The abjad-book Sphinx extension should only produce unique images.
Images can be found unique based on the md5 hash of the LilyPond file used to generate them. Once generated, the image should be saved using the md5 hash as all or part of its name. Subsequent attempts to generate an identical image can first test for the existence of an image whose name matches the hash, and then proceed to generate if it does not exist, or bail if it does.
This will allow multiple pages in the API to reference the same image, which is especially useful when illustrating inherited properties or methods, all of which may share an identical docstring. It will also save us space when uploading the documentation to projectabjad.org.
trevorbaca
commented
10 years ago
From josiah.oberholtzer on December 09, 2012 10:33:29
The abjad-book Sphinx extension should only produce unique images.
Images can be found unique based on the md5 hash of the LilyPond file used to generate them. Once generated, the image should be saved using the md5 hash as all or part of its name. Subsequent attempts to generate an identical image can first test for the existence of an image whose name matches the hash, and then proceed to generate if it does not exist, or bail if it does.
This will allow multiple pages in the API to reference the same image, which is especially useful when illustrating inherited properties or methods, all of which may share an identical docstring. It will also save us space when uploading the documentation to projectabjad.org.
Original issue: http://code.google.com/p/abjad/issues/detail?id=182