring the specific step where one wants to set the pull policy to false can it be configured.
As the bug stems from zero values being in the context config map, which is also addressed with this PR. That is the second part: Context config parameters are only set if they have a value.
* Don't set pull image if not configured
Otherwise, if the pull policy is not set explicitly for a step, dockerPullImage is set to true. Thus, before this change, it cannot be set in the general, or in dockerExecute or in dockerExecuteOnKubernetes configuration.
* Fix unit tests
* Add pullImage parameter test
* Do not place empty default values in context config
* Use putIfNotEmpty for sidecar container options
* Export common configuration
Keys that are set by both main and sidecar container can be exported
Co-authored-by: Stephan Aßmus <stephan.assmus@sap.com>
Without this change the default setting from dockerExecute
and dockerExecuteOnKubernetes are overruled.
With this change the user has to explicitly configure a step
if they want this behaviour.
Co-authored-by: Christopher Fenner <26137398+CCFenner@users.noreply.github.com>
* Move setting git info in cpe
Everything cpe related should be done in setupCommonPipelineEnvironment.
Thus a new parameter was introduced to accept the scmInfo object
returned by a checkout.
* Improve documentation
Only changes to master and PRs starts the travis job.
On PR Maven tests and mkdocs build for warnings and error check is done.
On change on master Maven is **not** run and the documentation is built
and deployed using a deploy key. The encrypted deploy key is not part of
this commit, it needs to be provided by an admin for this repository.
Before this commit the fileExists mock was not doing the same as the
real fileExists method. To stay close to reality we changed it construct
the absolute path and check for that when the fileExists method is
called.
Refactored to archiveName instead of archivePath, as this makes more
sense now.
When running on a slave we have to use the Pipeline method fileExists,
using the File class' exist on the absolute path fails.
The neo deployment uses the relative path as well.
The fileExists method is mocked with LesFurets.
To build the documentation mkdocs, a python package, is necessary. The
material theme is used and also needs to be installed.
Pass PEP370 --user to pip install, otherwise root privileges are needed.
Maven needs to be executed explicitely, as we are modifying the script
lifecycle phase.
In configurationLoader.md the reference to commonPipelineEnvironment.md
is wrong and will give you a 404, because commonPipelineEnvironment.md
lies in a separate folder called steps.
The change was already deployed to the gh-pages!
node() is not necessary in the LesFurets unit tests. This is especially
misleading for the pipelineExecute step, as the step itself opens a node
closure when loading a pipeline. If used like this in Jenkins
unnecessary executors will be used.
Mocking with LesFurets. Check that everything is at the right place.
1. Test with only the mandatory parameter 'repoUrl'.
1. Test setting all available parameters.
1. Test with no parameters throws Exception for missing mandatory
parameter 'repoUrl'.
A more convenient way to checkout the project sources. The idea is to
have a Jenkinsfile in the payload repository that only loads the shared
library and then runs this step. This step in turn loads a Jenkinsfile
from another repository.
The repo url, branch, Jenkinsfile path and credentials ID can be
provided to the step. The Jenkinsfile is first checked out to a
temporary folder, before it is loaded.
The payload repo must include a Jenkinsfile to be loaded with this
approach.
Oliver Feldmann
