This mostly reverts fd34771841 along with a
number of slight adjustments to avoid other typing luxuries which were
introduced with TypeScript 3.x.
While the generic `ServerOptionFunction` types and `HandlerArguments`
generic rest argument type were certainly welcome additions to the codebase,
they present a backwards compatibility problem for consumers of Apollo
Server who have not yet made the jump to TypeScript 3.x.
With any luck, when we bump the major version of Apollo Server to 3.x, and
in accordance with semantic versioning, we'll be in the position to
straight-up revert this commit.
This commit follows-up on #1795, which introduced the new request pipeline, and
implements the triggers for its new life-cycle hooks within the various
integration packages. Previously, the only implementation was within the
`apollo-server` package and didn't get triggered for those invoking the
`applyMiddleware` method on integrations (e.g. `apollo-server-express`,
`...-hapi` and `...-koa`).
While in an ideal world, ALL existing `applyMiddleware` functions would be
marked as `async` functions and allow us to `await ApolloServer#willStart`,
in practice the only `applyMiddleware` which is currently `async` is the one
within the Hapi implementation. Therefore, we'll instead kick off the
`willStart` lifecycle hook as soon as `applyMiddleware` is started, return
as quickly as we have before and then (essentially) yield the completion of
Apollo Server's `willStart` prior to serving the first request — thus
ensuring the completion of server-startup activities.
Similarly, we'll do the same for `createHandler` methods on integrations
which don't utilize Node.js server frameworks but don't have `async`
handlers (e.g. AWS, Lambda, etc.).
`esModuleInterop` was enabled in
e4164c8892
to help with importing from packages that use default exports.
Those changes were reverted in
https://github.com/apollographql/apollo-server/pull/1210
to work around a few reported issues. Those issues are no longer
relevant, so this commit re-enables `esModuleInterop`, and
updates all default imports to use the more common (standard)
import syntax.
* Add a wider diversity of `gui` options
Although I know we want to remain less tied to the GraphQL Playground
GUI options, we definitely want to support a wider variety of options to
be passed in. This adds support for specifying partial options either
statically or dynamically for the gui, which can be extended to allow
for a wider array of guis than only GraphQL playground.
* Add boolean option and configuration for tabs
* move gui setting into ApolloServer Constructor
* document playground configuration in the constructor
* update playground types and fixed micro + koa integrations
* change gui to playground
* docs: change gui to playground
* fix logic for playground creation
* add disableInterval option to engine reporting
* set engine options for lambda
* lambda: add comment on constructor
* lambda: update readme typo and include callout to use graphql.js
* disableInterval -> sendReportsImmediately
* use sendReportsImmediately correctly and fix compilation
* uses new Header to fix different header combination, fixes#1301
* add apollo-server-env dependency
* fix logic error in engine reporting
* initial implementation of aws lambda integration for apollo server 2
* ApolloServer class
* package.json cleanup
* parse json of body
* createHandler function, complete README
* cors options
* update readme
* lambdaApollo tests
* fix tests
Previously, we were planning to offer two kinds of APIs in "Apollo Server 2":
middleware functions just like in 1.0, and the new ApolloServer class. We have
not yet implemented ApolloServer for all of our supported web frameworks, so
this meant that the 2.0 version of modules such as apollo-server-koa (which does
not yet have an ApolloServer class) offered very little benefits over
apollo-server-koa@1.0 (as most of the benefits of 2.0 come from ApolloServer).
This is confusing. We are going to improve the current registerServer API so
that there's no real benefit to using a separately-exported middleware directly
rather than creating an ApolloServer and applying it to your web framework. So
the AS 2.0 API will just be ApolloServer. This means it doesn't make sense for
us to publish 2.x versions of the packages that don't yet support this API.
So this commit removes support for the following web frameworks: Adonis, Azure
Functions, Koa, Lambda, Micro, and Restify. (We leave in place Express, Hapi,
and Cloudflare workers.)
This isn't because we don't like these frameworks and don't want them to work
with Apollo Server 2.0! We would love to see each package resurrected and an
ApolloServer implementation built, either during this current 2.x beta phase or
after the official 2.0 release. Deleting these packages for now makes it more
clear which frameworks support 2.0 and which don't, rather than existing in a
state where apollo-server-express@2 has ApolloServer and apollo-server-koa@2
does not.
As originally reported in [0], this should make debugging performance issues
easier, since the name of the function will be properly displayed/shown in
flame-graphs.
[0]: https://github.com/apollographql/apollo-server/pull/827
