At 06:02 PM 7/28/00 -0400, you wrote:
>Sorry, this seems a bit off-topic, but I hope that you can see that this is
>somewhat related, and may even help JDE's software design in the long run.
>
>I've been thinking on-and-off about adding an Intellisense-like
>functionality for editing Python in Emacs, analogous to what JDE is doing
>for Java.
>
>Today I stumbled upon the following:
> http://www.lfw.org/python/
>which includes a Python->HTMLdoc generator consisting of the following two
>files:
> http://www.lfw.org/python/inspect.py
> http://www.lfw.org/python/htmldoc.py
>
>The existence of these source files (especially inspect.py) shows that
>Python's introspection capabilities are so great that one can easily obtain
>the appropriate class hierarchy for a set of Python files without resorting
>to complex tools like the semantic bovinator. It would be easy to perform
>a preprocessing step and produce a tree structure suitable for JDE's data
>structures, or any data structures which you can think of.
>
Hi Jonathan,
Before undertaking this project, you should consider the following points.
* The JDE's "intellisense"-like method and field completion factility uses
Java's "reflection,"
i.e., introspection capability. It does not use or depend on the semantic
bovinator in any
way.
* The idea of developing a pre-processor to create a methods/fields
database to serve as
the basis for a JDE intellisense capability comes up fairly frequently on
this list. I keep
pointing out that such a preprocessor and such a database already exist,
namely
javac and the set of Java class files specified that it produces. It is
this database that
is used by the JDE's existing intellisense facility.
* I have no objection to someone developing another type of
preprocessor/intellisense-database
capability. However, I don't see the benefit.
- Paul
