Functions/GenXdev.AI.Queries/Show-GenXdevScriptErrorFixInIde.ps1
|
################################################################################ <# .SYNOPSIS Parses error messages and fixes them using Github Copilot in VSCode. .DESCRIPTION This function analyzes error messages from any input source, identifies files that need attention, and opens them in Visual Studio Code with appropriate Github Copilot prompts to assist in fixing the issues. .PARAMETER Object The input to analyze for errors. This can be error messages, log entries, or any text containing error information. .PARAMETER Model The LM-Studio model to use for error analysis. Defaults to "qwen2.5-14b-instruct". .PARAMETER ModelLMSGetIdentifier Identifier used for getting specific model from LM Studio. .PARAMETER Temperature Temperature setting for response randomness (0.0-1.0). .PARAMETER MaxToken Maximum tokens allowed in response (-1 for default). .PARAMETER ShowWindow If specified, shows the LM Studio window during processing. .PARAMETER TTLSeconds Set a TTL (in seconds) for models loaded via API requests. .PARAMETER Gpu GPU offloading configuration: - "off": GPU offloading disabled - "max": All layers offloaded to GPU - 0-1: Fraction of layers offloaded - -1: LM Studio decides offloading - -2: Auto configure .PARAMETER Force Force stop LM Studio before initialization. .PARAMETER DontAddThoughtsToHistory Skip including model's thoughts in output. .PARAMETER ContinueLast Continue from last conversation. .PARAMETER Functions Array of function definitions. .PARAMETER ExposedCmdLets Array of PowerShell command definitions to use as tools. .PARAMETER NoConfirmationToolFunctionNames Array of command names that don't require confirmation. .PARAMETER Speak Enable text-to-speech for AI responses. .PARAMETER SpeakThoughts Enable text-to-speech for AI thought responses. .PARAMETER NoSessionCaching Don't store session in session cache. .PARAMETER ApiEndpoint Api endpoint url, defaults to http://localhost:1234/v1/chat/completions. .PARAMETER ApiKey The API key for authentication. .EXAMPLE Show-GenXdevScriptErrorFixInIde -Model "qwen2.5-14b-instruct" -Script { Run-ErrorousScript } .EXAMPLE $errorOutput | letsfixthis #> function Show-GenXdevScriptErrorFixInIde { [CmdletBinding()] [Alias("letsfixthis")] param ( ######################################################################## [Parameter( Position = 0, Mandatory = $true, ValueFromPipeline = $true, HelpMessage = "The script to execute and analyze for errors" )] [ScriptBlock] $Script, ######################################################################## [Parameter( Position = 1, Mandatory = $false, HelpMessage = "The LM-Studio model to use" )] [SupportsWildcards()] [string] $Model = "qwen2.5-14b-instruct", ######################################################################## [Parameter( Position = 2, Mandatory = $false, HelpMessage = "Identifier used for getting specific model from LM Studio" )] [string] $ModelLMSGetIdentifier = "qwen2.5-14b-instruct", ######################################################################## [Parameter( Mandatory = $false, HelpMessage = "Temperature for response randomness (0.0-1.0)")] [ValidateRange(0.0, 1.0)] [double] $Temperature = 0.2, ######################################################################## [Parameter( Mandatory = $false, HelpMessage = "Maximum tokens in response (-1 for default)")] [Alias("MaxTokens")] [int] $MaxToken = -1, ######################################################################## [Parameter( Mandatory = $false, HelpMessage = "Show the LM Studio window")] [switch] $ShowWindow, ######################################################################## [Alias("ttl")] [Parameter( Mandatory = $false, HelpMessage = "Set a TTL (in seconds) for models loaded via API requests")] [int] $TTLSeconds = -1, ######################################################################## [Parameter( Mandatory = $false, HelpMessage = "How much to offload to the GPU. If `"off`", GPU offloading is disabled. If `"max`", all layers are offloaded to GPU. If a number between 0 and 1, that fraction of layers will be offloaded to the GPU. -1 = LM Studio will decide how much to offload to the GPU. -2 = Auto " )] [int]$Gpu = -1, ######################################################################## [Parameter( Mandatory = $false, HelpMessage = "Force stop LM Studio before initialization" )] [switch]$Force, ######################################################################## [Parameter( Mandatory = $false, HelpMessage = "Include model's thoughts in output")] [switch] $DontAddThoughtsToHistory, ######################################################################## [Parameter( Mandatory = $false, HelpMessage = "Continue from last conversation")] [switch] $ContinueLast, ######################################################################## [Parameter( Mandatory = $false, HelpMessage = "Array of function definitions")] [hashtable[]] $Functions = @(), ######################################################################## [Parameter( Mandatory = $false, HelpMessage = "Array of PowerShell command definitions to use as tools")] [GenXdev.Helpers.ExposedCmdletDefinition[]] $ExposedCmdLets = $null, ######################################################################## # Array of command names that don't require confirmation [Parameter(Mandatory = $false)] [Alias("NoConfirmationFor")] [string[]] $NoConfirmationToolFunctionNames = @(), ########################################################################### [Parameter( HelpMessage = "Enable text-to-speech for AI responses", Mandatory = $false )] [switch] $Speak, ########################################################################### [Parameter( HelpMessage = "Enable text-to-speech for AI thought responses", Mandatory = $false )] [switch] $SpeakThoughts, ######################################################################## [Parameter( Mandatory = $false, HelpMessage = "Don't store session in session cache")] [switch] $NoSessionCaching, ######################################################################## [Parameter( Mandatory = $false, HelpMessage = "Api endpoint url, defaults to http://localhost:1234/v1/chat/completions")] [string] $ApiEndpoint = $null, ######################################################################## [Parameter( Mandatory = $false, HelpMessage = "The API key to use for the request")] [string] $ApiKey = $null ) begin { $llmPromptInfo = GenXdev.AI\Get-ScriptExecutionErrorFixPrompt @PSBoundParameters } process { # define function to process each issue identified by the LLM function processIssue($issue) { # copy parameters from this function to the LLM prompt function $invocationArgs = GenXdev.Helpers\Copy-IdenticalParamValues ` -BoundParameters $PSBoundParameters ` -FunctionName "GenXdev.AI\Get-ScriptExecutionErrorFixPrompt" $null = GenXdev.Coding\AssureVSCodeInstallation code (GenXdev.FileSystem\Expand-Path "$PSScriptRoot\..\..\..\..\..\") # wait for VSCode to open Microsoft.PowerShell.Utility\Start-Sleep 2 # send keystrokes to paste and execute the prompt in VSCode GenXdev.Windows\Send-Key -KeysToSend @("^``", "^``", "^+i", "^l", "^a", "{DELETE}") if ((-not $issue.Files) -or ($issue.Files.Count -eq 0)) { # preserve current clipboard content $oldClipboard = Microsoft.PowerShell.Management\Get-Clipboard try { # copy the AI prompt to clipboard $issue.Prompt | Microsoft.PowerShell.Management\Set-Clipboard # send keystrokes to paste and execute the prompt in VSCode GenXdev.Windows\Send-Key -KeysToSend @("^v", "{ENTER}", "^{ENTER}") # wait for VSCode to process the command Microsoft.PowerShell.Utility\Start-Sleep 2 } finally { # restore the original clipboard content $oldClipboard | Microsoft.PowerShell.Management\Set-Clipboard } } else { # iterate through each file mentioned in the issue $issue.Files | Microsoft.PowerShell.Core\ForEach-Object { # copy parameters for opening files in IDE $invocationArgs = GenXdev.Helpers\Copy-IdenticalParamValues ` -BoundParameters $PSBoundParameters ` -FunctionName "GenXdev.Coding\Open-SourceFileInIde" # set the path to the current file $invocationArgs.Path = $PSItem.Path # if line number is provided, include it in arguments if ($PSItem.LineNumber -is [int]) { $invocationArgs.LineNo = $PSItem.LineNumber } # enable code mode for IDE $invocationArgs.Code = $true # prepare key sequences for opening files in VSCode # first file gets different key sequence than subsequent files $invocationArgs.KeysToSend = @( @("{ESCAPE}", "^+%{F12}") ) # open the source file in the IDE with the given parameters GenXdev.Coding\Open-SourceFileInIde @invocationArgs } # preserve current clipboard content $oldClipboard = Microsoft.PowerShell.Management\Get-Clipboard try { # copy the AI prompt to clipboard $issue.Prompt | Microsoft.PowerShell.Management\Set-Clipboard # send keystrokes to paste and execute the prompt in VSCode GenXdev.Windows\Send-Key -KeysToSend @("^v", "{ENTER}", "^{ENTER}") # wait for VSCode to process the command Microsoft.PowerShell.Utility\Start-Sleep 2 } finally { # restore the original clipboard content $oldClipboard | Microsoft.PowerShell.Management\Set-Clipboard } } } # process each issue identified by the LLM $null = $llmpromptInfo | Microsoft.PowerShell.Core\ForEach-Object { if ($PSItem.StandardOutput -and ($PSItem.StandardOutput.Count -gt 0)) { return $PSItem.StandardOutput } # loop until user chooses to stop while ($true) { # process the current issue processIssue -issue $PSItem # prompt user for next action switch ($host.ui.PromptForChoice( "Make a choice", "What to do next?", @("&Stop", "Redo &Last"), 0)) { 0 { throw "Stopped"; return } } } } } end { } } ################################################################################ |