This change enables globbing (wildcard expansion) against the file system for native commands like '/bin/ls'. The expansion is only done in the file system. In non-filesystem drives expansion is not done and the pattern is returned unchanged.
Limitations of the fix:
Currently quoting is not honored so for a command like /bin/ls "*.txt", wildcard expansion will still be done. Adding support for bare word detection will come in a future PR. Use --% to suppress wildcard expansion e.g. git add --% *
This fixes issue #2607.
'RequiredModules' is a field in module manifest that can reference other modules using ModuleSpecification format.
The basic version of this format (just module name) was working fine, however, there was a problem when a more detailed version of the format was used (the one that uses module versions or/and GUIDs).
During module import, there is a check for cyclic references through 'RequiredModules' field. The bug was in this check for cyclic references, related to comparison rules for ModuleSpecification objects - as a result, the code was incorrectly reporting 'cyclic reference' error in cases when there was none.
Also, added tests for different ModuleSpecification formats and a test for error when there is actually a cyclic reference.
* fixes summary block typo
* Adds support for Port parameter for SSH PSSessions
* Reverted back to master, modified based on PR feedback
* Update exception message
* remove unused line
* Add existing constructor back in as to not break public contract
* remove port check
* pass nested inner exception straight to Should
* dispose runspace after each test
* Add SSHHostParameterSet attribute for Invoke-Command Port property
* Update ParseSSHConnectionHashTable method to accept Port value as integer
* Add helper method for validating port in range. Refactor port parameter constructor overload to use original constructor
* rename method
* Adds GetSSHConnectionStringParameter and GetSSHConnectionIntParameter methods for retrieving SSHConnection hashtable values
* Adds method comments
* Adds helper method comment
* Change methods to add C# 7 patterns
string. If the string was a uri with spaces, ToString() doesn't return the escaped version. The AbsoluteUri property
should be used instead which returns an escaped absolute uri (if valid).
Also renamed TestModuleManfest.ps1 to TestModuleManifest.Tests.ps1 so that it gets picked up correctly as Pester test.
Since HelpInfoUri is just a string, ensure it is a valid absolute uri and escaped correctly whereas before it was just
an opaque string that wasn't validated.
This makes it possibe to write for example
[ValidatePattern('[A-Z]:', ErrorMessage='The Drive should be specified as a single letter followed by a colon, for example "D:"')]
[string] $Drive,
The element being validated is also passed, so {0} can be used in the custom error message
This change moves powershell to .NET Core 2.0. Major changes are:
1. PowerShell assemblies are now targeting `netcoreapp2.0`. We are using `microsoft.netcore.app-2.0.0-preview1-001913-00`, which is from dotnet-core build 4/4/17. We cannot target `netstandard2.0` because the packages `System.Reflection.Emit` and `System.Reflection.Emit.Lightweight`, which are needed for powershell class, cannot be referenced when targeting `netstandard2.0`.
2. Refactor code to remove most CLR stub types and extension types.
3. Update build scripts to enable CI builds. The `-cache` section is specified to depend on `appveyor.yml`, so the cache will be invalidated if `appveyor.yml` is changed.
4. Ship `netcoreapp` reference assemblies with powershell to fix the issues in `Add-Type` (#2764). By default `Add-Type` will reference all those reference assemblies when compiling C# code. If `-ReferenceAssembly` is specified, then we search reference assemblies first, then the framework runtime assemblies, and lastly the loaded assemblies (possibly a third-party one that was already loaded).
5. `dotnet publish` generates executable on Unix platforms, but doesn't set "x" permission and thus it cannot execute. Currently, the "x" permission is set in the build script, `dotnet/cli` issue [#6286](https://github.com/dotnet/cli/issues/6286) is tracking this.
6. Replace the use of some APIs with the ones that take `SecureString`.
7. osx.10.12 is required to update to `netcoreapp2.0` because `dotnet-cli` 2.0.0-preview only works on osx.10.12.
8. Add dependency to `System.ValueTuple` to work around a ambiguous type identity issue in coreclr. The issue is tracked by `dotnet/corefx` [#17797](https://github.com/dotnet/corefx/issues/17797). When moving to newer version of `netcoreapp2.0`, we need to verify if this dependency is still needed.
There are cases like using PowerShell via Puppet where the account being
used does not have a home directory. Updated PowerShell to use a process
specific temporary directory if HOME, CONFIG, CACHE, and DATA directories
are not available. Temporary directory is removed when last runspace
is disposed.
Our assembly cache contains assemblies that are explicitly loaded by powershell egine, such as via module loading or the assembly entries from InitialSessionState. We should search it before searching all loaded assemblies to give preference to resolve a type against the assemblies contained in the cache, so that in case there is a conflict, we might have a preferred assembly to use for a type resolution.
Changes:
- Search from context.AssemblyCache.Values before search from all loaded assemblies.
- Skip assemblies that we already searched and found no matching type.
- Skip checking PS types kept in the scope and type accelerators when it's not necessary.
* PowerShell transcripts should include the configuration name in the transcript header
* adding test case for PowerShell transcripts should include the configuration name in the transcript header #2890
* corrected use of PSModulePath casing to be consistent with Windows PowerShell
addresses #3227
* addressing review feedback
make "PSModulePath" into const
fixed some test workarounds due to failures for external reasons that wasn't meant to be checked in
* addressing review feedback
make "PSModulePath" into const
fixed some test workarounds due to failures for external reasons that wasn't meant to be checked in
XDG profile directory creation can fail for accounts that do not have home directories.
The module analysis was trying to persist it's cache in an XDG profile directory.
The cache is less critical than it once was, so it's reasonable to not cache if there is no good place to do so.
Fixes#3011
Changed hard coded Windows directory separator and resolved path so the slashes are correct.
Throw if resolving file path returns more than one result
Fixes#2610
These changes provide the ability to debug remote running scripts started with the Invoke-Command cmdlet. The design is event based and provides new public events that allow subscribers to be notified when an Invoke-Command remote session is ready for debugging. Since Invoke-Command allows running scripts on multiple targets at once (fan-out) the notification event is raised for each remote session as it becomes ready for debugging. The subscriber to these events will be a script debugger implementation (such as PowerShell console, ISE, or VSCode) and will handle all debugging details such as simultaneously debugging multiple remote sessions at once in separate windows.
But these changes also include an internal implementation which is used by default if host debuggers don't want to handle the debugging details. This internal implementation is what PowerShell console, ISE uses so they can have this new behavior without having to modify their debugger implementations. The internal implementation serializes each remote session of Invoke-Command so that they can be debugged one at a time. The remote session debugger is "pushed" onto the internal debugger stack so that debugging transitions to the remote session. Existing debugging commands work so that the "quit" debugging command will stop the current remote session script from running and allow the next remote session to be debugged. Similarly the "continue" debugging command allows the script to continue running outside step mode and again go to the next remote session for debugging. The "stepout" debugging command steps out of all Invoke-Command remote sessions and lets the script continue to run for each remote session in parallel as they are normally run.
The purpose of Invoke-Command step-in remote debugging is allow seamless debugging of a local script that calls Invoke-Command on remote targets. But there is also a new Invoke-Command "-RemoteDebug" parameter that lets you Invoke-Command on the command line and have it drop directly into the debugger.
An example from the PowerShell command line looks like this:
```
PS C:\> C:\TestICM.ps1
Entering debug mode. Use h or ? for help.
Hit Command breakpoint on 'Invoke-Command'
At C:\TestICM.ps1:2 char:1
+ Invoke-Command -cn $computerName,paulhig-3 -File c:\LinuxScript.ps1
+ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
[DBG]: PS C:\>> list
1: $computerName = "localhost"
2:* Invoke-Command -cn $computerName,paulhig-3 -File c:\LinuxScript.ps1
3: "Test Complete!"
[DBG]: PS C:\>> stepin
At line:1 char:1
+ Write-Output "Running script on Linux!"
+ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
[paulhig-3]:[DBG]: [Process:14072]: [Runspace5]: PS C:\Users\paulhi\Documents>
```
Notice that the debugger "stepin" command transitioned from local script debugging to debugging the remote session on computer "paulhig-3", as can be seen by the change in the debugger prompt.
You can also do this from the command line to drop directly into the debugger
```
Invoke-Command -cn localhost -Script $scriptblock -RemoteDebug
```
These changes also remove an old behavior that was incompatible with this new step-in feature. Previously if a remote session running script hit a break point it would stop in the debugger and go to the "disconnected session" state. This was to allow the user to reconnect using Enter-PSSession and then interactively debug the remote session script. This behavior has been removed and now the user needs to attach a debugger using the newer Debug-Runspace cmdlet.
* Implement -version parameter in console host (address part of https://github.com/PowerShell/PowerShell/issues/1084)
This does not support providing a specific version to run, but
like most other *nix commands, -version will now return the version
of the PowerShell Engine. 'powershell' is prepended to the output to
match other *nix commands. We are using gitcommitid which includes more
info about the build.
When a TypeTable is created it includes the types from type files provided along with references to the type files. When the InitialSessionState (ISS) object processes these types it reads the type files again and ends up with duplicate type entries. PowerShell V5.1 ISS type processing was re-written to improve performance and no longer removes duplicate types, so that this scenario (runspace ISS reuse) results in errors causing a regression.
The fix is to copy only type data when a TypeTable is passed to the ISS.
Interactive hosts expect an `IncompleteParseException` to signal that more input is expected.
When detecting errors, the parser can report 2 positions:
* where the error should be reported
* where the error was detected
Typically these are the same, so most error reporting methods have a single parameter.
For missing braces, the pattern is supposed to be to report the error after the opening brace, but the error is typically detected at the end of the file.
There were a few places where we were not consistent in reporting such errors, this PR corrects those places.
Native argument completers were not invoked when the argument was a single dash.
The fix is to treat unbound command parameters as command arguments for the purposes of parameter/argument completion.
In rare cases, it was possible to cause high contention on a lock used
while compiling code to run in the interpreter.
This fixes the problem in a couple different ways:
* Use ConditionalWeakTable which has finer granularity in locking
* Avoid dictionary lookups in the most common cases
- There really aren't too many real cases, I could have covered them
all, but keeping the code generic is useful for the future.
Not quite related, but I noticed we didn't cache certain interpreter
instructions and it made sense to add a cache because we generate a lot
of array-init instructions.
Ilya