Add support for loading image streams

[Mark Lee]

Currently, pyexiv2 (both 0.1.x and 0.2~bzr) only support loading images from the filesystem. I have added basic support for loading "image streams" via file-like objects (i.e., objects with the "read()" method), including unit tests.

This enhancement depends on bug 517298.

[Olivier Tilloy (osomon)]

Hey Mark, thanks for your work on this one.
I didn't test it but I read the code, and it looks correct to me.

Supporting file-like objects sounds neat, but I was wondering what's the most common use-case, and it occurred to me that in general image manipulation libraries rather give you access to a data buffer than to a file-like interface. See for example PIL.Image, and its tostring() method (http://www.pythonware.com/library/pil/handbook/image.htm#Image.tostring). Anyway, I think this use-case ought to be considered, at least as seriously as supporting file-like objects.

If we are to support different "constructors" for ImageMetadata, I would rather go for static methods than a polymorphic constructor (which implies magic code and is harder to document), e.g.:

  ImageMetadata.open(filename_or_file)
  ImageMetadata.from_buffer(data)

If .open() (or .__init__()) is called with a file-like object, the file can be .read() from the python code and the data passed directly to .from_buffer(), which avoids the need for a tricky data stream constructor in C++.

I'm open to comments, critics and suggestions. Let's agree on the API before implementing it.

[Mark Lee (malept)]

My particular use case is retrieving metadata from an uploaded image file in a web application. In Django, at least, the default settings are that smaller files (for some definition of "smaller") are handled in-memory, whereas larger files are handled on disk (see <http://docs.djangoproject.com/en/dev/topics/http/file-uploads/#upload-handlers>). Because the upload handlers are customizable, Django provides convenient access to uploaded files via file-like objects.

If I recall correctly, when I was writing my implementation, I was initially thinking about static methods, but I thought that at least in C++, overloading the constructor felt more natural to the language (whatever the analogous term to "pythonic" is).

I don't have any objections to using @staticmethod-style constructors in the Python code.

[Olivier Tilloy (osomon)]

I linked a branch (lp:~osomon/pyexiv2/from_buffer) that implements instantiating an ImageMetadata from a buffer, and exposing the image data as a buffer. This also allows to easily instantiate an ImageMetadata from a file-like object, using the following syntax:

    m = ImageMetadata.from_buffer(f.read())

@Mark and others, your feedback and comments on this approach are welcome, if it suits your needs I'll merge it.

[Olivier Tilloy (osomon)]

In the absence of feedback, I assume the proposed solution is good enough.
Merged with revision 288.