Perforce Triggers - modify description text - perforce

I was looking through Perforce : Prefill changelist description which is supposed to describe 1:1 how to create a form-out change trigger, which I did.
p4 triggers file
suffix form-out change "powershell S:/p4_template.ps1 %formfile%"
template file:
(Get-Content $args[0]) |
Foreach-Object {$_ -replace "<enter description here>", "template text"} |
Set-Content $args[0]
However, I'm constantly getting the following error message:
'suffix' validation failed:
Execution Failed: powershell -ExecutionPolicy Bypass -File
S:/p4template.ps1 /tmp/tmp.379055.140083088879040.130: No such file or directory
I did restart the perforce server as well (just because it was mentioned in that thread but looks like it doesn't really need to be restarted). From what I see, we want to have access to the tmp file and replace the text, but that tmp file /tmp/tmp.379055.140083088879040.130 doesn't seem to exist? I've checked %TEMP% and set it as well with an ENV var. Further i'm getting this message in p4v when I click refresh and it does a p4 change -o operation.

Answering for posterity: the issue here is that the trigger runs on the server, not the client, and presumably the No such file refers to either the powershell executable or the script not existing on the server.
Documentation on Perforce triggers: https://www.perforce.com/manuals/p4sag/Content/P4SAG/chapter.scripting.triggers.html

Related

Cross talk threads to allow for file system access in powershell

Reference: Runspace for button event in powershell
https://www.foxdeploy.com/blog/part-v-powershell-guis-responsive-apps-with-progress-bars.html
So, I believe my issue is that PowerShell is unable to access the memory space of the file system, from within the memory block, of my thread, is there a way to solve this, to access the file system, from a multi-threaded application?
Back Story:
So, I run a program, that calls upon "code"/command, from the command prompt, (*.exe) (Robocopy) to copy files from a server, to a group of computers, at a time. We have a classroom environment, at my work, so I have my setup, in a way, that I have a folder, per room. I keep a list of all our addresses (static), for each room, in their perspective folders. We have an update, from our developers, that we need to push to all of the rooms. We need to run a slow push, as to not disturb the production environment. It's proprietary, so we can't use a/any typical solution(s), like Microsoft SCCM. So, I created a script to push to the rooms. while it does work, it's not a smooth operation. I'm not actually the one pushing the update, because of the slow process, of updating. I'm just trying to make a stable smooth-running package, for the person, who is going to be doing it. My code works, outside of the thread, (I) tested it, I know it works.
So how I came my conclusion of knowing, that my code works outside of the thread. (The picture) I followed the same setup, with my code, (A button click event inside of a thread, running the form). Placed the actual working code, (tried, and tested, before making a Thread, for the interface, after completing backend operation code testing.)
("Region Boe's Addition") referring to Boe Prox, (from the link)
In his, he is updating from a command line/powershell window, via a function run inside a thread. I'm running an event from a button, inside of a thread and trying to run a separate thread, for the click event(s). Outside of the thread. The event works fine, but inside, it doesn't work, at all..
Basic Code:
// Multi thread, thread for the $form, and thread for the event (as per referenced link)
$var = [PowerShell]::Create().AddScript({ button.Add_Click{
$var = [PowerShell]::Create().AddScript{<Thread><Robocopy></Thread>}
})
Needed the "Start-Process" -Wait command to allow for the listbox, to be updated in-between copies, to confirm installation, through each step in the loop.
$choice = $comboBox.SelectedItem
# $drive = Get-Location
if(!(Test-Path -PathType Container -Path "L:\$choice"))
{
# New-Item -ItemType "container" -Path . -Name $choice
New-Item -ItemType "Directory" -Path . -Name $choice
}
# $folder = $_
# Where is it being stored at?
[System.IO.File]::ReadLines("Y:\$choice\IPs.txt") | foreach {
ping -a -n 2 -w 2000 $_ | Out-Null
Test-Connection -Count 2 -TimeToLive 2 $_ | Out-Null
if($?)
{
RoboCopy /Log:"L:\$folder\$_.log" $source \\$_\c$\tools
RoboCopy /Log+:"L:\$folder\$folder-MovementLogs.log" $source \\$_\c$\tools
Start-Process -Wait "P:\psexec.exe" -ArgumentList "\\$_ -d -e -h -s cmd /c reg import C:\tools\dump.reg"
# Copy-Item -LiteralPath Y:\* -Destination \\$_\c$\tools
$listBox.Items.Add($_)
}
}

Azure Pipeline Extract Task 7zip

Unable to extract zip to destination with default usage of Extract Task its fails with error:
##[error]Unable to locate executable file: 'C:\azagent\A5\_work\_tasks\ExtractFiles_5e1e3830-fbfb-11e5-aab1-090c92bc4988\1.200.0\7zip\7z.exe'. Please verify either the file path exists or the file can be found within a directory specified by the PATH environment variable. Also verify the file has a valid extension for an executable file.
Stating its fails to locate default 7zip path. Tried to use custom PATH setting but also fails with the same error.
UPDATE
Issue seem to be caused by permissions of agent. Still haven't been able to execute Release with Admin privileges in service mode. When run in interactive mode as Admin the release executes successfully.
Task fails whenever admin permission is required.
From the error message, 7zip seems not installed on your self-hosted agent. Try to install the 7zip before you use the Extract Task.
Take Bash Task as an example:
brew install p7zip
For Windows, use the below PowerShell script to install:
$dlurl = 'https://7-zip.org/' + (Invoke-WebRequest -UseBasicParsing -Uri 'https://7-zip.org/' | Select-Object -ExpandProperty Links | Where-Object {($_.outerHTML -match 'Download')-and ($_.href -like "a/*") -and ($_.href -like "*-x64.exe")} | Select-Object -First 1 | Select-Object -ExpandProperty href)
# modified to work without IE
# above code from: https://perplexity.nl/windows-powershell/installing-or-updating-7-zip-using-powershell/
$installerPath = Join-Path $env:TEMP (Split-Path $dlurl -Leaf)
Invoke-WebRequest $dlurl -OutFile $installerPath
Start-Process -FilePath $installerPath -Args "/S" -Verb RunAs -Wait
Remove-Item $installerPath

How to use New-AzDataLakeGen2Item to send multiple files under a directory

Based on document from Microsoft
-Source
- Specify the local source file path which will be upload to a Datalake Gen2 file.
TABLE 13
Type: String
Position: Named
Default value: None
Accept pipeline input: True
Accept wildcard characters: False <-- no wildcard?
The doc also provided an example
PS C:\> $task = New-AzDataLakeGen2Item -FileSystem "testfilesystem" -Path "dir1/dir2/file1" -Source "c:\sourcefile.txt" -Force -asjob
Then what if I'd like to send all files under a directory, so that
-Source "c:\temp\*" <-- wildcard not allowed
-Source "c:\temp\" <-- "Could not find file..."
instead of
-Source "c:\sourcefile.txt"
I don't think you will be able to upload multiple files using New-AzDataLakeGen2Item Cmdlet.
You can however make use of azcopy tool to upload multiple files. Based on the documentation, it supports copy files from local to ADLS Gen2 account (and vice-versa) among other things.
Please see this link for more details: https://learn.microsoft.com/en-us/azure/storage/common/storage-ref-azcopy-copy.

p4python create and submit a new file

How to create and submit a new file using p4python?
create_and_submit_file(full_path_in_depot, new_file_text_content):
logging.basicConfig(level=logging.INFO, format=LOG_FORMAT)
p4 = get_p4() # implemented
try: # Catch exceptions with try/except
connect_and_login_p4(p4) # implemented
# .. implementation here .. p4.some_call()
LOGGER.info('done')
except P4Exception:
for e in p4.errors: # Display errors
LOGGER.error(e)
If the file already exists on the local filesystem within a workspace, all you need to do is p4.run_add(file) and p4.run_submit('-d', 'this is my awesome file').
If the file doesn't exist, you need to create it, and if you don't have a workspace, you need to create that too. For the sake of brevity, here's how you'd do that from the command line completely from scratch (this maps pretty directly to P4Python but I don't know enough about your environment to give you code that'll work out of the box so I won't attempt the translation):
echo "my awesome file content" > my_awesome_file
p4 set P4CLIENT=my_awesome_client
p4 --field "View=//depot/... //my_awesome_client/..." client -o | p4 client -i
p4 add my_awesome_file
p4 submit -d "this is my awesome file"
Check out the example for p4.save_client to see a simple example of how you can create/modify a client spec with P4Python and modify the fields to suit your environment (similar to how I used the --field flag to set the View such that the root of my_awesome_client corresponds to //depot/...):
https://www.perforce.com/perforce/r14.2/manuals/p4script/python.p4.html#python.p4.save_spectype

using team city to insert build number & perform string replacement operations

I am using team city 9.1.7 version on Windows 2012 server. As part of the build steps, I build a nodejs based application using command line. The output is bunch of Javascript and html files.
In the next step (after the build is over & output is generated), I want to perform following:
Take the current build number from team city and insert it into index.html (available in output folder) file. I want to add a meta tag which can tell me the build version.
In the same file (index.html), I want to perform string find and replace operations. I want to add time stamp to files.
Find this <script src="bundle.js"></script> and
replace with <script src="bundle.js?time=getTime()"></script>
will results in <script src="bundle.js?time=4324324324"></script>
Try the following
Add a PowerShell step and run the following as source code
$versionNumber = "%build.number%"
$filePath = "%teamcity.agent.work.dir%\path\file.txt"
(GC $filePath).Replace("<head>", "<head><meta http-equiv='X-Version-Number' content='$versionNumber'>").Replace("bundle.js", "bundle.js?time=getTime()") | Set-Content $filePath
This will read the file contents in and perform two replacements on them and then write back to the file.
Not sure what your file path is or what you want the header called, but you should be able to change this to suit your requirements.
Hope this helps
REVISION
To catch any exceptions, try wrapping the code in a try catch block
try {
(GC $filePath).Replace("<head>", "<head><meta http-equiv='X-Version-Number' content='$versionNumber'>").Replace("bundle.js", "bundle.js?time=getTime()") | Set-Content $filePath
}
catch [System.Exception] {
Write-Output $_
Exit 1
}
To break out of the cache you could use the version number as this will increment each build and thus be unique
try {
(GC $filePath).Replace("<head>", "<head><meta http-equiv='X-Version-Number' content='$versionNumber'>").Replace("bundle.js", "bundle.js?v=$versionNumber") | Set-Content $filePath
}
catch [System.Exception] {
Write-Output $_
Exit 1
}

Resources