A PowerShell module for downloading files.

• Invoke-WebRequest requires you to supply the output filename, which you might not know in advance, and it's not always possible to extract it from the URL.
• It also holds the entire file in memory whilst downloading, which is bad news if downloading large files.
• It will not retain the modified date of the original source file as many browsers do.
• Also (on Windows PowerShell with $ProgressPreference set as default) the progress bar gets updated so frequently that it drastically slows down the download.
• Start-BitsTransfer can determine the file name automatically, but it does not work for all URLs and is only supported on Windows.
• Some URLs require different user agents connect successfully.

This module solves these problems:

• It uses the .NET HttpClient class (which is now recommended for use by Microsoft over the now deprecated WebClient class).
• It will first attempt to grab the file name from the Content-Disposition header. Headers are obtained by a regular GET request (as some web servers have been shown to refuse HEAD requests). If this header is not present, it will extract the file name from the absolute URL (since the supplied URL may redirect elsewhere).
• The file is streamed directly to disk rather than holding it entirely in memory.
• The modified date will be updated once download has complete to match the Last-Modified header if found.
• The progress bar is limited to updating every 250ms to prevent overuse of system resources.
• User agent strings for Chrome and the Googlebot web crawler will be attempted by default.

Install from the Powershell Gallery by running the following command:

Install-Module -Name PsDownload

Invoke-Download -URL "https://aka.ms/vs/17/release/VC_redist.x64.exe" -Destination "$env:USERPROFILE\Downloads"

Pipeline input is also supported:

"https://aka.ms/vs/17/release/VC_redist.x64.exe","https://aka.ms/vs/17/release/VC_redist.x86.exe" | Invoke-Download -Destination "$env:USERPROFILE\Downloads"

URI is also accepted as an alias of URL.

Optional parameters:

• FileName
  • Use this to override the file name rather than trying to auto-detect.
• UserAgent
  • Override the default user agent. By default it will cycle through using:
    • Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/116.0.0.0 Safari/537.36
    • Googlebot/2.1 (+http://www.google.com/bot.html)
• TempPath
  • By default the download in progress will be saved to %TEMP% / $env:TEMP.
• IgnoreDate
  • Ignore the Date-Modified header, modified will be the date the file was downloaded instead.
• BlockFile
  • Mark the file as downloaded from the internet (by default it does not do this).
• NoClobber
  • Use this to prevent overwriting an existing file.
• NoProgress
  • Suppress progress bar.
• PassThru
  • Returns a FileInfo object to the pipeline for the downloaded file.

This has been tested against a large number of URLs, please submit an issue if it is unable to download a specific file. Note that some web pages may run some javascript to trigger the actual download. This type of URL is not supported, it must point directly at a resource without the need to execute any client-side scripts.