has anyone ever implemented something similar to postgresql_autodoc in Python?
TIA,
Sincerely,
Wolfgang
--
NO "Courtesy Copies" PLEASE!
Dunno - what is it?
--
Cheers,
Simon B.
Starting from the ugliest:
- pydoc
- epydoc
- sphinx
are 3 of them. I would suggest epydoc, the best look&feel/complexity
ratio. Sphinx is used to generate any kind of documentation (look at
http://docs.python.org/ it's build with Sphinx) but I never figured out
how to easily build documentation from code despite it states it is
posible. With epydoc you only need to press the GO button.
JM
I suspect he meant documenting postgres database structure.
One way would be reading the informaton_schema* and generating python
code out of it then use any of the above methods to finally document the
stuff :-)
*) http://www.alberton.info/postgresql_meta_info.html
Regards
Tino
I will re-precise my question:
Has anyone ever implemented a script in Python that generates documentation (especially diagrams, in a format such as e.g. Dia, Inkscape SVG or Tikz) for a PostgreSQL database either from an SQL script or by connecting to the database?
Postgresql_autodoc is unfortunately written in Perl. >;->
TIA,
And, btw., please respect my .sig,
Sincerely,
Wolfgang Keller
> Hello,
>
> I will re-precise my question:
>
> Has anyone ever implemented a script in Python that generates
> documentation (especially diagrams, in a format such as e.g. Dia, Inkscape
> SVG or Tikz) for a PostgreSQL database either from an SQL script or by
> connecting to the database?
I've written a schemadiff-tool for postgres based on SQLAlchemy reflection,
which is a branch of SQLAlchemy and which should become part of the
upcoming SA 0.6.
http://svn.sqlalchemy.org/sqlalchemy/branches/reflection
It allows you to extract information about the schema of the DB. This could
be the foundation of your tool.
>
> Postgresql_autodoc is unfortunately written in Perl. >;->
>
> TIA,
>
> And, btw., please respect my .sig,
Which is supposed to mean what? Never heard the term courtesy copy.
Diez
It appears to be an incorrect expansion of "Cc", which is actually
Carbon copy. Apparently he means all replies should be directly to the
list and not Cc his individual email address.
Which seems strange, because usually mailinglists are smart enough not
to send the author duplicate copies if they're named in the To or Cc
fields.
Cheers,
Chris
--
http://blog.rebertia.com