2018-05-22 19:43:36 +00:00
|
|
|
// Copyright 2016-2018, Pulumi Corporation.
|
|
|
|
//
|
|
|
|
// Licensed under the Apache License, Version 2.0 (the "License");
|
|
|
|
// you may not use this file except in compliance with the License.
|
|
|
|
// You may obtain a copy of the License at
|
|
|
|
//
|
|
|
|
// http://www.apache.org/licenses/LICENSE-2.0
|
|
|
|
//
|
|
|
|
// Unless required by applicable law or agreed to in writing, software
|
|
|
|
// distributed under the License is distributed on an "AS IS" BASIS,
|
|
|
|
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
|
|
// See the License for the specific language governing permissions and
|
|
|
|
// limitations under the License.
|
Add custom decoders to eliminate boilerplate
This change overhauls the approach to custom decoding. Instead of decoding
the parts of the struct that are "trivial" in one pass, and then patching up
the structure afterwards with custom decoding, the decoder itself understands
the notion of custom decoder functions.
First, the general purpose logic has moved out of pkg/pack/encoding and into
a new package, pkg/util/mapper. Most functions are now members of a new top-
level type, Mapper, which may be initialized with custom decoders. This
is a map from target type to a function that can decode objects into it.
Second, the AST-specific decoding logic is rewritten to use it. All AST nodes
are now supported, including definitions, statements, and expressions. The
overall approach here is to simply define a custom decoder for any interface
type that will occur in a node field position. The mapper, upon encountering
such a type, will consult the custom decoder map; if a decoder is found, it
will be used, otherwise an error results. This decoder then needs to switch
on the type discriminated kind field that is present in the metadata, creating
a concrete struct of the right type, and then converting it to the desired
interface type. Note that, subtly, interface types used only for "marker"
purposes don't require any custom decoding, because they do not appear in
field positions and therefore won't be encountered during the decoding process.
2017-01-16 17:41:26 +00:00
|
|
|
|
|
|
|
package mapper
|
|
|
|
|
|
|
|
import (
|
|
|
|
"reflect"
|
|
|
|
"strings"
|
|
|
|
|
2021-03-17 13:20:05 +00:00
|
|
|
"github.com/pulumi/pulumi/sdk/v3/go/common/util/contract"
|
Add custom decoders to eliminate boilerplate
This change overhauls the approach to custom decoding. Instead of decoding
the parts of the struct that are "trivial" in one pass, and then patching up
the structure afterwards with custom decoding, the decoder itself understands
the notion of custom decoder functions.
First, the general purpose logic has moved out of pkg/pack/encoding and into
a new package, pkg/util/mapper. Most functions are now members of a new top-
level type, Mapper, which may be initialized with custom decoders. This
is a map from target type to a function that can decode objects into it.
Second, the AST-specific decoding logic is rewritten to use it. All AST nodes
are now supported, including definitions, statements, and expressions. The
overall approach here is to simply define a custom decoder for any interface
type that will occur in a node field position. The mapper, upon encountering
such a type, will consult the custom decoder map; if a decoder is found, it
will be used, otherwise an error results. This decoder then needs to switch
on the type discriminated kind field that is present in the metadata, creating
a concrete struct of the right type, and then converting it to the desired
interface type. Note that, subtly, interface types used only for "marker"
purposes don't require any custom decoding, because they do not appear in
field positions and therefore won't be encountered during the decoding process.
2017-01-16 17:41:26 +00:00
|
|
|
)
|
|
|
|
|
Implement mapper.Encode "for real"
This change implements `mapper.Encode` "for real" (that is, in a way
that isn't a complete embarrassment). It uses the obvious reflection
trickery to encode a tagged struct and its values as a JSON-like
in-memory map and collection of keyed values.
During this, I took the opportunity to also clean up a few other things
that had been bugging me. Namely, the presence of `mapper.Object` was
always error prone, since it isn't a true "typedef" in the sence that
it carries extra RTTI. Instead of doing that, let's just use the real
`map[string]interface{}` "JSON-map-like" object type. Even better, we
no longer require resource providers to deal with the mapper
infrastructure. Instead, the `Check` function can simply return an
array of errors. It's still best practice to return field-specific errors
to facilitate better diagnostics, but it's no longer required; and I've
added `resource.NewFieldError` to eliminate the need to import mapper.
As of this change, we can also consistently emit RPC structs with `lumi`
tags, rather than `lumi` tags on the way in and `json` on the way out.
This completes pulumi/lumi#183.
2017-06-06 00:43:52 +00:00
|
|
|
// Mapper can map from weakly typed JSON-like property bags to strongly typed structs, and vice versa.
|
Add custom decoders to eliminate boilerplate
This change overhauls the approach to custom decoding. Instead of decoding
the parts of the struct that are "trivial" in one pass, and then patching up
the structure afterwards with custom decoding, the decoder itself understands
the notion of custom decoder functions.
First, the general purpose logic has moved out of pkg/pack/encoding and into
a new package, pkg/util/mapper. Most functions are now members of a new top-
level type, Mapper, which may be initialized with custom decoders. This
is a map from target type to a function that can decode objects into it.
Second, the AST-specific decoding logic is rewritten to use it. All AST nodes
are now supported, including definitions, statements, and expressions. The
overall approach here is to simply define a custom decoder for any interface
type that will occur in a node field position. The mapper, upon encountering
such a type, will consult the custom decoder map; if a decoder is found, it
will be used, otherwise an error results. This decoder then needs to switch
on the type discriminated kind field that is present in the metadata, creating
a concrete struct of the right type, and then converting it to the desired
interface type. Note that, subtly, interface types used only for "marker"
purposes don't require any custom decoding, because they do not appear in
field positions and therefore won't be encountered during the decoding process.
2017-01-16 17:41:26 +00:00
|
|
|
type Mapper interface {
|
Unmap properties on the receive side
This change makes progress on a few things with respect to properly
receiving properties on the engine side, coming from the provider side,
of the RPC boundary. The issues here are twofold:
1. Properties need to get unmapped using a JSON-tag-sensitive
marshaler, so that they are cased properly, etc. For that, we
have a new mapper.Unmap function (which is ultra lame -- see
pulumi/lumi#138).
2. We have the reverse problem with respect to resource IDs: on
the send side, we must translate from URNs (which the engine
knows about) and provider IDs (which the provider knows about);
similarly, then, on the receive side, we must translate from
provider IDs back into URNs.
As a result of these getting fixed, we can now properly marshal the
resulting properties back into the resource object during the plan
execution, alongside propagating and memoizing its ID.
2017-05-31 18:51:46 +00:00
|
|
|
// Decode decodes a JSON-like object into the target pointer to a structure.
|
Implement mapper.Encode "for real"
This change implements `mapper.Encode` "for real" (that is, in a way
that isn't a complete embarrassment). It uses the obvious reflection
trickery to encode a tagged struct and its values as a JSON-like
in-memory map and collection of keyed values.
During this, I took the opportunity to also clean up a few other things
that had been bugging me. Namely, the presence of `mapper.Object` was
always error prone, since it isn't a true "typedef" in the sence that
it carries extra RTTI. Instead of doing that, let's just use the real
`map[string]interface{}` "JSON-map-like" object type. Even better, we
no longer require resource providers to deal with the mapper
infrastructure. Instead, the `Check` function can simply return an
array of errors. It's still best practice to return field-specific errors
to facilitate better diagnostics, but it's no longer required; and I've
added `resource.NewFieldError` to eliminate the need to import mapper.
As of this change, we can also consistently emit RPC structs with `lumi`
tags, rather than `lumi` tags on the way in and `json` on the way out.
This completes pulumi/lumi#183.
2017-06-06 00:43:52 +00:00
|
|
|
Decode(obj map[string]interface{}, target interface{}) MappingError
|
2022-10-09 14:58:33 +00:00
|
|
|
// DecodeValue decodes a single JSON-like value (with a given type and name) into a target pointer to a structure.
|
Implement mapper.Encode "for real"
This change implements `mapper.Encode` "for real" (that is, in a way
that isn't a complete embarrassment). It uses the obvious reflection
trickery to encode a tagged struct and its values as a JSON-like
in-memory map and collection of keyed values.
During this, I took the opportunity to also clean up a few other things
that had been bugging me. Namely, the presence of `mapper.Object` was
always error prone, since it isn't a true "typedef" in the sence that
it carries extra RTTI. Instead of doing that, let's just use the real
`map[string]interface{}` "JSON-map-like" object type. Even better, we
no longer require resource providers to deal with the mapper
infrastructure. Instead, the `Check` function can simply return an
array of errors. It's still best practice to return field-specific errors
to facilitate better diagnostics, but it's no longer required; and I've
added `resource.NewFieldError` to eliminate the need to import mapper.
As of this change, we can also consistently emit RPC structs with `lumi`
tags, rather than `lumi` tags on the way in and `json` on the way out.
This completes pulumi/lumi#183.
2017-06-06 00:43:52 +00:00
|
|
|
DecodeValue(obj map[string]interface{}, ty reflect.Type, key string, target interface{}, optional bool) FieldError
|
Unmap properties on the receive side
This change makes progress on a few things with respect to properly
receiving properties on the engine side, coming from the provider side,
of the RPC boundary. The issues here are twofold:
1. Properties need to get unmapped using a JSON-tag-sensitive
marshaler, so that they are cased properly, etc. For that, we
have a new mapper.Unmap function (which is ultra lame -- see
pulumi/lumi#138).
2. We have the reverse problem with respect to resource IDs: on
the send side, we must translate from URNs (which the engine
knows about) and provider IDs (which the provider knows about);
similarly, then, on the receive side, we must translate from
provider IDs back into URNs.
As a result of these getting fixed, we can now properly marshal the
resulting properties back into the resource object during the plan
execution, alongside propagating and memoizing its ID.
2017-05-31 18:51:46 +00:00
|
|
|
// Encode encodes an object into a JSON-like in-memory object.
|
Implement mapper.Encode "for real"
This change implements `mapper.Encode` "for real" (that is, in a way
that isn't a complete embarrassment). It uses the obvious reflection
trickery to encode a tagged struct and its values as a JSON-like
in-memory map and collection of keyed values.
During this, I took the opportunity to also clean up a few other things
that had been bugging me. Namely, the presence of `mapper.Object` was
always error prone, since it isn't a true "typedef" in the sence that
it carries extra RTTI. Instead of doing that, let's just use the real
`map[string]interface{}` "JSON-map-like" object type. Even better, we
no longer require resource providers to deal with the mapper
infrastructure. Instead, the `Check` function can simply return an
array of errors. It's still best practice to return field-specific errors
to facilitate better diagnostics, but it's no longer required; and I've
added `resource.NewFieldError` to eliminate the need to import mapper.
As of this change, we can also consistently emit RPC structs with `lumi`
tags, rather than `lumi` tags on the way in and `json` on the way out.
This completes pulumi/lumi#183.
2017-06-06 00:43:52 +00:00
|
|
|
Encode(source interface{}) (map[string]interface{}, MappingError)
|
|
|
|
// EncodeValue encodes a value into its JSON-like in-memory value format.
|
|
|
|
EncodeValue(v interface{}) (interface{}, MappingError)
|
Add custom decoders to eliminate boilerplate
This change overhauls the approach to custom decoding. Instead of decoding
the parts of the struct that are "trivial" in one pass, and then patching up
the structure afterwards with custom decoding, the decoder itself understands
the notion of custom decoder functions.
First, the general purpose logic has moved out of pkg/pack/encoding and into
a new package, pkg/util/mapper. Most functions are now members of a new top-
level type, Mapper, which may be initialized with custom decoders. This
is a map from target type to a function that can decode objects into it.
Second, the AST-specific decoding logic is rewritten to use it. All AST nodes
are now supported, including definitions, statements, and expressions. The
overall approach here is to simply define a custom decoder for any interface
type that will occur in a node field position. The mapper, upon encountering
such a type, will consult the custom decoder map; if a decoder is found, it
will be used, otherwise an error results. This decoder then needs to switch
on the type discriminated kind field that is present in the metadata, creating
a concrete struct of the right type, and then converting it to the desired
interface type. Note that, subtly, interface types used only for "marker"
purposes don't require any custom decoding, because they do not appear in
field positions and therefore won't be encountered during the decoding process.
2017-01-16 17:41:26 +00:00
|
|
|
}
|
|
|
|
|
Implement mapper.Encode "for real"
This change implements `mapper.Encode` "for real" (that is, in a way
that isn't a complete embarrassment). It uses the obvious reflection
trickery to encode a tagged struct and its values as a JSON-like
in-memory map and collection of keyed values.
During this, I took the opportunity to also clean up a few other things
that had been bugging me. Namely, the presence of `mapper.Object` was
always error prone, since it isn't a true "typedef" in the sence that
it carries extra RTTI. Instead of doing that, let's just use the real
`map[string]interface{}` "JSON-map-like" object type. Even better, we
no longer require resource providers to deal with the mapper
infrastructure. Instead, the `Check` function can simply return an
array of errors. It's still best practice to return field-specific errors
to facilitate better diagnostics, but it's no longer required; and I've
added `resource.NewFieldError` to eliminate the need to import mapper.
As of this change, we can also consistently emit RPC structs with `lumi`
tags, rather than `lumi` tags on the way in and `json` on the way out.
This completes pulumi/lumi#183.
2017-06-06 00:43:52 +00:00
|
|
|
// New allocates a new mapper object with the given options.
|
2017-03-12 21:13:44 +00:00
|
|
|
func New(opts *Opts) Mapper {
|
|
|
|
var initOpts Opts
|
|
|
|
if opts != nil {
|
|
|
|
initOpts = *opts
|
2017-02-25 00:44:46 +00:00
|
|
|
}
|
2017-03-12 21:13:44 +00:00
|
|
|
if initOpts.CustomDecoders == nil {
|
|
|
|
initOpts.CustomDecoders = make(Decoders)
|
|
|
|
}
|
|
|
|
return &mapper{opts: initOpts}
|
|
|
|
}
|
|
|
|
|
Implement mapper.Encode "for real"
This change implements `mapper.Encode` "for real" (that is, in a way
that isn't a complete embarrassment). It uses the obvious reflection
trickery to encode a tagged struct and its values as a JSON-like
in-memory map and collection of keyed values.
During this, I took the opportunity to also clean up a few other things
that had been bugging me. Namely, the presence of `mapper.Object` was
always error prone, since it isn't a true "typedef" in the sence that
it carries extra RTTI. Instead of doing that, let's just use the real
`map[string]interface{}` "JSON-map-like" object type. Even better, we
no longer require resource providers to deal with the mapper
infrastructure. Instead, the `Check` function can simply return an
array of errors. It's still best practice to return field-specific errors
to facilitate better diagnostics, but it's no longer required; and I've
added `resource.NewFieldError` to eliminate the need to import mapper.
As of this change, we can also consistently emit RPC structs with `lumi`
tags, rather than `lumi` tags on the way in and `json` on the way out.
This completes pulumi/lumi#183.
2017-06-06 00:43:52 +00:00
|
|
|
// Opts controls the way mapping occurs; for default behavior, simply pass an empty struct.
|
2017-03-12 21:13:44 +00:00
|
|
|
type Opts struct {
|
2017-09-22 02:18:21 +00:00
|
|
|
Tags []string // the tag names to recognize (`json` and `pulumi` if unspecified).
|
Implement mapper.Encode "for real"
This change implements `mapper.Encode` "for real" (that is, in a way
that isn't a complete embarrassment). It uses the obvious reflection
trickery to encode a tagged struct and its values as a JSON-like
in-memory map and collection of keyed values.
During this, I took the opportunity to also clean up a few other things
that had been bugging me. Namely, the presence of `mapper.Object` was
always error prone, since it isn't a true "typedef" in the sence that
it carries extra RTTI. Instead of doing that, let's just use the real
`map[string]interface{}` "JSON-map-like" object type. Even better, we
no longer require resource providers to deal with the mapper
infrastructure. Instead, the `Check` function can simply return an
array of errors. It's still best practice to return field-specific errors
to facilitate better diagnostics, but it's no longer required; and I've
added `resource.NewFieldError` to eliminate the need to import mapper.
As of this change, we can also consistently emit RPC structs with `lumi`
tags, rather than `lumi` tags on the way in and `json` on the way out.
This completes pulumi/lumi#183.
2017-06-06 00:43:52 +00:00
|
|
|
OptionalTags []string // the tags to interpret to mean "optional" (`optional` if unspecified).
|
|
|
|
SkipTags []string // the tags to interpret to mean "skip" (`skip` if unspecified).
|
2017-03-12 21:13:44 +00:00
|
|
|
CustomDecoders Decoders // custom decoders.
|
|
|
|
IgnoreMissing bool // ignore missing required fields.
|
|
|
|
IgnoreUnrecognized bool // ignore unrecognized fields.
|
Add custom decoders to eliminate boilerplate
This change overhauls the approach to custom decoding. Instead of decoding
the parts of the struct that are "trivial" in one pass, and then patching up
the structure afterwards with custom decoding, the decoder itself understands
the notion of custom decoder functions.
First, the general purpose logic has moved out of pkg/pack/encoding and into
a new package, pkg/util/mapper. Most functions are now members of a new top-
level type, Mapper, which may be initialized with custom decoders. This
is a map from target type to a function that can decode objects into it.
Second, the AST-specific decoding logic is rewritten to use it. All AST nodes
are now supported, including definitions, statements, and expressions. The
overall approach here is to simply define a custom decoder for any interface
type that will occur in a node field position. The mapper, upon encountering
such a type, will consult the custom decoder map; if a decoder is found, it
will be used, otherwise an error results. This decoder then needs to switch
on the type discriminated kind field that is present in the metadata, creating
a concrete struct of the right type, and then converting it to the desired
interface type. Note that, subtly, interface types used only for "marker"
purposes don't require any custom decoding, because they do not appear in
field positions and therefore won't be encountered during the decoding process.
2017-01-16 17:41:26 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
type mapper struct {
|
2017-03-12 21:13:44 +00:00
|
|
|
opts Opts
|
Add custom decoders to eliminate boilerplate
This change overhauls the approach to custom decoding. Instead of decoding
the parts of the struct that are "trivial" in one pass, and then patching up
the structure afterwards with custom decoding, the decoder itself understands
the notion of custom decoder functions.
First, the general purpose logic has moved out of pkg/pack/encoding and into
a new package, pkg/util/mapper. Most functions are now members of a new top-
level type, Mapper, which may be initialized with custom decoders. This
is a map from target type to a function that can decode objects into it.
Second, the AST-specific decoding logic is rewritten to use it. All AST nodes
are now supported, including definitions, statements, and expressions. The
overall approach here is to simply define a custom decoder for any interface
type that will occur in a node field position. The mapper, upon encountering
such a type, will consult the custom decoder map; if a decoder is found, it
will be used, otherwise an error results. This decoder then needs to switch
on the type discriminated kind field that is present in the metadata, creating
a concrete struct of the right type, and then converting it to the desired
interface type. Note that, subtly, interface types used only for "marker"
purposes don't require any custom decoding, because they do not appear in
field positions and therefore won't be encountered during the decoding process.
2017-01-16 17:41:26 +00:00
|
|
|
}
|
|
|
|
|
Implement mapper.Encode "for real"
This change implements `mapper.Encode` "for real" (that is, in a way
that isn't a complete embarrassment). It uses the obvious reflection
trickery to encode a tagged struct and its values as a JSON-like
in-memory map and collection of keyed values.
During this, I took the opportunity to also clean up a few other things
that had been bugging me. Namely, the presence of `mapper.Object` was
always error prone, since it isn't a true "typedef" in the sence that
it carries extra RTTI. Instead of doing that, let's just use the real
`map[string]interface{}` "JSON-map-like" object type. Even better, we
no longer require resource providers to deal with the mapper
infrastructure. Instead, the `Check` function can simply return an
array of errors. It's still best practice to return field-specific errors
to facilitate better diagnostics, but it's no longer required; and I've
added `resource.NewFieldError` to eliminate the need to import mapper.
As of this change, we can also consistently emit RPC structs with `lumi`
tags, rather than `lumi` tags on the way in and `json` on the way out.
This completes pulumi/lumi#183.
2017-06-06 00:43:52 +00:00
|
|
|
// Map decodes an entire map into a target object, using an anonymous decoder and tag-directed mappings.
|
|
|
|
func Map(obj map[string]interface{}, target interface{}) MappingError {
|
|
|
|
return New(nil).Decode(obj, target)
|
|
|
|
}
|
Add custom decoders to eliminate boilerplate
This change overhauls the approach to custom decoding. Instead of decoding
the parts of the struct that are "trivial" in one pass, and then patching up
the structure afterwards with custom decoding, the decoder itself understands
the notion of custom decoder functions.
First, the general purpose logic has moved out of pkg/pack/encoding and into
a new package, pkg/util/mapper. Most functions are now members of a new top-
level type, Mapper, which may be initialized with custom decoders. This
is a map from target type to a function that can decode objects into it.
Second, the AST-specific decoding logic is rewritten to use it. All AST nodes
are now supported, including definitions, statements, and expressions. The
overall approach here is to simply define a custom decoder for any interface
type that will occur in a node field position. The mapper, upon encountering
such a type, will consult the custom decoder map; if a decoder is found, it
will be used, otherwise an error results. This decoder then needs to switch
on the type discriminated kind field that is present in the metadata, creating
a concrete struct of the right type, and then converting it to the desired
interface type. Note that, subtly, interface types used only for "marker"
purposes don't require any custom decoding, because they do not appear in
field positions and therefore won't be encountered during the decoding process.
2017-01-16 17:41:26 +00:00
|
|
|
|
Implement mapper.Encode "for real"
This change implements `mapper.Encode` "for real" (that is, in a way
that isn't a complete embarrassment). It uses the obvious reflection
trickery to encode a tagged struct and its values as a JSON-like
in-memory map and collection of keyed values.
During this, I took the opportunity to also clean up a few other things
that had been bugging me. Namely, the presence of `mapper.Object` was
always error prone, since it isn't a true "typedef" in the sence that
it carries extra RTTI. Instead of doing that, let's just use the real
`map[string]interface{}` "JSON-map-like" object type. Even better, we
no longer require resource providers to deal with the mapper
infrastructure. Instead, the `Check` function can simply return an
array of errors. It's still best practice to return field-specific errors
to facilitate better diagnostics, but it's no longer required; and I've
added `resource.NewFieldError` to eliminate the need to import mapper.
As of this change, we can also consistently emit RPC structs with `lumi`
tags, rather than `lumi` tags on the way in and `json` on the way out.
This completes pulumi/lumi#183.
2017-06-06 00:43:52 +00:00
|
|
|
// MapI decodes an entire map into a target object, using an anonymous decoder and tag-directed mappings. This variant
|
|
|
|
// ignores any missing required fields in the payload in addition to any unrecognized fields.
|
|
|
|
func MapI(obj map[string]interface{}, target interface{}) MappingError {
|
|
|
|
return New(&Opts{
|
|
|
|
IgnoreMissing: true,
|
|
|
|
IgnoreUnrecognized: true,
|
|
|
|
}).Decode(obj, target)
|
|
|
|
}
|
Add custom decoders to eliminate boilerplate
This change overhauls the approach to custom decoding. Instead of decoding
the parts of the struct that are "trivial" in one pass, and then patching up
the structure afterwards with custom decoding, the decoder itself understands
the notion of custom decoder functions.
First, the general purpose logic has moved out of pkg/pack/encoding and into
a new package, pkg/util/mapper. Most functions are now members of a new top-
level type, Mapper, which may be initialized with custom decoders. This
is a map from target type to a function that can decode objects into it.
Second, the AST-specific decoding logic is rewritten to use it. All AST nodes
are now supported, including definitions, statements, and expressions. The
overall approach here is to simply define a custom decoder for any interface
type that will occur in a node field position. The mapper, upon encountering
such a type, will consult the custom decoder map; if a decoder is found, it
will be used, otherwise an error results. This decoder then needs to switch
on the type discriminated kind field that is present in the metadata, creating
a concrete struct of the right type, and then converting it to the desired
interface type. Note that, subtly, interface types used only for "marker"
purposes don't require any custom decoding, because they do not appear in
field positions and therefore won't be encountered during the decoding process.
2017-01-16 17:41:26 +00:00
|
|
|
|
Implement mapper.Encode "for real"
This change implements `mapper.Encode` "for real" (that is, in a way
that isn't a complete embarrassment). It uses the obvious reflection
trickery to encode a tagged struct and its values as a JSON-like
in-memory map and collection of keyed values.
During this, I took the opportunity to also clean up a few other things
that had been bugging me. Namely, the presence of `mapper.Object` was
always error prone, since it isn't a true "typedef" in the sence that
it carries extra RTTI. Instead of doing that, let's just use the real
`map[string]interface{}` "JSON-map-like" object type. Even better, we
no longer require resource providers to deal with the mapper
infrastructure. Instead, the `Check` function can simply return an
array of errors. It's still best practice to return field-specific errors
to facilitate better diagnostics, but it's no longer required; and I've
added `resource.NewFieldError` to eliminate the need to import mapper.
As of this change, we can also consistently emit RPC structs with `lumi`
tags, rather than `lumi` tags on the way in and `json` on the way out.
This completes pulumi/lumi#183.
2017-06-06 00:43:52 +00:00
|
|
|
// MapIM decodes an entire map into a target object, using an anonymous decoder and tag-directed mappings. This variant
|
|
|
|
// ignores any missing required fields in the payload.
|
|
|
|
func MapIM(obj map[string]interface{}, target interface{}) MappingError {
|
|
|
|
return New(&Opts{IgnoreMissing: true}).Decode(obj, target)
|
|
|
|
}
|
Add custom decoders to eliminate boilerplate
This change overhauls the approach to custom decoding. Instead of decoding
the parts of the struct that are "trivial" in one pass, and then patching up
the structure afterwards with custom decoding, the decoder itself understands
the notion of custom decoder functions.
First, the general purpose logic has moved out of pkg/pack/encoding and into
a new package, pkg/util/mapper. Most functions are now members of a new top-
level type, Mapper, which may be initialized with custom decoders. This
is a map from target type to a function that can decode objects into it.
Second, the AST-specific decoding logic is rewritten to use it. All AST nodes
are now supported, including definitions, statements, and expressions. The
overall approach here is to simply define a custom decoder for any interface
type that will occur in a node field position. The mapper, upon encountering
such a type, will consult the custom decoder map; if a decoder is found, it
will be used, otherwise an error results. This decoder then needs to switch
on the type discriminated kind field that is present in the metadata, creating
a concrete struct of the right type, and then converting it to the desired
interface type. Note that, subtly, interface types used only for "marker"
purposes don't require any custom decoding, because they do not appear in
field positions and therefore won't be encountered during the decoding process.
2017-01-16 17:41:26 +00:00
|
|
|
|
Implement mapper.Encode "for real"
This change implements `mapper.Encode` "for real" (that is, in a way
that isn't a complete embarrassment). It uses the obvious reflection
trickery to encode a tagged struct and its values as a JSON-like
in-memory map and collection of keyed values.
During this, I took the opportunity to also clean up a few other things
that had been bugging me. Namely, the presence of `mapper.Object` was
always error prone, since it isn't a true "typedef" in the sence that
it carries extra RTTI. Instead of doing that, let's just use the real
`map[string]interface{}` "JSON-map-like" object type. Even better, we
no longer require resource providers to deal with the mapper
infrastructure. Instead, the `Check` function can simply return an
array of errors. It's still best practice to return field-specific errors
to facilitate better diagnostics, but it's no longer required; and I've
added `resource.NewFieldError` to eliminate the need to import mapper.
As of this change, we can also consistently emit RPC structs with `lumi`
tags, rather than `lumi` tags on the way in and `json` on the way out.
This completes pulumi/lumi#183.
2017-06-06 00:43:52 +00:00
|
|
|
// MapIU decodes an entire map into a target object, using an anonymous decoder and tag-directed mappings. This variant
|
|
|
|
// ignores any unrecognized fields in the payload.
|
|
|
|
func MapIU(obj map[string]interface{}, target interface{}) MappingError {
|
|
|
|
return New(&Opts{IgnoreUnrecognized: true}).Decode(obj, target)
|
|
|
|
}
|
2017-03-12 21:13:44 +00:00
|
|
|
|
Implement mapper.Encode "for real"
This change implements `mapper.Encode` "for real" (that is, in a way
that isn't a complete embarrassment). It uses the obvious reflection
trickery to encode a tagged struct and its values as a JSON-like
in-memory map and collection of keyed values.
During this, I took the opportunity to also clean up a few other things
that had been bugging me. Namely, the presence of `mapper.Object` was
always error prone, since it isn't a true "typedef" in the sence that
it carries extra RTTI. Instead of doing that, let's just use the real
`map[string]interface{}` "JSON-map-like" object type. Even better, we
no longer require resource providers to deal with the mapper
infrastructure. Instead, the `Check` function can simply return an
array of errors. It's still best practice to return field-specific errors
to facilitate better diagnostics, but it's no longer required; and I've
added `resource.NewFieldError` to eliminate the need to import mapper.
As of this change, we can also consistently emit RPC structs with `lumi`
tags, rather than `lumi` tags on the way in and `json` on the way out.
This completes pulumi/lumi#183.
2017-06-06 00:43:52 +00:00
|
|
|
// Unmap translates an already mapped target object into a raw, unmapped form.
|
|
|
|
func Unmap(obj interface{}) (map[string]interface{}, error) {
|
|
|
|
return New(nil).Encode(obj)
|
|
|
|
}
|
|
|
|
|
|
|
|
// structFields digs into a type to fetch all fields, including its embedded structs, for a given type.
|
|
|
|
func structFields(t reflect.Type) []reflect.StructField {
|
|
|
|
contract.Assertf(t.Kind() == reflect.Struct,
|
|
|
|
"StructFields only valid on struct types; %v is not one (kind %v)", t, t.Kind())
|
Add custom decoders to eliminate boilerplate
This change overhauls the approach to custom decoding. Instead of decoding
the parts of the struct that are "trivial" in one pass, and then patching up
the structure afterwards with custom decoding, the decoder itself understands
the notion of custom decoder functions.
First, the general purpose logic has moved out of pkg/pack/encoding and into
a new package, pkg/util/mapper. Most functions are now members of a new top-
level type, Mapper, which may be initialized with custom decoders. This
is a map from target type to a function that can decode objects into it.
Second, the AST-specific decoding logic is rewritten to use it. All AST nodes
are now supported, including definitions, statements, and expressions. The
overall approach here is to simply define a custom decoder for any interface
type that will occur in a node field position. The mapper, upon encountering
such a type, will consult the custom decoder map; if a decoder is found, it
will be used, otherwise an error results. This decoder then needs to switch
on the type discriminated kind field that is present in the metadata, creating
a concrete struct of the right type, and then converting it to the desired
interface type. Note that, subtly, interface types used only for "marker"
purposes don't require any custom decoding, because they do not appear in
field positions and therefore won't be encountered during the decoding process.
2017-01-16 17:41:26 +00:00
|
|
|
|
|
|
|
var fldinfos []reflect.StructField
|
Implement mapper.Encode "for real"
This change implements `mapper.Encode` "for real" (that is, in a way
that isn't a complete embarrassment). It uses the obvious reflection
trickery to encode a tagged struct and its values as a JSON-like
in-memory map and collection of keyed values.
During this, I took the opportunity to also clean up a few other things
that had been bugging me. Namely, the presence of `mapper.Object` was
always error prone, since it isn't a true "typedef" in the sence that
it carries extra RTTI. Instead of doing that, let's just use the real
`map[string]interface{}` "JSON-map-like" object type. Even better, we
no longer require resource providers to deal with the mapper
infrastructure. Instead, the `Check` function can simply return an
array of errors. It's still best practice to return field-specific errors
to facilitate better diagnostics, but it's no longer required; and I've
added `resource.NewFieldError` to eliminate the need to import mapper.
As of this change, we can also consistently emit RPC structs with `lumi`
tags, rather than `lumi` tags on the way in and `json` on the way out.
This completes pulumi/lumi#183.
2017-06-06 00:43:52 +00:00
|
|
|
fldtypes := []reflect.Type{t}
|
Add custom decoders to eliminate boilerplate
This change overhauls the approach to custom decoding. Instead of decoding
the parts of the struct that are "trivial" in one pass, and then patching up
the structure afterwards with custom decoding, the decoder itself understands
the notion of custom decoder functions.
First, the general purpose logic has moved out of pkg/pack/encoding and into
a new package, pkg/util/mapper. Most functions are now members of a new top-
level type, Mapper, which may be initialized with custom decoders. This
is a map from target type to a function that can decode objects into it.
Second, the AST-specific decoding logic is rewritten to use it. All AST nodes
are now supported, including definitions, statements, and expressions. The
overall approach here is to simply define a custom decoder for any interface
type that will occur in a node field position. The mapper, upon encountering
such a type, will consult the custom decoder map; if a decoder is found, it
will be used, otherwise an error results. This decoder then needs to switch
on the type discriminated kind field that is present in the metadata, creating
a concrete struct of the right type, and then converting it to the desired
interface type. Note that, subtly, interface types used only for "marker"
purposes don't require any custom decoding, because they do not appear in
field positions and therefore won't be encountered during the decoding process.
2017-01-16 17:41:26 +00:00
|
|
|
for len(fldtypes) > 0 {
|
|
|
|
fldtype := fldtypes[0]
|
|
|
|
fldtypes = fldtypes[1:]
|
|
|
|
for i := 0; i < fldtype.NumField(); i++ {
|
Implement mapper.Encode "for real"
This change implements `mapper.Encode` "for real" (that is, in a way
that isn't a complete embarrassment). It uses the obvious reflection
trickery to encode a tagged struct and its values as a JSON-like
in-memory map and collection of keyed values.
During this, I took the opportunity to also clean up a few other things
that had been bugging me. Namely, the presence of `mapper.Object` was
always error prone, since it isn't a true "typedef" in the sence that
it carries extra RTTI. Instead of doing that, let's just use the real
`map[string]interface{}` "JSON-map-like" object type. Even better, we
no longer require resource providers to deal with the mapper
infrastructure. Instead, the `Check` function can simply return an
array of errors. It's still best practice to return field-specific errors
to facilitate better diagnostics, but it's no longer required; and I've
added `resource.NewFieldError` to eliminate the need to import mapper.
As of this change, we can also consistently emit RPC structs with `lumi`
tags, rather than `lumi` tags on the way in and `json` on the way out.
This completes pulumi/lumi#183.
2017-06-06 00:43:52 +00:00
|
|
|
if fldinfo := fldtype.Field(i); fldinfo.Anonymous {
|
Add custom decoders to eliminate boilerplate
This change overhauls the approach to custom decoding. Instead of decoding
the parts of the struct that are "trivial" in one pass, and then patching up
the structure afterwards with custom decoding, the decoder itself understands
the notion of custom decoder functions.
First, the general purpose logic has moved out of pkg/pack/encoding and into
a new package, pkg/util/mapper. Most functions are now members of a new top-
level type, Mapper, which may be initialized with custom decoders. This
is a map from target type to a function that can decode objects into it.
Second, the AST-specific decoding logic is rewritten to use it. All AST nodes
are now supported, including definitions, statements, and expressions. The
overall approach here is to simply define a custom decoder for any interface
type that will occur in a node field position. The mapper, upon encountering
such a type, will consult the custom decoder map; if a decoder is found, it
will be used, otherwise an error results. This decoder then needs to switch
on the type discriminated kind field that is present in the metadata, creating
a concrete struct of the right type, and then converting it to the desired
interface type. Note that, subtly, interface types used only for "marker"
purposes don't require any custom decoding, because they do not appear in
field positions and therefore won't be encountered during the decoding process.
2017-01-16 17:41:26 +00:00
|
|
|
// If an embedded struct, push it onto the queue to visit.
|
Implement mapper.Encode "for real"
This change implements `mapper.Encode` "for real" (that is, in a way
that isn't a complete embarrassment). It uses the obvious reflection
trickery to encode a tagged struct and its values as a JSON-like
in-memory map and collection of keyed values.
During this, I took the opportunity to also clean up a few other things
that had been bugging me. Namely, the presence of `mapper.Object` was
always error prone, since it isn't a true "typedef" in the sence that
it carries extra RTTI. Instead of doing that, let's just use the real
`map[string]interface{}` "JSON-map-like" object type. Even better, we
no longer require resource providers to deal with the mapper
infrastructure. Instead, the `Check` function can simply return an
array of errors. It's still best practice to return field-specific errors
to facilitate better diagnostics, but it's no longer required; and I've
added `resource.NewFieldError` to eliminate the need to import mapper.
As of this change, we can also consistently emit RPC structs with `lumi`
tags, rather than `lumi` tags on the way in and `json` on the way out.
This completes pulumi/lumi#183.
2017-06-06 00:43:52 +00:00
|
|
|
if fldinfo.Type.Kind() == reflect.Struct {
|
|
|
|
fldtypes = append(fldtypes, fldinfo.Type)
|
|
|
|
}
|
Add custom decoders to eliminate boilerplate
This change overhauls the approach to custom decoding. Instead of decoding
the parts of the struct that are "trivial" in one pass, and then patching up
the structure afterwards with custom decoding, the decoder itself understands
the notion of custom decoder functions.
First, the general purpose logic has moved out of pkg/pack/encoding and into
a new package, pkg/util/mapper. Most functions are now members of a new top-
level type, Mapper, which may be initialized with custom decoders. This
is a map from target type to a function that can decode objects into it.
Second, the AST-specific decoding logic is rewritten to use it. All AST nodes
are now supported, including definitions, statements, and expressions. The
overall approach here is to simply define a custom decoder for any interface
type that will occur in a node field position. The mapper, upon encountering
such a type, will consult the custom decoder map; if a decoder is found, it
will be used, otherwise an error results. This decoder then needs to switch
on the type discriminated kind field that is present in the metadata, creating
a concrete struct of the right type, and then converting it to the desired
interface type. Note that, subtly, interface types used only for "marker"
purposes don't require any custom decoding, because they do not appear in
field positions and therefore won't be encountered during the decoding process.
2017-01-16 17:41:26 +00:00
|
|
|
} else {
|
|
|
|
// Otherwise, we will go ahead and consider this field in our decoding.
|
|
|
|
fldinfos = append(fldinfos, fldinfo)
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
Implement mapper.Encode "for real"
This change implements `mapper.Encode` "for real" (that is, in a way
that isn't a complete embarrassment). It uses the obvious reflection
trickery to encode a tagged struct and its values as a JSON-like
in-memory map and collection of keyed values.
During this, I took the opportunity to also clean up a few other things
that had been bugging me. Namely, the presence of `mapper.Object` was
always error prone, since it isn't a true "typedef" in the sence that
it carries extra RTTI. Instead of doing that, let's just use the real
`map[string]interface{}` "JSON-map-like" object type. Even better, we
no longer require resource providers to deal with the mapper
infrastructure. Instead, the `Check` function can simply return an
array of errors. It's still best practice to return field-specific errors
to facilitate better diagnostics, but it's no longer required; and I've
added `resource.NewFieldError` to eliminate the need to import mapper.
As of this change, we can also consistently emit RPC structs with `lumi`
tags, rather than `lumi` tags on the way in and `json` on the way out.
This completes pulumi/lumi#183.
2017-06-06 00:43:52 +00:00
|
|
|
return fldinfos
|
|
|
|
}
|
Add custom decoders to eliminate boilerplate
This change overhauls the approach to custom decoding. Instead of decoding
the parts of the struct that are "trivial" in one pass, and then patching up
the structure afterwards with custom decoding, the decoder itself understands
the notion of custom decoder functions.
First, the general purpose logic has moved out of pkg/pack/encoding and into
a new package, pkg/util/mapper. Most functions are now members of a new top-
level type, Mapper, which may be initialized with custom decoders. This
is a map from target type to a function that can decode objects into it.
Second, the AST-specific decoding logic is rewritten to use it. All AST nodes
are now supported, including definitions, statements, and expressions. The
overall approach here is to simply define a custom decoder for any interface
type that will occur in a node field position. The mapper, upon encountering
such a type, will consult the custom decoder map; if a decoder is found, it
will be used, otherwise an error results. This decoder then needs to switch
on the type discriminated kind field that is present in the metadata, creating
a concrete struct of the right type, and then converting it to the desired
interface type. Note that, subtly, interface types used only for "marker"
purposes don't require any custom decoding, because they do not appear in
field positions and therefore won't be encountered during the decoding process.
2017-01-16 17:41:26 +00:00
|
|
|
|
Implement mapper.Encode "for real"
This change implements `mapper.Encode` "for real" (that is, in a way
that isn't a complete embarrassment). It uses the obvious reflection
trickery to encode a tagged struct and its values as a JSON-like
in-memory map and collection of keyed values.
During this, I took the opportunity to also clean up a few other things
that had been bugging me. Namely, the presence of `mapper.Object` was
always error prone, since it isn't a true "typedef" in the sence that
it carries extra RTTI. Instead of doing that, let's just use the real
`map[string]interface{}` "JSON-map-like" object type. Even better, we
no longer require resource providers to deal with the mapper
infrastructure. Instead, the `Check` function can simply return an
array of errors. It's still best practice to return field-specific errors
to facilitate better diagnostics, but it's no longer required; and I've
added `resource.NewFieldError` to eliminate the need to import mapper.
As of this change, we can also consistently emit RPC structs with `lumi`
tags, rather than `lumi` tags on the way in and `json` on the way out.
This completes pulumi/lumi#183.
2017-06-06 00:43:52 +00:00
|
|
|
// defaultTags fetches the mapper's tag names from the options, or supplies defaults if not present.
|
|
|
|
func (md *mapper) defaultTags() (tags []string, optionalTags []string, skipTags []string) {
|
|
|
|
if md.opts.Tags == nil {
|
2017-09-22 02:18:21 +00:00
|
|
|
tags = []string{"json", "pulumi"}
|
Implement mapper.Encode "for real"
This change implements `mapper.Encode` "for real" (that is, in a way
that isn't a complete embarrassment). It uses the obvious reflection
trickery to encode a tagged struct and its values as a JSON-like
in-memory map and collection of keyed values.
During this, I took the opportunity to also clean up a few other things
that had been bugging me. Namely, the presence of `mapper.Object` was
always error prone, since it isn't a true "typedef" in the sence that
it carries extra RTTI. Instead of doing that, let's just use the real
`map[string]interface{}` "JSON-map-like" object type. Even better, we
no longer require resource providers to deal with the mapper
infrastructure. Instead, the `Check` function can simply return an
array of errors. It's still best practice to return field-specific errors
to facilitate better diagnostics, but it's no longer required; and I've
added `resource.NewFieldError` to eliminate the need to import mapper.
As of this change, we can also consistently emit RPC structs with `lumi`
tags, rather than `lumi` tags on the way in and `json` on the way out.
This completes pulumi/lumi#183.
2017-06-06 00:43:52 +00:00
|
|
|
} else {
|
|
|
|
tags = md.opts.Tags
|
Add custom decoders to eliminate boilerplate
This change overhauls the approach to custom decoding. Instead of decoding
the parts of the struct that are "trivial" in one pass, and then patching up
the structure afterwards with custom decoding, the decoder itself understands
the notion of custom decoder functions.
First, the general purpose logic has moved out of pkg/pack/encoding and into
a new package, pkg/util/mapper. Most functions are now members of a new top-
level type, Mapper, which may be initialized with custom decoders. This
is a map from target type to a function that can decode objects into it.
Second, the AST-specific decoding logic is rewritten to use it. All AST nodes
are now supported, including definitions, statements, and expressions. The
overall approach here is to simply define a custom decoder for any interface
type that will occur in a node field position. The mapper, upon encountering
such a type, will consult the custom decoder map; if a decoder is found, it
will be used, otherwise an error results. This decoder then needs to switch
on the type discriminated kind field that is present in the metadata, creating
a concrete struct of the right type, and then converting it to the desired
interface type. Note that, subtly, interface types used only for "marker"
purposes don't require any custom decoding, because they do not appear in
field positions and therefore won't be encountered during the decoding process.
2017-01-16 17:41:26 +00:00
|
|
|
}
|
Implement mapper.Encode "for real"
This change implements `mapper.Encode` "for real" (that is, in a way
that isn't a complete embarrassment). It uses the obvious reflection
trickery to encode a tagged struct and its values as a JSON-like
in-memory map and collection of keyed values.
During this, I took the opportunity to also clean up a few other things
that had been bugging me. Namely, the presence of `mapper.Object` was
always error prone, since it isn't a true "typedef" in the sence that
it carries extra RTTI. Instead of doing that, let's just use the real
`map[string]interface{}` "JSON-map-like" object type. Even better, we
no longer require resource providers to deal with the mapper
infrastructure. Instead, the `Check` function can simply return an
array of errors. It's still best practice to return field-specific errors
to facilitate better diagnostics, but it's no longer required; and I've
added `resource.NewFieldError` to eliminate the need to import mapper.
As of this change, we can also consistently emit RPC structs with `lumi`
tags, rather than `lumi` tags on the way in and `json` on the way out.
This completes pulumi/lumi#183.
2017-06-06 00:43:52 +00:00
|
|
|
if md.opts.OptionalTags == nil {
|
|
|
|
optionalTags = []string{"omitempty", "optional"}
|
|
|
|
} else {
|
|
|
|
optionalTags = md.opts.OptionalTags
|
Add custom decoders to eliminate boilerplate
This change overhauls the approach to custom decoding. Instead of decoding
the parts of the struct that are "trivial" in one pass, and then patching up
the structure afterwards with custom decoding, the decoder itself understands
the notion of custom decoder functions.
First, the general purpose logic has moved out of pkg/pack/encoding and into
a new package, pkg/util/mapper. Most functions are now members of a new top-
level type, Mapper, which may be initialized with custom decoders. This
is a map from target type to a function that can decode objects into it.
Second, the AST-specific decoding logic is rewritten to use it. All AST nodes
are now supported, including definitions, statements, and expressions. The
overall approach here is to simply define a custom decoder for any interface
type that will occur in a node field position. The mapper, upon encountering
such a type, will consult the custom decoder map; if a decoder is found, it
will be used, otherwise an error results. This decoder then needs to switch
on the type discriminated kind field that is present in the metadata, creating
a concrete struct of the right type, and then converting it to the desired
interface type. Note that, subtly, interface types used only for "marker"
purposes don't require any custom decoding, because they do not appear in
field positions and therefore won't be encountered during the decoding process.
2017-01-16 17:41:26 +00:00
|
|
|
}
|
Implement mapper.Encode "for real"
This change implements `mapper.Encode` "for real" (that is, in a way
that isn't a complete embarrassment). It uses the obvious reflection
trickery to encode a tagged struct and its values as a JSON-like
in-memory map and collection of keyed values.
During this, I took the opportunity to also clean up a few other things
that had been bugging me. Namely, the presence of `mapper.Object` was
always error prone, since it isn't a true "typedef" in the sence that
it carries extra RTTI. Instead of doing that, let's just use the real
`map[string]interface{}` "JSON-map-like" object type. Even better, we
no longer require resource providers to deal with the mapper
infrastructure. Instead, the `Check` function can simply return an
array of errors. It's still best practice to return field-specific errors
to facilitate better diagnostics, but it's no longer required; and I've
added `resource.NewFieldError` to eliminate the need to import mapper.
As of this change, we can also consistently emit RPC structs with `lumi`
tags, rather than `lumi` tags on the way in and `json` on the way out.
This completes pulumi/lumi#183.
2017-06-06 00:43:52 +00:00
|
|
|
if md.opts.SkipTags == nil {
|
|
|
|
skipTags = []string{"skip"}
|
|
|
|
} else {
|
|
|
|
skipTags = md.opts.SkipTags
|
2017-03-12 21:13:44 +00:00
|
|
|
}
|
2017-12-28 01:10:12 +00:00
|
|
|
return tags, optionalTags, skipTags
|
Add custom decoders to eliminate boilerplate
This change overhauls the approach to custom decoding. Instead of decoding
the parts of the struct that are "trivial" in one pass, and then patching up
the structure afterwards with custom decoding, the decoder itself understands
the notion of custom decoder functions.
First, the general purpose logic has moved out of pkg/pack/encoding and into
a new package, pkg/util/mapper. Most functions are now members of a new top-
level type, Mapper, which may be initialized with custom decoders. This
is a map from target type to a function that can decode objects into it.
Second, the AST-specific decoding logic is rewritten to use it. All AST nodes
are now supported, including definitions, statements, and expressions. The
overall approach here is to simply define a custom decoder for any interface
type that will occur in a node field position. The mapper, upon encountering
such a type, will consult the custom decoder map; if a decoder is found, it
will be used, otherwise an error results. This decoder then needs to switch
on the type discriminated kind field that is present in the metadata, creating
a concrete struct of the right type, and then converting it to the desired
interface type. Note that, subtly, interface types used only for "marker"
purposes don't require any custom decoding, because they do not appear in
field positions and therefore won't be encountered during the decoding process.
2017-01-16 17:41:26 +00:00
|
|
|
}
|
|
|
|
|
Implement mapper.Encode "for real"
This change implements `mapper.Encode` "for real" (that is, in a way
that isn't a complete embarrassment). It uses the obvious reflection
trickery to encode a tagged struct and its values as a JSON-like
in-memory map and collection of keyed values.
During this, I took the opportunity to also clean up a few other things
that had been bugging me. Namely, the presence of `mapper.Object` was
always error prone, since it isn't a true "typedef" in the sence that
it carries extra RTTI. Instead of doing that, let's just use the real
`map[string]interface{}` "JSON-map-like" object type. Even better, we
no longer require resource providers to deal with the mapper
infrastructure. Instead, the `Check` function can simply return an
array of errors. It's still best practice to return field-specific errors
to facilitate better diagnostics, but it's no longer required; and I've
added `resource.NewFieldError` to eliminate the need to import mapper.
As of this change, we can also consistently emit RPC structs with `lumi`
tags, rather than `lumi` tags on the way in and `json` on the way out.
This completes pulumi/lumi#183.
2017-06-06 00:43:52 +00:00
|
|
|
// structFieldTags includes a field's information plus any parsed tags.
|
|
|
|
type structFieldTags struct {
|
Tidy up more lint
This change fixes a few things:
* Most importantly, we need to place a leading "." in the paths
to Gometalinter, otherwise some sub-linters just silently skip
the directory altogether. errcheck is one such linter, which
is a very important one!
* Use an explicit Gometalinter.json file to configure the various
settings. This flips on a few additional linters that aren't
on by default (line line length checking). Sadly, a few that
I'd like to enable take waaaay too much time, so in the future
we may consider a nightly job (this includes code similarity,
unused parameters, unused functions, and others that generally
require global analysis).
* Now that we're running more, however, linting takes a while!
The core Lumi project now takes 26 seconds to lint on my laptop.
That's not terrible, but it's long enough that we don't want to
do the silly "run them twice" thing our Makefiles were previously
doing. Instead, we shall deploy some $$($${PIPESTATUS[1]}-1))-fu
to rely on the fact that grep returns 1 on "zero lines".
* Finally, fix the many issues that this turned up.
I think(?) we are done, except, of course, for needing to drive
down some of the cyclomatic complexity issues (which I'm possibly
going to punt on; see pulumi/lumi#259 for more details).
2017-06-22 19:09:46 +00:00
|
|
|
Info reflect.StructField // the struct field info.
|
Implement mapper.Encode "for real"
This change implements `mapper.Encode` "for real" (that is, in a way
that isn't a complete embarrassment). It uses the obvious reflection
trickery to encode a tagged struct and its values as a JSON-like
in-memory map and collection of keyed values.
During this, I took the opportunity to also clean up a few other things
that had been bugging me. Namely, the presence of `mapper.Object` was
always error prone, since it isn't a true "typedef" in the sence that
it carries extra RTTI. Instead of doing that, let's just use the real
`map[string]interface{}` "JSON-map-like" object type. Even better, we
no longer require resource providers to deal with the mapper
infrastructure. Instead, the `Check` function can simply return an
array of errors. It's still best practice to return field-specific errors
to facilitate better diagnostics, but it's no longer required; and I've
added `resource.NewFieldError` to eliminate the need to import mapper.
As of this change, we can also consistently emit RPC structs with `lumi`
tags, rather than `lumi` tags on the way in and `json` on the way out.
This completes pulumi/lumi#183.
2017-06-06 00:43:52 +00:00
|
|
|
Optional bool // true if this can be missing.
|
|
|
|
Skip bool // true to skip a field.
|
Tidy up more lint
This change fixes a few things:
* Most importantly, we need to place a leading "." in the paths
to Gometalinter, otherwise some sub-linters just silently skip
the directory altogether. errcheck is one such linter, which
is a very important one!
* Use an explicit Gometalinter.json file to configure the various
settings. This flips on a few additional linters that aren't
on by default (line line length checking). Sadly, a few that
I'd like to enable take waaaay too much time, so in the future
we may consider a nightly job (this includes code similarity,
unused parameters, unused functions, and others that generally
require global analysis).
* Now that we're running more, however, linting takes a while!
The core Lumi project now takes 26 seconds to lint on my laptop.
That's not terrible, but it's long enough that we don't want to
do the silly "run them twice" thing our Makefiles were previously
doing. Instead, we shall deploy some $$($${PIPESTATUS[1]}-1))-fu
to rely on the fact that grep returns 1 on "zero lines".
* Finally, fix the many issues that this turned up.
I think(?) we are done, except, of course, for needing to drive
down some of the cyclomatic complexity issues (which I'm possibly
going to punt on; see pulumi/lumi#259 for more details).
2017-06-22 19:09:46 +00:00
|
|
|
Key string // the JSON key name.
|
Add custom decoders to eliminate boilerplate
This change overhauls the approach to custom decoding. Instead of decoding
the parts of the struct that are "trivial" in one pass, and then patching up
the structure afterwards with custom decoding, the decoder itself understands
the notion of custom decoder functions.
First, the general purpose logic has moved out of pkg/pack/encoding and into
a new package, pkg/util/mapper. Most functions are now members of a new top-
level type, Mapper, which may be initialized with custom decoders. This
is a map from target type to a function that can decode objects into it.
Second, the AST-specific decoding logic is rewritten to use it. All AST nodes
are now supported, including definitions, statements, and expressions. The
overall approach here is to simply define a custom decoder for any interface
type that will occur in a node field position. The mapper, upon encountering
such a type, will consult the custom decoder map; if a decoder is found, it
will be used, otherwise an error results. This decoder then needs to switch
on the type discriminated kind field that is present in the metadata, creating
a concrete struct of the right type, and then converting it to the desired
interface type. Note that, subtly, interface types used only for "marker"
purposes don't require any custom decoding, because they do not appear in
field positions and therefore won't be encountered during the decoding process.
2017-01-16 17:41:26 +00:00
|
|
|
}
|
|
|
|
|
Implement mapper.Encode "for real"
This change implements `mapper.Encode` "for real" (that is, in a way
that isn't a complete embarrassment). It uses the obvious reflection
trickery to encode a tagged struct and its values as a JSON-like
in-memory map and collection of keyed values.
During this, I took the opportunity to also clean up a few other things
that had been bugging me. Namely, the presence of `mapper.Object` was
always error prone, since it isn't a true "typedef" in the sence that
it carries extra RTTI. Instead of doing that, let's just use the real
`map[string]interface{}` "JSON-map-like" object type. Even better, we
no longer require resource providers to deal with the mapper
infrastructure. Instead, the `Check` function can simply return an
array of errors. It's still best practice to return field-specific errors
to facilitate better diagnostics, but it's no longer required; and I've
added `resource.NewFieldError` to eliminate the need to import mapper.
As of this change, we can also consistently emit RPC structs with `lumi`
tags, rather than `lumi` tags on the way in and `json` on the way out.
This completes pulumi/lumi#183.
2017-06-06 00:43:52 +00:00
|
|
|
// structFieldsTags digs into a type to fetch all fields, including embedded structs, plus any associated tags.
|
|
|
|
func (md *mapper) structFieldsTags(t reflect.Type) []structFieldTags {
|
|
|
|
// Fetch the tag names to use.
|
|
|
|
tags, optionalTags, skipTags := md.defaultTags()
|
|
|
|
|
|
|
|
// Now walk the field infos and parse the tags to create the requisite structures.
|
|
|
|
var fldtags []structFieldTags
|
|
|
|
for _, fldinfo := range structFields(t) {
|
|
|
|
for _, tagname := range tags {
|
|
|
|
if tag := fldinfo.Tag.Get(tagname); tag != "" {
|
|
|
|
var key string // the JSON key name.
|
|
|
|
var optional bool // true if this can be missing.
|
|
|
|
var skip bool // true if we should skip auto-marshaling.
|
|
|
|
|
|
|
|
// Decode the tag.
|
|
|
|
tagparts := strings.Split(tag, ",")
|
|
|
|
contract.Assertf(len(tagparts) > 0,
|
|
|
|
"Expected >0 tagparts on field %v.%v; got %v", t.Name(), fldinfo.Name, len(tagparts))
|
|
|
|
key = tagparts[0]
|
|
|
|
if key == "-" {
|
|
|
|
skip = true // a name of "-" means skip
|
Add custom decoders to eliminate boilerplate
This change overhauls the approach to custom decoding. Instead of decoding
the parts of the struct that are "trivial" in one pass, and then patching up
the structure afterwards with custom decoding, the decoder itself understands
the notion of custom decoder functions.
First, the general purpose logic has moved out of pkg/pack/encoding and into
a new package, pkg/util/mapper. Most functions are now members of a new top-
level type, Mapper, which may be initialized with custom decoders. This
is a map from target type to a function that can decode objects into it.
Second, the AST-specific decoding logic is rewritten to use it. All AST nodes
are now supported, including definitions, statements, and expressions. The
overall approach here is to simply define a custom decoder for any interface
type that will occur in a node field position. The mapper, upon encountering
such a type, will consult the custom decoder map; if a decoder is found, it
will be used, otherwise an error results. This decoder then needs to switch
on the type discriminated kind field that is present in the metadata, creating
a concrete struct of the right type, and then converting it to the desired
interface type. Note that, subtly, interface types used only for "marker"
purposes don't require any custom decoding, because they do not appear in
field positions and therefore won't be encountered during the decoding process.
2017-01-16 17:41:26 +00:00
|
|
|
}
|
Implement mapper.Encode "for real"
This change implements `mapper.Encode` "for real" (that is, in a way
that isn't a complete embarrassment). It uses the obvious reflection
trickery to encode a tagged struct and its values as a JSON-like
in-memory map and collection of keyed values.
During this, I took the opportunity to also clean up a few other things
that had been bugging me. Namely, the presence of `mapper.Object` was
always error prone, since it isn't a true "typedef" in the sence that
it carries extra RTTI. Instead of doing that, let's just use the real
`map[string]interface{}` "JSON-map-like" object type. Even better, we
no longer require resource providers to deal with the mapper
infrastructure. Instead, the `Check` function can simply return an
array of errors. It's still best practice to return field-specific errors
to facilitate better diagnostics, but it's no longer required; and I've
added `resource.NewFieldError` to eliminate the need to import mapper.
As of this change, we can also consistently emit RPC structs with `lumi`
tags, rather than `lumi` tags on the way in and `json` on the way out.
This completes pulumi/lumi#183.
2017-06-06 00:43:52 +00:00
|
|
|
for _, part := range tagparts[1:] {
|
|
|
|
var match bool
|
|
|
|
for _, optionalTag := range optionalTags {
|
|
|
|
if part == optionalTag {
|
|
|
|
optional = true
|
|
|
|
match = true
|
|
|
|
break
|
|
|
|
}
|
Add custom decoders to eliminate boilerplate
This change overhauls the approach to custom decoding. Instead of decoding
the parts of the struct that are "trivial" in one pass, and then patching up
the structure afterwards with custom decoding, the decoder itself understands
the notion of custom decoder functions.
First, the general purpose logic has moved out of pkg/pack/encoding and into
a new package, pkg/util/mapper. Most functions are now members of a new top-
level type, Mapper, which may be initialized with custom decoders. This
is a map from target type to a function that can decode objects into it.
Second, the AST-specific decoding logic is rewritten to use it. All AST nodes
are now supported, including definitions, statements, and expressions. The
overall approach here is to simply define a custom decoder for any interface
type that will occur in a node field position. The mapper, upon encountering
such a type, will consult the custom decoder map; if a decoder is found, it
will be used, otherwise an error results. This decoder then needs to switch
on the type discriminated kind field that is present in the metadata, creating
a concrete struct of the right type, and then converting it to the desired
interface type. Note that, subtly, interface types used only for "marker"
purposes don't require any custom decoding, because they do not appear in
field positions and therefore won't be encountered during the decoding process.
2017-01-16 17:41:26 +00:00
|
|
|
}
|
Implement mapper.Encode "for real"
This change implements `mapper.Encode` "for real" (that is, in a way
that isn't a complete embarrassment). It uses the obvious reflection
trickery to encode a tagged struct and its values as a JSON-like
in-memory map and collection of keyed values.
During this, I took the opportunity to also clean up a few other things
that had been bugging me. Namely, the presence of `mapper.Object` was
always error prone, since it isn't a true "typedef" in the sence that
it carries extra RTTI. Instead of doing that, let's just use the real
`map[string]interface{}` "JSON-map-like" object type. Even better, we
no longer require resource providers to deal with the mapper
infrastructure. Instead, the `Check` function can simply return an
array of errors. It's still best practice to return field-specific errors
to facilitate better diagnostics, but it's no longer required; and I've
added `resource.NewFieldError` to eliminate the need to import mapper.
As of this change, we can also consistently emit RPC structs with `lumi`
tags, rather than `lumi` tags on the way in and `json` on the way out.
This completes pulumi/lumi#183.
2017-06-06 00:43:52 +00:00
|
|
|
if !match {
|
|
|
|
for _, skipTag := range skipTags {
|
|
|
|
if part == skipTag {
|
|
|
|
skip = true
|
|
|
|
match = true
|
|
|
|
break
|
|
|
|
}
|
|
|
|
}
|
Add custom decoders to eliminate boilerplate
This change overhauls the approach to custom decoding. Instead of decoding
the parts of the struct that are "trivial" in one pass, and then patching up
the structure afterwards with custom decoding, the decoder itself understands
the notion of custom decoder functions.
First, the general purpose logic has moved out of pkg/pack/encoding and into
a new package, pkg/util/mapper. Most functions are now members of a new top-
level type, Mapper, which may be initialized with custom decoders. This
is a map from target type to a function that can decode objects into it.
Second, the AST-specific decoding logic is rewritten to use it. All AST nodes
are now supported, including definitions, statements, and expressions. The
overall approach here is to simply define a custom decoder for any interface
type that will occur in a node field position. The mapper, upon encountering
such a type, will consult the custom decoder map; if a decoder is found, it
will be used, otherwise an error results. This decoder then needs to switch
on the type discriminated kind field that is present in the metadata, creating
a concrete struct of the right type, and then converting it to the desired
interface type. Note that, subtly, interface types used only for "marker"
purposes don't require any custom decoding, because they do not appear in
field positions and therefore won't be encountered during the decoding process.
2017-01-16 17:41:26 +00:00
|
|
|
}
|
Implement mapper.Encode "for real"
This change implements `mapper.Encode` "for real" (that is, in a way
that isn't a complete embarrassment). It uses the obvious reflection
trickery to encode a tagged struct and its values as a JSON-like
in-memory map and collection of keyed values.
During this, I took the opportunity to also clean up a few other things
that had been bugging me. Namely, the presence of `mapper.Object` was
always error prone, since it isn't a true "typedef" in the sence that
it carries extra RTTI. Instead of doing that, let's just use the real
`map[string]interface{}` "JSON-map-like" object type. Even better, we
no longer require resource providers to deal with the mapper
infrastructure. Instead, the `Check` function can simply return an
array of errors. It's still best practice to return field-specific errors
to facilitate better diagnostics, but it's no longer required; and I've
added `resource.NewFieldError` to eliminate the need to import mapper.
As of this change, we can also consistently emit RPC structs with `lumi`
tags, rather than `lumi` tags on the way in and `json` on the way out.
This completes pulumi/lumi#183.
2017-06-06 00:43:52 +00:00
|
|
|
contract.Assertf(match, "Unrecognized tagpart on field %v.%v: %v", t.Name(), fldinfo.Name, part)
|
Add custom decoders to eliminate boilerplate
This change overhauls the approach to custom decoding. Instead of decoding
the parts of the struct that are "trivial" in one pass, and then patching up
the structure afterwards with custom decoding, the decoder itself understands
the notion of custom decoder functions.
First, the general purpose logic has moved out of pkg/pack/encoding and into
a new package, pkg/util/mapper. Most functions are now members of a new top-
level type, Mapper, which may be initialized with custom decoders. This
is a map from target type to a function that can decode objects into it.
Second, the AST-specific decoding logic is rewritten to use it. All AST nodes
are now supported, including definitions, statements, and expressions. The
overall approach here is to simply define a custom decoder for any interface
type that will occur in a node field position. The mapper, upon encountering
such a type, will consult the custom decoder map; if a decoder is found, it
will be used, otherwise an error results. This decoder then needs to switch
on the type discriminated kind field that is present in the metadata, creating
a concrete struct of the right type, and then converting it to the desired
interface type. Note that, subtly, interface types used only for "marker"
purposes don't require any custom decoding, because they do not appear in
field positions and therefore won't be encountered during the decoding process.
2017-01-16 17:41:26 +00:00
|
|
|
}
|
|
|
|
|
Implement mapper.Encode "for real"
This change implements `mapper.Encode` "for real" (that is, in a way
that isn't a complete embarrassment). It uses the obvious reflection
trickery to encode a tagged struct and its values as a JSON-like
in-memory map and collection of keyed values.
During this, I took the opportunity to also clean up a few other things
that had been bugging me. Namely, the presence of `mapper.Object` was
always error prone, since it isn't a true "typedef" in the sence that
it carries extra RTTI. Instead of doing that, let's just use the real
`map[string]interface{}` "JSON-map-like" object type. Even better, we
no longer require resource providers to deal with the mapper
infrastructure. Instead, the `Check` function can simply return an
array of errors. It's still best practice to return field-specific errors
to facilitate better diagnostics, but it's no longer required; and I've
added `resource.NewFieldError` to eliminate the need to import mapper.
As of this change, we can also consistently emit RPC structs with `lumi`
tags, rather than `lumi` tags on the way in and `json` on the way out.
This completes pulumi/lumi#183.
2017-06-06 00:43:52 +00:00
|
|
|
fldtags = append(fldtags, structFieldTags{
|
|
|
|
Key: key,
|
|
|
|
Optional: optional,
|
|
|
|
Skip: skip,
|
|
|
|
Info: fldinfo,
|
|
|
|
})
|
Add custom decoders to eliminate boilerplate
This change overhauls the approach to custom decoding. Instead of decoding
the parts of the struct that are "trivial" in one pass, and then patching up
the structure afterwards with custom decoding, the decoder itself understands
the notion of custom decoder functions.
First, the general purpose logic has moved out of pkg/pack/encoding and into
a new package, pkg/util/mapper. Most functions are now members of a new top-
level type, Mapper, which may be initialized with custom decoders. This
is a map from target type to a function that can decode objects into it.
Second, the AST-specific decoding logic is rewritten to use it. All AST nodes
are now supported, including definitions, statements, and expressions. The
overall approach here is to simply define a custom decoder for any interface
type that will occur in a node field position. The mapper, upon encountering
such a type, will consult the custom decoder map; if a decoder is found, it
will be used, otherwise an error results. This decoder then needs to switch
on the type discriminated kind field that is present in the metadata, creating
a concrete struct of the right type, and then converting it to the desired
interface type. Note that, subtly, interface types used only for "marker"
purposes don't require any custom decoding, because they do not appear in
field positions and therefore won't be encountered during the decoding process.
2017-01-16 17:41:26 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
Implement mapper.Encode "for real"
This change implements `mapper.Encode` "for real" (that is, in a way
that isn't a complete embarrassment). It uses the obvious reflection
trickery to encode a tagged struct and its values as a JSON-like
in-memory map and collection of keyed values.
During this, I took the opportunity to also clean up a few other things
that had been bugging me. Namely, the presence of `mapper.Object` was
always error prone, since it isn't a true "typedef" in the sence that
it carries extra RTTI. Instead of doing that, let's just use the real
`map[string]interface{}` "JSON-map-like" object type. Even better, we
no longer require resource providers to deal with the mapper
infrastructure. Instead, the `Check` function can simply return an
array of errors. It's still best practice to return field-specific errors
to facilitate better diagnostics, but it's no longer required; and I've
added `resource.NewFieldError` to eliminate the need to import mapper.
As of this change, we can also consistently emit RPC structs with `lumi`
tags, rather than `lumi` tags on the way in and `json` on the way out.
This completes pulumi/lumi#183.
2017-06-06 00:43:52 +00:00
|
|
|
return fldtags
|
2017-03-12 21:13:44 +00:00
|
|
|
}
|