Hi all, my web page parsing program stopped working today because they started gzipping their content. I use the PageReceived event and my fix for now is to write it out to a file, gunzip and read back in; but I'd rather not have to go out through a file. Is that possible? The man pages say it'll process standard input to standard output but I don't know how to use those, and maybe I'm misreading :) (for mac if that makes a difference)

This is the current file round-tripping. If I have to use this is it sound?

Sub PageReceived(..., content as string)

  //content is gzipped
  dim f As FolderItem = SpecialFolder.Temporary.Child("agcontent.gz")
  dim bs As BinaryStream = BinaryStream.Create(f, true)
  bs.Write(content)
  bs.close
  bs = nil

  dim sh As new Shell
  sh.Execute("gunzip " + f.ShellPath)
  sh = nil

  f = SpecialFolder.Temporary.Child("agcontent")
  bs = BinaryStream.Open(f)
  content = bs.Read(bs.Length, Encodings.UTF8)
  bs.close
  bs = nil
  f.Delete
  f = nil
  
  //content now html, parse as before
  //...

End Sub

Thanks guys. What a rabbit hole. I don't really much at all understand how gzip, zlib, command line and declaring all work together. From a snippet posted by Eli Ott a couple weeks ago I think the declare lib would be "libz.dylib". There's actually 4 libz.*.dylib files in my ~usr/lib/ folder. Then it would be translating this code which I very tentatively think maybe I follow. Looks like a lot of work.

In terminal I was able to get gzip compressing standard input to a file
$ gzip > test.gz [return]
apples [return][control-d]
$

and then decompress the file to standard output
$ gzip -c -d test.gz [return]
apples
$

Doing them together seems to be prohibited: "If no files are specified, gzip will compress from standard input, OR decompress to standard output."

So now I look for Thomas Tempelmanns code... voila . ZLibInflator.Inflate look like the lot-of-work :) Will work it tomorrow, very tired. I use Thomas' global hotkey code too, need to keep these links handier.