In multi-action workflows the override check should be performed by
action so that different groups of users can override for different
actions. Additionally, abstracting the request_commands() method provides
a flexible base for additional commands to be added by ReviewBots.
Not only does this expose previously hidden messages on multi-action
requests, but also provides clarity as to which action triggered a
specific response. Since the keys are generated in a standard way and
the data formatted as YAML it can also be retrieved.
Ignore is used to wait for a change to occur while patchinfo should never
affect the result. Since all actions must be accepted for the request to
be accepted setting patchinfo to accept is the correct way to ignore it.
Their is very little overlap in the types of reviews one typically wants
to perform on delete requests. Already all of the existing code will crash
if a delete project request is reviewed.
Distinct copyrights were left as I do not wish to track down commit
history to ensure it properly documents the copyright holders. Also left
non-GPLv2 licenses and left bs_copy untouched as a mirror from OBS.
Already have a mix of with and without headers and even OBS does not place
on majority of files. If SUSE lawyers have an issue it will come up in
legal review for Factory.
The owner logic surrounding a package removed from Factory does not appear
to make sense as the current behavior of OBS never returns another owner
pair for such packages. As such the existing devel project lookup makes
more sense and is more straight forward.
Half the bots already utilize the Cache since it is used by StagingAPI
which they call, but really no reason for all of them not to use it by
default. With planned repo_checker changes the StagingAPI is not always
utilized, but it is desirable for the cache to always be used.
No sense calling Cache.init() in ReviewBot constructor as useless extra
calls when bots embed each other.
As the remote config is no longer optional for SLE and is utilized by
openSUSE to the point were it is dangerous not to load the remote config
it should be required. Currently only certain users call apply_remote()
while this will make it built-in during construction and thus makes the
usage consistent and no longer require StagingAPI.
Allows tool to be used on multi-action requests while still enforcing
the rule for Factory and Leap which should reject such requests due to
staging process.
Otherwise, memoize(session=True) caches are kept until the process is
stopped which is obviously not how they were intended to be used. This
can cause incorrect behavior and continued memory growth.
self.requests is only the incidents to be reviewed, but for testing we
need to take into account all open incidents (i.e. those we already approved
but not yet release migth interfere with those newly in testing)
