Add C++-to-JS test for Smart Card Connector API #869
Comments
emaxx-google
added a commit
that referenced
this issue
Aug 1, 2023
Add build scripts and an empty test for testing Smart Card Connector. This boilerplate can be used for adding functional tests in follow-ups as tracked by #869.
emaxx-google
added a commit
that referenced
this issue
Aug 1, 2023
Add build scripts and an empty test for testing Smart Card Connector. This boilerplate can be used for adding functional tests in follow-ups as tracked by #869.
emaxx-google
added a commit
that referenced
this issue
Aug 1, 2023
Add build scripts and an empty test for testing Smart Card Connector. This boilerplate can be used for adding functional tests in follow-ups as tracked by #869.
emaxx-google
added a commit
that referenced
this issue
Aug 1, 2023
Add build scripts and an empty test for testing Smart Card Connector. This boilerplate can be used for adding functional tests in follow-ups as tracked by #869.
emaxx-google
added a commit
that referenced
this issue
Aug 1, 2023
Change IntegrationTestHelper::SetUp() to return its result asynchronously via a callback. These asynchronous helpers will be used in follow-up PRs for doing startup/shutdown of the PC/SC server in JS-to-C++ tests of Smart Card Connector, tracked by #869.
emaxx-google
added a commit
that referenced
this issue
Aug 1, 2023
Change IntegrationTestHelper::SetUp() to return its result asynchronously via a callback. These asynchronous helpers will be used in follow-up PRs for doing startup/shutdown of the PC/SC server in JS-to-C++ tests of Smart Card Connector, tracked by #869.
This was referenced Aug 1, 2023
emaxx-google
added a commit
that referenced
this issue
Aug 1, 2023
Assert (and crash if it fails) that by the time IntegrationTestService::Deactivate() is called all previously set up helpers have been torn down. This is anyway guaranteed by the current behavior of the test controller. Also this assumption greatly simplifies transition to the asynchronous teardown model, because it'll be hard to make Deactivate() asynchronous as well (it's called in the "entry point" class destructor). This commit contributes to #869.
emaxx-google
added a commit
that referenced
this issue
Aug 1, 2023
Change IntegrationTestHelper::TearDown() to report its completion asynchronously via a callback. This change is similar to the previous change for SetUp(), except that teardown doesn't need to return any value, and that we have to properly accoutn for callbacks in the bulk "tear down all helpers" operation. This commit contributes to #869.
emaxx-google
added a commit
that referenced
this issue
Aug 2, 2023
Add SmartCardConnectorApplicationTestHelper that allows the JS side to start the instance of the application (i.e., the PC/SC daemon) in tests. The helper currently only supports the most basic operations: startup and shutdown. Its usages and additional operations will be added in follow-ups tracked by #869.
emaxx-google
added a commit
that referenced
this issue
Aug 2, 2023
Add SmartCardConnectorApplicationTestHelper that allows the JS side to start the instance of the application (i.e., the PC/SC daemon) in tests. The helper currently only supports the most basic operations: startup and shutdown. Its usages and additional operations will be added in follow-ups tracked by #869.
emaxx-google
added a commit
that referenced
this issue
Aug 2, 2023
Add a test that verifies the PC/SC server successfully started up. Technically, the JS side does this by asking the C++ side to set up the "SmartCardConnectorApplicationTestHelper", and then waiting until the C++ side sends the "ready" message to the JS. Follow-ups will use this approach for testing actual PC/SC functionality, as tracked by #869.
emaxx-google
added a commit
that referenced
this issue
Aug 2, 2023
Closure Library's standard test timeout - 1 second - is insufficient for JS-to-C++ tests that spend multiple seconds alone on setup/teardown of the executable module. Increase the limit to 60 seconds for now. This commit contributes to #869.
emaxx-google
added a commit
that referenced
this issue
Aug 2, 2023
Change IntegrationTestHelper::SetUp() to return its result asynchronously via a callback. These asynchronous helpers will be used in follow-up PRs for doing startup/shutdown of the PC/SC server in JS-to-C++ tests of Smart Card Connector, tracked by #869.
emaxx-google
added a commit
that referenced
this issue
Aug 2, 2023
Assert (and crash if it fails) that by the time IntegrationTestService::Deactivate() is called all previously set up helpers have been torn down. This is anyway guaranteed by the current behavior of the test controller. Also this assumption greatly simplifies transition to the asynchronous teardown model, because it'll be hard to make Deactivate() asynchronous as well (it's called in the "entry point" class destructor). This commit contributes to #869.
emaxx-google
added a commit
that referenced
this issue
Aug 2, 2023
Assert (and crash if it fails) that by the time IntegrationTestService::Deactivate() is called all previously set up helpers have been torn down. This is anyway guaranteed by the current behavior of the test controller. Also this assumption greatly simplifies transition to the asynchronous teardown model, because it'll be hard to make Deactivate() asynchronous as well (it's called in the "entry point" class destructor). This commit contributes to #869.
emaxx-google
added a commit
that referenced
this issue
Aug 2, 2023
Change IntegrationTestHelper::TearDown() to report its completion asynchronously via a callback. This change is similar to the previous change for SetUp(), except that teardown doesn't need to return any value, and that we have to properly accoutn for callbacks in the bulk "tear down all helpers" operation. This commit contributes to #869.
emaxx-google
added a commit
that referenced
this issue
Aug 2, 2023
Change IntegrationTestHelper::TearDown() to report its completion asynchronously via a callback. This change is similar to the previous change for SetUp(), except that teardown doesn't need to return any value, and that we have to properly accoutn for callbacks in the bulk "tear down all helpers" operation. This commit contributes to #869.
emaxx-google
added a commit
that referenced
this issue
Aug 2, 2023
Add SmartCardConnectorApplicationTestHelper that allows the JS side to start the instance of the application (i.e., the PC/SC daemon) in tests. The helper currently only supports the most basic operations: startup and shutdown. Its usages and additional operations will be added in follow-ups tracked by #869.
emaxx-google
added a commit
that referenced
this issue
Aug 2, 2023
Add SmartCardConnectorApplicationTestHelper that allows the JS side to start the instance of the application (i.e., the PC/SC daemon) in tests. The helper currently only supports the most basic operations: startup and shutdown. Its usages and additional operations will be added in follow-ups tracked by #869.
emaxx-google
added a commit
that referenced
this issue
Aug 2, 2023
Add a test that verifies the PC/SC server successfully started up. Technically, the JS side does this by asking the C++ side to set up the "SmartCardConnectorApplicationTestHelper", and then waiting until the C++ side sends the "ready" message to the JS. Follow-ups will use this approach for testing actual PC/SC functionality, as tracked by #869.
emaxx-google
added a commit
that referenced
this issue
Aug 2, 2023
Add a test that verifies the PC/SC server successfully started up. Technically, the JS side does this by asking the C++ side to set up the "SmartCardConnectorApplicationTestHelper", and then waiting until the C++ side sends the "ready" message to the JS. Follow-ups will use this approach for testing actual PC/SC functionality, as tracked by #869.
emaxx-google
added a commit
that referenced
this issue
Aug 2, 2023
Specify a generous timeout when running JS-to-C++ tests in Emscripten mode (via run-js-tests.py, which in turn uses Chromedriver). The default timeout of 60 seconds is insufficient for anything more than a single test case with the PC/SC server startup+shutdown. This contributes to #869.
emaxx-google
added a commit
that referenced
this issue
Aug 29, 2023
Verify the JavaScript API's SCardGetStatusChange() behavior on reader plug/unplug events. This commit contributes to #869.
emaxx-google
added a commit
that referenced
this issue
Aug 29, 2023
Verify the JavaScript API's SCardGetStatusChange() behavior on reader plug/unplug events. This commit contributes to #869.
emaxx-google
added a commit
that referenced
this issue
Aug 29, 2023
Verify the JavaScript API's SCardGetStatusChange() behavior on card insertion/removal events. This commit contributes to #869.
emaxx-google
added a commit
that referenced
this issue
Aug 29, 2023
Verify the JavaScript API's SCardGetStatusChange() behavior on card insertion/removal events. This commit contributes to #869.
emaxx-google
added a commit
that referenced
this issue
Aug 29, 2023
Verify the JavaScript API's SCardCancel() behavior on canceling a blocking SCardGetStatusChange() call. This commit contributes to #869.
emaxx-google
added a commit
that referenced
this issue
Aug 29, 2023
Verify the JavaScript API's SCardCancel() behavior on canceling a blocking SCardGetStatusChange() call. This commit contributes to #869.
emaxx-google
added a commit
that referenced
this issue
Aug 29, 2023
Test various edge cases of receiving incorrect PC/SC requests. The only thing to verify we don't crash for them. This commit contributes to #869.
emaxx-google
added a commit
that referenced
this issue
Aug 29, 2023
Test various edge cases of receiving incorrect PC/SC requests. The only thing to verify we don't crash for them. This commit contributes to #869.
emaxx-google
added a commit
that referenced
this issue
Aug 29, 2023
Verify the JavaScript API's SCardGetStatusChange() behavior on card insertion/removal events. This commit contributes to #869.
emaxx-google
added a commit
that referenced
this issue
Aug 29, 2023
Create a separate static library containing the Connector's application.cc logic, as opposed to compiling this file 3 times for each target that needed it (executable_module, cpp_unittests, js_to_cxx_tests). This reduces duplication across different makefiles and should slightly improve overall build times. This contributes to #869 as it introduced one extra recompilation of the same source.
emaxx-google
added a commit
that referenced
this issue
Aug 29, 2023
Create a separate static library containing the Connector's application.cc logic, as opposed to compiling this file 3 times for each target that needed it (executable_module, cpp_unittests, js_to_cxx_tests). This reduces duplication across different makefiles and should slightly improve overall build times. This contributes to #869 as it introduced one extra recompilation of the same source.
emaxx-google
added a commit
that referenced
this issue
Aug 29, 2023
Create a separate static library containing the Connector's application.cc logic, as opposed to compiling this file 3 times for each target that needed it (executable_module, cpp_unittests, js_to_cxx_tests). This reduces duplication across different makefiles and should slightly improve overall build times. This contributes to #869 as it introduced one extra recompilation of the same source.
emaxx-google
added a commit
that referenced
this issue
Aug 30, 2023
Make JS-to-C++ tests also work in NaCl builds, by initializing the nacl_io library (and hence allowing to perform file operations). This contributes to effort tracked by #869.
emaxx-google
added a commit
that referenced
this issue
Aug 30, 2023
Verify the JavaScript API's SCardGetStatusChange() behavior on card insertion/removal events. This commit contributes to #869.
emaxx-google
added a commit
that referenced
this issue
Aug 30, 2023
Create a separate static library containing the Connector's application.cc logic, as opposed to compiling this file 3 times for each target that needed it (executable_module, cpp_unittests, js_to_cxx_tests). This reduces duplication across different makefiles and should slightly improve overall build times. This contributes to #869 as it introduced one extra recompilation of the same source.
emaxx-google
added a commit
that referenced
this issue
Aug 30, 2023
Make JS-to-C++ tests also work in NaCl builds, by initializing the nacl_io library (and hence allowing to perform file operations). This contributes to effort tracked by #869.
emaxx-google
added a commit
that referenced
this issue
Aug 30, 2023
Test various edge cases of receiving incorrect PC/SC requests. The only thing to verify we don't crash for them. This commit contributes to #869.
emaxx-google
added a commit
that referenced
this issue
Aug 30, 2023
Test various edge cases of receiving incorrect PC/SC requests. The only thing to verify is that we don't crash. This commit contributes to #869.
This was referenced Sep 8, 2023
This was referenced Sep 12, 2023
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
With #816 implemented, we have a framework that allows writing JavaScript tests that exercise both JavaScript and C/C++ code.
We should employ this to write basic tests for verifying the PC/SC API exposed by the Connector.
The text was updated successfully, but these errors were encountered: