I just changed the semantics of names for imports, so
now if you give a directory name, you get that package
and all descendant packages & modules. If you want
just the package, then you have to either give its
module name, or the filename of its __init__.py file.
This change should be sufficient to cover the common
case of wanting to document an entire package. But it's
not quite the same as recursive behavior. In particular,
it won't recurse into directories that don't include
__init__.py files; and it only picks up .py? files, not .so
files. That kind of recursion is occasionally useful (e.g.,
for documenting the entire standard library), but isn't
usually what people want. Unless you can give a
compelling reason for including it, I'm inclined to leave
it out, and let people resort to other mechanisms if they
need true recursion (e.g., the xNIX find command).
(Note: this change affects the gui as well as the
command line interface.)
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Logged In: YES
user_id=195958
I just changed the semantics of names for imports, so
now if you give a directory name, you get that package
and all descendant packages & modules. If you want
just the package, then you have to either give its
module name, or the filename of its __init__.py file.
This change should be sufficient to cover the common
case of wanting to document an entire package. But it's
not quite the same as recursive behavior. In particular,
it won't recurse into directories that don't include
__init__.py files; and it only picks up .py? files, not .so
files. That kind of recursion is occasionally useful (e.g.,
for documenting the entire standard library), but isn't
usually what people want. Unless you can give a
compelling reason for including it, I'm inclined to leave
it out, and let people resort to other mechanisms if they
need true recursion (e.g., the xNIX find command).
(Note: this change affects the gui as well as the
command line interface.)