Using powershell to open an MTHML document and same embedded images as a separate files

Good find, Christopher! The BITS stuff is indeed something very smart to use.
Also, the way approaching how to use PS code is exactly what I would have used.

naleo96, if that is not sufficient to get the images, we will need an example mshtml file to try ourselves. Sadly, using and parsing HTML pages for particular info can get tricky sometimes.
The linked script downloads every image into the same folder, not keeping any folder hierarchy. Usually that should be fine, but not if you need to keep some source info.

What is still missing from that is the FTP upload, but you told that is not important ;-). Do you only upload the images? If you need the report itself to be transferred, you'll have to adapt pathes, of course.

Sadly it seems that you cannot use BITS with FTP, so you would have to do that using a System.Net.WebClient object if you want to run that in PS (or use a cmdline FTP tool for that).

Hi All.

Thanks for your replies. I've been away a couple of days and will be able to review your suggestions later today.

Thanks for your help so far

Thanks for your help.

The theory is all great but I've hit a bit of a brick wall. The BITS transfer method works great on a sample site. In my case however the path to the images is actually a parameterised query (I think thats the right term .. see the example below) and as result the Start-BitsTransfer fails with a 'FileNotFoundException'.

IMG Element Name):

<IMG BORDER="0" style="top:0px;left:0px;position:relative;" SRC="http://hansendbserver/ReportServer?%2FReservoir%20Level%20Reporting%2FReservoir%20Storage%20Level%20Report&rs%3ASessionID=zvai5c550yxoi1454paeu2yd&rs%3AFormat=HTML4.0&rs%3AImageID=IMGCON_1_0"/>

Select allOpen in new window

Error Message:

One alternative I have is to load a MHTML file. However this too is unsuccessful and again the images don't actually exist as files ... they're embedded in the file. The IMG source looks like this:

Once loaded into IE, the properties of the image look like this:


Once again, the Bits-Transfer method cannot be used to copy the file since the physical file does not exist.

My next strategy will be to get the report output as a Word file, and try to extract the images from that.

Any other suggestions would be most welcome.

Correct, BITS needs static content, and the image sources are indeed dynamic reports.
I assume you are not able to post an example MHTML?

Qlemo, thanks for your help.

Attached is a sample MHTML file - the one used in the previous example but renamed.

Don't waste too much time on this as I've actually moved on to a method that involves  using PowerShell to open the file using MSWord and perform a SaveAs to the HTML format (forcing the images to be saved as individual files). From there the script copies the individual files to the relevant external website using FTP. This is working ok but does have the drawback that it needs to run on a PC with Word installed but I can live with that at the moment.

Thanks again
sample.zip