This will change the `AbstractWebview` to dispose its tracked objects
(using `this.push`) when the panel is disposed rather than when the
view is disposed. This makes `this.push` actually useful in a view.
Before, the objects would only get disposed when the extension itself
was disposed.
This updates the model editor queries to the version that will be merged
into the CodeQL repository. There are some slight changes to the output
format, so we slightly need to change the BQRS decoding of those
queries.
The queries themselves were copied from the two PRs with some minor
additions at the end since these were changes in core CodeQL library
files.
After the upgrade to the correct types for js-yaml, the return type
of `load` is correctly typed as `unknown`. This means that we can't
use the return value directly, but need to validate it first.
This adds such validation by generating a JSON schema for a newly
created typed. The JSON schema generation is very similar to how we do
it in https://github.com/github/codeql-variant-analysis-action.
Before this change and starting with CLI v2.14.3, if you wanted to run
a variant analysis query and the pack it is contained in has at least
one query that contains an extensible predicate, this would be an error.
The reason is that v2.14.3 introduced deep validation for data
extensions. We are not copying the query that contains an extensible
predicate to the synthetic pack we are generating. This means that deep
validation will fail because there will be extensions that target the
missing extensible predicate.
This change avoids the problem by copying any query files that contain
extensible predicates to the synthetic pack. It uses the internal
`generate extensible-predicate-metadata` command to discover which
query files contain extensible predicates and copies them.
* Don't download artifacts for repos with no results
* Remove getVariantAnalysisRepoResult requests for repos with no results
* Run fix-scenario-file-numbering for mrva-problem-query-success scenario
* Update CHANGELOG
When running tests using `--runTestsByPath <some-path>`, the tests were
being run twice because jest-runner-vscode
[resolves test paths](0c98dc12ad/packages/jest-runner-vscode/src/runner.ts (L57-L66)),
while the original arguments were also still passed to Jest. So, the
arguments Jest would receive would look something like
`test/vscode-tests/no-workspace/databases/local-databases/locations.test.ts /Users/koesie10/github/vscode-codeql/extensions/ql-vscode/test/vscode-tests/no-workspace/databases/local-databases/locations.test.ts`
which would cause Jest to run the tests twice. This fixes this by
resolving the paths to their absolute paths, and then removing any
duplicates.
This commit fixes a bug in the extension where the qhelp preview was not
being refreshed after the first time the preview was rendered. The
reason is that vscode will not refresh the markdown preview unless the
original file with the markdown in it is already open in the editor.
This fix will briefly open the raw markdown, refresh the preview and
close the raw markdown.
An upcoming change in the CLI will require that the extensible
predicates that are targeted by a data extension needs to be available
in order for the `resolve extensions` command to succeed.
There are a handful of tests that are failing with this new CLI. This
change will update the tests so that the `codeql/java-all` pack is
available in the tests and ensures they pass.
pick 2db42e3e Pull out createDataExtensionYamls into yaml.ts
pick 52f7cac0 Move saveModeledMethods to a separate file
pick ba27230e Move loadModeledMethods to a separate file
pick c512a11e Split out listModelFiles from loadModeledMethods
pick 752cf8ab Add some tests of listModelFiles
This removes the call type as shown for an unmodelable method. We still
need to decide how to show this information, so this may be added back
in later.
* Move code-tour.ts to /code-tour
* Move qlpack-generator.ts to /local-queries
* Move query-status.ts to /query-history
* Move skeleton-query-wizard.ts to /local-queries
* Add version constraint for Check errors
* Refactor parts of the ideserver out of extension.ts
* Give visibility information to the ide-server.
This allows it to report errors on visible files
eagerly.
This adds a refresh button to the data extensions editor when the
framework mode feature flag is enabled. If you are using framework mode,
you can have multiple tabs of the data extensions editor open in which
you are modeling the library separately from the application. When you
save the library in framework mode, the application mode will not
refresh and show that these calls have been modeled. Rather than using
apply, which might also save all modeled methods, you can now use the
refresh button to refresh the external API usages and whether they are
supported.
This excludes methods defined in tests in framework mode, significantly
cutting down on the number of methods shown that would need to be
modeled.
For C#, this just checks that the file is not a test file, as defined
by the QL library.
For Java, this makes a copy of the internal
[`ModelExclusions.qll`](249f9f863d/java/ql/lib/semmle/code/java/dataflow/internal/ModelExclusions.qll)
file to avoid the use of internal modules. This module will tell us
whether a method is "interesting" to model or not. Not all of the checks
in this module need to happen for framework mode, but these checks might
be useful for telling a user whether a method is interesting to model
in application mode.
This makes the method name and parameters in framework mode a link to
the definition of the method. In framework mode, the `usages` contains
1 element, which is the location of the definition of the method.
Therefore, we can simply use `jumpToUsage` to jump to the definition.
Similar to https://github.com/github/vscode-codeql/pull/2553, this
changes the C# query to correctly report the name of nested types. I
couldn't find a `nestedName` method for C#, so this adds one in the
`AutomodelVsCode` library.
C# seems to use `+` as a separator for nested types, as reported by
`getQualifiedName()`:
```
GitHub.Nested.MyFirstClass+NestedClass
```
The `getApiName()` will now report:
```
GitHub.Nested#MyFirstClass+NestedClass.Test()
```
Adds a command to run all queries in a certain folder.
This uses the existing `runQueries` command, which lets you run multiple queries against the selected local database.
We don't have a corresponding command for running multiple variant analysis queries, so I haven't implemented that.
This changes the Java `CallableMethod.getApiName()` to use `nestedName`
instead of `getSourceDeclaration`. `getSourceDeclaration` would return a
`RefType`, on which the `toString()` method returns its `getName`().
However, for nested types this wouldn't work and wouldn't include the
enclosing type. This fixes it by using `nestedName` which matches the
method that is also used for determining whether a type matches an
extensible predicate.
* Add version constraint for quick-eval-count
* Add quick eval count context.
* Add support for running quick-eval-count from the command pallete
* Adjust name for quick-eval-count-queries
* Add changenote for quick-eval-count.
* QuickEval:Address review comments
* Fix rebase conflict in changelog
This refactors the data extensions editor queries to use a new
`AutomodelVsCode` module. This module is based on the `ExternalApi`
module, but is more general and can be used for retrieving public
methods from the source as well. The actual conditions are now in the
queries themselves.
This reduces the duplicated module in the framework mode query and will
mean that when we update the `ExternalApi` module, we will just have to
port it to the `AutomodelVsCode` module, and not to the `ExternalApi`
and a separate framework mode query.
This changes the sorting of the methods sent to LLM to match the order
shown in the data extensions editor. This will ensure that the methods
which are shown first in the data extensions editor are also modeled
first.
This will allow users to override the default directory in which
extension packs are created by using the
`codeQL.dataExtensions.extensionsDirectory` setting. This setting can be
overriden per language, so the user could create the following
configuration to set the extension pack setting for Java only:
```json
"[java]": {
"codeQL.dataExtensions.extensionsDirectory": "/Users/user/github/vscode-codeql-starter/codeql-custom-queries-java",
}
```
This will remove the user input for a model file and will instead create
1 model file per library (JAR/DLL). The model filename will be based on
the JAR/DLL name, but will remove the version number and the file
extension. It will also normalize the name.
These files will be created automatically, and the editor now also reads
in all files contained in an extension pack to read the modeled methods.
This could result in duplicates if the user has created a different file
to contain the same modeled methods, but this is an edge-case that we're
explicitly not handling.
The `logging` part of `common` was exported in `common/index.ts` and
could be imported by importing `common`. I don't think this makes a lot
of sense, so I removed it.
The `common/logging/index.ts` also contained exports of the
`common/logging/vscode` directory, which means that importing
`common/logging` automatically brings in the `vscode` module. This
removes that export, so now there are two separate imports needed for
importing the common part and the `vscode` specific part. This should
make it easier to keep them separate and be more explicit about what
you're importing.
This will change how extension packs are named in the data extensions
editor. Before, the user had to pick a workspace folder and a name for
the extension pack. Now, the workspace folder will be picked
automatically if we can detect it (i.e. it follows the naming structure
we expect), or the user will still need to select it. The extension pack
name is always auto-generated based on the database name and the
database language.
This adds a new `codeQL.dataExtensions.disableAutoNameExtensionPack`
setting to disable this behavior while we are still working on changing
how the data extensions editor works.
This adds a new library column to the data extensions editor containing
the JAR or DLL file the method is defined in. This will be used to group
methods by library in the future. For now, it just shows in a column.
See https://github.com/github/vscode-codeql/pull/2490#discussion_r1226437598
for more explanation. This will make the class more useful for future usecases
where we don't want the behaviour of only calling update when there isn't
another refresh scheduled. I also think it doesn't negatively affect other
users such as the query test discovery. The effect should be that we'll see
more updates to the UI. These updates will get overwritten quickly, but they
are all genuine snapshots of the filesystem at the point the discovery process
ran, so they aren't incorrect, or aren't more incorrect than continuing to show
the old state before any discovery ran.
To increase the use of the `app` logger, this replaces the direct use of
`extLogger` by the `app.logger` where possible. This should not change
the behavior since the `extLogger` is the logger used by the `app`.
This moves the `showAndLog` family of functions to the `common/logging`
directory. It explicitly moves the `showAndLogExceptionWithTelemetry`
function to the `common/vscode/logging.ts` file because it still has a
dependency on the `telemetryListener`, which depends on the `vscode`
module.
It seems like some CLI tests are hanging and only completing after 6
hours when they run into the default timeout. This updates the timeout
to 30 minutes. All CLI tests should complete in 30 minutes, so this
should ensure that they are cancelled when they are stuck.
This moves the Webview HTML generation used by `AbstractWebview` out of
`interface-utils.ts` and into a new file `webview-html.ts` in the
`common/vscode` directory.
When packaging an extension pack, unscoped extension pack names are not
allowed and calling `codeql pack bundle` will fail with an error. This
command will be called when running a variant analysis, so these packs
will not work for a variant analysis.
To improve the user experience, we now only allow scoped extension pack
names. This means that the user will now have to enter a scope when
creating a new extension pack.
This option was used to ignore source archives for `.testproj`
databases. It is only set to `true` or `false` when creating the
database and could not be changed, so I don't think we need this option.
It can simply be derived from the database URI. This simplifies handling
of databases a bit.
This moves the `refresh` method from `DatabaseItem` to `DatabaseManager`
and makes it private. This makes the `DatabaseItem` interface smaller
and more focused and ensures that `refresh` cannot be called from
outside of the `DatabaseManager`.
This will allow us to implement specific behavior on the
`DatabaseItemImpl` which is not available on the `DatabaseItem`. This
will allow us to make the surface area of the `DatabaseItem` smaller.
Generated variant analysis packs will use the original name
of the pack that the query is located in. This is to support
some future work where we do extra validation of data extensions.
If the query is not in a pack, the default name is used.
This adds a warning when the user is using an unsupported version of the
CodeQL CLI. The warning is shown once per session, and only if the
version is older than the oldest supported version.
When the user filters the repositories, the buttons should reflect that
the results are filtered and that the user is not exporting or copying
all the results. If the user has selected repositories, the buttons
should still say that they are exporting selected results.
This changes the text of the export/copy buttons on a variant analysis
when at least one repository is selected. This makes it more clear that
the user is only exporting/copying the results of the selected
repositories.
The compare view typically works by matching the results sets of
queries. It only allows the results sets of queries with identical
names to be compared.
Manually run queries use `#select` as the default result set. However,
quick eval queries use a different, generated, name. Therefore, these
two kinds of queries cannot be compared.
This commit changes the heuristics so that if there are no identical;y
named results sets, the first result set of each query that is prefixed
with `#` is used (this is the default result set).
It also makes a slightly nicer error message if there are no comparable
results sets.
The data extensions editor was always setting the `provenance` field of
MaD to `manual`. This will change the `provenance` to be either
`editor-manual` (for models which were added by the user),
`df-generated` (for models generated by the dataflow generator), and
`df-manual` (for models generated and then edited). This makes it easier
to trace the origin of a model.
This will change the data extensions editor generator to resolve the
queries based on the `modelgenerator` tag. This removes the requirement
of having a `ql` folder in the workspace.
I chose to use the path instead of the `id` for now to avoid having to
resolve the query 4 times. This also avoids the need to map the language
names to the language ID in the tag (i.e. `csharp` -> `cs`).
* Log stdout when servers are terminated with errors.
This logs the last stdout chunk (probabaly the last line) if things
went wrong. This can sometimes be useful for debugging.
It also prints the signal when killed by a signal
(rather than printing null)
* Restart/Abort the queryserver if the process dies.
This cancels any running tasks and gives a limited number of restarts.
* Update extensions/ql-vscode/src/codeql-cli/cli.ts
Co-authored-by: Andrew Eisenberg <aeisenberg@github.com>
* Update extensions/ql-vscode/src/query-server/query-server-client.ts
Co-authored-by: Andrew Eisenberg <aeisenberg@github.com>
---------
Co-authored-by: Andrew Eisenberg <aeisenberg@github.com>
`url.parse` is deprecated and should be replaced by the WHATWG URL API,
so this makes that change. The `protocol` and `host` properties are
unchanged, so no other changes are needed.
This will make both the `askForLanguage` and `findLanguage` functions
return a `QueryLanguage` instead of a `string`. This will make it harder
to make mistakes when using these functions.
There are also some other changes with regards to `QueryLanguage` such
that we never need to use `as QueryLanguage` explicitly anymore, except
for the new `isQueryLanguage` function. The only remaining place that I
know of where we're using a `string` to represent the `QueryLanguage`
is in a database item's language, but this is harder to change and may
be relied upon by language authors.
This will suppress showing the same error message when monitoring the
variant analysis fails multiple times in a row. This is useful when
e.g. the internet is disconnected or the endpoint is non-functional for
any other reason.
This will show the error message at least once, and then only show it
if there has been a successful attempt in between or when the error
message is different. This should result in a much less noisy
experience.
For `@kind problem` queries, it seems like the BQRS contains four
result sets: `edges`, `nodes`, `subpaths`, and `#select`. The user is
probably interested in the results of `#select` since that contains the
direct result of the query. This changes the extraction of raw (BQRS)
results to always prefer `#select` over any other result set. If it
can't find a result set named `#select`, it will fall back to the first
result set in the BQRS like before.
After some bit of refactoring, the templates are no longer being
passed to the CFG viewer and this is preventing the viewer from
running.
This change fixes it.
An internal option to help library authours to run and debug
the find references and find dependencies contetextual queries
without relying on the implicit cache.
I initially started defining an enum, but I'm not able to import it in package.json,
where the list of options is defined (under `codeQL.createQuery.autogenerateQlPacks`).
This was only a problem for the new test UI, because the third-party Test Explorer extension we used before must have had this filter in its implementation.
When running "Create Query", if the user escapes out of the language quickpick we don't need to throw an error. Instead we can just show a `UserCancellationException` notice.
We were only clearing the pack cache when a pack file was modified. When
using a script to create a new pack, the pack file is created at once
without a change event firing. This will change the behavior to also
clear the pack cache when a pack file is created or deleted.
And add tests to check this.
I've had to adapt the existing `findExistingDatabaseItem` method
so receive params so that I'm better able to send it a language
and a list of database items.
Offer non-codespace users the option to configure their storage folder for skeleton packs.
Suggested here:
https://github.com/github/vscode-codeql/pull/2310#issuecomment-1507428217
At the moment we're choosing to create our skeleton packs in the first
folder in the workspace.
This is fine for the codespace template because we can control the
folder structure in that repo.
For users outside of this we'd like to offer them the option to choose
where to save their skeleton packs.
After some testing of the wizard with a single rooted workspace
(`github/code-scanning`) we discovered a general VSCode extension
bug whereby after we download a database, we don't select it.
This isn't an issue in the wizard, but it does affect us as it means
we'll generate the QL pack, download the db for you but then you won't
know that we haven't selected your database.
So let's make sure our flow works for this case by explicitly selecting
the database once it's downloaded.
We've noticed by testing that we need to set the current database item
before we call `addDatabaseSourceArchiveFolder()`. Once that method is
called, the call to `setCurrentDatabaseItem()` is ignored.
We've had to make some changes to the openDatabase() method to select
a database item by default, since most places where we call `openDatabase()`
also immediately select the item.
There is one exception [1] in the test-runner.ts file, where we set the
current database item under special conditions.
For this reason, we've made the behaviour configurable and tried to add
some descriptive naming to the params so that it's easy to understand
what the config is doing.
[1]: 4170e7f7a7/extensions/ql-vscode/src/test-runner.ts (L120-L124)
This removes the link to the model file when it does not exist. It will
still show the filename of the model file. When clicking on "Apply", it
will refresh whether the model file exists after writing the file.
This fixes an issue where we would download a database every time
we create a new QL pack. Sometimes, a database is already available
so let's always check for it, regardless of the existence of a QL
pack.
This will filter the extension packs shown to the user when selecting an
extension pack to use in the data extension editor to only include the
extension packs that are compatible with the language of the database
item.
Unfortunately, this required quite some changes to the tests to ensure
the extension packs are actually setup properly since it's now reading
the extension pack files.
This will add the extension pack name to the data extensions editor and
allow the user to click on it to go to the folder of the extension pack
in the explorer panel.
This adds the model filename to the data extensions editor and will also
allow the user to go to the model file by clicking on the filename.
This also updates the general UI to be somewhat more compact by moving
the modeled percentages to be below the header in 1 line.
When you have just created a model file using the quick picker/input
box, the data extension editor will try to read it and fail with an
error message. This adds a check to ensure the model file exists and if
it doesn't, it will not try to read in the file.
This should always be safe since the model file picker will only allow
you to select existing files.
This will not allow the user to open the data extensions editor for a
database if it is not one of the supported languages. The supported
languages is a list of `string` rather than a list of `QueryLanguage`
because a database item's language is also a `string`.
There were still some places where we were hardcoding Java in the data
extension editor. This changes these places to use the database item
language instead.
This commit addresses various test flakiness:
1. Bump timeouts for queries tests
2. Add a dispose handler to queryserver-client. This will help us during
tests because if there is a test that timesout while a query is
running, the query's progress callback won't be invoked. We will
still get a timeout error in the first test, but the second test will
not get a spurious error.
3. Handle a disposed query server in `deregisterDatabase`. This method
will remove the database from the currently running query server.
If there is no query server, then there is nothing to remove. So,
this error is safe to ignore.
4. Explicitly `end()` a connection `ServerProcess`. I'm not 100% sure if
this is necessary, but it seems like it prevents responses from being
handled and erroring out.
5. Better handling of ideServer restarts. Previously, if you quickly
called `CodeQL: Restart Query Server` twice in a row, you would get
an error from the ideServer restart. Restart fails if the server is
not already started. So, in this case just call a start.
We've made an exception to fetch the parent folder when we're
in the vscode-codeql-starter workspace.
We'd like to make this more specific so that it doesn't interfere
with other repos.
When running Create Query in the codespaces-codeql repo, it successfully
creates codeql-custom-queries-xxx as a subfolder of the first workspace
folder, and then adds a database. After the database gets added, we get
prompted with this message:
```
We've noticed you don't have a CodeQL pack available to analyze this
database. Can we set up a query pack for you?
```
which would try to create another QL pack.
Since we're no longer pushing QL packs as top level folders in the
workspace when we use the new "Create Query" flow, we also need to adapt
the original flow to take into account subfolders.
Just as a reminder, the original flow is:
- Be in the codespace template
- Download a database from GitHub
- The extension will offer to create a QL pack for the database
The new flow:
- Run the "Create Query" command
- Choose a language
- Create a QL pack
- Download a database for it
In the new flow the last step of downloading a database would trigger
the extension to offer to create a QL pack.
Let's fix this by detecting subfolders as well and exiting early.
This sets a default timeout of 3 minutes on CLI integration tests. This
is because these tests call into the CLI and execute queries, so these
are expected to take a lot longer than the default 5 seconds. This
allows us to remove all the individual `jest.setTimeout` calls with
different values from the test files.
Set the CLI version in the telemetry listener whenever the version
changes.
A few things to note here:
1. In `CliServer::getVersion()`, avoid calling `supportsPerQueryEvalLog`
directly. This avoids a recursive call to `CliServer::getVersion()`.
Currently, it's always safe to do this, but I thought that it would
be good to avoid recursion here in case we change things in the
future.
2. Now, we are sending the CLI version with all telemetry events.
This adds the external API query text to the extension directly to avoid
users having to copy the query to their local `codeql` submodule or
having to checkout a specific branch.
This is a temporary solution until the queries are stabilized. Once they
are, we will upstream these to `github/codeql` and use them like other
contextual queries.
Since this is just a temporary solution, this is not the prettiest code
and is not intended to be a long-term solution. It does not do proper
caching and will create a new temporary directory for every query run.
The performance hit of this is acceptable and expected.
When we create a skeleton query, we check whether you already have an
existing database with the same name (e.g. `github/codeql`). If we can't
find one, we also check for an existing database with the same language.
If we find one, we select it instead of downloading a new database.
Here we're filtering out databases with errors.
In the original flow for creating skeleton packs, we were starting out
by choosing a database (e.g. github/codeql) and having the extension
create the QL pack for us.
At that point, we were storing the QL pack together with the database in
the extension storage because we weren't interested in committing it to
the repo.
This means we weren't able to see it in the file explorer so in order to
make it visible, we decided to push it as a top-level folder in the
workspace.
Hindsight is 20/20.
Let's change this original flow by just creating the folder in the
workspace storage instead of the extension storage (which will make it
visible in the file explorer) and stop pushing it as an extra top level
folder in the workspace.
NB: For this flow, we exit early in the `createSkeletonPacks` method if
the folder already exists so we don't need to check this again.
This adds the CLI version to telemetry command-usage events.
Note that the CLI server is created after the telemetry listener is
created. The first few telemetry events may have a "not-set" value for
the CLI version.
This changes the kind input from a text field to a dropdown in the data
extensions editor. The supported values for each extensible predicate
are based on what is currently in-scope for the documentation. Other
kinds are not supported.
The supported kinds are now stored on the
`extensiblePredicateDefinitions` to make it easier to add new kinds in
the future.
We now use `fsPath` instead of `path`.
Note: I haven't yet fixed the tests, nor checked manually on mac/linux
Tangential change: we now use the `dirname` method, instead of manually splitting paths to get a parent folder.
The `getFirstStoragePath()` method would break on windows:
```
Path contains invalid characters: /c:/git-repo/codespaces-codeql (codeQL.createSkeletonQuery)
```
This makes sense, since we're looking to get the parent folder by splitting for `/`.
In windows, paths use `\` instead of `/`.
So let's detect the platform and add a test for this case.
To be consistent with other database item search methods, we're renaming
ours:
`digForDatabaseItem` -> `findDatabaseItemByNwo`
and
`digForDatabaseItemSameLanguage` -> `findDatabaseItemByLanguage`
At the moment, we're always deciding which database to download for the
user for an example query.
We'd like to give them a chance to change the database, so here we're
adding a step where we're showing the user a selection box with the
suggested database pre-filled.
They can choose to type in a different database before continuing the
skeleton generation process.
We'd like to select an existing database for our query, if on is
already downloaded and matches the query language.
Previously we were re-using the database if the language and name
matched (e.g. the name would be `github/codeql`).
When we try to determine the next file name for our example query,
we only look at `example<n>.ql` files.
e.g. if the files in the folder are:
- `example.ql`
- `example2.ql`
- `MyQuery.ql`
we will create an `example3.ql` file.
Previously we were counting all existing `.ql` files.
We've now added more tests and pushed the total duration over 5 seconds
for all the tests in this file.
This limitation seems to be a recent development where files with tests
that last longer than 5 seconds start failing in jest.
We're bumping the timeout limit to 40 seconds for now.
We initially defined the default database to download as one from the
`github/codeql` repo as it was convenient.
However, this repo doesn't have a lot of vulnerabilities to discover.
Let's use repos that are in our MRVA top 10 list to allow users to
write more interesting queries.
We set up the "Create Query" command with the assumption that
the first folder in the workspace is the parent folder.
This is true for the `codespaces-codeql` repo where we expect
to use this command.
However, for the `vscode-codeql-starter` repo, the top level
folders are QL packs:
- codeql-custom-queries-cpp
- codeql-custom-queries-ruby
... etc.
In order to make the command work for people using the starter
repo, we'll need to introduce a check for these QL packs when
we decide the storage path.
The end goal is to replace the starter workspace completely
with the codespaces-codeql repo, so this code can be removed
in the future when we retire the repo.
Until then, the command will need this to be able to work in
both starter workspaces.
We offer `github/codeql` as a repo to use for downloading databases
for our skeleton pack.
Once the repo is specified, the user is prompted to choose a language.
At this point, we already know what language the user wants, so let's
change the `downloadGitHubDatabase` and `convertGithubNwoToDatabaseUrl`
methods to accept a language parameter.
We check if the language is in the list of languages received in the
response. If it isn't, we still prompt the user.
This will be triggered by a "Create Query" command.
It will:
- prompt the user for a language
- create a skeleton pack based on the language chosen
- download a database for the QL pack
- open the new query file
If the skeleton pack already exists, we just create a new query file
in the existing folder.
If the database is already downloaded, we just re-use it.
We introduced this QlPackGenerator a while ago. It always creates an `example.ql` query file as part of the skeleton pack.
We'd like to set the name of the query file, since we'll allow the user to create queries multiple times in the same skeleton pack folder.
The folder will be named `codeql-custom-queries-${language}` and will first receive an `example.ql` file.
If the user then tries to create a new query for the same language, we'll just create an `example2.ql`, `example3.ql` etc. file in the existing folder.
We'll use this to check whether a database for our ql pack already exists.
While there are other methods that search for a database item by URI, we
only have a language chosen by the user and an nwo ("github/codeql").
So let's introduce a way to search for the db based on the information we
have.
We plan to ask the user to choose a language, before attempting
to download a corresponding database for them.
The functionality already exists, so let's re-use it.
This was nested in a method that included prompting the user for a
github repo.
We'd like to re-use this to download a database of our choice from
GitHub, based on which language a user chooses.
This will allow a user to create a new model file in an existing data
extension when opening the data extension editor. There is some
validation on the name of the model file, which depends on reading in
the qlpack.
This adds a pickable model filename from an existing extension pack to
the data extensions editor. This allows the user to edit one of their
existing data extensions. This does not yet add the ability to create
new extension packs and/or new model files.
This uses the `codeql resolve extensions` command to get the list of
available model files. This should be available in all CLI versions
which the data extensions editor supports.
The data extension editor was only using the default data extensions
found in the `ql` submodule to find external API calls. This will add
support for using data extensions found in the workspace.
Rather than using the `codeQL.runningQueries.useExtensionPacks` setting,
this will always include data extensions since the editor doesn't make
sense to use without data extensions. We will also forbid the user from
opening this view unless they are using a CLI which supports data
extension packs.
We were using a single-use class for generating the flow model, while we
are actually able to do it using two functions. This is more in line
with our existing codebase.
This fixes the "Webview is disposed" error which occurs when the user
closes the variant analysis webview while the extension is still
activating. We will now check whether the webview is disposed before
restoring the view.
This was pointed out by CodeQL: when calling `setState` and using
`this.props`, it may not be up-to-date because `setState` may run
concurrently. Therefore, we should use the `setState` callback variant
to ensure we get the latest props.
This refactors the code a bit to ensure we're not using `this.props`
anywhere, including in the `getResultSets` function which is called
in the `setState` callback.
When updating to React 18, we removed the loading step from the updating
of the state of the result view since React would batch the updates
anyway. However, this caused a bug where the result view would be empty
when switching between queries. This is because the result view would
retain the old selected result set name. This would not happen
previously because React would re-render the view at least once, which
would cause the result view to be unmounted and re-created.
This fixes it by resetting the selected result set if we can't find the
result set in the new result sets.
The default Storybook Babel config did not recognize the `public`
keyword in our custom errors (e.g. `ExhaustivityCheckingError`
and `RedactableError`). To fix this, we can use Storybook's V7 mode to
supply a custom Babel config. This fixes the compilation error.
See: https://storybook.js.org/docs/react/configure/babel
This adds tests for the external API query and retrieving of results. It
does not use the "real" CLI integration, but instead mocks the CLI
server and query runner.
To make mocking easier and require less type casting, I've narrowed some
of the arguments of some other functions. They now use `Pick` to only
require the properties they need.
The version of ts-jest we were using was given a warning that it did
not support TypeScript 5, even though it was working fine. The latest
version of ts-jest adds official support for TypeScript and removes the
warning.
This loads in the existing data extension YAML file for the selected
database. It only supports the filename we save it to, and will not load
it from any other data extension YAML files.
This adds the ability to save the modeled methods in the data extensions
editor to a YAML file named after the database name. It will save it to
the `ql` submodule for now. Support for data extension packs will be
added later.
This updates the view of the data extensions editor to show a table of
possible sources/sinks/flow summaries that can be edited. It's not yet
possible to save the changes or load the existing file.
This upgrades Storybook to an alpha version to ensure we can start
Storybook when using TypeScript 5.0. This is temporary until we can
upgrade to Storybook 7 or to a released version of 6.5.17.
This adds a really simple regression unit test for the results view
which checks that the results view can render a SARIF file. This is
in preparation for the upgrade to React 18 to ensure that we don't
break the basic functionality of the results view.
This will improve the error message shown when monitoring fails. Instead
of showing "Error while monitoring variant analysis: Not Found", this
will now show "Error while monitoring variant analysis "Empty block
(javascript) [29/3/2023 10:45:10]: Not Found". This should make it
easier for the user to figure out which query history item is
problematic.
We're not using the full query history item label here because that
would require access to the query history item, which we don't yet have
here. Adding it here would add a dependency on the query history, which
seems undesirable.
We were ignoring errors coming from `vsce publish` and this was causing
the workflow to succeed even when the publish failed. This will remove
the `||` and let the workflow fail if the publish fails.
The `glob` package now uses promises in version 9, so we don't need the
separate `glob-promise` package anymore.
This also updates one call site to use `cwd` instead of `join` to
avoid possible issues due to a breaking change in version 8 which treats
Windows path separators differently. By changing the `cwd`, we should
not run into these issues.
Removing the `resolutions` of `glob-parent` did not change the
`package-lock.json`, so this does not have any effect on the package
version we are using.
We introduced this change to help with reducing flakiness in CI [1].
This has a slightly different effect locally, where every failed test
will output three times.
This in turn makes it harder to read, especially when you have multiple
failing tests.
Since the original intent for this behaviour was to be used in CI, I'm
proposing we disable it when the CI env variable isn't set.
I've opted to set it for all jobs involving tests, just for consistency.
I'm happy to limit it to just the places where it's required.
[1]: https://github.com/github/vscode-codeql/pull/2059
This adds support for detecting the `CommandManager.execute` method in
the unique command use query.
This may not be the best way to implement this. There's a method
`hasUnderlyingType` on `this.getReceiver().getType()`, but I couldn't
really figure out how to get it recognize `CommandManager`. It might be
possible if we can construct the type of `CommandManager`, but this will
probably include the filepath to the `CommandManager` class, which might
not neccessarily be something we want: moving the `CommandManager` class
should not require updating the query. I'm very happy to hear other
suggestions.
The `commandRunner` name doesn't really make sense since it doesn't
"run" a command, but rather registers a command. This renames it
to `registerCommandWithErrorHandling` and moves it to the
`common/vscode` directory.
The `codeQL.checkForUpdatesToCLI` command is registered pre-activation,
and we don't really want to create the command manager before
activation, so this will just add the correct type without registering
it using the command manager.
The command manager types didn't fully support commands defined with
`Partial` because it deduced that the command function was `undefined`
when the function was not defined. However, if the command is not
present, the command registration will not be called. This fixes the
types by specifying that the command definition will never be
`undefined`.
This class seems to have been introduced at some point to reduce the
dependency on VS Code from the test UI service. However, none of its
methods are being used anymore, and by using typed commands we have
already reduced the dependency on VS Code. Therefore, we can simply
remove this class.
We attempted to specify exactly which URI we're expecting here.
However, `Uri.parse` behaves differently in the test than it does in
the code so we've inadvertently created a flakey test [1]. The URI we
generate in the test has a `scheme: 'c'` while the one in the code has
a `scheme: 'C'` property.
This only happens on windows, not ubuntu.
Let's narrow the comparison to just the path of the URI.
[1]: https://github.com/github/vscode-codeql/actions/runs/4478429334/jobs/7871178529#step:7:231
Some of the methods in the `DatabaseUI` were public because they were
used in the `extension.ts` file. We have moved these method calls into
this file, so they do not need to be public anymore. We can also get rid
of the separation between some of these methods, so I've moved them into
the function that calls them.
Now that we fixed our expectation in the previous commit, we could see we
were stubbing this to false instead of true.
So now the test is checking the right scenario.
Our expectation was quite narrow: we expect to not call an `openFolder`
command. We didn't specify any params for it, which might mean this
expectation wasn't working like it should.
Let's just check that `executeCommand` isn't called at all.
The local query commands are using a separate logger, and this is not
supported by the command manager because it is quite specific to this
extension. Therefore, we create a separate command manager which uses
a different logger to separate the commands.
We're running this at the extension start-up. We don't want it to block the extension
from completing activation, so let's swallow any errors from the code tour and output
them, instead of letting this affect the rest of the extension activation.
When opening https://github.com/github/codespaces-codeql/ in a
codespace, it's easy to miss the prompt that tells you to open the
tutorial.code-workspace file.
In fact people actively dismiss the alert to get it out of the way.
If you miss that prompt, you end up with a single-rooted workspace,
which causes various other problems.
While there is an open issue to allow VS Code to open a default
workspace [1], there doesn't seem to have been any progress on it
in the last two years.
So we're taking matters into our own hands and forcing the extension
to open the tutorial workspace, if it detects it.
This will only happen if the following three conditions are met:
- the .tours folder exists
- the tutorial.code-workspace file exists
- the CODESPACES_TEMPLATE setting hasn't been set
NB: the `CODESPACES_TEMPLATE` setting can only be found if the
tutorial.code-workspace has already been opened. So it's a good
indicator that we're in the folder, but the user has ignored the prompt.
[1]: https://github.com/microsoft/vscode-remote-release/issues/3665
The local databases UI was essentially the only class which was defining
methods using assignment to a class property rather than using function
definitions and binding them. This switches it to use function
definitions and binding, which is more consistent with the rest of the
codebase.
This change allows the `codeQL.runningQueries.useExtensionPacks`
setting to be respected when running variant analysis queries. When
set to `all`, before uploading the generated query pack, all extension
packs in the workspace will be injected as dependencies into the qlpack
file.
In the results view, `setState` was used to set some state, and then
`loadResults` was called to set some other state. However, `setState`
is asynchronous, so the `this.state` in `loadResults` was not the state
that was set before the call.
This commit fixes it by combining the two `setState` calls into one. The
logic was hard to follow, so I'm not sure if this is the correct fix.
The `shouldKeepOldResultsWhileRendering` state seems to be unnecessary
now since everything is being done in `setState` call, but I'm not sure
about that.
The version of `@primer/react` we were using didn't have the correct
types for the `ThemeProvider`. By upgrading these packages to their
latest versions, all types are correct.
This enables React strict mode which will print extra warnings to the
console when we use certain constructs incorrectly. This does not affect
production builds.
See https://beta.reactjs.org/reference/react/StrictMode
The extension Webpack config does not use `babel-loader`, so we can
remove it as a direct dependency. `babel-loader` is still included in
our `node_modules` because Storybook depends on it, but the version is
now completely managed by Storybook rather than us.
This will make the progress options passed to `withProgress` optional by
moving it to be the second argument and setting a default value for the
`location`. This will make it much easier to use from a variety of
commands.
This removes the `args` from the `ProgressTask` passed to
`withProgress`. The `args` is only used by the
`commandRunnerWithProgress` and can easily be replaced by an anonymous
function that passes the `args` instead. This will simplify the
`ProgressTask` interface and make it easier to use.
The `commandRunnerWithProgress` implementation isn't actually any
different from `commandRunner`, except for the call to `withProgress`
and support for an `outputLogger` argument. Therefore, this will simply
make `commandRunnerWithProgress` a wrapper around `commandRunner`,
removing quite some duplication in the process.
If the user requests that extension packs be included in their MRVA run,
then do the following:
1. Search the workspace for all extension packs
2. Add each extension pack as an explicit and direct dependency on
the generated pack.
It is ok to use `*` as a dependency since we are guaranteed that
exactly one version of each injected extension pack dependency is
available when the pack is being compiled.
If we find multiple paths to an extension pack of the same name, this
is an error since it is ambiguous which path to use.
This will add a check to ensure that `showAndLogExceptionWithTelemetry`
is not called when downloading packs. This expectation is placed before
the check for `showAndLogInformationMessage` so that when the test
fails, the error message will be shown.
We are no longer including our dependencies in the VSIX package, so we
can tell VSCE that we don't want it to look at dependencies using
`--no-dependencies`. If we do this, VSCE doesn't require the
`node_modules` directory anymore and we can skip that step, which will
make building significantly faster.
I've confirmed that there are no changes between the two options by
building the extension both without and with the change. This is the
diff of the two outputs (using `diff -r`):
```diff
diff --color -r vscode-codeql-1.8.0-dev.2023.3.8.15.10.13/extension/package.json vscode-codeql-old/extension/package.json
7c7
< "version": "1.8.0-dev.2023.3.8.15.10.13",
---
> "version": "1.8.0-dev.2023.3.8.15.6.51",
diff --color -r vscode-codeql-1.8.0-dev.2023.3.8.15.10.13/extension.vsixmanifest vscode-codeql-old/extension.vsixmanifest
4c4
< <Identity Language="en-US" Id="vscode-codeql" Version="1.8.0-dev.2023.3.8.15.10.13" Publisher="GitHub" />
---
> <Identity Language="en-US" Id="vscode-codeql" Version="1.8.0-dev.2023.3.8.15.6.51" Publisher="GitHub" />
```
The only difference is the version number, which is expected.
When deleting a query history item and the "next" query is still
running, the `completedQuery` is `undefined`. This commit fixes it by
using optional chaining to ensure that the `completedQuery` is defined
before accessing its `successful` property.
This is definitely not a perfect solution since we're essentially just
moving the place where we're casting. However, because we have manually
made the types similar, this provides some type assurances where there
were none before. This also has the cast in only one place, which makes
it easier to find and fix in the future.
I was initially trying to understand why this method was failing due
to an unrelated error [1] so I ended up over-engineering the path
parsing.
We can use the path from the first workspace folder, like we do in
other places in the extension.
[1]: https://github.com/github/vscode-codeql/pull/2104
When a user goes through the Code Tour, we select a dummy `csv` database
for them to get them up and running.
Once they complete the code tour and would like to continue writing
queries, they will need to add their own database.
After they do that, we check the language of their new database and
generate a skeleton QL pack for them so that they don't need to create
these files by hand. See [1] for details.
This skeleton pack folder will be called
`codeql-custom-queries-<language>` and it comes with its own example
query: `example.ql`.
When we try to run this example query, the query gets confused about
which `dbscheme` to pick, as it sees a `qlpack.yml` file in the new
skeleton pack folder, as well as one in the existing `tutorial-lib`
folder.
So we'll need to get rid of the `tutorial-lib` folder in order to make room
for new queries to be run once the tour is complete.
This commit introduces a `handleTourDependencies` step which will
trigger a `codeql pack install` command in order to install real library
dependencies for `tutorial-queries`, since we no longer have the dummy
library in `tutorial-lib`.
Unfortunately `Object.defineProperty` doesn't work on proxies, so I've
added an options object to `mockedObject` which allows passing in
methods that will return a value for a specific property.
In https://github.com/github/codespaces-codeql/pull/12 we moved
the source for our tutorial database into `.tours` in order to
avoid confusing the user when we load the database into the
workspace, since they'd see two databases.
Since this is just the source, we'd like to hide it.
We mark this query as `@kind problem`.
We'll need to change the query a bit to make it fit this type of ...
erm... kind.
This means the results view will be formatted to display the file name
next to each of the results.
We're also getting rid of any mentions of an empty block query since
that's no longer what it checks.
This will remove some instances where we're using `as unknown as T` and
replace them by a call to `mockedObject<T>()`. The `mockedObject`
function is a bit more explicit about what it does and has types which
ensure that the methods that are set on the object actually exist.
Unfortunately, we can't fully get rid of `as unknown as T` in the
`mockedObject` function. However, this construct is more localized and
does not need to be used in as many places. If we do enable an ESLint
rule to prevent the use of `as unknown as T`, I would feel comfortable
with disabling the rule for the `mockedObject` function.
This also removes the type for the `runTests` `options` argument since
it's only used in this definition and don't actually use the type
anywhere else.
It seems that when we added the CSP policy to the webview, we did not
take into account that `d3-graphviz` uses `@hpcc-js/wasm` to load
Graphviz as a WASM module. This commit adds `'wasm-unsafe-eval'` to the
CSP policy to allow this.
This is step 3 in the Code Tour. At this point we don't need to create
the skeleton pack so let's disable that functionality.
Co-authored-by: Shati Patel <shati-patel@github.com>
By default, this is added when we call `runJsonCodeQlCliCommandWithAuthentication`.
However, `codeql pack add` doesn't support this option so we need to turn it off.
Co-authored-by: Shati Patel <shati-patel@github.com>
On windows, the `Uri.path` will return an extra folder, as we can see in the tests:
```
ENOENT: no such file or directory, open 'D:\C:\Users\RUNNER~1\AppData\Local\Temp\tmp-4784XPDQPb5jM6IW\test-ql-pack-ruby\qlpack.yml'
```
Let's use `Uri.fsPath` instead.
We were initially always removing the last folder in the workspace as
we assumed that would be the directory we use.
Now that we've switched to using a temporary directory, this is no longer
the case so we need to find the index of the directory in the list of
workspace folders and then use that index to remove the directory.
`Uri.parse` will not work with Windows paths as it will consider `C:\path`
to indicate a file scheme (the "C:" part) and will complain about it.
With `Uri.file` we can build the URI without hitting this complication.
We're checking that the skeleton QL pack doesn't exist as a workspace
folder, so we should be creating this folder in the workspace as well.
Initially this was being created in VSCode's local storage.
This will receive a folder name and language.
It will generate:
- a `codeql-pack.yml` file
- an `example.ql` file
- a `codeql-pack.lock.yml` file
It will also install dependencies listed in `codeql-pack.lock.yml`
file.
We were initially planning to call the `packInstall` command once
we generate `codeql-pack.yml` in order to install dependencies.
However, the `packAdd` command does this for us, as well as
generating a lock file.
Rather than trying to craft the lock file by hand, we're opting
to use the cli command.
NB: We're introducing a new `QueryLanguage` type which is identical
to the `VariantAnalysisQueryLanguage`. In a subsequent PR we'll
unify these two types.
Similar to what we do with `codeql pack install`.
Tnis will simulate us running `codeql pack add codeql/<language>-all`.
We're going to need in order to:
- generate a lock file (codeql-pack.lock.yaml)
- install the correct packages for our skeleton QL pack based on the
lock file.
This removes the remote queries history item as a supported history
item. This allows us to delete almost all code related to remote
queries except for the React view code which will be removed separately.
In the query serialization code, we now ignore remote queries.
This would run the unit, view, integration and CLI integration tests in
parallel, which would cause problems with multiple VSCode instances and
use a lot of memory.
- Add a new config that toggles between using all data extensions or
none.
- If using all data extensions, then before a query evaluation, run a
`codeql resolve qlpacks` command with the new `--kind` option. This
will return a list of extension packs in the workspace.
- Pass these packs to the cli before evaluation/
- This will only work with CLI v2.12.3 (not yet released) or later.
- Also include some CLI tests to ensure this works.
This will add support for the `codeql-pack.yml` filename in all places
where we currently support `qlpack.yml`. It centralizes the list of the
supported filenames in a single place and a method that can figure out
the correct filename to use.
This download download the sourcemaps from the release asset if it is
available. Unfortunately, I'm not able to test this yet because we don't
yet have any releases with sourcemaps attached. As a fallback, it will
still try to download from the workflow run.
Most of the warnings that are currently being shown in CI are coming
from VSCode. We can hide the VSCode output to make the CI logs more
readable. This should not influence the tests; the output of tests
(in particular using `console.log`/`console.error` etc.) will still be
shown.
See: 0c98dc12ad/packages/jest-runner-vscode/src/public-types.ts (L58-L68)
We'd like to add test coverage for the openDatabase function (which is
public).
At the moment, this relies on `resolveDatabaseContents` which is just
a standalone function.
This means we're unable to mock it using Jest.
So let's move it into its own class.
This method in turn depends on a `resolveDatabase` function, which we've
also moved into the new class.
The only usages I could find for there functions were from within
the `databases.ts` file.
This is unrelated to the changes in this PR but it's causing CI to fail.
```
config listeners › CliConfigListener › should listen for changes to 'codeQL.runningTests.numberOfThreads'
expect(jest.fn()).toHaveBeenCalledTimes(expected)
Expected number of calls: 1
Received number of calls: 2
109 | const newValue = listener[setting.property as keyof typeof listener];
110 | expect(newValue).toEqual(setting.values[1]);
> 111 | expect(onDidChangeConfiguration).toHaveBeenCalledTimes(1);
| ^
112 | });
113 | });
114 | });
```
We don't need to check that the callback is triggered a certain number of times, just that it works
so we can change this test to be more permissive.
We'd like to make it easier for a user going through the CodeQL Tour to
write their queries.
To help them along, we can generate skeleton QL packs once we know which
database they're using, instead of expecting them to know how to create
this themselves.
We're then able to download the necessary dependencies for their CodeQL
queries.
This checks that we're running the CodeTour by looking for the
`codeQL.codespacesTemplate` setting.
This will upload the sourcemaps produced as part of the release as a
release asset. This allows the sourcemaps to be downloaded and used
for decoding stack traces beyond the 90 day artifact limit of GitHub
Actions.
We run `npm run lint` every time we do a `git push`.
This takes quite a long time, and the lint command has already been run
when we created the commit in the first place.
Could we instead skip this and rely on CI to tell us if we've failed
to address a linting issue?
It seems like the `onDidChangeConfiguration` is being called multiple
times. It doesn't actually matter that it's being called twice, so we
just need to ensure it's called at least once.
This is blocking us from merging new PRs so while we figure out
how to fix them, let's skip the tests that are failing on our
`main` branch.
For full context: the tests started failing when a new version of
VSCode was released (1.75.0).
This adds support for mapping full stacktraces in the source map
script. This allows you to pass a full stacktrace to the script and get
back a stacktrace with all original positions.
This adds a script that can be used for retrieving the original source
location when given a location in the released extension. It will
download the source map from the Actions workflow run of the release and
use the `source-map` library to extract the original location.
In our tests, we were writing settings files to disk because we were
using the VSCode configuration API which writes settings to files. This
results in flaky tests because concurrency can cause the VSCode API to
misbehave.
This will switch the tests to use a mocked API by default. For some
tests the real implementation is used, but the large majority of tests
is now using a mocked version which only keeps track of the
configuration in memory. This makes it easier to reset the state between
tests since we can just empty out the in-memory configuration.
We have a codespace template which houses our CodeQL tour:
https://github.com/github/codespaces-codeql
This contains a repo with a default databases already loaded
for the user so that they can start writing queries more quickly.
At the moment we're asking the user to manually right click on
the database folder ('codeql-tutorial-database') and set it as
the current database.
We can take this one step further by defining a command that gets
triggered when we arrive at the step for setting up the database.
The command ("codeQL.setDefaultTourDatabase") will build the URI
pointing to our preloaded database and set it as the current one.
We initially considered whether we can re-use the setCurrentDatabase
command and pass the URI of the database from the codespace itself,
but the URI would be hardcoded as:
```
file://0-62/workspaces/codespaces-codeql/codeql-tutorial-database
```
as we can only pass the codeTour extension a command and string
parameters.
This would have been brittle as the filepath for a codespace might
change in the future.
Instead we can define a custom tour command ("setDefaultTourDatabase")
to look at the current workspace folder and build the path to the
database in the CodeQL extension.
Co-authored-by: Shati Patel <shati-patel@github.com>
Instead of using the third party `peter-evans/create-pull-request`
action for creating a PR for releases, we can use the already present
`./.github/actions/create-pr` action which is also used for the PR for
bumping CLI versions.
Instead of having different ESLint configuration files in each
directory which don't seem to inherit the configuration correctly, this
will add `overrides` in the root file.
This will add a welcome view to the database panel which is shown when
the controller repository is not setup. This welcome view will show a
button which can be used to set up the controller repository.
3 seconds is a really long time to wait for downloads since a
significant percentage of downloads will complete within 3 seconds. This
changes the update delay to 500ms which should still give us good
performance, but also make the download feel more responsive.
This removes all CodeQL CLI version constraints for unsupported CLI
versions (< 2.7.6). The oldest supported CLI version is 2.7.6 since GHES
3.3 recommends using CodeQL CLI 2.7.6.
Tests whether we choose "Yes" / "No" in the new modal window.
"Yes" -> remove the item and show you a toast notification
"No" -> don't remove item
This only shows up for in progress items.
We now have special behaviour for removing an "in progress" query so the
tests will be different.
Let's have a separate section for "in progress" queries. We'll add extra
behaviour testing in the next commit.
When you attempt to delete a query that's still in progress from your
query history, you get a prompt asking if you're sure.
If you pick "Yes", the item is removed and you see a toast notification
with a link towards the GitHub Action.
This will supply the GitHub access token to certain CodeQL CLI commands
such that private packages can be resolved. It will only do so if the
user has an existing auth session. If they don't, they will now get a
prompt to login. However, this will only happen for commands which
actually use authentication, which is limited to packaging commands.
This will set the `mode` of Webpack to `production` for release builds.
It will also stop inlining the sourcemap and instead produce a separate
file which is excluded by `.vscodeignore`.
In terms of the bundled extension, this will add 1 file
(`out/webview.js.LICENSE.txt`). It decreases the size of the VSIX file
from 4.28MB to 1.77MB.
VSCode was not able to find the original source of the bundled
extension because it was looking for the source in the `out` directory.
By setting the `sourceRoot` to the `extensions/ql-vscode` directory
which is located at `..` from the `out` directory, VSCode is able to
find the original source and breakpoints are hit.
This will copy the WASM file from source-map to the output directory.
This makes the source-map package work. See the comment in the code for
more details.
This bundles the extension files using esbuild, as recommended by
VSCode. This reduces the size of the extension from 34MB to less than
5MB.
Gulp will still run TypeScript to check types, but will not use the
TypeScript compiler output in the bundle.
Tests are now run separately, outside of Gulp, so their data doesn't
need to be copied anymore.
See: https://code.visualstudio.com/api/working-with-extensions/bundling-extension
Some checkouts of the github/codeql repo, such as the
internal submodule, may be named `ql` rather than
`codeql`. Allow this folder name when running tests.
I am removing these assertions so that our internal integration tests
can pass. They are currently failing because the number of dependencies
of the `codeql/javascript-all` pack has changed. It no longer makes
sense to test this value as newer versions of this pack will have more
dependencies and we expect this value will continue to go up.
This was initially added [here][1] but wasn't quite in the right place
to have the intended effect.
Let's move it up to the root of the project.
[1]: f515663640
This moves our existing test plan under a "Required testing" section.
We're also adding the scenarios used for testing live results under an "Optional testing" section.
I believe this doesn't change the user-visible behaviour at all. The user
won't be prompted to log in any more or less often than they would have
done before.
One benefit of this is that we can remove the registerListeners method
because we no longer need to know if the cached octokit is still valid.
Instead we just call vscode.authentication.getSession every time and it
will return the current session, which might be different from the last
time we called it. This might prompt the user to log in, but that would
have happened anyway because when the session changed we would have
overwritten our cached octokit instance.
Another benefit is that we no longer need the extension context and this
removed a surprisingly large amount of code where we are passing this
parameter around because we need it for the credentials.
The only downside I can see is that we call getSession more often and
create more javascript objects in general. I believe the performance
impact of this will be negligible and not worth worrying about.
I argue that calling createOctokit(false) adds no benefit. If an
authenticated session already exists then this silently create an
octokit, which makes getOctokit() a no-op just returning the field.
However if there is already an authenticated session then getOctokit()
would already be able to create an octokit without prompting the user.
On the other hand if there isn't an authenticated session then we
won't be able to pre-populate an octokit, so getOctokit() will have
to prompt the user anyway.
Not calling createOctokit(false) in registerListeners also doesn't
change behaviour. If the user is authenticated in the new session then
we would be able to create an octokit instance wihtout prompting in
getOctokit anyway. If the user is not authenticated in the new session
then we won't be able to create an instance without prompting either way.
The only benefit I can think of is that it moves a tiny amount of
computation earlier in the pipeline, but the amount of computation is
tiny and it isn't any more async than it would be if it happened in
getOctokit(). I don't think this is worth making the code more complex.
This was only used from initializeWithToken and only added a completely
separate case to the start of the method, effectively turning it into
two separate implementations. Therefore we can make things simpler by
inlining this case in the one place it is used.
It is true by default and no place in the codebase sets it to false. We can
simplify the code by removing this case we aren't using. If we want this
behaviour in the future we can always implement it again, but I think it's
likely to be unnecessary and if you don't want authenticated requests then
you likely won't be initializing a Credentials object.
This cannot happen already, even before the other changes in this PR.
The Credentials.initialize method can never return undefined, so these
checks would never return true. The real place that checks that we are
authenticated is in the vscode.authentication.getSession method, and
it will reject the promise if the user declines to authenticate or
anything else means we can't get an authenticated session.
I feel justified in removing the tests for these cases because the
code was never actually called in production, and we are covered by the
vscode authentication library rejecting the promise. Any exception
thrown from Credentials.initialize would behave the same as the one I'm
deleting.
This will sort the files in an exported Gist by the user-defined sort
order. It does so by prefixing the files with `result-{index}-` where
the `index` is the 1-based index of the repository in the sort order.
It will automatically pad the index with leading zeros to ensure that
the files are sorted in the correct order.
Unfortunately, we can't just use `{index}-` because numbers sort before
the `_` character, which is used in the summary filename to place it
first.
There are also some changes in how we determine which repositories to
export since we need to know in advance how many zeroes we need to pad
the index with. There should be no functional changes in which
repositories are actually exported.
@@ -67,10 +67,7 @@ members of the project's leadership.
## Attribution
This Code of Conduct is adapted from the [Contributor Covenant][homepage], version 1.4,
available at https://www.contributor-covenant.org/version/1/4/code-of-conduct.html
[homepage]: https://www.contributor-covenant.org
This Code of Conduct is adapted from the [Contributor Covenant, version 1.4](https://www.contributor-covenant.org/version/1/4/code-of-conduct.html).
For answers to common questions about this code of conduct, see
https://www.contributor-covenant.org/faq
[the Contributor Covenant FAQ](https://www.contributor-covenant.org/faq). For more about Contributor Covenant, see [the Contributor Covenant website](https://www.contributor-covenant.org).
Hi there! We're thrilled that you'd like to contribute to this project. Your help is essential for keeping it great.
@@ -23,7 +23,9 @@ Please note that this project is released with a [Contributor Code of Conduct][c
Here are a few things you can do that will increase the likelihood of your pull request being accepted:
* Follow the [style guide][style].
* Write tests. Tests that don't require the VS Code API are located [here](extensions/ql-vscode/test). Integration tests that do require the VS Code API are located [here](extensions/ql-vscode/src/vscode-tests).
* Write tests:
* [Tests that don't require the VS Code API are located here](extensions/ql-vscode/test).
* [Integration tests that do require the VS Code API are located here](extensions/ql-vscode/src/vscode-tests).
* Keep your change as focused as possible. If there are multiple changes you would like to make that are not dependent upon each other, consider submitting them as separate pull requests.
* Write a [good commit message](https://tbaggery.com/2008/04/19/a-note-about-git-commit-messages.html).
@@ -93,202 +95,7 @@ More information about Storybook can be found inside the **Overview** page once
### Testing
We have several types of tests:
* Unit tests: these live in the `tests/pure-tests/` directory
* View tests: these live in `src/view/variant-analysis/__tests__/`
* VSCode integration tests: these live in `src/vscode-tests/no-workspace` and `src/vscode-tests/minimal-workspace`
* CLI integration tests: these live in `src/vscode-tests/cli-integration`
The CLI integration tests require an instance of the CodeQL CLI to run so they will require some extra setup steps. When adding new tests to our test suite, please be mindful of whether they need to be in the cli-integration folder. If the tests don't depend on the CLI, they are better suited to being a VSCode integration test.
Any test data you're using (sample projects, config files, etc.) must go in a `src/vscode-tests/*/data` directory. When you run the tests, the test runner will copy the data directory to `out/vscode-tests/*/data`.
#### Running the tests
Pre-requisites:
1. Run `npm run build`.
2. You will need to have `npm run watch` running in the background.
##### 1. From the terminal
Then, from the `extensions/ql-vscode` directory, use the appropriate command to run the tests:
* Unit tests: `npm run test:unit`
* View Tests: `npm test:view`
* VSCode integration tests: `npm run integration`
###### CLI integration tests
The CLI integration tests require the CodeQL standard libraries in order to run so you will need to clone a local copy of the `github/codeql` repository.
1. Set the `TEST_CODEQL_PATH` environment variable: running from a terminal, you _must_ set the `TEST_CODEQL_PATH` variable to point to a checkout of the `github/codeql` repository. The appropriate CLI version will be downloaded as part of the test.
2. Run your test command:
```shell
cd extensions/ql-vscode && npm run cli-integration
```
##### 2. From VSCode
Alternatively, you can run the tests inside of VSCode. There are several VSCode launch configurations defined that run the unit and integration tests.
You will need to run tests using a task from inside of VS Code, under the "Run and Debug" view:
* Unit tests: run the _Launch Unit Tests - React_ task
* View Tests: run the _Launch Unit Tests_ task
* VSCode integration tests: run the _Launch Unit Tests - No Workspace_ and _Launch Unit Tests - Minimal Workspace_ tasks
###### CLI integration tests
The CLI integration tests require the CodeQL standard libraries in order to run so you will need to clone a local copy of the `github/codeql` repository.
1. Set the `TEST_CODEQL_PATH` environment variable: running from a terminal, you _must_ set the `TEST_CODEQL_PATH` variable to point to a checkout of the `github/codeql` repository. The appropriate CLI version will be downloaded as part of the test.
2. Set the codeql path in VSCode's launch configuration: open `launch.json` and under the _Launch Integration Tests - With CLI_ section, uncomment the `"${workspaceRoot}/../codeql"` line. If you've cloned the `github/codeql` repo to a different path, replace the value with the correct path.
3. Run the VSCode task from the "Run and Debug" view called _Launch Integration Tests - With CLI_.
#### Running a single test
##### 1. From the terminal
The easiest way to run a single test is to change the `it` of the test to `it.only` and then run the test command with some additional options
to only run tests for this specific file. For example, to run the test `src/vscode-tests/cli-integration/run-queries.test.ts`:
```shell
npm run cli-integration -- --runTestsByPath src/vscode-tests/cli-integration/run-queries.test.ts
```
You can also use the `--testNamePattern` option to run a specific test within a file. For example, to run the test `src/vscode-tests/cli-integration/run-queries.test.ts`:
```shell
npm run cli-integration -- --runTestsByPath src/vscode-tests/cli-integration/run-queries.test.ts --testNamePattern "should create a QueryEvaluationInfo"
```
##### 2. From VSCode
Alternatively, you can run a single test inside VSCode. To do so, install the [Jest Runner](https://marketplace.visualstudio.com/items?itemName=firsttris.vscode-jest-runner) extension. Then,
you will have quicklinks to run a single test from within test files. To run a single unit or integration test, click the "Run" button. Debugging a single test is currently only supported
for unit tests by default. To debug integration tests, open the `.vscode/settings.json` file and uncomment the `jestrunner.debugOptions` lines. This will allow you to debug integration tests.
Please make sure to revert this change before committing; with this setting enabled, it is not possible to debug unit tests.
Without the Jest Runner extension, you can also use the "Launch Selected Unit Test (vscode-codeql)" launch configuration to run a single unit test.
#### Using a mock GitHub API server
Multi-Repo Variant Analyses (MRVA) rely on the GitHub API. In order to make development and testing easy, we have functionality that allows us to intercept requests to the GitHub API and provide mock responses.
##### Using a pre-recorded test scenario
To run a mock MRVA scenario, follow these steps:
1. Enable the mock GitHub API server by adding the following in your VS Code user settings (which can be found by running the `Preferences: Open User Settings (JSON)` VS Code command):
```json
"codeQL.mockGitHubApiServer":{
"enabled":true
}
```
1. Run the `CodeQL: Mock GitHub API Server: Load Scenario` command from the command pallet, and choose one of the scenarios to load.
1. Execute a normal MRVA. At this point you should see the scenario being played out, rather than an actual MRVA running.
1. Once you're done, you can stop using the mock scenario with `CodeQL: Mock GitHub API Server: Unload Scenario`
If you want to replay the same scenario you should unload and reload it so requests are replayed from the start.
##### Recording a new test scenario
To record a new mock MRVA scenario, follow these steps:
1. Enable the mock GitHub API server by adding the following in your VS Code user settings (which can be found by running the `Preferences: Open User Settings (JSON)` VS Code command):
```json
"codeQL.mockGitHubApiServer":{
"enabled":true
}
```
1. Run the `CodeQL: Mock GitHub API Server: Start Scenario Recording` VS Code command from the command pallet.
1. Execute a normal MRVA.
1. Once what you wanted to record is done (e.g. the MRVA has finished), then run the `CodeQL: Mock GitHub API Server: Save Scenario` command from the command pallet.
1. The scenario should then be available for replaying.
If you want to cancel recording, run the `CodeQL: Mock GitHub API Server: Cancel Scenario Recording` command.
Once the scenario has been recorded, it's often useful to remove some of the requests to speed up the replay, particularly ones that fetch the variant analysis status. Once some of the request files have manually been removed, the [fix-scenario-file-numbering script](./extensions/ql-vscode/scripts/fix-scenario-file-numbering.ts) can be used to update the number of the files. See the script file for details on how to use.
#### Scenario data location
Pre-recorded scenarios are stored in `./src/mocks/scenarios`. However, it's possible to configure the location, by setting the `codeQL.mockGitHubApiServer.scenariosPath` configuration property in the VS Code user settings.
## Releasing (write access required)
1. Double-check the `CHANGELOG.md` contains all desired change comments and has the version to be released with date at the top.
* Go through all recent PRs and make sure they are properly accounted for.
* Make sure all changelog entries have links back to their PR(s) if appropriate.
1. Double-check that the node version we're using matches the one used for VS Code. If it doesn't, you will then need to update the node version in the following files:
*`.nvmrc` - this will enable `nvm` to automatically switch to the correct node version when you're in the project folder
*`.github/workflows/main.yml` - all the "node-version: <version>" settings
*`.github/workflows/release.yml` - the "node-version: <version>" setting
1. Double-check that the extension `package.json` and `package-lock.json` have the version you intend to release. If you are doing a patch release (as opposed to minor or major version) this should already be correct.
1. Create a PR for this release:
* This PR will contain any missing bits from steps 1 and 2. Most of the time, this will just be updating `CHANGELOG.md` with today's date.
* Create a new branch for the release named after the new version. For example: `v1.3.6`
* Create a new commit with a message the same as the branch name.
* Create a PR for this branch.
* Wait for the PR to be merged into `main`
1. Switch to `main` and add a new tag on the `main` branch with your new version (named after the release), e.g.
```bash
git checkout main
git tag v1.3.6
```
If you've accidentally created a badly named tag, you can delete it via
```bash
git tag -d badly-named-tag
```
1. Push the new tag up:
a. If you're using a fork of the repo:
```bash
git push upstream refs/tags/v1.3.6
```
b. If you're working straight in this repo:
```bash
git push origin refs/tags/v1.3.6
```
This will trigger [a release build](https://github.com/github/vscode-codeql/releases) on Actions.
* **IMPORTANT** Make sure you are on the `main` branch and your local checkout is fully updated when you add the tag.
* If you accidentally add the tag to the wrong ref, you can just force push it to the right one later.
1. Monitor the status of the release build in the `Release` workflow in the Actions tab.
* DO NOT approve the "publish" stages of the workflow yet.
1. Download the VSIX from the draft GitHub release at the top of [the releases page](https://github.com/github/vscode-codeql/releases) that is created when the release build finishes.
1. Unzip the `.vsix` and inspect its `package.json` to make sure the version is what you expect,
or look at the source if there's any doubt the right code is being shipped.
1. Install the `.vsix` file into your vscode IDE and ensure the extension can load properly. Run a single command (like run query, or add database).
1. Go to the actions tab of the vscode-codeql repository and select the [Release workflow](https://github.com/github/vscode-codeql/actions?query=workflow%3ARelease).
- If there is an authentication failure when publishing, be sure to check that the authentication keys haven't expired. See below.
1. Approve the deployments of the correct Release workflow. This will automatically publish to Open VSX and VS Code Marketplace.
1. Go to the draft GitHub release in [the releases tab of the repository](https://github.com/github/vscode-codeql/releases), click 'Edit', add some summary description, and publish it.
1. Confirm the new release is marked as the latest release at <https://github.com/github/vscode-codeql/releases>.
1. If documentation changes need to be published, notify documentation team that release has been made.
1. Review and merge the version bump PR that is automatically created by Actions.
## Secrets and authentication for publishing
Repository administrators, will need to manage the authentication keys for publishing to the VS Code marketplace and Open VSX. Each requires an authentication token. The VS Code marketplace token expires yearly.
To regenerate the Open VSX token:
1. Log in to the [user settings page on Open VSX](https://open-vsx.org/user-settings/namespaces).
1. Make sure you are a member of the GitHub namespace.
1. Go to the [Access Tokens](https://open-vsx.org/user-settings/tokens) page and generate a new token.
1. Update the secret in the `publish-open-vsx` environment in the project settings.
To regenerate the VSCode Marketplace token, please see our internal documentation. Note that Azure DevOps PATs expire every 90 days and must be regenerated.
[Information about testing can be found here](./docs/testing.md).
@@ -7,7 +7,7 @@ The extension is released. You can download it from the [Visual Studio Marketpla
To see what has changed in the last few versions of the extension, see the [Changelog](https://github.com/github/vscode-codeql/blob/main/extensions/ql-vscode/CHANGELOG.md).
[](https://github.com/github/vscode-codeql/actions?query=workflow%3A%22Build+Extension%22+branch%3Amaster)
@@ -15,6 +15,7 @@ To see what has changed in the last few versions of the extension, see the [Chan
* Shows the flow of data through the results of path queries, which is essential for triaging security results.
* Provides an easy way to run queries from the large, open source repository of [CodeQL security queries](https://github.com/github/codeql).
* Adds IntelliSense to support you writing and editing your own CodeQL query and library files.
* Supports you running CodeQL queries against thousands of repositories on GitHub using multi-repository variant analysis.
## Project goals and scope
@@ -24,8 +25,8 @@ This project will track new feature development in CodeQL and, whenever appropri
This extension depends on the following two extensions for required functionality. They will be installed automatically when you install VS Code CodeQL.
The CodeQL for VS Code extension defines the version of Node.js that it is intended to run with. This Node.js version is used when running most CI and unit tests.
When running in production (i.e. as an extension for a VS Code application) it will use the Node.js version provided by VS Code. This can mean a different Node.js version is used by different users with different versions of VS Code.
We should make sure the CodeQL for VS Code extension works with the Node.js version supplied by all versions of VS Code that we support.
## Checking the version of Node.js supplied by VS Code
You can find this info by seleting "About Visual Studio Code" from the top menu.

## Updating the Node.js version
The following files will need to be updated:
-`.github/workflows/cli-test.yml` - the "node-version: '[VERSION]'" setting
-`.github/workflows/main.yml` - all the "node-version: '[VERSION]'" settings
-`.github/workflows/release.yml` - the "node-version: '[VERSION]'" setting
-`extensions/ql-vscode/.nvmrc` - this will enable nvm to automatically switch to the correct node version when you're in the project folder
-`extensions/ql-vscode/package-lock.json` - the "engines.node: '[VERSION]'" setting
-`extensions/ql-vscode/package.json` - the "engines.node: '[VERSION]'" setting
## Node.js version used in tests
Unit tests will use whatever version of Node.js is installed locally. In CI this will be the version specified in the workflow.
Integration tests download a copy of VS Code and then will use whatever version of Node.js is provided by VS Code. Our integration tests are currently pinned to an older version of VS Code. See [VS Code version used in tests](./vscode-version.md#vs-code-version-used-in-tests) for more information.
1. Run the ["Run CLI tests" workflow](https://github.com/github/vscode-codeql/actions/workflows/cli-test.yml) and make sure the tests are green. If there were no merges between the time the workflow ran (it runs daily), and the release, you can skip this step.
1. Double-check the `CHANGELOG.md` contains all desired change comments and has the version to be released with date at the top.
* Go through all recent PRs and make sure they are properly accounted for.
* Make sure all changelog entries have links back to their PR(s) if appropriate.
* For picking the new version number, we default to increasing the patch version number, but make our own judgement about whether a change is big enough to warrant a minor version bump. Common reasons for a minor bump could include:
* Making substantial new features available to all users. This can include lifting a feature flag.
* Breakage in compatibility with recent versions of the CLI.
* Minimum required version of VS Code is increased.
* New telemetry events are added.
* Deprecation or removal of commands.
* Accumulation of many changes, none of which are individually big enough to warrant a minor bump, but which together are. This does not include changes which are purely internal to the extension, such as refactoring, or which are only available behind a feature flag.
1. Double-check that the node version we're using matches the one used for VS Code. See the [Node.js version instructions](./node-version.md) for more information.
1. Double-check that the extension `package.json` and `package-lock.json` have the version you intend to release. If you are doing a patch release (as opposed to minor or major version) this should already be correct.
1. Create a PR for this release:
* This PR will contain any missing bits from steps 1, 2 and 3. Most of the time, this will just be updating `CHANGELOG.md` with today's date.
* Create a new branch for the release named after the new version. For example: `v1.3.6`
* Create a new commit with a message the same as the branch name.
* Create a PR for this branch.
* Wait for the PR to be merged into `main`
1. Switch to `main` branch and pull latest changes
1. Lock the `main` branch.
* Go to the [branch protection rules for the `main` branch](https://github.com/github/vscode-codeql/settings/branch_protection_rules/16447115)
* Select "Lock branch"
* Click "Save changes"
1. Ensure that no PRs have been merged since the release PR that you merged. If there were, you might need to unlock `main` temporarily and update the CHANGELOG again.
1. Build the extension `npm run build` and install it on your VS Code using "Install from VSIX".
1. Go through [our test plan](./test-plan.md) to ensure that the extension is working as expected.
1. Switch to `main` and add a new tag on the `main` branch with your new version (named after the release), e.g.
```bash
git checkout main
git tag v1.3.6
```
If you've accidentally created a badly named tag, you can delete it via
```bash
git tag -d badly-named-tag
```
1. Unlock the main branch
* Go to the [branch protection rules for the `main` branch](https://github.com/github/vscode-codeql/settings/branch_protection_rules/16447115)
* Deselect "Lock branch"
* Click "Save changes"
1. Push the new tag up:
a. If you're using a fork of the repo:
```bash
git push upstream refs/tags/v1.3.6
```
b. If you're working straight in this repo:
```bash
git push origin refs/tags/v1.3.6
```
This will trigger [a release build](https://github.com/github/vscode-codeql/releases) on Actions.
* **IMPORTANT** Make sure you are on the `main` branch and your local checkout is fully updated when you add the tag.
* If you accidentally add the tag to the wrong ref, you can just force push it to the right one later.
1. Monitor the status of the release build in the `Release` workflow in the Actions tab.
* DO NOT approve the "publish" stages of the workflow yet.
1. Download the VSIX from the draft GitHub release at the top of [the releases page](https://github.com/github/vscode-codeql/releases) that is created when the release build finishes.
1. Unzip the `.vsix` and inspect its `package.json` to make sure the version is what you expect,
or look at the source if there's any doubt the right code is being shipped.
1. Install the `.vsix` file into your vscode IDE and ensure the extension can load properly. Run a single command (like run query, or add database).
1. Go to the actions tab of the vscode-codeql repository and select the [Release workflow](https://github.com/github/vscode-codeql/actions?query=workflow%3ARelease).
* If there is an authentication failure when publishing, be sure to check that the authentication keys haven't expired. See below.
1. Approve the deployments of the correct Release workflow. This will automatically publish to Open VSX and VS Code Marketplace.
1. Go to the draft GitHub release in [the releases tab of the repository](https://github.com/github/vscode-codeql/releases), click 'Edit', add some summary description, and publish it.
1. Confirm the new release is marked as the latest release at <https://github.com/github/vscode-codeql/releases>.
1. If documentation changes need to be published, notify documentation team that release has been made.
1. Review and merge the version bump PR that is automatically created by Actions.
## Secrets and authentication for publishing
Repository administrators, will need to manage the authentication keys for publishing to the VS Code marketplace and Open VSX. Each requires an authentication token. The VS Code marketplace token expires yearly.
To regenerate the Open VSX token:
1. Log in to the [user settings page on Open VSX](https://open-vsx.org/user-settings/namespaces).
1. Make sure you are a member of the GitHub namespace.
1. Go to the [Access Tokens](https://open-vsx.org/user-settings/tokens) page and generate a new token.
1. Update the secret in the `publish-open-vsx` environment in the project settings.
To regenerate the VSCode Marketplace token, please see our internal documentation. Note that Azure DevOps PATs expire every 90 days and must be regenerated.
This document describes the manual test plan for the QL extension for Visual Studio Code.
The plan will be executed manually to start with but the goal is to eventually automate parts of the process (based on
effort vs value basis).
## What this doesn't cover
We don't need to test features (and permutations of features) that are covered by automated tests.
## Before releasing the VS Code extension
- Run at least one local query and MRVA using the existing version of the extension (to generate "old" query history items).
- Go through the required test cases listed below.
- Check major PRs since the previous release for specific one-off things to test. Based on that, you might want to
choose to go through some of the Optional Test Cases.
## Required Test Cases
### Local databases
#### Test case 1: Download a database from GitHub
1. Click "Download Database from GitHub" and enter `angular-cn/ng-nice` and select the javascript language if prompted
#### Test case 2: Import a database from an archive
1. Obtain a javascript database for `babel/babel`
- You can do `gh api "/repos/babel/babel/code-scanning/codeql/databases/javascript" -H "Accept: application/zip" > babel.zip` to fetch a database from GitHub.
2. Click "Choose Database from Archive" and select the file you just downloaded above.
### Local queries
#### Test case 1: Running a path problem query and viewing results
1. Open the [javascript UnsafeJQueryPlugin query](https://github.com/github/codeql/blob/main/javascript/ql/src/Security/CWE-079/UnsafeJQueryPlugin.ql).
2. Select the `angular-cn/ng-nice` database (or download it if you don't have one already)
3. Run a local query.
4. Once the query completes:
- Check that the result messages are rendered
- Check that the paths can be opened and paths are rendered correctly
- Check that alert locations can be clicked on
#### Test case 2: Running a problem query and viewing results
1. Open the [javascript ReDoS query](https://github.com/github/codeql/blob/main/javascript/ql/src/Performance/ReDoS.ql).
2. Select the `babel/babel` database (or download it if you don't have one already)
3. Run a local query.
4. Once the query completes:
- Check that the result messages are rendered
- Check that alert locations can be clicked on
#### Test case 3: Running a non-problem query and viewing results
1. Open the [cpp FunLinesOfCode query](https://github.com/github/codeql/blob/main/cpp/ql/src/Metrics/Functions/FunLinesOfCode.ql).
2. Select the `google/brotli` database (or download it if you don't have one already)
3. Run a local query.
4. Once the query completes:
- Check that the results table is rendered
- Check that result locations can be clicked on
#### Test case 4: Can use AST viewer
1. Click on any code location from a previous query to open a source file from a database
2. Open the AST viewing panel and click "View AST"
3. Once the AST is computed:
- Check that it can be navigated
### MRVA
#### Test Case 1: Running a path problem query and viewing results
1. Open the [javascript UnsafeJQueryPlugin query](https://github.com/github/codeql/blob/main/javascript/ql/src/Security/CWE-079/UnsafeJQueryPlugin.ql).
2. Run a MRVA against the following repo list:
```json
{
"name": "test-repo-list",
"repositories": [
"angular-cn/ng-nice",
"apache/hadoop",
"apache/hive"
]
}
```
3. Check that a notification message pops up and the results view is opened.
4. Check the query history. It should:
- Show that an item has been added to the query history
- The item should be marked as "in progress".
5. Once the query starts:
- Check the results view
- Check the code paths view, including the code paths drop down menu.
- Check that the repository filter box works
- Click links to files/locations on GitHub
- Check that the query history item is updated to show the number of results
6. Once the query completes:
- Check that the query history item is updated to show the query status as "complete"
#### Test Case 2: Running a problem query and viewing results
1. Open the [javascript ReDoS query](https://github.com/github/codeql/blob/main/javascript/ql/src/Performance/ReDoS.ql).
2. Run a MRVA against the "Top 10" repositories.
3. Check that a notification message pops up and the results view is opened.
4. Check the query history. It should:
- Show that an item has been added to the query history
- The item should be marked as "in progress".
5. Once the query completes:
- Check that the results are rendered with an alert message and a highlighted code snippet:
#### Test Case 3: Running a non-problem query and viewing results
1. Open the [cpp FunLinesOfCode query](https://github.com/github/codeql/blob/main/cpp/ql/src/Metrics/Functions/FunLinesOfCode.ql).
2. Run a MRVA against a single repository (e.g. `google/brotli`).
3. Check that a notification message pops up and the results view is opened.
4. Check the query history. It should:
- Show that an item has been added to the query history
- The item should be marked as "in progress".
5. Once the query completes:
- Check that the results show up in a table:

#### Test Case 4: Interacting with query history
1. Click a history item (for MRVA):
- Check that exporting results works
- Check that sorting results works
- Check that copying repo lists works
2. Click "Open Results Directory":
- Check that the correct directory is opened and there are results in it
3. Click "View Logs":
- Check that the correct workflow is opened
#### Test Case 5: Canceling a variant analysis run
Run one of the above MRVAs, but cancel it from within VS Code:
- Check that the query is canceled and the query history item is updated.
- Check that the workflow run is also canceled.
- Check that any available results are visible in VS Code.
### CodeQL Model Editor
#### Test Case 1: Opening the model editor
1. Download the `sofastack/sofa-jraft` java database from GitHub.
2. Open the Model Editor with the "CodeQL: Open CodeQL Model Editor" command from the command palette.
- Check that the editor loads and shows methods to model.
- Check that methods are grouped per library (e.g. `rocksdbjni@7.7.3` or `asm@6.0`)
- Check that the "Open database" link works.
#### Test Case 2: Model methods
1. Expand one of the libraries.
- Change the model type and check that the other dropdowns change.
2. Save the modeled methods.
3. Click "Open extension pack"
- Check that the file explorer opens a directory with a "models" directory
4. Open the ".model.yml" file corresponding to the library that was changed.
- Check that the file contrains the entries that was modeled.
#### Test Case 3: Model with AI
Note that this test requires the feature flag: `codeQL.model.llmGeneration`
1. Click "Model with AI".
- Check that rows change to "Thinking".
- Check that results come back and rows get filled out.
#### Test Case 4: Model as dependency
1. Click "Model as dependency"
- Check that grouping are now per package (e.g. `com.alipay.sofa.rraft.option` or `com.google.protobuf`)
2. Click "Generate".
- Check that rows are filled out.
### General
#### Test case 1: Change to a different colour theme
Open at least one of the above MRVAs and at least one local query, then try changing to a different colour theme and check that everything looks sensible.
Are there any components that are not showing up?
## Optional Test Cases
These are mostly aimed at MRVA, but some of them are also applicable to non-MRVA queries.
### Selecting repositories to run on
#### Test case 1: Running a query on a single repository
1. When the repository exists and is public
1. Has a CodeQL database for the correct language
2. Has a CodeQL database for another language
3. Does not have any CodeQL databases
2. When the repository exists and is private
1. Is accessible and has a CodeQL database
2. Is not accessible
3. When the repository does not exist
#### Test case 2: Running a query on a custom repository list
1. The repository list is non-empty
1. All repositories in the list have a CodeQL database
2. Some but not all repositories in the list have a CodeQL database
3. No repositories in the list have a CodeQL database
2. The repository list is empty
#### Test case 3: Running a query on all repositories in an organization
1. The org exists
1. The org contains repositories that have CodeQL databases
2. The org contains repositories of the right language but without CodeQL databases
3. The org contains repositories not of the right language
4. The org contains private repositories that are inaccessible
2. The org does not exist
### Using different types of controller repos
#### Test case 1: Running a query when the controller repository is public
1. Can run queries on public repositories
2. Can not run queries on private repositories
#### Test case 2: Running a query when the controller repository is private
1. Can run queries on public repositories
2. Can run queries on private repositories
#### Test case 3: Running a query when the controller repo exists but you do not have write access
1. Cannot run queries
#### Test case 4: Running a query when the controller repo doesn’t exist
1. Cannot run queries
#### Test case 5: Running a query when the "config field" for the controller repo is not set
1. Cannot run queries
### Query History
This requires running a MRVA query and viewing the query history.
The first test case specifies actions that you can do when the query is first run and is in "pending" state. We start
with this since it has quite a limited number of actions you can do.
#### Test case 1: When variant analysis state is "pending"
1. Starts monitoring variant analysis
2. Cannot open query history item
3. Can delete a query history item
1. Item is removed from list in UI
2. Files on dist are deleted (can get to files using "open query directory")
4. Can sort query history items
1. By name
2. By query date
3. By result count
5. Cannot open query directory
6. Can open query that produced these results
1. When the file still exists and has not moved
2. When the file does not exist
7. Cannot view logs
8. Cannot copy repository list
9. Cannot export results
10. Cannot select to create a gist
11. Cannot select to save as markdown
12. Cannot cancel analysis
#### Test case 2: When the variant analysis state is not "pending"
1. Query history is loaded when VSCode starts
2. Handles when action workflow was canceled while VSCode was closed
3. Can open query history item
1. Manually by clicking on them
2. Automatically when VSCode starts (if they were open when VSCode was last used)
4. Can delete a query history item
1. Item is removed from list in UI
2. Files on dist are deleted (can get to files using "open query directory")
5. Can sort query history items
1. By name
2. By query date
3. By result count
6. Can open query directory
7. Can open query that produced these results
1. When the file still exists and has not moved
2. When the file does not exist
8. Can view logs
9. Can copy repository list
1. Text is copied to clipboard
2. Text is a valid repository list
10. Can export results
11. Can select to create gist
1. A gist is created
2. The first thing in the gist is a summary
3. Contains a file for each repository with results
4. A popup links you to the gist
12. Can select to save as markdown
1. A directory is created on disk
2. Contains a summary file
3. Contains a file for each repository with results
4. A popup allows you to open the directory
#### Test case 3: When variant analysis state is "in_progress"
1. Starts monitoring variant analysis
1. Ready results are downloaded
2. Can cancel analysis
1. Causes the actions run to be canceled
#### Test case 4: When variant analysis state is in final state ("succeeded"/"failed"/"canceled")
1. Stops monitoring variant analysis
1. All results are downloaded if state is succeeded
2. Otherwise, ready results are downloaded, if any are available
2. Cannot cancel analysis
### MRVA results view
This requires running a MRVA query and seeing the results view.
<!-- markdownlint-disable-next-line MD024 -->
#### Test case 1: When variant analysis state is "pending"
1. Can open a results view
2. Results view opens automatically
- When starting variant analysis run
- When VSCode opens (if view was open when VSCode was closed)
3. Results view is empty
#### Test case 2: When variant analysis state is not "pending"
1. Can open a results view
2. Results view opens automatically
1. When starting variant analysis run
2. When VSCode opens (if view was open when VSCode was closed)
3. Can copy repository list
1. Text is copied to clipboard
2. Text is a valid repository list
4. Can export results
1. Only includes repos that you have selected (also see section from query history)
5. Can cancel analysis
6. Can open query file
1. When the file still exists and has not moved
2. When the file does not exist
7. Can open query text
8. Can sort repos
1. Alphabetically
2. By number of results
3. By popularity
9. Can filter repos
10. Shows correct statistics
1. Total number of results
2. Total number of repositories
3. Duration
11. Can see live results
1. Results appear in extension as soon as each query is completed
12. Can view interpreted results (i.e. for a "problem" query)
1. Can view non-path results
2. Can view code paths for "path-problem" queries
13. Can view raw results (i.e. for a non "problem" query)
1. Renders a table
14. Can see skipped repositories
1. Can see repos with no db in a tab
1. Shown warning that explains the tab
2. Can see repos with no access in a tab
1. Shown warning that explains the tab
3. Only shows tab when there are skipped repos
15. Result downloads
1. All results are downloaded automatically
2. Download status is indicated by a spinner (Not currently any indication of progress beyond "downloading" and "not downloading")
3. Only 3 items are downloaded at a time
4. Results for completed queries are still downloaded when
1. Some but not all queries failed
2. The variant analysis was canceled after some queries completed
#### Test case 3: When variant analysis state is in "succeeded" state
1. Can view logs
2. All results are downloaded
#### Test case 4: When variant analysis is in "failed" or "canceled" state
1. Can view logs
1. Results for finished queries are still downloaded.
### MRVA repositories panel
1. Add a list
1. Add a database at the top level
1. Add a database to a list
1. Add a the same database at a top-level and in a list
1. Delete a list
1. Delete a database from the top level
1. Delete a database from a list
1. Add an owner
1. Remove an owner
1. Rename a list
1. Open on GitHub
1. Select a list (via "Select" button and via context menu action)
1. Run MRVA against a user-defined list
1. Run MRVA against a top-N list
1. Run MRVA against an owner
1. Run MRVA against a database
1. Copy repo list
1. Open config file
1. Make changes via config file (ensure JSON schema is helping out)
1. Close and re-open VS Code (ensure lists are there)
1. Collapse/expand tree nodes
1. Create a new list, right click and select "Add repositories with GitHub Code Search". Enter the language 'python' and the query "UserMixin". This should show a rate limiting notification after a while but eventually populate the list with roughly 770 items.
Error cases that trigger an error notification:
1. Try to add a list with a name that already exists
1. Try to add a top-level database that already exists
1. Try to add a database in a list that already exists in the list
Error cases that show an error in the panel (and only the edit button should be visible):
1. Edit the db config file directly and save invalid JSON
1. Edit the db config file directly and save valid JSON but invalid config (e.g. add an unknown property)
1. Edit the db config file directly and save two lists with the same name
Cases where there the welcome view is shown:
1. No controller repo is set in the user's settings JSON.
* Unit tests: these live in the `tests/unit-tests/` directory
* View tests: these live in `src/view/variant-analysis/__tests__/`
* VSCode integration tests:
*`test/vscode-tests/activated-extension` tests: These are intended to cover functionality that require the full extension to be activated but don't require the CLI. This suite is not run against multiple versions of the CLI in CI.
*`test/vscode-tests/no-workspace` tests: These are intended to cover functionality around not having a workspace. The extension is not activated in these tests.
*`test/vscode-tests/minimal-workspace` tests: These are intended to cover functionality that need a workspace but don't require the full extension to be activated.
* CLI integration tests: these live in `test/vscode-tests/cli-integration`
* These tests are intended to cover functionality that is related to the integration between the CodeQL CLI and the extension. These tests are run against each supported versions of the CLI in CI.
The CLI integration tests require an instance of the CodeQL CLI to run so they will require some extra setup steps. When adding new tests to our test suite, please be mindful of whether they need to be in the cli-integration folder. If the tests don't depend on the CLI, they are better suited to being a VSCode integration test.
Any test data you're using (sample projects, config files, etc.) must go in a `test/vscode-tests/*/data` directory. When you run the tests, the test runner will copy the data directory to `out/vscode-tests/*/data`.
## Running the tests
Pre-requisites:
1. Run `npm run build`.
2. You will need to have `npm run watch` running in the background.
### 1. From the terminal
Then, from the `extensions/ql-vscode` directory, use the appropriate command to run the tests:
* Unit tests: `npm run test:unit`
* View Tests: `npm test:view`
* VSCode integration tests: `npm run test:vscode-integration`
#### Running CLI integration tests from the terminal
The CLI integration tests require the CodeQL standard libraries in order to run so you will need to clone a local copy of the `github/codeql` repository.
1. Set the `TEST_CODEQL_PATH` environment variable: running from a terminal, you _must_ set the `TEST_CODEQL_PATH` variable to point to a checkout of the `github/codeql` repository. The appropriate CLI version will be downloaded as part of the test.
2. Run your test command:
```shell
cd extensions/ql-vscode && npm run test:cli-integration
```
### 2. From VSCode
Alternatively, you can run the tests inside of VSCode. There are several VSCode launch configurations defined that run the unit and integration tests.
You will need to run tests using a task from inside of VS Code, under the "Run and Debug" view:
* Unit tests: run the _Launch Unit Tests - React_ task
* View Tests: run the _Launch Unit Tests_ task
* VSCode integration tests: run the _Launch Unit Tests - No Workspace_ and _Launch Unit Tests - Minimal Workspace_ tasks
#### Running CLI integration tests from VSCode
The CLI integration tests require the CodeQL standard libraries in order to run so you will need to clone a local copy of the `github/codeql` repository.
1. Set the `TEST_CODEQL_PATH` environment variable: running from a terminal, you _must_ set the `TEST_CODEQL_PATH` variable to point to a checkout of the `github/codeql` repository. The appropriate CLI version will be downloaded as part of the test.
2. Set the codeql path in VSCode's launch configuration: open `launch.json` and under the _Launch Integration Tests - With CLI_ section, uncomment the `"${workspaceRoot}/../codeql"` line. If you've cloned the `github/codeql` repo to a different path, replace the value with the correct path.
3. Run the VSCode task from the "Run and Debug" view called _Launch Integration Tests - With CLI_.
## Running a single test
### 1. Running a single test from the terminal
The easiest way to run a single test is to change the `it` of the test to `it.only` and then run the test command with some additional options
to only run tests for this specific file. For example, to run the test `test/vscode-tests/cli-integration/run-queries.test.ts`:
```shell
npm run test:cli-integration -- --runTestsByPath test/vscode-tests/cli-integration/run-queries.test.ts
```
You can also use the `--testNamePattern` option to run a specific test within a file. For example, to run the test `test/vscode-tests/cli-integration/run-queries.test.ts`:
```shell
npm run test:cli-integration -- --runTestsByPath test/vscode-tests/cli-integration/run-queries.test.ts --testNamePattern "should create a QueryEvaluationInfo"
```
### 2. Running a single test from VSCode
Alternatively, you can run a single test inside VSCode. To do so, install the [Jest Runner](https://marketplace.visualstudio.com/items?itemName=firsttris.vscode-jest-runner) extension. Then,
you will have quicklinks to run a single test from within test files. To run a single unit or integration test, click the "Run" button. Debugging a single test is currently only supported
for unit tests by default. To debug integration tests, open the `.vscode/settings.json` file and uncomment the `jestrunner.debugOptions` lines. This will allow you to debug integration tests.
Please make sure to revert this change before committing; with this setting enabled, it is not possible to debug unit tests.
Without the Jest Runner extension, you can also use the "Launch Selected Unit Test (vscode-codeql)" launch configuration to run a single unit test.
## Using a mock GitHub API server
Multi-Repo Variant Analyses (MRVA) rely on the GitHub API. In order to make development and testing easy, we have functionality that allows us to intercept requests to the GitHub API and provide mock responses.
### Using a pre-recorded test scenario
To run a mock MRVA scenario, follow these steps:
1. Enable the mock GitHub API server by adding the following in your VS Code user settings (which can be found by running the `Preferences: Open User Settings (JSON)` VS Code command):
```json
"codeQL.mockGitHubApiServer":{
"enabled":true
}
```
1. Run the `CodeQL: Mock GitHub API Server: Load Scenario` command from the command pallet, and choose one of the scenarios to load.
1. Execute a normal MRVA. At this point you should see the scenario being played out, rather than an actual MRVA running.
1. Once you're done, you can stop using the mock scenario with `CodeQL: Mock GitHub API Server: Unload Scenario`
If you want to replay the same scenario you should unload and reload it so requests are replayed from the start.
### Recording a new test scenario
To record a new mock MRVA scenario, follow these steps:
1. Enable the mock GitHub API server by adding the following in your VS Code user settings (which can be found by running the `Preferences: Open User Settings (JSON)` VS Code command):
```json
"codeQL.mockGitHubApiServer":{
"enabled":true
}
```
1. Run the `CodeQL: Mock GitHub API Server: Start Scenario Recording` VS Code command from the command pallet.
1. Execute a normal MRVA.
1. Once what you wanted to record is done (e.g. the MRVA has finished), then run the `CodeQL: Mock GitHub API Server: Save Scenario` command from the command pallet.
1. The scenario should then be available for replaying.
If you want to cancel recording, run the `CodeQL: Mock GitHub API Server: Cancel Scenario Recording` command.
Once the scenario has been recorded, it's often useful to remove some of the requests to speed up the replay, particularly ones that fetch the variant analysis status. Once some of the request files have manually been removed, the [fix-scenario-file-numbering script](../extensions/ql-vscode/scripts/fix-scenario-file-numbering.ts) can be used to update the number of the files. See the script file for details on how to use.
### Scenario data location
Pre-recorded scenarios are stored in `./src/common/mock-gh-api/scenarios`. However, it's possible to configure the location, by setting the `codeQL.mockGitHubApiServer.scenariosPath` configuration property in the VS Code user settings.
The CodeQL for VS Code extension specifies the versions of VS Code that it is compatible with. VS Code will only offer to install and upgrade the extension when this version range is satisfied.
## Where is the VS Code version specified
1. Hard limit in [`package.json`](https://github.com/github/vscode-codeql/blob/606bfd7f877d9fffe4ff83b78015ab15f8840b12/extensions/ql-vscode/package.json#L16)
This is the value that VS Code understands and respects. If a user does not meet this version requirement then VS Code will not offer to install the CodeQL for VS Code extension, and if the extension is already installed then it will silently refuse to upgrade the extension.
1. Soft limit in [`extension.ts`](https://github.com/github/vscode-codeql/blob/606bfd7f877d9fffe4ff83b78015ab15f8840b12/extensions/ql-vscode/src/extension.ts#L307)
This value is used internally by the CodeQL for VS Code extension and is used to provide a warning to users without blocking them from installing or upgrading. If the extension detects that this version range is not met it will output a warning message to the user prompting them to upgrade their VS Code version to ge the latest features of CodeQL.
## When to update the VS Code version
Generally we should aim to support as wide a range of VS Code versions as we can, so unless there is a reason to do so we do not update the minimum VS Code version requirement.
Reasons for updating the minimum VS Code version include:
- A new feature is included in VS Code. We may want to ensure that it is available to use so we do not have to provide an alternative code path.
- A breaking change has happened in VS Code, and it is not possible to support both new and old versions.
Also consider what percentage of our users are using each VS Code version. This information is available in our telemetry.
## How to update the VS Code version
To provide a good experience to users, it is recommented to update the `MIN_VERSION` in `extension.ts` first and release, and then update the `vscode` version in `package.json` and release again. By stagging this update across two releases it gives users on older VS Code versions a chance to upgrade before it silently refuses to upgrade them.
## VS Code version used in tests
Our integration tests are currently pinned to use an older version of VS Code due to <https://github.com/github/vscode-codeql/issues/2402>.
This version is specified in [`jest-runner-vscode.config.base.js`](https://github.com/github/vscode-codeql/blob/d93f2b67c84e79737b0ce4bb74e31558b5f5166e/extensions/ql-vscode/test/vscode-tests/jest-runner-vscode.config.base.js#L17).
Until this is resolved this will limit us updating our minimum supported version of VS Code.
- Add warning when using a VS Code version older than 1.82.0. [#2854](https://github.com/github/vscode-codeql/pull/2854)
- Fix a bug when parsing large evaluation log summaries. [#2858](https://github.com/github/vscode-codeql/pull/2858)
- Right-align and format numbers in raw result tables. [#2864](https://github.com/github/vscode-codeql/pull/2864)
- Remove rate limit warning notifications when using Code Search to add repositories to a variant analysis list. [#2812](https://github.com/github/vscode-codeql/pull/2812)
## 1.9.0 - 19 September 2023
- Release the [CodeQL model editor](https://codeql.github.com/docs/codeql/codeql-for-visual-studio-code/using-the-codeql-model-editor) to create CodeQL model packs for Java frameworks. Open the editor using the "CodeQL: Open CodeQL Model Editor (Beta)" command. [#2823](https://github.com/github/vscode-codeql/pull/2823)
## 1.8.12 - 11 September 2023
- Fix a bug where variant analysis queries would fail for queries in the `codeql/java-queries` query pack. [#2786](https://github.com/github/vscode-codeql/pull/2786)
## 1.8.11 - 7 September 2023
- Update how variant analysis results are displayed. For queries with ["path-problem" or "problem" `@kind`](https://codeql.github.com/docs/writing-codeql-queries/metadata-for-codeql-queries/#metadata-properties), you can choose to display the results as rendered alerts or as a table of raw results. For queries with any other `@kind`, the results are displayed as a table. [#2745](https://github.com/github/vscode-codeql/pull/2745) & [#2749](https://github.com/github/vscode-codeql/pull/2749)
- When running variant analyses, don't download artifacts for repositories with no results. [#2736](https://github.com/github/vscode-codeql/pull/2736)
- Group the extension settings, so that they're easier to find in the Settings UI. [#2706](https://github.com/github/vscode-codeql/pull/2706)
## 1.8.10 - 15 August 2023
- Add a code lens to make the `CodeQL: Open Referenced File` command more discoverable. Click the "Open referenced file" prompt in a `.qlref` file to jump to the referenced `.ql` file. [#2704](https://github.com/github/vscode-codeql/pull/2704)
## 1.8.9 - 3 August 2023
- Remove "last updated" information and sorting from variant analysis results view. [#2637](https://github.com/github/vscode-codeql/pull/2637)
- Links to code on GitHub now include column numbers as well as line numbers. [#2406](https://github.com/github/vscode-codeql/pull/2406)
- No longer highlight trailing commas for jump to definition. [#2615](https://github.com/github/vscode-codeql/pull/2615)
- Fix a bug where the QHelp preview page was not being refreshed after changes to the underlying `.qhelp` file. [#2660](https://github.com/github/vscode-codeql/pull/2660)
## 1.8.8 - 17 July 2023
- Remove support for CodeQL CLI versions older than 2.9.4. [#2610](https://github.com/github/vscode-codeql/pull/2610)
- Implement syntax highlighting for the `additional` and `default` keywords. [#2609](https://github.com/github/vscode-codeql/pull/2609)
## 1.8.7 - 29 June 2023
- Show a run button on the file tab for query files, that will start a local query. This button will only show when a local database is selected in the extension. [#2544](https://github.com/github/vscode-codeql/pull/2544)
- Add a `CodeQL: Quick Evaluation Count` command to generate the count summary statistics of the results set
without spending the time to compute locations and strings. [#2475](https://github.com/github/vscode-codeql/pull/2475)
## 1.8.6 - 14 June 2023
- Add repositories to a variant analysis list with GitHub Code Search. [#2439](https://github.com/github/vscode-codeql/pull/2439) and [#2476](https://github.com/github/vscode-codeql/pull/2476)
## 1.8.5 - 6 June 2023
- Add settings `codeQL.variantAnalysis.defaultResultsFilter` and `codeQL.variantAnalysis.defaultResultsSort` for configuring how variant analysis results are filtered and sorted in the results view. The default is to show all repositories, and to sort by the number of results. [#2392](https://github.com/github/vscode-codeql/pull/2392)
- Fix bug to ensure error messages have complete stack trace in message logs. [#2425](https://github.com/github/vscode-codeql/pull/2425)
- Fix bug where the `CodeQL: Compare Query` command did not work for comparing quick-eval queries. [#2422](https://github.com/github/vscode-codeql/pull/2422)
- Update text of copy and export buttons in variant analysis results view to clarify that they only copy/export the selected/filtered results. [#2427](https://github.com/github/vscode-codeql/pull/2427)
- Add warning when using unsupported CodeQL CLI version. [#2428](https://github.com/github/vscode-codeql/pull/2428)
- Retry variant analysis results download if connection times out. [#2440](https://github.com/github/vscode-codeql/pull/2440)
## 1.8.4 - 3 May 2023
- Avoid repeated error messages when unable to monitor a variant analysis. [#2396](https://github.com/github/vscode-codeql/pull/2396)
- Fix bug where a variant analysis didn't display the `#select` results set correctly when the [query metadata](https://codeql.github.com/docs/writing-codeql-queries/about-codeql-queries/#query-metadata) didn't exactly match the query results. [#2395](https://github.com/github/vscode-codeql/pull/2395)
- On the variant analysis results page, show the count of successful analyses instead of completed analyses, and indicate the reason why analyses were not successful. [#2349](https://github.com/github/vscode-codeql/pull/2349)
- Fix bug where the "CodeQL: Set Current Database" command didn't always select the database. [#2384](https://github.com/github/vscode-codeql/pull/2384)
## 1.8.3 - 26 April 2023
- Added ability to filter repositories for a variant analysis to only those that have results [#2343](https://github.com/github/vscode-codeql/pull/2343)
- Add new configuration option to allow downloading databases from http, non-secure servers. [#2332](https://github.com/github/vscode-codeql/pull/2332)
- Remove title actions from the query history panel that depended on history items being selected. [#2350](https://github.com/github/vscode-codeql/pull/2350)
## 1.8.2 - 12 April 2023
- Fix bug where users could end up with the managed CodeQL CLI getting uninstalled during upgrades and not reinstalled. [#2294](https://github.com/github/vscode-codeql/pull/2294)
- Fix bug that was causing code flows to not get updated when switching between results. [#2288](https://github.com/github/vscode-codeql/pull/2288)
- Restart the CodeQL language server whenever the _CodeQL: Restart Query Server_ command is invoked. This avoids bugs where the CLI version changes to support new language features, but the language server is not updated. [#2238](https://github.com/github/vscode-codeql/pull/2238)
- Avoid requiring a manual restart of the query server when the [external CLI config file](https://docs.github.com/en/code-security/codeql-cli/using-the-codeql-cli/specifying-command-options-in-a-codeql-configuration-file#using-a-codeql-configuration-file) changes. [#2289](https://github.com/github/vscode-codeql/pull/2289)
## 1.8.1 - 23 March 2023
- Show data flow paths of a variant analysis in a new tab. [#2172](https://github.com/github/vscode-codeql/pull/2172) & [#2182](https://github.com/github/vscode-codeql/pull/2182)
- Show labels of entities in exported CSV results. [#2170](https://github.com/github/vscode-codeql/pull/2170)
## 1.8.0 - 9 March 2023
- Send telemetry about unhandled errors happening within the extension. [#2125](https://github.com/github/vscode-codeql/pull/2125)
- Enable collection of telemetry concerning interactions with UI elements, including buttons, links, and other inputs. [#2114](https://github.com/github/vscode-codeql/pull/2114)
- Prevent the automatic installation of CodeQL CLI version 2.12.3 to avoid a bug in the language server. CodeQL CLI 2.12.2 will be used instead. [#2126](https://github.com/github/vscode-codeql/pull/2126)
## 1.7.10 - 23 February 2023
- Fix bug that was causing unwanted error notifications.
## 1.7.9 - 20 February 2023
No user facing changes.
## 1.7.8 - 2 February 2023
- Renamed command "CodeQL: Run Query" to "CodeQL: Run Query on Selected Database". [#1962](https://github.com/github/vscode-codeql/pull/1962)
- Remove support for CodeQL CLI versions older than 2.7.6. [#1788](https://github.com/github/vscode-codeql/pull/1788)
## 1.7.7 - 13 December 2022
- Increase the required version of VS Code to 1.67.0. [#1662](https://github.com/github/vscode-codeql/pull/1662)
@@ -28,7 +28,7 @@ For information about other configurations, see the separate [CodeQL help](https
1. [Run a query](#running-a-query).
---
<!-- markdownlint-disable-next-line MD024 -->
## Quick start: Installing and configuring the extension
### Installing the extension
@@ -71,6 +71,7 @@ in the starter workspace directory.
If you're using your own clone of the CodeQL standard libraries, you can do a `git pull` from where you have the libraries checked out.
<!-- markdownlint-disable-next-line MD024 -->
## Quick start: Using CodeQL
You can find all the commands contributed by the extension in the Command Palette (**Ctrl+Shift+P** or **Cmd+Shift+P**) by typing `CodeQL`, many of them are also accessible through the interface, and via keyboard shortcuts.
@@ -94,7 +95,7 @@ The instructions below assume that you're using the CodeQL starter workspace, or
1. Expand the `ql` folder and locate a query to run. The standard queries are grouped by target language and then type, for example: `ql/java/ql/src/Likely Bugs`.
1. Open a query (`.ql`) file.
1. Right-click in the query window and select **CodeQL: Run Query**. Alternatively, open the Command Palette (**Ctrl+Shift+P** or **Cmd+Shift+P**), type `Run Query`, then select **CodeQL: Run Query**.
1. Right-click in the query window and select **CodeQL: Run Query on Selected Database**. Alternatively, open the Command Palette (**Ctrl+Shift+P** or **Cmd+Shift+P**), type `Run Query`, then select **CodeQL: Run Query on Selected Database**.
The CodeQL extension runs the query on the current database using the CLI and reports progress in the bottom right corner of the application.
When the results are ready, they're displayed in the CodeQL Query Results view. Use the dropdown menu to choose between different forms of result output.
<pathd="M 35.300905,316.97546 H 93.308718 V 116.76062 L 30.203249,129.41687 V 97.07312 L 92.957155,84.41687 h 35.507815 v 232.55859 h 58.00781 v 29.88282 H 35.300905 Z"fill="#C5C5C5"/>
<pathd="M 35.300905,316.97546 H 93.308718 V 116.76062 L 30.203249,129.41687 V 97.07312 L 92.957155,84.41687 h 35.507815 v 232.55859 h 58.00781 v 29.88282 H 35.300905 Z"/>
"Query language is unsupported. Select language manually.",
);
}catch(e){
voidextLogger.log(
"Could not autodetect query language. Select language manually.",
);
}
}
// will be undefined if user cancels the quick pick.
returnawaitaskForLanguage(cliServer,false);
}
exportasyncfunctionaskForLanguage(
cliServer: CodeQLCliServer,
throwOnEmpty=true,
):Promise<QueryLanguage|undefined>{
constlanguage=awaitwindow.showQuickPick(
awaitcliServer.getSupportedLanguages(),
{
placeHolder:"Select target language for your query",
ignoreFocusOut: true,
},
);
if(!language){
// This only happens if the user cancels the quick pick.
if(throwOnEmpty){
thrownewUserCancellationException("Cancelled.");
}else{
voidshowAndLogErrorMessage(
extLogger,
"Language not found. Language must be specified manually.",
);
}
returnundefined;
}
if(!isQueryLanguage(language)){
voidshowAndLogErrorMessage(
extLogger,
`Language '${language}' is not supported. Only languages ${Object.values(
QueryLanguage,
).join(", ")} are supported.`,
);
returnundefined;
}
returnlanguage;
}
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.