* Do not apply content-type header if no request body
* Set content-type header if there are form params
* Add tests
* Run update script
* Regenerated sample clients
* Initial Commit
* Removed unneeded class, Updated names as per naming conventions, Removed licensing info from generated code
* Moved samples from client folder to server
* Fixed naming issue with scripts in the bin
* improved code format for api
* fixed compile issue and improved formatting
* fixed compile issue and improved formatting
* [csharp] Convert "false" properties to booleans
It appears as though "false" strings in additionalProperties are no
longer treated as false booleans. This may be an issue elsewhere, but a
simple fix is to always explicitly set the boolean value in a generator
class back to the additionalProperties map to convert boolean Strings to
boolean Objects.
* [nancyfx] Clean up async default option handling
* [nancyfx] Include asyncServer=false in sample script
* [csharp] Regenerate samples
* [csharp] Resolve .net 4 generation issues
Some functionality is missing from .NET 4.0, such as IReadonlyDictionary
and Type.GetTypeInfo().
This commit resolves compilation of generated .NET 4.0 code, requiring
no conditional versioning of Newtonsoft.Json.
* [csharp] Regenerate .net 4.0 sample
* [csharp] Resolve .NET 4.0 sample compile
Sample build.sh wasn't accounting for targeting different FCL correctly.
That is, when passing "net40" to the -sdk option, it would use the
default -sdk:4 and -langversion:6. These don't necessarily match with
what is installed on a machine with only .NET 4.0 (which is our targeted
use case here).
To resolve, we need to define another version-specific value for passing
to the mcs -sdk option (see man mcs for details).
This option currently isn't overridable in the client codegen class.
Also, langversion is set specifically to the version of C# available to
the targeted SDK version. If there is need, we may extend this to
something like:
langversion=${MCS_LANG_VERSION:-6}
To allow users to run as:
env MCS_LANG_VERSION=5 sh build.sh
I haven't done this because I doubt there's much of a use case via this
script. I'm assuming most consumers will build via IDE or MSBuild.
* [csharp] Revert bin/csharp-petstore.sh to 3.5
* [csharp] Regenerate .NET 3.5 sample
* [csharp] Resolve nuget issue with existing files
* [csharp] Update -all.sh, regenerate samples
* [csharp] Treat enum models consistently
C# works differently from most languages in that enums are not
considered objects. This means default(EnumType) will choose a default
of the first enum option. This isn't desirable because it breaks the
required = false functionality of swagger specs, which defines a
property which isn't required to exist in the message body.
Rather than force consumers to use enum values such as UNSPECIFIED, UNKNOWN,
NOT_SET, etc... we can treat enums as primitives. This means any
non-required enum will become Nullable<EnumType> regardless of whether
it is defined as an inline enum or a referenced enum model.
* Categorizing C# integration test for enums as general
* [csharp] Remove enum-ref integration test
* [csharp] Clean up general enum support integration test, validate different enum usage cases.
* [csharp][all] Assign one-based int to string enums
The EmitDefaultValue=false for string based enums will prevent the first
enum value from being serialized, because as 0 it is considered the
default.
This commit assigns an explicit numerical value to all non-integer
enums. This assignment has no effect on the
serialization/deserialization values, and only assigns the compiled
integer.
NOTE: This will have an effect of requiring recompilation of any code
that references the client/server models. This is because:
public enum Pet { Available }
Source files referencing Pet.Available as defined above will have a
constant 0 in place of the enum value.
public enum Pet { Available = 1 }
Source files referencing Pet.Available as defined above will have a
constant 1 in place of the enum value.
After compilation, Pet.Available in both instances lose their semantic
values and refer to the byte representation of their integral values.
For more info, see
https://docs.microsoft.com/en-us/dotnet/csharp/language-reference/keywords/enum
* [csharp] Treat enum models consistently
C# works differently from most languages in that enums are not
considered objects. This means default(EnumType) will choose a default
of the first enum option. This isn't desirable because it breaks the
required = false functionality of swagger specs, which defines a
property which isn't required to exist in the message body.
Rather than force consumers to use enum values such as UNSPECIFIED, UNKNOWN,
NOT_SET, etc... we can treat enums as primitives. This means any
non-required enum will become Nullable<EnumType> regardless of whether
it is defined as an inline enum or a referenced enum model.
* Categorizing C# integration test for enums as general
* [csharp] Remove enum-ref integration test
* [csharp] Clean up general enum support integration test, validate different enum usage cases.
* [kotlin] Nested enum naming fix, and naming options via CLI
* [kotlin] Add option test for enum property naming
* [kotlin] Escape all reserved/keywords for enums
flatten methods replaces original ObjectProperty objects with new
RefProperty objects which required variables are false as default.
Newly created objects' required variables are set as the original
property.
* revert changes to fakes yaml
the yaml currently includes invalid definitions that do not exist: ('#/definitions/xmlObject').
revert to known good yaml state.
* update template creators
* Rust API client/server generator
* `Future::boxed()` has been deprecated - replace with `Box::new(...)`
* rebasing to rust
* MMMLS211 use empty vec over none
* MMMLS211 rebuild after merge from rust
* MMMLS211 YAML array examples not wrapped in Some()
* MMMLS211 Array parameters bad mustache fix
* MMMLS211 don't parse map containers
* MMMLS211 Tidy container types
* MMMLS-211 rebuild example
* MMMLS211 mvn rebuild
* Percent-decode parameters contained in the path
* Produce warnings when unknown fields are present
We still accept unknown fields and discard them. However, to improve
diagnosability, we now write a warning log and return a `Warning`
header.
Note that this is server-only
* Markup
* MMMLS211: Make optional arrays Options again
* 211 markups
* Temporary attempt at tweaking Cow ownership tweak while merging changes from rust branch
* Remove to_string call while parsing path parameters, which requires definining a temporary var in a block because rust can't tell where a Cow reference gets dropped
* Fix rustfmt to the correct version
* Fix rustfmt to the correct version
* Add more response information to ApiError in client
* Re-add missing brace
* Code review markups
* Allow converting out of wrapper types
* Store arrays in new-types too
* Use a new hyper_client every request
* Add vec-like traits to array types
* Xml support - new branch
* Moved conversion from serde_xml_rs::Error to ApiError from swagger-rs to client code until upstream PR is accepted
* MMSUB-172 Don't set Content-Type when there's no body.
If we don't have a body we've serialized, then don't declare
any content type for the nonexistent body.
This is really important for 204 No Content responses, but it's
also morally important for all other non-bodied responses.
* MMSUB-172 Move to swagger 0.6.
* Manually implement debug for the client
* Allow `Context` to be bound to `Api`, and not passed on every function call
* Support "." in parameter names
* Support generate's "--reserved-words-mappings" option
* Support "." in parameter names
* Support generate's "--reserved-words-mappings" option
* Added support for enums in Dart.
* Pick non-private names for enum values.
The _ prefix denotes a private member in Dart, so avoid generating enum values starting with this character.
* Properly encode enum values into query paramters.
* Various cleanups.
* Add support for x-enum-values extension.
Use class instead of enum for better ergonomy.
Better generated enum names.
* Fixed test.
* Support enum descriptions.
Fixes#5985
We were experiencing syntax issues when generating the Petstore example. See some of the examples below:
StoreApiInterface.php
namespace Swagger\Server\Api;
use Symfony\Component\HttpKernel\Exception\NotFoundHttpException;
use Swagger\Server\Model\Order;
**use maparray<string,int>;**
use Symfony\Component\HttpKernel\Exception\BadRequestHttpException;
UserApiInterface.php
public function createUsersWithArrayInput(**User[]** $body);
public function createUsersWithListInput(User[] $body);
As far as I know, it is not possible to use array of objects in this way.
PetApiInterface.php
namespace Swagger\Server\Api;
use Symfony\Component\HttpKernel\Exception\NotFoundHttpException;
use Swagger\Server\Model\Pet;
use Symfony\Component\HttpKernel\Exception\MethodNotAllowedHttpException;
use Swagger\Server\Model\ApiResponse;
**use string[];**
use Symfony\Component\HttpKernel\Exception\BadRequestHttpException;
public function findPetsByStatus(string[] $status);
public function findPetsByTags(string[] $tags);
* Add addiitional files from upstream
* Remove mis-added files
* Fix compilation issue with Swift4 inline enums.
This change fixes this issue: https://github.com/swagger-api/swagger-codegen/issues/6607
The problem was that I was using "datatype" instead of "datatypeWithEnum" in the model.mustache file.
When you have a the following model property:
"myInlineStringEnum": {
"type": "string",
"enum": [
"inlineStringEnumValue1",
"inlineStringEnumValue2",
"inlineStringEnumValue3"
]
}
Then we were generating:
public enum MyInlineStringEnum: String, Codable {
case inlinestringenumvalue1 = "inlineStringEnumValue1"
case inlinestringenumvalue2 = "inlineStringEnumValue2"
case inlinestringenumvalue3 = "inlineStringEnumValue3"
}
However, when we decode this, we were using type of the enum ("datatype") rather than the enum type itself ("datatypeWithEnum"). So we were generating:
myInlineStringEnum = try container.decodeIfPresent(String.self, forKey: "myInlineStringEnum")
rather than:
myInlineStringEnum = try container.decodeIfPresent(MyInlineStringEnum.self, forKey: "myInlineStringEnum")
* Adds the ability to create code for an interface-based jaxrs client.
* Adds shell script and sample files for jaxrs-spec-interface
* rebase into adds shell
* Fixes bug in creation of Produces/Consumes in method annotation. Allows for instance "application/json; charset=utf-8"
* Fixes generated pom.xml
* Generate pom.xml by default
* Prettier output from api.mustache
* Fixes bug in mediatype, allowing charset-specification in swagger.yaml.
* Merges generation of interface-based jaxrs client/api into jaxrs-spec.
* Moves jaxrs-spec server interface to match location of jaxrs-spec server
* Makes Generated-annotation in genereated classes slightly prettier.
* Add test which reproduce the warning #5338https://github.com/swagger-api/swagger-codegen/issues/5338#issuecomment-292502895
`[main] WARN io.swagger.codegen.languages.PhpClientCodegen - Type object not handled properly in setParameterExampleValue`
* Fix Type object not handled properly in setParameterExampleValue
* Update samples
- /bin/php-petstore.sh
- /bin/security/php-petstore.sh
A type alias in this context is where a model is simply another name for a
primitive type, such as `MyString` in the following model definitions:
MyList:
type: array
items:
$ref: '#/definitions/MyString'
MyString:
type: string
It is valid to use a type alias as a property in another object or array model,
even if the object/array is defined before the alias is, as in the example
above. However, the current alias logic only looks "back" in list of previously
defined models, meaning that `MyList` would not know that `MyString` is an
alias. This change fixes the incorrect behavior by pre-calculating the list of
aliases before any models are processed. It also changes the test endpoint to
verify the correct behavior even when an object is defined before an alias it
uses.
* fix compilation error in eclipse
by updating package declarations in moved files
(eclipse validates that package and folder names match)
* permit specifying the full angular version
simplifying the templates by moving trivial case splits to the model
* remove dead code
this method is never called ...
* support HttpClient in addition to Http, clean up generated code
Fixes#6080
* added new sample, and regenerated existing samples
* updated samples
this time with the freshly build generator ...
* improve formatting
* updated samples
* fix compilation error in generated code
the overload for blobs does not have a type parameter
* added the first test for the utils package
* fix extra trainling commas in function argument lists
* regenerated samples
* Fix dependencies and generate model classes
* Better elixir client generation.
Responses are parsed and serialized by Poison into the model structs.
Use shared helper functions to generate the request.
Extract client connection configuration from api calls.
Elixir client can sanitize the operationId
Correctly output the model variables. Fix typos
Fix path replacement when there are multiple replacements
Cannot separate globally shared parameters from operations
Error handling for the tesla response
update templates
Can generate clients that compile
Can make requests - parse optional params, build query
Add oauth to connection. Fix connection directory
Add basic auth helper for creating a connection
Fix map types. Fix guard clauses for creaing connections
Add licenceInfo template. Parse config for moduleName via standard invokerPackage option
Can provide and inject a license header into all source files
fix location of connection.ex
Move shared code into reusable modules
Elixir filenames should be underscored
Fix visibility of helper functions
Parse the packageName from config options
Handle date and datetime fields with DateTime.from_iso8601
Fix indentation
Update documentation, add typespecs
Generate a standard elixir .gitignore
typespec is calculated recursively in java
Use the JSON middleware and using Poison.Decoder.decode on already parsed structs
move decoded struct into java
Fix handling of non-json responses
Switch basic auth to use the provided Tesla.Middleware.BasicAuth
Update README template to include the appDescription
Update sample elixir client
remove junk client models that don't belong with petstore
Only implement Poison.Decoder protocol if needed
Update samples with skipped Poison.Deocder impl
* Handle multipart file uploads
Handle building form params in the body
Files are handled as strings for input
* Requests with no defined return type will return the Tesla.Env response
* Run the bin/elixir-petstore.sh
Previously, we had implemented the Codable protocol by simply claiming conformance, and making sure that each of our internal classes also implemented the Codable protocol. So our model classes looked like:
class MyModel: Codable {
var propInt: Int
var propString: String
}
class MyOtherModel: Codable {
var propModel: MyModel
}
Previously, our additionalProperties implementation would have meant an object schema with an additionalProperties of Int type would have looked like:
class MyModelWithAdditionalProperties: Codable {
var additionalProperties: [String: Int]
}
But the default implementation of Codable would have serialized MyModelWithAdditionalProperties like this:
{
"additionalProperties": {
"myInt1": 1,
"myInt2": 2,
"myInt3": 3
}
}
The default implementation would put the additionalProperties in its own dictionary (which would be incorrect), as opposed to the desired serialization of:
{
"myInt1": 1,
"myInt2": 2,
"myInt3": 3
}
So therefore, the only way to support this was to do our own implementation of the Codable protocol. The Codable protocol is actually two protocols: Encodable and Decodable.
So therefore, this change generates implementations of Encodable and Decodable for each generated model class. So the new generated classes look like:
class MyModel: Codable {
var propInt: Int
var propString: String
// Encodable protocol methods
public func encode(to encoder: Encoder) throws {
var container = encoder.container(keyedBy: String.self)
try container.encode(propInt, forKey: "propInt")
try container.encode(propString, forKey: "propString")
}
// Decodable protocol methods
public required init(from decoder: Decoder) throws {
let container = try decoder.container(keyedBy: String.self)
propInt = try container.decode(Int.self, forKey: "propInt")
propString = try container.decode(String.self, forKey: "propString")
}
}
class MyOtherModel: Codable {
var propModel: MyModel
// Encodable protocol methods
public func encode(to encoder: Encoder) throws {
var container = encoder.container(keyedBy: String.self)
try container.encode(propModel, forKey: "propModel")
}
// Decodable protocol methods
public required init(from decoder: Decoder) throws {
let container = try decoder.container(keyedBy: String.self)
propModel = try container.decode(MyModel.self, forKey: "propModel")
}
}
* Implemented support for Play! 2.5.x / java8
* Added petstore sample for play25
* implemented play25 support
implemented version switch via playVersion config option
* fixed compilation failure after merge
* Disabled generation of CustomInstantDeserializer if dateLibrary equals to java8
* Rrename for angular and angularjs to reflect naming guidelines
see http://angularjs.blogspot.com.au/2017/01/branding-guidelines-for-angular-and.html
- move makes typescript-angular -> typescript-angularjs
- move makes typescript-angular2 -> typescript-angular
Language string with generate was already this so generation shouldnt change
but class naming follows this convention now
* Fixing merge problems
* Support additionalproperties and more.
- Upgrade to TypeScript 2
- Use type definition from npm instead of typings, typings is deprecation
- Use Enum instead of String Literal Types
- Use typescript es6 lib for target es5
- Support additionalproperties
- Support JSDoc
- Add snapshot and npmRepository option
- Update typescript-fetch run script for linux
- Create typescript-fetch run script for windows
* Update and fix
- Fix circle run script
- Fix duplicate query parameter
* Rename typescript-fetch folder to lowercase
* Fix for review and update new line end of file
* Fix end of file
* rename script to {lang}-petstore-all.sh and fix test
* Fix override query string
https://stackoverflow.com/a/7517673/1077943
* trying an approach with providing asyncio as a framework.
* adding example of asyncio.
* removing sample client to help PR look more manageable.
* possibly fixing a unit test
* getting unit test to pass again.
* addressing comments.
* add rust generator (1st release)
* update based on feedback
* fix reserved keyword
* fix string parameter
* Convert String to &str in trait definition
* Only pass pathParams to uri builder
* Fixed the html escaping in return type
* Fixed the hashmap constructor
* Added models into API scope
* removed models subimport, reference from super
* update returntype in method signature
* Fixed the remaining templates inconsistencies
* Fixed issues that floated up in kubernetes swagger file
* add hash support, fix docstring
* fix map parameter, update api.mustache
* use baseName for parameter
* use fully-qualfiied model name
* add rust tests
* fix test cases
* Rust gen slightly more idiomatic (#6247)
* Go -> Rust in README
* Remove leftover go file in rust sample
* rust: Regenerate sample
* rust: Rename *Impl -> *Client
* rust: one-line use line
More in line with common style
* rust: Replace tabs (in java) with 4 spaces
* Added trivial getter implementation (#6249)
* update rust petstore samples
