Commit Graph

23980 Commits

Author SHA1 Message Date
James Bardin af9dacb9f9 add failing test for diff.Apply
Add a diff test using a shcema with ConfigModeAttr.

It's in the test provider, because that is what is mostly responsible
for exercising diff.Apply, and where the other tests are.
2019-04-08 16:45:06 -04:00
Sander van Harmelen f2d9821edf
Update CHANGELOG.md 2019-04-08 19:33:52 +02:00
Sander van Harmelen 8f8df16d2b
Merge pull request #20952 from hashicorp/svh/b-slug-uploads
gomod: update `go-tfe`
2019-04-08 19:32:04 +02:00
Pam Selle c514867367
Merge pull request #20944 from pselle/registry-link
Only display status link if public registry used
2019-04-08 10:25:10 -04:00
Sander van Harmelen 8a430113f6 gomod: update `go-tfe` 2019-04-08 11:13:14 +02:00
Martin Atkins ffcf9fc51b lang/blocktoattr: Apply fixup within already-fixed-up blocks too
When a top-level list-of-object contains an attribute that is also
list-of-object we need to do the fixup again inside the nested body (using
our synthetic attributes-only schema) so that the attr-as-blocks mechanism
can apply within the nested blocks too.
2019-04-05 14:22:41 -07:00
Pam Selle 346e341ff2 Only display status link if public registry used
Updates to throw a specific error if using the defined public
registry vs. another registry.
2019-04-05 16:49:27 -04:00
Matthew Sanabria 0d0a1128e4
Clarifying behavior for destroy-time provisioners (#20819) 2019-04-05 14:22:16 -04:00
Matthew Sanabria be04e70a0f
Adding documentation for TF_CLI_CONFIG_FILE environment variable (#20834)
Fixes #15849
2019-04-05 14:21:40 -04:00
Martin Atkins 26c1e40ad7 configs/configupgrade: Normalize number literals to decimal
The v0.12 language supports numeric constants only in decimal notation, as
a simplification. For rare situations where a different base is more
appropriate, such as unix-style file modes, we've found it better for
providers to accept a string containing a representation in the
appropriate base, since that way the interpretation can be validated and
it will be displayed in the same way in the rendered plan diff, in
outputs, etc.

We use tv.Value() here to mimick how HCL 1 itself would have interpreted
these, and then format them back out in the canonical form, which
implicitly converts any non-decimal constants to decimal on the way
through.
2019-04-04 18:09:44 -07:00
Martin Atkins ac2052f0bb website: correct the title of the "matchkeys" function page 2019-04-04 15:35:13 -07:00
Martin Atkins 99f2623d10 website: Documentation of the special "attributes as blocks" mode
The re-introduction of some of the ambiguity between argument and nested
block syntax (for compatibility with existing provider patterns)
unfortunately leads to some interesting consequences for attributes using
this mode.

While the behavior is generally as before in straightforward cases, this
page aims to spell out some of the different usage patterns explicitly
for the benefit of those writing more complex configurations, such as
generic re-usable modules where using argument vs. block syntax leads to
some real differences.

This page is intentionally not linked from anywhere in the part of the
website maintained in the Terraform repository. Instead, it can be linked
from the provider documentation for any argument where this pattern is
used, to help users understand the ways in which that argument might
deviate from the usual behaviors of arguments vs. nested blocks.
2019-04-04 13:49:37 -07:00
Jeff 3110dde1d1 Fix docs typo (#20924)
* fix docs typo

* restore trailing newline
2019-04-04 11:21:36 -04:00
Martin Atkins 4d52999538 lang: EvalBlock should use ReferencesInBlock
Previously it was calling directly to hcldec.Variables, and thus missing
the special fixups we do inside ReferencesInBlock to deal with
Terraform-specific concerns such as our attribute-as-blocks preprocessing.
2019-04-03 15:57:17 -07:00
James Bardin 90b70bba08
Merge pull request #20923 from hashicorp/jbardin/planned-computed
Allow all planned computed diff fields
2019-04-03 18:40:32 -04:00
James Bardin 7b67105407 don't strip new-computeds from plan diffs
Stripping these was a patch for some provider behavior which was fixed
in other ways, and is no longer needed.
Removing this allows us to implement correct CusomizeDiffFuncs in
providers so that they can mark fields with empty values as computed
during a plan.
2019-04-03 17:37:58 -04:00
James Bardin f52a6630f5 create a downstream failure from a computed value
These are the largest source of the old "diffs didn't match after apply"
errors. It's almost always an upstream dependency that caused the final
error.
2019-04-03 17:36:08 -04:00
Kristin Laemmert 3aa4ac43f9
configs/configupgrade: return if hil.Parse() produces an error. (#20920)
Fixes #20917
2019-04-03 14:20:59 -04:00
James Bardin 51ddc554f5
Merge pull request #20909 from hashicorp/jbardin/validate-config-nulls
Validate configuration shims for nulls in lists
2019-04-03 14:06:22 -04:00
Kristin Laemmert f8a5e17d3d
command/format: improve "source" of error messages regarding missing arguments (#20907)
* vendor: update hcl2 dependency
* command/format: revert diagnostic format behavior if snippet or highlight range is empty
2019-04-03 14:04:59 -04:00
James Bardin e024960c74 Revert "filter nulls when shimming a config"
This reverts commit 97bde5467c.
2019-04-03 13:52:56 -04:00
James Bardin f5395bd98a validate null values in shimmed configs
A list-like attribute containing null values will present a list to
helper/schema with nils, which can cause panics. Since null values were
not possible in configuration before HCL2 and not supported by the
legacy SDK, return an error to the user.
2019-04-03 11:10:24 -04:00
Brian Flad 87f141000b
Update CHANGELOG for #20891 2019-04-03 08:54:39 -04:00
Brian Flad d5a8545a15
Merge pull request #20891 from hashicorp/b-helper-schema-setSet-panic
helper/schema: Prevent setSet() panic with typed nil
2019-04-03 08:52:38 -04:00
James Bardin 7ced46425e
Merge pull request #20906 from hashicorp/jbardin/grpc-recv-limit
increase grpc recv limit
2019-04-02 16:46:30 -04:00
James Bardin 2ebc2beda0 add 3rd param to mock call 2019-04-02 16:11:32 -04:00
James Bardin 746aac8bda increase grpc recv limit
Some providers may generate quite large schemas, and the internal
default grpc response size limit is 4MB. 64MB should cover most any use
case, and if we get providers nearing that we may want to consider a
finer-grained API to fetch individual resource schemas.
2019-04-02 15:52:54 -04:00
James Bardin 34358254ec
Merge pull request #20904 from hashicorp/jbardin/filter-shim-nulls
filter nulls when shimming a config
2019-04-02 15:10:17 -04:00
James Bardin 97bde5467c filter nulls when shimming a config
Nulls can't exist in HCL, and the legacy sdk will panic when
encountering them. The map shim already did this, just copy the same
pattern in the list case.
2019-04-02 14:27:31 -04:00
Brian Flad a8e3787afc
helper/schema: Prevent setSet() panic with typed nil
References:

* https://github.com/hashicorp/terraform/issues/14418
* v0.9.5 (original bug report): a59ee0b30e/helper/schema/field_writer_map.go (L311)
* v0.11.12 (Terraform AWS Provider discovery): 057286e522/helper/schema/field_writer_map.go (L343)

When creating flatten functions in Terraform Providers that return *schema.Set, its possible to return a typed `nil`, e.g.

```go
func flattenHeaders(h *cloudfront.Headers) *schema.Set {
	if h.Items != nil {
		return schema.NewSet(schema.HashString, flattenStringList(h.Items))
	}
	return nil
}
```

This previously could cause a panic, e.g.

```
panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x1 addr=0x8 pc=0x1881911]

goroutine 1325 [running]:
github.com/hashicorp/terraform/helper/schema.(*MapFieldWriter).setSet(0xc00054bf00, 0xc00073efa0, 0x5, 0x5, 0x5828140, 0x0, 0xc0002cea50, 0xc000e996a8, 0xc001026e40)
	/Users/bflad/go/pkg/mod/github.com/hashicorp/terraform@v0.11.12/helper/schema/field_writer_map.go:343 +0x211
```

Here we catch the typed `nil` and return an empty list flatmap result instead. Unit testing result prior to code update:

```
--- FAIL: TestMapFieldWriter (0.00s)
panic: runtime error: invalid memory address or nil pointer dereference [recovered]
  panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x1 addr=0x8 pc=0x1777cdc]

goroutine 913 [running]:
testing.tRunner.func1(0xc00045b800)
  /usr/local/Cellar/go/1.12.1/libexec/src/testing/testing.go:830 +0x392
panic(0x192cf20, 0x2267ca0)
  /usr/local/Cellar/go/1.12.1/libexec/src/runtime/panic.go:522 +0x1b5
github.com/hashicorp/terraform/helper/schema.(*MapFieldWriter).setSet(0xc0004648a0, 0xc0004408d0, 0x1, 0x1, 0x19e3de0, 0x0, 0xc00045c600, 0x30, 0x19e0080)
  /Users/bflad/src/github.com/hashicorp/terraform/helper/schema/field_writer_map.go:344 +0x68c
github.com/hashicorp/terraform/helper/schema.(*MapFieldWriter).set(0xc0004648a0, 0xc0004408d0, 0x1, 0x1, 0x19e3de0, 0x0, 0x1, 0x18)
  /Users/bflad/src/github.com/hashicorp/terraform/helper/schema/field_writer_map.go:107 +0x28b
github.com/hashicorp/terraform/helper/schema.(*MapFieldWriter).WriteField(0xc0004648a0, 0xc0004408d0, 0x1, 0x1, 0x19e3de0, 0x0, 0x0, 0x0)
  /Users/bflad/src/github.com/hashicorp/terraform/helper/schema/field_writer_map.go:89 +0x504
github.com/hashicorp/terraform/helper/schema.TestMapFieldWriter(0xc00045b800)
  /Users/bflad/src/github.com/hashicorp/terraform/helper/schema/field_writer_map_test.go:337 +0x2ddd
testing.tRunner(0xc00045b800, 0x1a44f90)
  /usr/local/Cellar/go/1.12.1/libexec/src/testing/testing.go:865 +0xc0
created by testing.(*T).Run
  /usr/local/Cellar/go/1.12.1/libexec/src/testing/testing.go:916 +0x35a
```
2019-04-01 20:10:32 -04:00
Chris Griggs 16778fea92
Merge pull request #20887 from cgriggs01/cgriggs01-fix-typo
[Website] Update provider name
2019-04-01 14:35:46 -07:00
Martin Atkins 6c5819f910 configs/configupgrade: Prefer block syntax for list-of-object attributes
In order to preserve pre-v0.12 idiom for list-of-object attributes, we'll
prefer to use block syntax for them except for the special situation where
the user explicitly assigns an empty list, where attribute syntax is
required in order to allow existing provider logic to differentiate from
an implicit lack of blocks.
2019-04-01 13:30:24 -07:00
cgriggs01 cfd271a040 update provider name 2019-04-01 08:41:45 -07:00
Kristin Laemmert 39ef97beff
website: adding examples of complex variables on the command line (#20871) 2019-04-01 07:51:17 -04:00
Bruno Tavares 6bad319187 go.mod: remove unneeded indirect for "mgo" and "gocheck"
We don't appear to actually need to constrain these, and removing these decls may help avoid the need for `bzr` to install all the dependencies.
2019-03-29 16:35:22 -07:00
Martin Atkins 1f9a8aa75e
Update CHANGELOG.md 2019-03-29 14:54:01 -07:00
Frederic 0f1f504c22 command/format: indicate in diff when adding an attribute forces replacement 2019-03-29 14:52:13 -07:00
James Bardin e893bae5da
Merge pull request #20874 from hashicorp/jbardin/shim-unknown-values
Better handling of unknowns in sdk shims
2019-03-29 15:03:29 -04:00
James Bardin 64c76be804 fixup lost collections containing unknowns
It turns out that collections containing only unknowns could be lost,
meaning there wasn't a direct correlation between the unknown and null
value which would have otherwise been restored.
2019-03-29 14:54:54 -04:00
James Bardin 6868ddd085 don't set NewRemoved values as empty strings
NewRemoved diff values were somtimes left in maps and lists.
2019-03-29 13:56:43 -04:00
James Bardin 86e30add98 fix unknowns added to maps by schemaMap
The legacy diff process inserts unknown values into an optional+computed
map. Fix these up in post-plan normalization process, by looking for
known strings that were changed to unknown.
2019-03-29 13:56:43 -04:00
James Bardin 009df443f7 restore lost unknowns during a planned update.
Because schema.ResourceDiff can't differentiate between unknown
values and new computed values, unknowns can be lost during an update.
If a planned value converted an unknown to a null, restore the unknown
so that it can be correctly replaced in the final plan.
2019-03-29 13:56:43 -04:00
James Bardin 8b9fa6d05f add test provider coverage around unknown vals 2019-03-29 13:56:42 -04:00
Alex Somesan ef681e527d Rephrase for clarity 2019-03-29 16:53:13 +01:00
Sander van Harmelen 137974c4a8
Update CHANGELOG.md 2019-03-29 08:34:49 +01:00
Sander van Harmelen b94ddab3a7
Merge pull request #20857 from hashicorp/svh/b-console-vars
backend/remote: correctly load remote variables
2019-03-29 08:34:01 +01:00
Kristin Laemmert da52170797
configupgrade: fix test (#20863) 2019-03-28 18:00:33 -04:00
Pam Selle 4c1bb8abd7
Update CHANGELOG.md 2019-03-28 15:06:41 -04:00
Kristin Laemmert 1baa1b907e
configs/configupgrade: detect invalid resource names and print a TODO (#20856)
* configs/configupgrade: detect invalid resource names and print a TODO
message

In terraform 0.11 and prior it was possible to start a resource name
with a number. This is no longer valid, as the resource name would would
be ambiguous with number values in HCL expressions.

Fixes #19919

* Update configs/configupgrade/test-fixtures/valid/invalid-resource-name/want/resource.tf

Co-Authored-By: mildwonkey <mildwonkey@users.noreply.github.com>
2019-03-28 13:48:35 -04:00
Martin Atkins 003317d7c8 lang: Detect references when a list/set attr is defined using blocks
For compatibility with documented patterns from existing providers we are
now allowing (via a pre-processing step) any attribute whose type is a
list-of-object or set-of-object type to optionally be assigned using one
or more blocks whose type is the attribute name.

The pre-processing functionality was implemented in previous commits but
we were not correctly detecting references within these blocks that are,
from the perspective of the primary schema, invalid. Now we'll use an
alternative implementation of variable detection that is able to apply the
same schema rewriting technique we used to implement the transform and
thus can find all of the references as if they were already in their
final locations.
2019-03-28 10:41:01 -07:00