Instead of inserting the entire global context into the iframe's HTML,
where we need to encode it to and decode it from JSON and base64, we
only insert a stripped global context, i.e. without the file tree and
the utils. If these are needed in the iframe, for example when a script
changes the DOM and needs to load an image, we use messages to
communicate with the parent document and retrieve the file in question.
It confuses scripts when it is triggered twice. Instead of passing the
global context, we simply write it into the HTML of the iframe.
We need to encode it in base64 to avoid context confusion (e.g. `</script>`
inside the JSON of the global context).
With this commit, the virtual HTML file is processed before it's fed to
the iframe and thus rendered. With the previous approach, script tags
with inline JavaScript were executed first and script tags with a
virtual `src` were executed last. This messed up the execution flow of
the document.
We don't need to know about the `ImportError` every time we guess a mime
type. Because an `ImportError` is also raised if `python-magic` is
installed but `libmagic1` is not present, we can handle this at the
beginning of the file.
This fixes a bug that appears if the document is built on a system
without libmagic1. The python library python-magic uses it to guess the
mime type. We treat files with unknown mime type as `octet-stream`,
which causes them to be downloaded rather than displayed in the main
iframe. This results in lots of popups when opening the resulting HTML
file in a browser.
In this commit we add a fallback to the built-in library `mimetypes`
which gueses the mime type from the filename alone. This should work in
most cases but could still fail.
Close#2.
