Going back a long time we've had a special magic behavior which tries to
recognize a situation where a module author either added or removed the
"count" argument from a resource that already has instances, and to
silently rename the zeroth or no-key instance so that we don't plan to
destroy and recreate the associated object.
Now we have a more general idea of "move statements", and specifically
the idea of "implied" move statements which replicates the same heuristic
we used to use for this behavior, we can treat this magic renaming rule as
just another "move statement", special only in that Terraform generates it
automatically rather than it being written out explicitly in the
configuration.
In return for wiring that in, we can now remove altogether the
NodeCountBoundary graph node type and its associated graph transformer,
CountBoundaryTransformer. We handle moves as a preprocessing step before
building the plan graph, so we no longer need to include any special nodes
in the graph to deal with that situation.
The test updates here are mainly for the graph builders themselves, to
acknowledge that indeed we're no longer inserting the NodeCountBoundary
vertices. The vertices that NodeCountBoundary previously depended on now
become dependencies of the special "root" vertex, although in many cases
here we don't see that explicitly because of the transitive reduction
algorithm, which notices when there's already an equivalent indirect
dependency chain and removes the redundant edge.
We already have plenty of test coverage for these "count boundary" cases
in the context tests whose names start with TestContext2Plan_count and
TestContext2Apply_resourceCount, all of which continued to pass here
without any modification and so are not visible in the diff. The test
functions particularly relevant to this situation are:
- TestContext2Plan_countIncreaseFromNotSet
- TestContext2Plan_countDecreaseToOne
- TestContext2Plan_countOneIndex
- TestContext2Apply_countDecreaseToOneCorrupted
The last of those in particular deals with the situation where we have
both a no-key instance _and_ a zero-key instance in the prior state, which
is interesting here because to exercises an intentional interaction
between refactoring.ImpliedMoveStatements and refactoring.ApplyMoves,
where we intentionally generate an implied move statement that produces
a collision and then expect ApplyMoves to deal with it in the same way as
it would deal with all other collisions, and thus ensure we handle both
the explicit and implied collisions in the same way.
This does affect some UI-level tests, because a nice side-effect of this
new treatment of this old feature is that we can now report explicitly
in the UI that we're assigning new addresses to these objects, whereas
before we just said nothing and hoped the user would just guess what had
happened and why they therefore weren't seeing a diff.
The backend/local plan tests actually had a pre-existing bug where they
were using a state with a different instance key than the config called
for but getting away with it because we'd previously silently fix it up.
That's still fixed up, but now done with an explicit mention in the UI
and so I made the state consistent with the configuration here so that the
tests would be able to recognize _real_ differences where present, as
opposed to the errant difference caused by that inconsistency.
Per our rule that the content of the state can never make a move statement
invalid, our behavior for two objects trying to occupy the same address
will be to just ignore that and let the object already at the address
take priority.
For the moment this is silent from an end-user perspective and appears
only in our internal logs. However, I'm hoping that our future planned
adjustment to the interface of this function will include some way to
allow reporting these collisions in some end-user-visible way, either as
a separate warning per collision or as a single warning that collects
together all of the collisions into a single message somehow.
This situation can arise both because the previous run state already
contained an object at the target address of a move and because more than
one move ends up trying to target the same location. In the latter case,
which one "wins" is decided by our depth-first traversal order, which is
in turn derived from our chaining and nesting rules and is therefore
arbitrary but deterministic.
This new function complements the existing function FindMoveStatements
by potentially generating additional "implied" move statements that aren't
written explicit in the configuration but that we'll infer by comparing
the configuration and te previous run state.
The goal here is to infer only enough to replicate the effect of the
"count boundary fixup" graph node (terraform.NodeCountBoundary) that we
currently use to deal with this concern of preserving the zero-instance
when switching between "count" and not "count".
This is just dead code for now. A subsequent commit will introduce this
into the "terraform" package while also removing
terraform.NodeCountBoundary, thus achieving the same effect as before but
in a way that'll get reported in the UI as a move, using the same language
that we'd use for an explicit move statement.
This is similar to the existing SelectsModule method, returning true if
the reciever selects either a particular resource as a whole or any of the
instances of that resource.
We don't need this test in the normal case, but we will need it in a
subsequent commit when we'll be possibly generating _implied_ move
statements between instances of resources, but only if there aren't
explicit move statements mentioning those resources already.
The set of drifted resources now includes move-only changes, where the
object value is identical but a move has been executed. In normal
operation, we previousl displayed these moves twice: once as part of
drift output, and once as part of planned changes.
As of this commit we omit move-only changes from drift display, except
for refresh-only plans. This fixes the redundant output.
Previously, drifted resources included only updates and deletes. To
correctly display the full changes which would result as part of a
refresh-only apply, the drifted resources must also include move-only
changes.
Colorizing the result of an interpolated string can result in
incorrect output, if the values used to generate the string happen to
include color codes such as `[red]` or `[bold]`. Instead we should
always colorize the format string before calling functions like
`Sprintf`. This commit fixes all instances in this file.
Rather than delaying resource drift detection until it is ready to be
presented, here we perform that computation after the plan walk has
completed. The resulting drift is represented like planned resource
changes, using a slice of ResourceInstanceChangeSrc values.
Because "moved" blocks produce changes that span across more than one
resource instance address at the same time, we need to take extra care
with them during planning.
The -target option allows for restricting Terraform's attention only to
a subset of resources when planning, as an escape hatch to recover from
bugs and mistakes.
However, we need to avoid any situation where only one "side" of a move
would be considered in a particular plan, because that'd create a new
situation that would be otherwise unreachable and would be difficult to
recover from.
As a compromise then, we'll reject an attempt to create a targeted plan if
the plan involves resolving a pending move and if the source address of
that move is not included in the targets.
Our error message offers the user two possible resolutions: to create an
untargeted plan, thus allowing everything to resolve, or to add additional
-target options to include just the existing resource instances that have
pending moves to resolve.
This compromise recognizes that it is possible -- though hopefully rare --
that a user could potentially both be recovering from a bug or mistake at
the same time as processing a move, if e.g. the bug was fixed by upgrading
a module and the new version includes a new "moved" block. In that edge
case, it might be necessary to just add the one additional address to
the targets rather than removing the targets altogether, if creating a
normal untargeted plan is impossible due to whatever bug they're trying to
recover from.
When originally filling out these test cases we didn't yet have the logic
in place to detect chained moves and so this test couldn't succeed in
spite of being correct.
We now have chain-detection implemented and so consequently we can also
detect cyclic chains. This commit largely just enables the original test
unchanged, although it does include the text of the final error message
for reporting cyclic move chains which wasn't yet finalized when we were
stubbing out this test case originally.
The CoerceValue code was not updated to handle NestedTypes, and while
none of the new codepaths make use of this method, there are still some
internal uses.
The original intent of this test was to verify that we properly release
the state lock if terraform.NewContext fails. This was in response to a
bug in an earlier version of Terraform where that wasn't true.
In the recent refactoring that made terraform.NewContext no longer
responsible for provider constraint/checksum verification, this test began
testing a failed plan operation instead, which left the error return path
from terraform.NewContext untested.
An invalid parallelism value is the one remaining case where
terraform.NewContext can return an error, so as a localized fix for this
test I've switched it to just intentionally set an invalid parallelism
value. This is still not ideal because it's still testing an
implementation detail, but I've at least left a comment inline to try to
be clearer about what the goal is here so that we can respond in a more
appropriate way if future changes cause this test to fail again.
In the long run I'd like to move this last remaining check out to be the
responsibility of the CLI layer, with terraform.NewContext either just
assuming the value correct or panicking when it isn't, but the handling
of this CLI option is currently rather awkwardly spread across the
command and backend packages so we'll save that refactoring for a later
date.
The presence of this field was confusing because in practice the local
backend doesn't use it for anything and the remote backend was using it
only to return an error if it's set to anything other than the default,
under the assumption that it would always match ContextOpts.Parallelism.
The "command" package is the one actually responsible for handling this
option, and it does so by placing it into the partial ContextOpts which it
passes into the backend when preparing for a local operation. To make that
clearer, here we remove Operation.Parallelism and change the few uses of
it to refer to ContextOpts.Parallelism instead, so that everyone is
reading and writing this value from the same place.
In order to handle optional attributes, the Variable type needs to keep
track of the type constraint for decoding and conversion, as well as the
concrete type for creating values and type comparison.
Since the Type field is referenced throughout the codebase, and for
future refactoring if the handling of optional attributes changes
significantly, the constraint is now loaded into an entirely new field
called ConstraintType. This prevents types containing
ObjectWithOptionalAttrs from escaping the decode/conversion codepaths
into the rest of the codebase.
Objects with optional attributes are only used for the decoding of
HCL, and those types should never be exposed elsewhere within
terraform.
Separate the external ImpliedType method from the cty.Type generated
internally for the decoder spec. This unfortunately causes our
ImpliedType method to return a different type than the
hcldec.ImpliedType function, but the former is only used within
terraform for concrete values, while the latter is used to decode
HCL. Renaming the ImpliedType methods could be done to further
differentiate them, but that does cause fairly large diff in the
codebase that does not seem worth the effort at this time.
Thus far our various interactions with the bits of state we keep
associated with a working directory have all been implemented directly
inside the "command" package -- often in the huge command.Meta type -- and
not managed collectively via a single component.
There's too many little codepaths reading and writing from the working
directory and data directory to refactor it all in one step, but this is
an attempt at a first step towards a future where everything that reads
and writes from the current working directory would do so via an object
that encapsulates the implementation details and offers a high-level API
to read and write all of these session-persistent settings.
The design here continues our gradual path towards using a dependency
injection style where "package main" is solely responsible for directly
interacting with the OS command line, the OS environment, the OS working
directory, the stdio streams, and the CLI configuration, and then
communicating the resulting information to the rest of Terraform by wiring
together objects. It seems likely that eventually we'll have enough wiring
code in package main to justify a more explicit organization of that code,
but for this commit the new "workdir.Dir" object is just wired directly in
place of its predecessors, without any significant change of code
organization at that top layer.
This first commit focuses on the main files and directories we use to
find provider plugins, because a subsequent commit will lightly reorganize
the separation of concerns for plugin launching with a similar goal of
collecting all of the relevant logic together into one spot.
Previously our graph walker expected to recieve a data structure
containing schemas for all of the provider and provisioner plugins used in
the configuration and state. That made sense back when
terraform.NewContext was responsible for loading all of the schemas before
taking any other action, but it no longer has that responsiblity.
Instead, we'll now make sure that the "contextPlugins" object reaches all
of the locations where we need schema -- many of which already had access
to that object anyway -- and then load the needed schemas just in time.
The contextPlugins object memoizes schema lookups, so we can safely call
it many times with the same provider address or provisioner type name and
know that it'll still only load each distinct plugin once per Context
object.
As of this commit, the Context.Schemas method is now a public interface
only and not used by logic in the "terraform" package at all. However,
that does leave us in a rather tenuous situation of relying on the fact
that all practical users of terraform.Context end up calling "Schemas" at
some point in order to verify that we have all of the expected versions
of plugins. That's a non-obvious implicit dependency, and so in subsequent
commits we'll gradually move all responsibility for verifying plugin
versions into the caller of terraform.NewContext, which'll heal a
long-standing architectural wart whereby the caller is responsible for
installing and locating the plugin executables but not for verifying that
what's installed is conforming to the current configuration and dependency
lock file.
Previously the graph builders all expected to be given a full manifest
of all of the plugin component schemas that they could need during their
analysis work. That made sense when terraform.NewContext would always
proactively load all of the schemas before doing any other work, but we
now have a load-as-needed strategy for schemas.
We'll now have the graph builders use the contextPlugins object they each
already hold to retrieve individual schemas when needed. This avoids the
need to prepare a redundant data structure to pass alongside the
contextPlugins object, and leans on the memoization behavior inside
contextPlugins to preserve the old behavior of loading each provider's
schema only once.
By tolerating ProviderSchema and ProvisionerSchema potentially returning
errors, we can slightly simplify EvalContextBuiltin by having it retrieve
individual schemas when needed directly from the "Plugins" object.
EvalContextBuiltin already needs to be holding a contextPlugins instance
for other reasons anyway, so this allows us to get the same result with
fewer moving parts.
The responsibility for actually instantiating a single plugin and reading
out its schema now belongs to the contextPlugins type, which memoizes the
results by each plugin's unique identifier so that we can avoid retrieving
the same schemas multiple times when working with the same context.
This doesn't change the API of Context.Schemas but it does restore the
spirit of an earlier version of terraform.Context which did all of the
schema loading proactively inside terraform.NewContext. In an earlier
commit we reduced the scope of terraform.NewContext, making schema loading
a separate step, but in the process of doing that removed the effective
memoization of the schema results that terraform.NewContext was providing.
The memoization here will play out in a different way than before, because
we'll be treating each plugin call as separate rather than proactively
loading them all up front, but this is effectively the same because all
of our operation methods on Context call Context.Schemas early in their
work and thus end up forcing all of the necessary schemas to load up
front nonetheless.
In the v0.12 timeframe we made contextComponentFactory an interface with
the expectation that we'd write mocks of it for tests, but in practice we
ended up just always using the same "basicComponentFactory" implementation
throughout.
In the interests of simplification then, here we replace that interface
and its sole implementation with a new concrete struct type
contextPlugins.
Along with the general benefit that this removes an unneeded indirection,
this also means that we can add additional methods to the struct type
without the usual restriction that interface types prefer to be small.
In particular, in a future commit I'm planning to add methods for loading
provider and provisioner schemas, working with the currently-unused new
fields this commit has included in contextPlugins, as compared to its
predecessor basicComponentFactory.
In earlier Terraform versions we used the set of all available plugins of
each type to make graph-building decisions, but in modern Terraform we
make those decisions based entirely on the configuration.
Consequently, we no longer need the methods which can enumerate all of the
known plugin components of a given type. Instead, we just try to
instantiate each of the plugins that the configuration refers to and then
handle the error when that fails, which typically means that the user
needs to run "terraform init" to install some new plugins.
In earlier incarnations of these transformers we used the set of all
available providers for tasks such as generating implied provider
configuration nodes.
However, in modern Terraform we can extract all of the information we need
from the configuration itself, and so these transformers weren't actually
using this set of provider addresses.
These also ended up getting left behind as sets of string rather than sets
of addrs.Provider in our earlier refactoring work, which didn't really
matter because the result wasn't used anywhere anyway.
Rather than updating these to use addrs.Provider instead, I've just
removed the unused arguments entirely in the hope of making it easier to
see what inputs these transformers use to make their decisions.
The public interface for loading schemas is Context.Schemas, which can
take into account the context's records of which plugin versions and
checksums we're expecting. loadSchemas is an implementation detail of
that, representing the part we run only after we've verified all of the
plugins.
For resources which are planned to move, render the previous run address
as additional information in the plan UI. For the case of a move-only
resource (which otherwise is unchanged), we also render that as a
planned change, but without any corresponding action symbol.
If all changes in the plan are moves without changes, the plan is no
longer considered "empty". In this case, we skip rendering the action
symbols in the UI.
This package level variable can be overridden at link time to allow
temporarily disabling the UI warning when experimental features are
enabled. This makes it easier to understand how UI will render when the
feature is no longer experimental.
This change is only for those developing Terraform.
This is the first test exercising the basic functionality of config-driven
move. We previously had it skipped because Terraform's previous design
of treating all three of the state artifacts as mutable attributes of
terraform.Context meant that it was too late during planning to deal with
the move operations, and thus this test was failing.
Thanks to the previous commit, which changes the terraform.Context API
such that we can defer creating the three state artifacts until we're
already doing planning, this test now works and shows Terraform correctly
handling a resource that was formerly called "a" and is now called "b",
with a "moved" block recording that renaming.
Previously terraform.Context was built in an unfortunate way where all of
the data was provided up front in terraform.NewContext and then mutated
directly by subsequent operations. That made the data flow hard to follow,
commonly leading to bugs, and also meant that we were forced to take
various actions too early in terraform.NewContext, rather than waiting
until a more appropriate time during an operation.
This (enormous) commit changes terraform.Context so that its fields are
broadly just unchanging data about the execution context (current
workspace name, available plugins, etc) whereas the main data Terraform
works with arrives via individual method arguments and is returned in
return values.
Specifically, this means that terraform.Context no longer "has-a" config,
state, and "planned changes", instead holding on to those only temporarily
during an operation. The caller is responsible for propagating the outcome
of one step into the next step so that the data flow between operations is
actually visible.
However, since that's a change to the main entry points in the "terraform"
package, this commit also touches every file in the codebase which
interacted with those APIs. Most of the noise here is in updating tests
to take the same actions using the new API style, but this also affects
the main-code callers in the backends and in the command package.
My goal here was to refactor without changing observable behavior, but in
practice there are a couple externally-visible behavior variations here
that seemed okay in service of the broader goal:
- The "terraform graph" command is no longer hooked directly into the
core graph builders, because that's no longer part of the public API.
However, I did include a couple new Context functions whose contract
is to produce a UI-oriented graph, and _for now_ those continue to
return the physical graph we use for those operations. There's no
exported API for generating the "validate" and "eval" graphs, because
neither is particularly interesting in its own right, and so
"terraform graph" no longer supports those graph types.
- terraform.NewContext no longer has the responsibility for collecting
all of the provider schemas up front. Instead, we wait until we need
them. However, that means that some of our error messages now have a
slightly different shape due to unwinding through a differently-shaped
call stack. As of this commit we also end up reloading the schemas
multiple times in some cases, which is functionally acceptable but
likely represents a performance regression. I intend to rework this to
use caching, but I'm saving that for a later commit because this one is
big enough already.
The proximal reason for this change is to resolve the chicken/egg problem
whereby there was previously no single point where we could apply "moved"
statements to the previous run state before creating a plan. With this
change in place, we can now do that as part of Context.Plan, prior to
forking the input state into the three separate state artifacts we use
during planning.
However, this is at least the third project in a row where the previous
API design led to piling more functionality into terraform.NewContext and
then working around the incorrect order of operations that produces, so
I intend that by paying the cost/risk of this large diff now we can in
turn reduce the cost/risk of future projects that relate to our main
workflow actions.
Martin Atkins