|
Originally Posted by OUBeaver
|
Yeah. It only happens with certain sites.
Is it a problem with the site or PIE?
|
It is a problem with both, the site is not set up properly to tell PIE that the file is a binary (rather than plain text) and PIE being too stupid to look at the file extension and decide intelligently what it should do.
For those who are interested in why some sites work and others don't:
It's all to do with mimetypes, as a file is downloaded from any site the site will send back a header for the file describing it's size and type i.e. text/html for a webpage and image/jpeg for a jpeg picture, many sites are set up badly and will default to text/plain for undefined filetypes (.cab is very likely to be left off the list of filetypes as it is not very common) and so PIE will moronically trust that what is being sent truly is viewable text and render it as such. The generally used mime type for unknown filetypes intended for download is application/octet-stream, I can only assume that sites intended for storing .cab files have properly set this up and so PIE will know that it is a data file it won't be able to handle and so saves it instead.