Use objchange.NormalizeObjectFromLegacySDK to ensure that all objects
returned from the provider match what is expected based on the
configuration according to the schemas.
Providers were not strict (and were not forced to be) about customizing
the diff when a computed attribute needed to be updated during apply.
The fix we have in place to prevent loss of information during the
helper/schema apply process would add in single missing value back in.
The first place this was caught was when we attempt to fix up the
flatmapped attributes. The 1->0 count error is now better handled by our
cty.Value normalization step, so we can remove the special apply case
here altogether
The next place is in normalizeNullValues, and since the intent was to
re-insert missing zero-value lists and sets, adding a check for a length
of 0 protects us from adding in extra elements.
The new test fixture emulated common provider behavior of re-computing
values without customizing the diff. Since we can work around it, and
core will provider appropriate warnings, the shims should try to
maintain the legacy behavior.
The NewExtra values are stored outside the diff from plan, and the
original keys may not contain the ~ prefix. Adding the NewExtra back
into the diff with the mismatched key was causing an entire new set
element to be populated. Since this symbol isn't used to apply the diff
in helper/schema, we can simply strip them out.
The hcl2shims will always add in the timeouts block, because there's no
way to differentiate a null single block from an empty one in the
flatmapped state. Since we are only concerned with keeping the prior
timeouts value, always set the new value to null, and then copy over the
prior value if it exists.
When the user aborts input, it may end up as an unknown value, which
needs to be converted to null for PrepareConfig.
Allow PrepareConfig to accept null config values in order to fill in
missing defaults.
The new normalization should make preventing those changes unnecessary,
and will also prevent extra empty elements from being added when
resources are refreshed.
This mirrors the change made for providers, so that default values can
be inserted into the config by the backend implementation. This is only
the interface and method name changes, it does not yet add any default
values.
cty.Value.AsValueMap can return nil if called on an empty map or object.
The logic above was dealing with that case for maps, but object types
were falling through into this codepath and panicking when trying to
assign a new key into the nil dstMap.
This also includes a bonus fix where we were calling ty.ElementType in
a switch case that accepts object types. Object types don't have a single
element type, so we can't call ElementType on those (that also panics)
but we _can_ use the type of the value we selected from src to construct
our placeholder null value.
Due to the inprecision of our shimming from the legacy SDK type system to
the new Terraform Core type system, the legacy SDK produces a number of
inconsistencies that produce only minor quirky behavior or broken
edge-cases. To retain compatibility with those existing weird behaviors,
the legacy SDK opts out of our safety checks.
The intent here is to allow existing providers to continue to do their
previous unsafe behaviors for now, accepting that this will allow certain
quirky bugs from previous releases to persist, and then gradually migrate
away from the legacy SDK and remove this opt-out on a per-resource basis
over time.
As with the apply-time safety check opt-out, this is reserved only for
the legacy SDK and must not be used in any new SDK implementations. We
still include any inconsistencies as warnings in the logs as an aid to
anyone debugging weird behavior, so that they can see situations where
blame may be misplaced in the user-visible error messages.
Terraform core expects a sane state even when the provider returns an
error. Make sure at the prior state is always the default value to
return, and then alway attempt to process any state returned by
provider.Apply.
This was changed in the single attribute test cases, but the AttrPair
test is used a lot for data source. As far as tests are concerned, 0 and
unset should be treated equally for flatmapped collections.
Check attributes on null objects, and fill in unknowns. If we're
evaluating the object, it either means we are at the top level, or a
NestingSingle block was present, and in either case we need to treat the
attributes as null rather than the entire object.
Switch on the block types rather than Nesting, so we don't need add any
logic to change between List/Tuple or Map/Object when DynamicPseudoType
is involved.
The shim layer for the legacy SDK type system is not precise enough to
guarantee it will produce identical results between plan and apply. In
particular, values that are null during plan will often become zero-valued
during apply.
To avoid breaking those existing providers while still allowing us to
introduce this check in the future, we'll introduce a rather-hacky new
flag that allows the legacy SDK to signal that it is the legacy SDK and
thus disable the check.
Once we start phasing out the legacy SDK in favor of one that natively
understands our new type system, we can stop setting this flag and thus
get the additional safety of this check without breaking any
previously-released providers.
No other SDK is permitted to set this flag, and we will remove it if we
ever introduce protocol version 6 in future, assuming that any provider
supporting that protocol will always produce consistent results.
If set elements are computed, we can't be certain that they are actually
equal. Catch identical computed set hashes when they are added to the
set, and alter the set key slightly to keep the set counts correct.
In previous versions the interpolation string would be included in the
set, and different string values would cause the set to hash
differently, so this is change is only activated for the new protocol.
This turns it on at the last moment, and in one place for all uses of
helper/schema. There's no way to use the new protocol without calling
GetSchema, so we can be sure that any subsequent api calls have this set
when required.
Sets rely on diffs being complete for all elements, even when they are
unchanged. When encountering a DiffSuppressFunc inside a set the diffs
were being dropped entirely, possible causing set elements to be lost.
Previously we were using the type name requested in the import to select
the schema, but a provider is free to return additional objects of other
types as part of an import result, and so it's important that we perform
schema selection separately for each returned object.
If we don't do this, we get confusing downstream errors where the
resulting object decodes to the wrong type and breaks various invariants
expected by Terraform Core.
The testResourceImportOther test in the test provider didn't catch this
previously because it happened to have an identical schema to the other
resource type being imported. Now the schema is changed and also there's
a computed attribute we can set as part of the refresh phase to make sure
we're completing the Read call properly during import. Refresh was working
correctly, but we didn't have any tests for it as part of the import flow.
This checking helper is frequently used in provider tests for data
sources, as a shorthand to verify that an attribute of the data source
matches with the corresponding attribute on a managed resource.
Since we now leave empty collections null in more cases, this function is
sometimes effectively asked to verify that a given attribute is _unset_
in both the data source and the resource, so here we slightly adjust the
definition of the check to consider two nulls to be equal to one another,
which at this layer manifests as the keys not being present in the state
attributes map at all.
This check function didn't previously have tests, so this commit also adds
a basic suite of tests, including coverage for the new behavior.
While copyMissingValues was meant to re-insert empty values that were
null after apply, it turns out plan is sometimes not predictable as
well.
normalizeNullValue is meant to fix up any null/empty transitions between
to values, and be useful during plan as well. For plan the function only
concerns itself with individual, known values, and skips sets entirely.
The result of running with plan == true is that only changes between
empty and null collections should be fixed.
The new decoder is more precise, and unpacks the timeout block into a
single map, which ResourceTimeout.ConfigDecode was updated to handle.
We however still need to work with legacy versions of terraform, with
the old decoder.
With the new diff.Apply we can keep the diff mostly intact, but we need
turn off all RequiresNew flags so that the prior state is not removed
from the apply.
One quirky aspect of our import feature is that we allow the importer to
produce additional resources alongside the one that was imported, such as
to create separate rules for each rule of an imported security group.
Providers need to be able to set the types of these other resources since
they may not match the "main" resource type. They do this by calling
ResourceData.SetType, which in turn sets InstanceState.Ephemeral.Type.
In our shims here we therefore need to copy that out into our new TypeName
field so that the new core import code can see it and create the right
type in the state.
Testing this required a minor change to the test harness to allow the
ImportStateCheck function to see the resource type.
Historically helper/schema did not support non-primitive map attributes
because they cannot be represented unambiguously in flatmap. When we
initially implemented CoreConfigSchema here we mapped that situation to
a nested block of mode NestingMap, even though that'd never worked until
now, assuming that it'd be harmless because providers wouldn't be using
it.
It turns out that some providers are, in fact, incorrectly populating
a TypeMap schema with Elem: &schema.Resource, apparently under the false
assumption that it would constrain the keys allowed in the map. In
practice, helper/schema has just been ignoring this and treating such
attributes as map of string. (#20076)
In order to preserve the behavior of these existing incorrectly-specified
attribute definitions, here we mimic the helper/schema behavior by
presenting as an attribute of type map(string).
These attributes have also been shown in some documentation as nested
blocks (with no equals sign), so that'll need to be fixed in user
configurations as they upgrade to Terraform 0.12. However, the existing
upgrade tool rules will take care of that as a natural consequence of the
name being indicated as an attribute in the schema, rather than as a block
type.
This fixes#20076.
Our new diff handling no longer requires stripping the empty diffs out,
and provider may be relying on some of the empty-value quirks in
helper/schema.
Due to various inprecisions in the old SDK implementation, applying the
generated diff can potentially make changes to the data structure that
have no real effect, such as replacing an empty list with a null list or
vice-versa.
Although we can't totally eliminate such diff noise, here we attempt to
avoid it in situations where there are _only_ meaningless changes -- where
the prior state and planned state are equivalent -- by just echoing back
the prior state verbatim to ensure that Terraform will treat it as a noop
change.
If there _are_ some legitimate changes then the result may still contain
meaningless changes alongside it, but that is just a cosmetic problem for
the diff renderer, because the meaningless changes will be ignored
altogether during a subsequent apply anyway. The primary goal here is just
to ensure we can converge on a fixpoint when there are no explicit changes
in the configuration.
This adds unexpected values in some cases, and since the case this
handles is only within set objects, we'll deal woth this when tackling
the sets themselves.
Cycle through the shim operations after Apply, to ensure that we can
converge on a stable value for for Plan. While the shims produce valid
values in both directions, helper/schema sometimes does not agree on
which containers should be empty or null.
There are a few constructs from 0.11 and prior that cause 0.12 parsing to
fail altogether, which previously created a chicken/egg problem because
we need to install the providers in order to run "terraform 0.12upgrade"
and thus fix the problem.
This changes "terraform init" to use the new "early configuration" loader
for module and provider installation. This is built on the more permissive
parser in the terraform-config-inspect package, and so it allows us to
read out the top-level blocks from the configuration while accepting
legacy HCL syntax.
In the long run this will let us do version compatibility detection before
attempting a "real" config load, giving us better error messages for any
future syntax additions, but in the short term the key thing is that it
allows us to install the dependencies even if the configuration isn't
fully valid.
Because backend init still requires full configuration, this introduces a
new mode of terraform init where it detects heuristically if it seems like
we need to do a configuration upgrade and does a partial init if so,
before finally directing the user to run "terraform 0.12upgrade" before
running any other commands.
The heuristic here is based on two assumptions:
- If the "early" loader finds no errors but the normal loader does, the
configuration is likely to be valid for Terraform 0.11 but not 0.12.
- If there's already a version constraint in the configuration that
excludes Terraform versions prior to v0.12 then the configuration is
probably _already_ upgraded and so it's just a normal syntax error,
even if the early loader didn't detect it.
Once the upgrade process is removed in 0.13.0 (users will be required to
go stepwise 0.11 -> 0.12 -> 0.13 to upgrade after that), some of this can
be simplified to remove that special mode, but the idea of doing the
dependency version checks against the liberal parser will remain valuable
to increase our chances of reporting version-based incompatibilities
rather than syntax errors as we add new features in future.
Sander van Harmelen