For testing it is very helpful to isolate sandbox changes to `~/.aws` from the host

…ound thread (aws#4946)

Co-authored-by: Richard Li <[email protected]>

…es (aws#4928)

Co-authored-by: Richard Li <[email protected]>

… to the service (aws#4884)

…rd events for telemetry. (aws#4951)

* Adding Acceptance metrics for UTG

* fix(dev): add error messages for upload URL expiry

* Update .changes/next-release/bugfix-caf370b9-c943-4ece-a4c0-81bcefa7a4f5.json

Co-authored-by: Richard Li <[email protected]>

---------

Co-authored-by: Richard Li <[email protected]>

* Add signout metric

* Fix broken sign out

…n OS CPU load is high (aws#4952)

* upgrade lsp to 0.1.13

* Update .changes/next-release/feature-ac28b77d-4b47-4312-afe7-ece71de355ed.json

Co-authored-by: Richard Li <[email protected]>

* Update .changes/next-release/feature-ac28b77d-4b47-4312-afe7-ece71de355ed.json

Co-authored-by: Richard Li <[email protected]>

---------

Co-authored-by: Richard Li <[email protected]>

* Add metadata to authAddConnection metrics

* detekt

* raise max indexing size to 4GB

* change log

* Update bugfix-e76aa3d8-3d9a-4887-8ce6-018ed4c7c443.json

---------

Co-authored-by: Richard Li <[email protected]>

…servability (aws#4953)

**Problem:** Feature dev instrumentation 1) records all failures as `FeatureDevException`, losing information about the specific exception type, and 2) lacks annotation of `reasonDesc` and the operation related to the error. (By operation, although we do capture `amazonq_codeGenerationInvoke`,  we're missing the subprocess name such as `StartCodeGeneration`.) As a result, we do not have insight into feature dev failures in JetBrains.


**Solution:** 1) Update exception modeling pattern for feature dev functionality to derive subclasses of `FeatureDevException`, providing a meaningful class name when inspected as error `reason`. 2) On `FeatureDevException`, accept an `operation` and `desc` to provide further annotation (merged when logged as `reasonDesc`).

This change will have conflicts with aws#4938 and aws#4949. I'll rebase on top of those and reconcile any drift in a subsequent revision.

…4938)

* Fix lint errors

* Fix lint errors for plugin-amazonq:chat

* Fix lint errors for plugin-toolkit:jetbrains-core

`CodeWhispererStatusBarWidget#getSelectedValue` is called under EDT, but may cause `ProfileWatcher` to be indirectly initialized on the background through `ToolkitConnectionManager#activeConnectionForFeature`. However `ProfileWatcher` requires EDT to initialize correctly since it relies on VFS events to propagate on EDT before it returns

Required for private forks

Incomplete rebase

…insertAtCursor and CopyAtClipboard events (aws#4959)

* Adding Acceptance metrics for UTG

… folders for certain project setup (aws#4975)

* fix wrong project root for @workspace for multi-folder project setup

* change log

* log

* Update .changes/next-release/bugfix-9e318f1e-baa7-4d10-98de-edfa70fcd7e0.json

Co-authored-by: Richard Li <[email protected]>

* collect file from different paths

---------

Co-authored-by: Richard Li <[email protected]>

* fix memory report telemetry

* rm extra change log

* auto-trigger a/b

* fix compile error

* detekt fix

* Keep one topic for all the listeners and changes around coroutine

* test fix

* detekt fix

* change force accept to shift+tab and remove opt+ enter

* keybinding change

* change feature value to be string as service config has

* test fix

* Revert "test fix"

This reverts commit b932c2c.

* test fix

aws#4970)

…to amazonq/shared (aws#4971)

…4972)

preparation work for aws#4976
depends on aws#4970, aws#4971

…ws#4988)

* refactor ProjectContextProvider and add tests

* add more test

* lint

* format json blob

* revert wrong merge conflict resolve

* lint

* should call updateIndex regardless index is complete or not which should be handled by LSP

…4990)

Jackson's class resolver, if not defined on its `TypeFactory`, first tries to use `Thread#getContextClassLoader` of the current thread to attempt to load the requested type. On failure it will then attempt to use its own classloader. Since by design the default classloader is not defined, the first load attempt will fail (since the context classloader is the root IDE classpath) and fall back to using the classloader that loaded Jackson.

Effectively, this means that if we want to avoid setting the type classloader for every Jackson `ObjectMapper` instance, we must always have Jackson in the plugin classloader, rather than using the version bundled by the IDE so that it can find the classes it needs.

At the moment we only include Jackson in plugin-core, but it introduces a question on what happens if a leaf plugin (i.e. plugin-amazonq/plugin-toolkit) loads Jackson through core. Will Jackson successfully resolve classes correctly? Inspecting the conetents of v3.20, when we fixed our last Jackson issue, the answer seems to be yes.

Interestingly, though this change brings our packaging back to v3.19, we do not seem to be facing the same deserialization issue on 2023.3 as last time (aws#4732)
> In Jackson 2.16.x, the Kotlin module uses reflection on the class annotations to determine if it should step in to deserialize a Kotlin class.
> However in Jackson 2.17.x, they switched to the standard isAnnotationPresent call, which ultimately performs an identity check on the annotation.
> In <241, the plugin classloader for the Q plugin INCORRECTLY loads the Kotlin metadata class from the Kotlin plugin, which results in isKotlinClass returning false (because the identity check fails if loaded from different classloaders) and then we get the cryptic cannot deserialize from Object value (no delegate- or property-based Creator) from the KotlinModule noop-ing.

And as always, unit testing of classloader issues is not possible at the moment because we are not set up for E2E tests and the classloader in headless tests differs from that of a running IDE.

aws#5003)

Currently /dev don't support stop code generation. This PR introduces this functionality (watch the video below).

Mynah UI provides onStopChatResponse API which we can hook in the cancellation token provided on VS Code, sharing across an active session, aborting current progress.

* Add accetpedCharCount into UserTriggerDecision STE

* detekt

* test fix

* Add an extra . before the file extensions

* addressed feedback

* detekt

* feedback

* Updated tests

* Add metric for location of reauth

* fixed tests

* Removed comment

… code changes + Increase polling rate for Q agent code summary (aws#5015)

* feat(amazonq) Increase polling rate for Q agent code summary

* Adjusted CodeGenerationStateTest to be aligned with new polling rate

* Removed forced Read-Only view for before diff on Q code changes

* Reverting remove read-only view change

* fix(amazonq): Fixed syntax error in CodeGenerationStateTest

* feat(Amazonq): Remove forced read-only view from before diff on agent code changess

---------

Co-authored-by: JuusticeT <[email protected]>
Co-authored-by: manodnyab <[email protected]>

* Add try-catch blocks for refresh steps.

* add metric that indicates when token was successfully written to the json & stored in cache

* remove comment

* add more metrics to catch blocks.

* Added reauth for expired client registration

* fix try-catch Exception blocks. remove bad reauth logic

* add metrics to saveClientRegistration

* make metrics same in saveClientRegistration

* Add modifyCredentials to invalidateClientRegistration

* Move save/modify credentials to diskCache

* linter

* saveCredentials moved to writeKey

* saveCredentials moved to writeKey

* capture saveCredentials per saveAccessToken/ClientRegistration calls

* add modifyCredentials + load stages to LoadClientRegistration flow

* remove success metrics

* add logs

* add auth_ModifyConnection to Override

* Fix Tests

* nit fixes

* consistent metrics

* null e message, clean up unused enums

* add more modifyConnection

* nits

* add enums to diskCache stages

* try block success metric

* detekt

---------

Co-authored-by: Manodnya Jaydeep Bhoite <[email protected]>

* fix perceivedLatency being calculated incorrectly

1. perceivedLatency was previously calculated at
the time of sending STE, but the timeOfLastCharTyped value
is no longer accurate, so calculate the perceivedLatency value
at suggestion showing time to get the most accurate timeOfLastCharTyped

* test fix

* temporarily disable test

* fix(amazonq): default keybinding for inline chat conflicts
---------

Co-authored-by: manodnyab <[email protected]>
Co-authored-by: Richard Li <[email protected]>

aws#5029)

Builds are still running out of disk space

…tion and onCopyCodeToClipboard function (aws#5031)

)

* hide inline chat shortcut hint on remote/gateway

* add change log

* Update .changes/next-release/bugfix-bc45640b-9877-43e5-b6cd-8b7dc35c570b.json

Co-authored-by: Richard Li <[email protected]>

---------

Co-authored-by: Richard Li <[email protected]>

* fix(dev): add state reassign to keep consistent across

* chore(dev): added changelog

* Update bugfix-8abe8650-ac4a-40ea-9243-071fd3104029.json

Co-authored-by: manodnyab <[email protected]>

---------

Co-authored-by: manodnyab <[email protected]>

…5037)

* chore(dev): added changelog

* fix: Update changelog message

---------

Co-authored-by: Thiago Verney <[email protected]>

* lambda LOG.warn

* typo

---------

Co-authored-by: manodnyab <[email protected]>

The OpenTelemetry SDK can automatically attach the same trace ID to spans recorded under the same 'context'. This is somewhat challenging because the JetBrains platform frequently requires switching threads and coroutine scope contexts, which loses the `ThreadLocal` state needed to connect parent->child when spans are started.

This on its own is not too difficult, but becomes a significant ergonomic challenge when we also want to bring some type-safety to how span metadata is built. This PR allows spans generated in aws/aws-toolkit-common#899 to automatically perform context propagation.

See internal documentation for details. A contributor-friendly version of the doc will be written after we validate this pattern with some initial metrics.

In CI, the override just drops all the source because all paths contain `/codebuild`

```
> Task :plugin-amazonq:codewhisperer:jetbrains-community:detektMain NO-SOURCE
Skipping task ':plugin-amazonq:codewhisperer:jetbrains-community:detektMain' as it has no source files and no previous output files.
Resolve mutations for :plugin-core:jetbrains-community:compileTestFixturesJava (Thread[aws#144,Execution worker Thread 4,5,main]) started.
:plugin-core:jetbrains-community:compileTestFixturesJava (Thread[aws#144,Execution worker Thread 4,5,main]) started.
```

…ws#5027)

* Depends on aws#4982
* Depends on aws/aws-toolkit-common#899
* Rename Unit -> MetricUnit directly in SDK codegen
* Use newer provider convention syntax for input/output in generator task
* expose `passive`/`unit`/`value` as special fields
* add smoke test for `validateRequiredAttributes`

This PR transforms OTel spans to the format expected by the current telemetry service

* improve indexing auto start stop

* lsp 0.1.26

* Add session_start telemetry to startup activities

* use runOnceUtil

* remove hardcoding of class name

* remove order first

* internal class

* Fix popup not init in EDT for remote

1. BackendComponentBePopupBuilder.createPopup somehow needs
to be executed in EDT while recent changes made our initPopup()
to happen in BGT.

* Update plugins/amazonq/codewhisperer/jetbrains-community/src/software/aws/toolkits/jetbrains/services/codewhisperer/service/CodeWhispererService.kt

Co-authored-by: Richard Li <[email protected]>

* change log

---------

Co-authored-by: Richard Li <[email protected]>

* add authScopes to emitUserState telemetry

* add authScopes to emitUserState telemetry

* Replace userState function with direct TelemetryService record function to capture authScopes

* add value parameter

* fix whitespace

* Update plugins/core/jetbrains-community/src/software/aws/toolkits/jetbrains/core/gettingstarted/GettingStartedAuthUtils.kt

Co-authored-by: Richard Li <[email protected]>

* Check current explorerConnection for scopes only

* Add function to grab scopes from each activeConnection used that is connected/expired.

* remove TelemetryOverride definition and use common's telemetry function instead

* remove unused imports

* Change the way explorer connection scope is checked

* remove check for CodeWhispererConnection: merged into qConnection & qScopes

* add common logic to one function

* Update telemetryOverride.json

missed {

---------

Co-authored-by: Richard Li <[email protected]>

* change log to info

* lazyLog

* Scrub names from messages in logging/telemetry

* removing commented code

* feedback

* detekt

…stion (aws#5067)

Co-authored-by: Richard Li <[email protected]>

…ws#5071)

* initial commit

* add source to loading metrics

* detekt

* fix tests

* fix more tests

…hen indexing finished (aws#5062)

* adjust LSP API timeout

* update time

Per product request. Tested in internal branch