From jh@web.de  Tue Apr  1 23:08:31 2003
From: jh@web.de (Juergen Hermann)
Date: Wed, 02 Apr 2003 01:08:31 +0200
Subject: [Doc-SIG] Using docutils on source code
Message-ID: <E190Urr-0008LM-00@smtp.web.de>

Hi!

Maybe=20I'm=20thick=20here,=20but=20from=20a=20more=20than=20cursory=20loo=
k=20at=20the=20docs=20I=20
got=20the=20impression=20that=20docutils=20is=20still=20"just"=20a=20docum=
ent=20processor,=20
and=20that=20there=20is=20no=20front-end=20to=20extract=20docstrings=20fro=
m=20source=20and=20
format=20them.

Can=20somebody=20enlighten=20me,=20and=20shove=20me=20in=20the=20right=20d=
irection?


Ciao,=20J=FCrgen




From smerten@oekonux.de  Wed Apr  2 19:47:22 2003
From: smerten@oekonux.de (Stefan Merten)
Date: Wed, 02 Apr 2003 21:47:22 +0200
Subject: rst-mode.el (was: [Doc-SIG] Re: rest-mode.el and more)
In-Reply-To: Your message of "Thu, 27 Mar 2003 22:36:30 EST"
 <3E83C33E.2070801@python.org>
Message-ID: <200304021947.h32JlMf18969@rosalu.merten-home.de>

-----BEGIN PGP SIGNED MESSAGE-----

Hi David and all!

5 days ago David Goodger wrote:
> [Stefan Merten]
>> I coded a font-lock mode which recognizes most reST constructs.
>=20
> It looks quite good!

:-)

I'm glad it's helping people :-) .

> I hope you continue to develop it.

I already did. It now supports comments and literal blocks. However,
this involves multi-line font-locking which is slightly beyond simple
font-lock. Particularly typing gets **slooooow** when point is after
such a block. Because of that the mode switches to `lazy-lock-mode'
which one might configure independently. There is a customizable
variable `rst-mode-lazy' which triggers this behavior so this can be
switched off if it is more disturbing than useful.

> Could it be made a derivative of indented-text-mode, so that it
> inherits that mode's behavior?  Or is it already, except for keymaps?
> (How does one tell?)  Or could this become a minor mode of
> indented-text-mode?

<disclaimer>
I'm actually nor an Emacs lisp hacker. I learned some Emacs lisp a
long time ago and added to that over time but I did not follow the
modern developments such as customization.
</disclaimer>

In modern Emacsen I once saw a technique to derive a new mode from an
old one. However, this seems not be supported everywhere. I'll check
out how ``rst-mode`` can be integrated with ``indented-text-mode``
(which at least in 20.7 is just an alias for ``text-mode`` anyway).

>> If you want to include it in an official CVS tree that's fine.
>=20
> Thank you.  I think I'll include it as tools/editors/emacs/rst-mode.el

Meanwhile you added me to the developers. Thanks :-) .

I just added::

	docutils/tools/editors/emacs/rst-mode.el

to the CVS repository. You may download it from::

	http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/*checkout*/docutils/docut=
ils/tools/editors/emacs/rst-mode.el?rev=3DHEAD&content-type=3Dtext/plain

> ("rst" not "rest", see below).

I changed all the names to "rst" instead of "rest". I just needed a
name to start with.


						Mit Freien Gr=FC=DFen

						Stefan

-----BEGIN PGP SIGNATURE-----
Version: 2.6.3in
Charset: noconv
Comment: Processed by Mailcrypt 3.5.7, an Emacs/PGP interface

iQCVAwUBPos+SgnTZgC3zSk5AQHy3QQAqgyAcANHpAdpcAdegJC4wpKQbWf9N7+N
HEHjuKPntw0trDv0PWx6sVrtxcEs/y8Kaf/+Z7/a2/AeSAp5bvk25hKxwcQI2jIM
2CQaPQqH19dkYwsVv59sldnBTMSmiDdpDnBEAsbSFqOwybHW9foAIUQckvJ97J1O
7npyJh7EoF8=3D
=3DfH1D
-----END PGP SIGNATURE-----


From smerten@oekonux.de  Wed Apr  2 19:53:59 2003
From: smerten@oekonux.de (Stefan Merten)
Date: Wed, 02 Apr 2003 21:53:59 +0200
Subject: [Doc-SIG] Re: rst-mode.el
In-Reply-To: Your message of "Wed, 02 Apr 2003 21:47:22 +0200"
 <200304021947.h32JlMf18969@rosalu.merten-home.de>
Message-ID: <200304021953.h32Jrxf19089@rosalu.merten-home.de>

-----BEGIN PGP SIGNED MESSAGE-----

Hi all!

I forgot this.

4 minutes ago Stefan Merten wrote:
> I just added::
>=20
> 	docutils/tools/editors/emacs/rst-mode.el
>=20
> to the CVS repository. You may download it from::
>=20
> 	http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/*checkout*/docutils/doc=
utils/tools/editors/emacs/rst-mode.el?rev=3DHEAD&content-type=3Dtext/plai=
n

I'm keen to know of any problems with that (besides the ones noted in
the source already). Feel free to write me private mail. However,
please don't expect to get a reply immediately.


						Mit Freien Gr=FC=DFen

						Stefan

-----BEGIN PGP SIGNATURE-----
Version: 2.6.3in
Charset: noconv
Comment: Processed by Mailcrypt 3.5.7, an Emacs/PGP interface

iQCVAwUBPos/1wnTZgC3zSk5AQGB9gQAkPTnJG4aIerRwgsF5DVvBFggHJNz2NvV
Lh26JSYUZdQ4QR6f+L5xYijMzp1And+f3KSDr0QTijWUuzBBxoX58JCqTjYjHxOy
144/HNIekHAzbF3CQVJLsgc0kHATPixhicEMqSOnBkbyeRUv2EsKc7OQANgUI7D+
fT3p1FsNrNA=3D
=3DObKA
-----END PGP SIGNATURE-----


From goodger@python.org  Fri Apr  4 23:51:18 2003
From: goodger@python.org (David Goodger)
Date: Fri, 04 Apr 2003 18:51:18 -0500
Subject: [Doc-SIG] Using docutils on source code
In-Reply-To: <E190Urr-0008LM-00@smtp.web.de>
References: <E190Urr-0008LM-00@smtp.web.de>
Message-ID: <3E8E1A76.1020106@python.org>

Juergen Hermann wrote:
> Maybe I'm thick here, but from a more than cursory look at the docs I 
> got the impression that docutils is still "just" a document processor, 
> and that there is no front-end to extract docstrings from source and 
> format them.
 >
> Can somebody enlighten me, and shove me in the right direction?

Docstring extraction is still under development (you're not thick, at 
least not in this instance ;-).  I have most of a source code parsing 
module in docutils/readers/python/moduleparser.py.  I haven't worked on 
it in a while, but I do plan to finish it eventually.  Ian Bicking wrote 
an initial front end for my moduleparser module, in 
sandbox/ianb/extractor/extractor.py.

-- 
David Goodger    http://starship.python.net/~goodger

Programmer/sysadmin for hire: http://starship.python.net/~goodger/cv



From ianb@colorstudy.com  Sat Apr  5 00:44:18 2003
From: ianb@colorstudy.com (Ian Bicking)
Date: 04 Apr 2003 18:44:18 -0600
Subject: [Doc-SIG] Re: Wikis using reST
Message-ID: <1049503458.3784.43.camel@lothlorien>

Someone pointed this thread out to me.  I thought I'd note that the
example application for the Web Framework Shootout is a Wiki using reST:
  http://colorstudy.com/docs/shootout.html

It uses a generic Wiki module that is in the docutils sandbox CVS
repository.  It is certainly fairly primitive -- I wanted an example
that could fit in a paper (actually fit in many times over) -- but the
core Wiki concepts are delightfully simple, and it lives up to that at
least.

I'm interested in extending the module, as I like the basic architecture
with a firm separation of content from interface.  But there's lots of
things I'm interested in extending...

  Ian



From aahz@pythoncraft.com  Tue Apr 15 02:14:50 2003
From: aahz@pythoncraft.com (Aahz)
Date: Mon, 14 Apr 2003 21:14:50 -0400
Subject: [Doc-SIG] Python book ideas
Message-ID: <20030415011450.GA25720@panix.com>

[Cross-posted to both doc-sig and marketing-python]

Last week I went to the Waterside Conference, which is a conference for
writers and publishers put on by my agent's company.  I picked up some
good leads for book ideas, some of which I'm holding to myself for a bit
while I talk with my agent, but I'd like to throw out one idea that's
been mentioned before as a joke: _Python for VB6 Programmers_

According to a presentation I saw, one survey claims that more than 50%
of VB6 programmers currently refuse to consider upgrading to VB.net.
That's a huge market for us even if we only get a tiny slice of it.  I'll
be glad to assist someone else to find a publisher.  Any takers?

Another idea that I and others have been kicking around for a while is a
data structures and algorithms book using Python.  Based on what I saw
at the conference, I think now would be a good time to start working on
that, possibly beginning it as an Open Source project of some kind.
-- 
Aahz (aahz@pythoncraft.com)           <*>         http://www.pythoncraft.com/

This is Python.  We don't care much about theory, except where it intersects 
with useful practice.  --Aahz, c.l.py, 2/4/2002


From altis@semi-retired.com  Tue Apr 15 03:04:46 2003
From: altis@semi-retired.com (Kevin Altis)
Date: Mon, 14 Apr 2003 19:04:46 -0700
Subject: [Doc-SIG] RE: [marketing-python] Python book ideas
In-Reply-To: <20030415011450.GA25720@panix.com>
Message-ID: <KJEOLDOPMIDKCMJDCNDPIEGFDEAA.altis@semi-retired.com>

> From: Aahz
>
> [Cross-posted to both doc-sig and marketing-python]
>
> Last week I went to the Waterside Conference, which is a conference for
> writers and publishers put on by my agent's company.  I picked up some
> good leads for book ideas, some of which I'm holding to myself for a bit
> while I talk with my agent, but I'd like to throw out one idea that's
> been mentioned before as a joke: _Python for VB6 Programmers_

Actually, I've mentioned it several times to some authors and O'Reilly and I
definitely wasn't joking, though nobody has chosen to pursue the project as
far as I know. The last person I brought this up with was Todd Mezzulo at
O'Reilly.

> According to a presentation I saw, one survey claims that more than 50%
> of VB6 programmers currently refuse to consider upgrading to VB.net.
> That's a huge market for us even if we only get a tiny slice of it.  I'll
> be glad to assist someone else to find a publisher. Any takers?

Regardless of the exact number, there are definitely a lot of VB 6 folks
that feel abandoned. MS has essentially told them to switch to .NET or die.
It probably represents the single largest market segment that is ripe for a
switch to Python, if the right environment is in place so that they can
still leverage a simple to learn and use GUI IDE and that can leverage
existing COM components and build standalones. The GUI IDE part is a bit
rough, but win32all and py2exe take care of the latter.

VBA and VBScript are other good "switch" opportunities. Unlike VB6, you can
convert VBScript code to Python code without having to deal with the GUI
issue. You can already use Python within Windows Script Host (WSH) but even
better you can just avoid using WSH, Python does it better on its own. You
can drive Outlook and write ASP code. Pretty much everything that is in the
"VBScript in a Nutshell" book can be done better with Python.

  http://windows.oreilly.com/

As a bonus, many scripts can be written to work cross-platform which is
something that Windows admins and web developers are starting to think about
more and more.

The first sections of diveintopython make some attempt at reaching the VB
and VBScript user by mentioning the data structures and language elements
they would be familiar with, but it is quite brief.

  http://diveintopython.org/

Some combination of "Python Programming on Win32" and "Learning Python" is
probably the right answer for a "Python for VBA/VBScript Programmers" book.

So, I guess I'm agreeing that there is at least a book or two here with a
potentially huge audience; from hundreds of thousands to the low millions.

ka

> Another idea that I and others have been kicking around for a while is a
> data structures and algorithms book using Python.  Based on what I saw
> at the conference, I think now would be a good time to start working on
> that, possibly beginning it as an Open Source project of some kind.
> --
> Aahz (aahz@pythoncraft.com)           <*>
> http://www.pythoncraft.com/
>
> This is Python.  We don't care much about theory, except where it
> intersects
> with useful practice.  --Aahz, c.l.py, 2/4/2002
> _______________________________________________
> marketing-python mailing list
> marketing-python@wingide.com
> http://pythonology.org/mailman/listinfo/marketing-python
>



From magnus@thinkware.se  Tue Apr 15 20:59:43 2003
From: magnus@thinkware.se (Magnus =?iso-8859-1?Q?Lyck=E5?=)
Date: Tue, 15 Apr 2003 21:59:43 +0200
Subject: [Doc-SIG] Python book ideas
In-Reply-To: <20030415011450.GA25720@panix.com>
Message-ID: <5.2.1.1.0.20030415215112.00bd4d40@www.thinkware.se>

At 21:14 2003-04-14 -0400, Aahz wrote:
>[Cross-posted to both doc-sig and marketing-python]
>
>Last week I went to the Waterside Conference, which is a conference for
>writers and publishers put on by my agent's company.  I picked up some
>good leads for book ideas, some of which I'm holding to myself for a bit
>while I talk with my agent, but I'd like to throw out one idea that's
>been mentioned before as a joke: _Python for VB6 Programmers_

I've recently done some VBA programming, and it struck me
how many similarities there are between Python and VBA. I
feel that Python syntax is VBA done right!

When I thought about this I realized how rare it is that you
see things like Python tutorials for VB programmers or
comparisions between VB and Python sysntax etc. You can find
side by side comparisions between Python and almost every
programming language available as open source out there, but
I've never seen any text (or Python book, and I have more
than a dozen of them) explaining Python from a VB perspective.
Not even Hammond & Robinson does this.

Considering the amount of Visual Basic programmers out there,
I think this is a great book idea. Probably one of the most
needed Python books right now!


--
Magnus Lycka, magnus@thinkware.se
Thinkware AB, www.thinkware.se  



From guido@python.org  Tue Apr 15 21:09:46 2003
From: guido@python.org (Guido van Rossum)
Date: Tue, 15 Apr 2003 16:09:46 -0400
Subject: [Doc-SIG] Python book ideas
In-Reply-To: Your message of "Tue, 15 Apr 2003 21:59:43 +0200."
 <5.2.1.1.0.20030415215112.00bd4d40@www.thinkware.se>
References: <5.2.1.1.0.20030415215112.00bd4d40@www.thinkware.se>
Message-ID: <200304152009.h3FK9ka29775@odiug.zope.com>

> I've recently done some VBA programming, and it struck me
> how many similarities there are between Python and VBA. I
> feel that Python syntax is VBA done right!
> 
> When I thought about this I realized how rare it is that you
> see things like Python tutorials for VB programmers or
> comparisions between VB and Python sysntax etc. You can find
> side by side comparisions between Python and almost every
> programming language available as open source out there, but
> I've never seen any text (or Python book, and I have more
> than a dozen of them) explaining Python from a VB perspective.
> Not even Hammond & Robinson does this.
> 
> Considering the amount of Visual Basic programmers out there,
> I think this is a great book idea. Probably one of the most
> needed Python books right now!

I can get you in touch with a publisher if you need one.

--Guido van Rossum (home page: http://www.python.org/~guido/)


From aahz@pythoncraft.com  Tue Apr 15 21:37:09 2003
From: aahz@pythoncraft.com (Aahz)
Date: Tue, 15 Apr 2003 16:37:09 -0400
Subject: [Doc-SIG] Python book ideas
In-Reply-To: <200304152009.h3FK9ka29775@odiug.zope.com>
References: <5.2.1.1.0.20030415215112.00bd4d40@www.thinkware.se> <200304152009.h3FK9ka29775@odiug.zope.com>
Message-ID: <20030415203707.GB8393@panix.com>

On Tue, Apr 15, 2003, Guido van Rossum wrote:
>Magnus:
>>
>> I've recently done some VBA programming, and it struck me
>> how many similarities there are between Python and VBA. I
>> feel that Python syntax is VBA done right!
>> 
>> When I thought about this I realized how rare it is that you
>> see things like Python tutorials for VB programmers or
>> comparisions between VB and Python sysntax etc. You can find
>> side by side comparisions between Python and almost every
>> programming language available as open source out there, but
>> I've never seen any text (or Python book, and I have more
>> than a dozen of them) explaining Python from a VB perspective.
>> Not even Hammond & Robinson does this.
>> 
>> Considering the amount of Visual Basic programmers out there,
>> I think this is a great book idea. Probably one of the most
>> needed Python books right now!
> 
> I can get you in touch with a publisher if you need one.

Heh.  That was my line, but you're obviously a better person if you've
got the bandwidth.

Magnus, if you want a co-author and nobody else responds to this thread
here, I'd suggest posting to c.l.py and c.l.py.announce.
-- 
Aahz (aahz@pythoncraft.com)           <*>         http://www.pythoncraft.com/

This is Python.  We don't care much about theory, except where it intersects 
with useful practice.  --Aahz, c.l.py, 2/4/2002


From jordanm@uts.cc.utexas.edu  Tue Apr 15 23:04:30 2003
From: jordanm@uts.cc.utexas.edu (Jordan McCoy)
Date: Tue, 15 Apr 2003 17:04:30 -0500
Subject: [Doc-SIG] Greetings
Message-ID: <000101c3039a$fd583850$2301a8c0@telos>

Greetings and good day.

Having just joined the DOC-sig, I wanted to introduce myself and
determine the status of the work the SIG is doing, perhaps see what I
could contribute. I've been working with Python (and programming in
general) for quite a while, but only recently starting to involve myself
with the community.

I'm going to review the archives of the list, but would appreciate a
quick summary of the current work of the SIG and what contributions the
group is currently seeking. Thanks.

Jordan McCoy
jordanm@uts.cc.utexas.edu



From goodger@python.org  Sat Apr 19 04:23:11 2003
From: goodger@python.org (David Goodger)
Date: Fri, 18 Apr 2003 23:23:11 -0400
Subject: [Doc-SIG] Greetings
In-Reply-To: <000101c3039a$fd583850$2301a8c0@telos>
References: <000101c3039a$fd583850$2301a8c0@telos>
Message-ID: <3EA0C11F.4070001@python.org>

Jordan McCoy wrote:
> I'm going to review the archives of the list, but would appreciate a
> quick summary of the current work of the SIG and what contributions the
> group is currently seeking. Thanks.

The SIG has two main focuses: documentation *of* Python, and 
documentation *with* Python.  I'm sure Fred Drake, who's in charge of 
the documentation of Python, would welcome help on that front.  There 
are several projects that use Python for documentation; Docutils 
(http://docutils.sf.net) is one of them, and has been monopolizing the 
Doc-SIG list for a while.  We'd certainly welcome your help.  What kind 
of contributions are you looking to make?

-- 
David Goodger    http://starship.python.net/~goodger

Programmer/sysadmin for hire: http://starship.python.net/~goodger/cv



From aahz@pythoncraft.com  Fri Apr 25 14:58:44 2003
From: aahz@pythoncraft.com (Aahz)
Date: Fri, 25 Apr 2003 09:58:44 -0400
Subject: [Doc-SIG] Re: [marketing-python] 2.3 release - proposed plan
In-Reply-To: <m24r4mel88.fsf@orbtech.com>
References: <Pine.LNX.4.50.0304180933410.3101-100000@hedgehog> <1050684637.12570.49.camel@ida.dylanreinhardt.com> <Pine.LNX.4.50.0304231627230.3347-100000@hedgehog> <1051146156.22265.11.camel@ida.dylanreinhardt.com> <200304240048.h3O0mQR31446@pcp02138704pcs.reston01.va.comcast.net> <1051149264.22265.70.camel@ida.dylanreinhardt.com> <m2he8or7a1.fsf@orbtech.com> <1051155688.22265.120.camel@ida.dylanreinhardt.com> <16039.64072.657758.943971@montanaro.dyndns.org> <m24r4mel88.fsf@orbtech.com>
Message-ID: <20030425135844.GA11633@panix.com>

[cc'ing doc-sig in case anyone is interested]

On Fri, Apr 25, 2003, Patrick K. O'Brien wrote:
>
> When I gave my presentation to the ACM I just created an html page,
> increased the font size in the browser, and scrolled through the
> items.  I've done my share of PowerPoint slides, but I'd rather have a
> way to generate slides using reST.  Kevin Altis said he's toying
> around with something to do this.

If you'd like, you can have my front-end to PythonPoint.  (I created a
dead-simple half-reSTed syntax to avoid the drudgery of writing XML.)
Once Kevin does his work, it should be simple to modify my front-end to
write out reST.
-- 
Aahz (aahz@pythoncraft.com)           <*>         http://www.pythoncraft.com/

"In many ways, it's a dull language, borrowing solid old concepts from
many other languages & styles:  boring syntax, unsurprising semantics,
few automatic coercions, etc etc.  But that's one of the things I like
about it."  --Tim Peters on Python, 16 Sep 93


From pobrien@orbtech.com  Fri Apr 25 15:15:21 2003
From: pobrien@orbtech.com (Patrick K. O'Brien)
Date: 25 Apr 2003 09:15:21 -0500
Subject: [Doc-SIG] Re: [marketing-python] 2.3 release - proposed plan
In-Reply-To: <20030425135844.GA11633@panix.com>
References: <Pine.LNX.4.50.0304180933410.3101-100000@hedgehog>
 <1050684637.12570.49.camel@ida.dylanreinhardt.com>
 <Pine.LNX.4.50.0304231627230.3347-100000@hedgehog>
 <1051146156.22265.11.camel@ida.dylanreinhardt.com>
 <200304240048.h3O0mQR31446@pcp02138704pcs.reston01.va.comcast.net>
 <1051149264.22265.70.camel@ida.dylanreinhardt.com>
 <m2he8or7a1.fsf@orbtech.com>
 <1051155688.22265.120.camel@ida.dylanreinhardt.com>
 <16039.64072.657758.943971@montanaro.dyndns.org>
 <m24r4mel88.fsf@orbtech.com> <20030425135844.GA11633@panix.com>
Message-ID: <m2vfx2d5om.fsf@orbtech.com>

Aahz <aahz@pythoncraft.com> writes:

> [cc'ing doc-sig in case anyone is interested]
> 
> On Fri, Apr 25, 2003, Patrick K. O'Brien wrote:
> >
> > When I gave my presentation to the ACM I just created an html page,
> > increased the font size in the browser, and scrolled through the
> > items.  I've done my share of PowerPoint slides, but I'd rather have a
> > way to generate slides using reST.  Kevin Altis said he's toying
> > around with something to do this.
> 
> If you'd like, you can have my front-end to PythonPoint.  (I created a
> dead-simple half-reSTed syntax to avoid the drudgery of writing XML.)
> Once Kevin does his work, it should be simple to modify my front-end to
> write out reST.

Sure, I'd like to take a look at that.  Thanks.  I also see that
Richard Jones has a pythonpoint writer in his docutils sandbox.  Has
anyone used it?

-- 
Patrick K. O'Brien
Orbtech      http://www.orbtech.com/web/pobrien
-----------------------------------------------
"Your source for Python programming expertise."
-----------------------------------------------



From priest@sfu.ca  Fri Apr 25 15:38:48 2003
From: priest@sfu.ca (David Priest)
Date: Fri, 25 Apr 2003 07:38:48 -0700
Subject: [Doc-SIG] Re: [marketing-python] 2.3 release - proposed plan
In-Reply-To: <20030425135844.GA11633@panix.com>
References: <Pine.LNX.4.50.0304180933410.3101-100000@hedgehog> <1050684637.12570.49.camel@ida.dylanreinhardt.com> <Pine.LNX.4.50.0304231627230.3347-100000@hedgehog> <1051146156.22265.11.camel@ida.dylanreinhardt.com> <200304240048.h3O0mQR31446@pcp02138704pcs.reston01.va.comcast.net> <1051149264.22265.70.camel@ida.dylanreinhardt.com> <m2he8or7a1.fsf@orbtech.com> <1051155688.22265.120.camel@ida.dylanreinhardt.com> <16039.64072.657758.943971@montanaro.dyndns.org> <m24r4mel88.fsf@orbtech.com> <20030425135844.GA11633@panix.com>
Message-ID: <oprn6waysx8v1t4r@mail.telus.net>

On Fri, 25 Apr 2003 09:58:44 -0400, Aahz <aahz@pythoncraft.com> wrote:

> "In many ways, it's a dull language, borrowing solid old concepts
> from many other languages & styles:  boring syntax, unsurprising
> semantics, few automatic coercions, etc etc.  But that's one of
> the things I like about it."  --Tim Peters on Python, 16 Sep 93

This is, I think, a quote that should be used in marketing Python.  It's a 
familiar, powerful, *comfortable* language, with all the power required to 
create full, powerful applications quickly and easily.  Take away the fear 
("it's a dull language") and promote its power.


From aahz@pythoncraft.com  Fri Apr 25 16:24:16 2003
From: aahz@pythoncraft.com (Aahz)
Date: Fri, 25 Apr 2003 11:24:16 -0400
Subject: [Doc-SIG] PythonPoint front-end
In-Reply-To: <m2vfx2d5om.fsf@orbtech.com>
References: <Pine.LNX.4.50.0304231627230.3347-100000@hedgehog> <1051146156.22265.11.camel@ida.dylanreinhardt.com> <200304240048.h3O0mQR31446@pcp02138704pcs.reston01.va.comcast.net> <1051149264.22265.70.camel@ida.dylanreinhardt.com> <m2he8or7a1.fsf@orbtech.com> <1051155688.22265.120.camel@ida.dylanreinhardt.com> <16039.64072.657758.943971@montanaro.dyndns.org> <m24r4mel88.fsf@orbtech.com> <20030425135844.GA11633@panix.com> <m2vfx2d5om.fsf@orbtech.com>
Message-ID: <20030425152415.GC13187@panix.com>

--0OAP2g/MAC+5xKAE
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline

On Fri, Apr 25, 2003, Patrick K. O'Brien wrote:
> Aahz <aahz@pythoncraft.com> writes:
>> 
>> If you'd like, you can have my front-end to PythonPoint.  (I created a
>> dead-simple half-reSTed syntax to avoid the drudgery of writing XML.)
>> Once Kevin does his work, it should be simple to modify my front-end to
>> write out reST.
> 
> Sure, I'd like to take a look at that.  Thanks.  I also see that
> Richard Jones has a pythonpoint writer in his docutils sandbox.  Has
> anyone used it?

Here you go.  I'm including my PyCon objects presentation so you can see
how it works.  (Haven't tried using Richard's bit; I wrote mine a year
ago when reST was really in the Dark Ages. ;-)

(It'll be available on my web site Real Soon Now....)
-- 
Aahz (aahz@pythoncraft.com)           <*>         http://www.pythoncraft.com/

"In many ways, it's a dull language, borrowing solid old concepts from
many other languages & styles:  boring syntax, unsurprising semantics,
few automatic coercions, etc etc.  But that's one of the things I like
about it."  --Tim Peters on Python, 16 Sep 93

--0OAP2g/MAC+5xKAE
Content-Type: text/plain; charset=us-ascii
Content-Disposition: attachment; filename="stx2xml.py"

import sys
import os
import re
import imp
import cgi

#cfg = imp.new_module('cfg')
class cfg: pass

tagDict = {}

class Tag:
    closes = []
    def __init__(self):
        self._attr = {}
        self.elements = []
    def default(self, value):
        pass
    def end(self):
        raise NotImplementedError
    def output(self):
        raise NotImplementedError
    def get_attrs(self):
        tmp = []
        for key,value in self._attr.iteritems():
            tmp.append('%s="%s"' % (key,value))
        return ' '.join(tmp)



class Presentation(Tag):
    def __init__(self):
        Tag.__init__(self)
        tagStack = []

    def pragma_outputFile(self, value):
        self.outputFile = value

    def pragma_style(self, value):
        mod, func = value.split(',')
        self.styleMod, self.styleFunc = mod, func
        f, p, d = imp.find_module(mod)
        mod = imp.load_module(mod, f, p, d)
        func = getattr(mod, func)
        cfg.styles = func()

    def pragma_frame(self, value):
        cfg.pageFrame = tuple(value.split(','))

    def pragma_footer(self, value):
        cfg.footerFrame = tuple(value.split(','))

    def pragma_codeRatio(self, value):
        cfg.codeRatio = float(value)

    def output(self):
        self.setFooters()
        tmp = []
        tmp.append('<presentation %s>' % self.get_attrs())
        mod, func = self.styleMod, self.styleFunc
        tmp.append('<stylesheet module="%s" function="%s"/>' % (mod, func))
        tmp.append('<section name="Main">')
        for slide in self.elements:
            tmp.append(slide.output())
        tmp.append('</section>\n</presentation>')
        return '\n\n'.join(tmp)

    def setFooters(self):
        numPages = len(self.elements)
        pageNo = 0
        for slide in self.elements:
            pageNo += 1
            slide.pageNo = pageNo
            slide.numPages = numPages

tagDict['presentation'] = Presentation



class Para(Tag):
    tagName = 'para'
    styleName = None
    reCode = re.compile("``([^`]+)``")

    def __init__(self):
        Tag.__init__(self)
        self.codeSize = None

    def getCodeSize(self):
        fontSize = cfg.styles[self.styleName].fontSize
        self.codeSize = round(fontSize * cfg.codeRatio, 1)

    def parse(self, line):
        if self.codeSize is None:
            self.getCodeSize()
        repl = r'<font name="Courier" size="%s">\1</font>' % self.codeSize
        return self.reCode.sub(repl, line)

    def default(self, value):
        value = self.parse(value)
        self.elements.append(value)

    def end(self):
        pass

    def output(self):
        outdict = {
            'tagName': self.tagName,
            'style': 'style="%s"' % self.styleName,
            'data': '\n'.join(self.elements)
            }
        tmp = '<%(tagName)s %(style)s>%(data)s</%(tagName)s>' % outdict
        return tmp

Para.closes = [Para]



class PreFmt(Para):
    tagName = 'prefmt'
    styleName = 'Normal'
tagDict['prefmt'] = PreFmt



class Slide(Tag):
    def __init__(self):
        Tag.__init__(self)

    def pragma_title(self, value):
        self._attr['title'] = value
        tmp = Title()
        tmp.default(value)
        self.elements.append(tmp)

    def pragma_head(self, value):
        self._attr['title'] = value.replace("``", "")
        tmp = Head()
        tmp.default(value)
        self.elements.append(tmp)

    def end(self):
        pass

    def output(self):
        tmp = []
        tmp.append('<slide %s>' % self.get_attrs())
        tmp.append(make_frame(cfg.pageFrame))
        for tag in self.elements:
            tmp.append(tag.output())
        tmp.append('</frame>')
        tmp.append(make_frame(cfg.footerFrame))
        pageNo, numPages = self.pageNo, self.numPages
        f = '<para style="Footer">%s</para>' % pageNo
        tmp.append(f)
        tmp.append('</frame>')
        tmp.append('</slide>')
        return '\n'.join(tmp)

Slide.closes = [Para, Slide]
tagDict['slide'] = Slide



class Title(Para):
    styleName = 'title'
tmp = Title
tagDict[tmp.styleName] = tmp



class Head(Para):
    styleName = 'head'
tmp = Head
tagDict[tmp.styleName] = tmp



class SmallTitle(Para):
    styleName = 'smalltitle'
tmp = SmallTitle
tagDict[tmp.styleName] = tmp



class Normal(Para):
    styleName = 'normal'
tmp = Normal
tagDict[tmp.styleName] = tmp



class Bullet(Para):
    styleName = 'Bullet'                # PythonPoint needs cap :-(
tagDict['bullet'] = Bullet



class Indent(Para):
    styleName = 'indent'
tmp = Indent
tagDict[tmp.styleName] = tmp



class URL(Para):
    styleName = 'url'
tmp = URL
tagDict[tmp.styleName] = tmp



class Code(PreFmt):
    styleName = 'code'
    def default(self, value):
    	value = cgi.escape(value)
	PreFmt.default(self, value)
    def pragma_include(self, value):
        data = open(value).read()
        data = cgi.escape(data)
        self.elements.append(data)
tmp = Code
tagDict[tmp.styleName] = tmp



class Spacer(Para):
    def output(self):
        return '<spacer %s/>' % self.get_attrs()
tagDict['spacer'] = Spacer



class NumberedList(Para):
    def default(self, value):
        self.elements.append(value)
    def output(self):
        i = 1
        tmp = []
        for item in self.elements:
            s = '<para style="NumList">%s. %s</para>' % (i, item)
            tmp.append(s)
	    i += 1
        return '\n'.join(tmp)
tagDict['numlist'] = NumberedList



def make_frame(frame):
    return '<frame x="%s" y="%s" width="%s" height="%s">' % frame



def dispatch_tag(tag, root):
    newTag = tagDict[tag]()
    tagStack = root.tagStack
    #print tag, newTag.closes
    for tagType in newTag.closes:
        currTag = tagStack[-1]
        #print isinstance(currTag, tagType)
        if isinstance(currTag, tagType):
            currTag.end()
            tagStack.pop()
    tagStack[-1].elements.append(newTag)
    tagStack.append(newTag)
    #print tagStack



def dispatch_attrib(line, root):
    currTag = root.tagStack[-1]
    key, value = line.split(' ', 1)
    currTag._attr[key] = value



def dispatch_pragma(line, root):
    currTag = root.tagStack[-1]
    pragma, value = line.split(' ', 1)
    meth = getattr(currTag, 'pragma_' + pragma)
    meth(value)

def dispatch_comment(line, root):
    pass



dispatch = {
    '.': dispatch_tag,
    '=': dispatch_attrib,
    '@': dispatch_pragma,
    '#': dispatch_comment,
    }

def parse(input):
    root = Presentation()
    root.tagStack = [root]
    try:
        for line in input:
            line = line.rstrip()
            if not line:
                root.tagStack[-1].default(line)
                continue
            ch = line[0]
            if ch in dispatch:
                # handle continuation lines in interactive mode
                if line[1] == '.':
                    root.tagStack[-1].default(line)
                else:
                    dispatch[ch](line[1:], root)
            else:
                root.tagStack[-1].default(line)
    except KeyError:
        print line
        raise
    return root



def write(root):
    f = open(root.outputFile, 'w')
    f.write(root.output())
    # print root.output()



def process(input):
    input = open(input)
    root = parse(input)
    write(root)



if __name__ == '__main__':
    input = sys.argv[1]
    process(input)

--0OAP2g/MAC+5xKAE
Content-Type: text/plain; charset=us-ascii
Content-Disposition: attachment; filename="pyobj.stx"

=filename pyobj.pdf
@outputFile pyobj.xml
@style style,Styles
@frame 0,0,782,612
@footer 0,5,787,25
@codeRatio 0.85

.slide
@title Python Objects and New-style Classes
.spacer 
=height 40
.smalltitle
<font size="60">Aahz</font>
.spacer 
=height 20
.smalltitle
``aahz@pythoncraft.com``
.smalltitle
``http://pythoncraft.com/``
.spacer 
=height 60
.smalltitle
Powered by PythonPoint
.spacer 
=height 10
.smalltitle
``http://www.reportlab.com/``

.slide
=outlinelevel 1
@head Before we begin...
.bullet
I'm hearing-impaired
.indent
Please speak slowly and clearly

.slide
@head Overall
.bullet
Goal: understand objects
.indent
Use new-style classes to illuminate
.url
http://www.python.org/2.2.2/descrintro.html
.indent
No classic classes
.bullet
Timing
.indent
30 minutes core material
.indent
15 minutes optional material
.indent
# Deal with stupid ReportLab entity bug
15-30 minutes Q&amp;amp;A

.slide
@head Contents
.bullet
Object Basics
.bullet
Type/class tangle
.bullet
Memory management
.bullet
Optional Material
.indent
Mutable/immutable

.slide
@head Object Basics
.bullet
Objects
.bullet
Targets
.bullet
Namespace/Scope

.slide
=outlinelevel 1
@head Objects Defined
.bullet
Object is structured memory
.bullet
Object has:
.indent
Type
.indent
Namespace (attributes)
.indent
Value (optional: int/string objects have value, files don't)

.slide
=outlinelevel 1
@head Everything an Object 
.bullet
All data is object
.indent
Functions, types, classes, class instances, iterators, generators,
modules, and files all first-class objects
.bullet
Capabilities
.indent
Objects all have same potential capabilities, but objects
choose which capabilities to expose.
.indent
The type determines the capabilities.
.indent
("Capability" in its generic sense, essentially 
"set of data/methods" -- not the python-dev thread. ;-)

.slide
=outlinelevel 1
@head Accessing objects
.bullet
Targets contain bindings to objects
.indent
"Binding" used instead of "reference" because you can only deal with
objects directly, never the references themselves (except in C API).

.slide
=outlinelevel 1
@head Kinds of Targets
.bullet
Names
.bullet
Attributes
.indent
Names in object space, accessed with dot notation
.bullet
Computed
.indent
Indexes / Keys
.bullet
Anonymous
.indent
E.g. ``return`` or function defaults
.indent
Important for understanding refcounts

.slide
=outlinelevel 1
@head Names
.bullet
What's a name?
.indent
Bare word at builtin, module global, or function scope (more on scope
shortly)
.bullet
"Name" instead of "variable"
.indent
Names in Python don't contain data, only bindings; they're used like
variables, mostly, but use "names" instead to remember the semantics

.slide
=outlinelevel 1
@head Each Object a Namespace
.bullet
Attributes
.indent
Objects contain dict that maps key/value pairs to names and objects.
For some objects (primarily built-in types, function locals, and some
new-style classes), dict gets mapped to vector for reduced memory
consumption and speed.

.slide
=outlinelevel 1
@head Binding Operations
.bullet
Creating names
.indent
``=``, ``for`` (and list comprehensions), ``def``, 
``import`` (all forms), ``class``, function/method parameters,
``except`` block (second parameter)
.bullet
Other targets
.indent
``=`` (sequences/maps), ``return``, ``yield``, ``print``, 
function/method default parameters

.slide
=outlinelevel 1
@head Namespace Demo
.bullet
``namespace.py``

.slide
=outlinelevel 1
@head Scope
.bullet
Two scope hierarchies:
.indent
Execution scope
.indent
Class inheritance scope
.bullet
Read-only
.indent
Non-local scopes are read-only unless made explicit

.slide
=outlinelevel 1
@head Execution Scope
.bullet
Searches three namespaces in order:
.indent
Function local
.indent
Module global
.indent
Builtins
.bullet
Shadowing
.indent
Locals shadow globals shadow builtins

.slide
=outlinelevel 1
@head Nested Scopes
.bullet
Functions inside functions
.bullet
Lexical scope
.indent
Not dynamic scope
.indent
Allows static analysis of source code
.bullet
Python 2.1
.indent
``from __future__ import nested_scopes``

.slide
=outlinelevel 1
@head Class Inheritance Scope
.bullet
Non-standard "scope" usage
.indent
Some people disagree with me that class attribute/method inheritance
should be discussed as a "scope", but it's a convenient way to explain
the similarities with execution scope
.bullet
``self``
.indent
Used in methods to distinguish between execution scope (implicit as
in other functions) and the class inheritance scope (which is made
explicit through ``self``)

.slide
=outlinelevel 1
@head Scope Demo
.bullet
``scope.py``

.slide
@head Type/Class Tangle
.bullet
Inheritance vs. creation
.bullet
Class vs. instance

.slide
=outlinelevel 1
@head Inheritance vs. creation
.bullet 
Inheritance
.indent
Attribute search path
.bullet
Type
.indent
Types are objects that create other objects
.indent
Metaclass is type that generates classes; class is type that
generates class instances
.indent
``type(Foo) == Foo.__class__``
.bullet
Instance
.indent
Instance is an object created by a type

.slide
=outlinelevel 1
@head Class vs. Instance
.bullet
Instance attribute
.indent
Access with ``self``
.bullet
Class attribute
.indent
Access with class name
.indent
But remember that a class is also an instance...

.slide
=outlinelevel 1
@head More Inheritance
.bullet
Bases vs. Types
.indent
An instance's inheritance searches its type, then its type's base
classes.  Inheritance does NOT search the type's type.

.slide
=outlinelevel 1
@head Type/Class Demo
.bullet
``typeclass.py``

.slide
@head Memory Management
.bullet
Reference counts
.bullet
Garbage collection
.bullet
``__slots__``

.slide
=outlinelevel 1
@head Reference Counts
.bullet
Object deletion
.indent
Objects are deleted when no bindings reference object
.bullet
Binding
.indent
Each binding increases refcount
.bullet
Decrease refcount
.indent
Rebinding
.indent
Name going out of scope
.indent
``del``

.slide
=outlinelevel 1
@head ``del``
.bullet
Deletes bindings, not objects
.bullet
Can ``del`` parts of objects:
.spacer
=height 10
.code
>>> L = [1,2,3,4,5]
>>> del L[1:3]
>>> L
[1, 4, 5]
>>> d = {'a': 1, 'b': 2, 'c': 5}
>>> del d['b']
>>> d
{'a': 1, 'c': 5}

.slide
=outlinelevel 1
@head Refcount Cascades
.bullet
Objects refer to other objects
.indent
When an object gets deleted, all objects it refers to get decremented and
may be deleted -- leading to a cascade of object deletions
.bullet
Large lists/dicts
.indent
Deleting a large list or dict can take a long time as each object in the
list/dict needs to be walked

.slide
=outlinelevel 1
@head GC
.bullet
Garbage collection
.indent
In addition to refcount, not instead of
.bullet
Cycles
.indent
Occur when two or more objects refer to each other; refcount
can't go to zero.  If no external targets refer to cycle,
cycle is "garbage" and GC deletes it.
.bullet
``__del__()``
.indent
Objects with ``__del__()`` can't be GC'd because of ordering problem

.slide
=outlinelevel 1
@head Refcount/GC Demo
.bullet
``refcount_gc.py``

.slide
=outlinelevel 1
@head ``__slots__``
.bullet
Attributes
.indent
Normally stored in dict
.indent
``__slots__`` creates indexed vector
.indent
Attribute names stored with class, not instance

.slide
@head Mutable vs. Immutable
.bullet
Immutable objects
.indent
Strings, numbers, tuples, some class instances, and combinations
.bullet
Mutable objects
.indent
Lists, dicts, most classes and class instances
.bullet
Why immutable?
.indent
Dict keys, space efficiency

.slide
=outlinelevel 1
@head Mutable Subtleties
.bullet
Lists embedded in tuples
.spacer
=height 10
.code
>>> a = (1, ['foo'], 's')
>>> a[1] = {'foo': 'bar'}
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
TypeError: object doesn't support item assignment
>>> a[1].append('bar')
>>> a
(1, ['foo', 'bar'], 's')
.spacer
=height 5
.bullet
Classes
.indent
Cannot force class immutable; it's an implied contract in code and
documentation (must use C API to force immutable)

.slide
=outlinelevel 1
@head Copy Semantics (for function calls)
.bullet
Immutables don't need copy
.bullet
Mutables must be copied
.indent
If you don't want them modified, that is

.slide
=outlinelevel 1
@head Copy Example
.spacer
=height 15
.normal
``list.sort()`` is an in-place operation; here we create a function that
returns the list
.spacer
=height 10
.code
>>> def sort(L):
...     L.sort()
...     return L
...
>>> L = [11, 7, 12, 5]
>>> sort(L[:])
[5, 7, 11, 12]
>>> L
[11, 7, 12, 5]
>>> sort(L)
[5, 7, 11, 12]
>>> L
[5, 7, 11, 12]
.spacer
=height 15
.normal
Notice how failing to make copy modifies original list

.slide
=outlinelevel 1
@head Mutability Again
.bullet
``=`` vs. ``+=`` (and family)
.indent
``=`` always [re]binds
.indent
mutable vs. immutable
.indent
``+=`` may rebind ``self`` to original object if mutable (but not
required)

.slide
=outlinelevel 1
@head Tuples and ``+=``
.spacer
=height 10
.code
>>> a = (1, ['foo'], 'xyzzy')
>>> a[1].append('bar')
>>> a
(1, ['foo', 'bar'], 'xyzzy')
>>> a[1] = 9
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
TypeError: object doesn't support item assignment
>>> a
(1, ['foo', 'bar'], 'xyzzy')
>>> a[1] += ['spam']
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
TypeError: object doesn't support item assignment
>>> a
(1, ['foo', 'bar', 'spam'], 'xyzzy')

.slide
=outlinelevel 1
@head ``+=`` expansion
.spacer
=height 10
.code
a.__setitem__(1, 
    a.__getitem__(1).__iadd__(['spam']))
.spacer
=height 15
.normal
Rewrite with temp vars:
.spacer
=height 10
.code
>>> a = (1, ['foo', 'bar'], 'xyzzy')
>>> x = a.__getitem__(1)
>>> x
['foo', 'bar']
>>> x.__iadd__(['spam'])
['foo', 'bar', 'spam']
>>> a.__setitem__(1, x)
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
AttributeError: 'tuple' object has 
    no attribute '__setitem__'
>>> a
(1, ['foo', 'bar', 'spam'], 'xyzzy')


--0OAP2g/MAC+5xKAE--


From Nicolas.Chauvat@logilab.fr  Fri Apr 25 20:20:29 2003
From: Nicolas.Chauvat@logilab.fr (Nicolas Chauvat)
Date: Fri, 25 Apr 2003 21:20:29 +0200
Subject: [Doc-SIG] Re: [marketing-python] 2.3 release - proposed plan
In-Reply-To: <20030425135844.GA11633@panix.com>
References: <1050684637.12570.49.camel@ida.dylanreinhardt.com> <Pine.LNX.4.50.0304231627230.3347-100000@hedgehog> <1051146156.22265.11.camel@ida.dylanreinhardt.com> <200304240048.h3O0mQR31446@pcp02138704pcs.reston01.va.comcast.net> <1051149264.22265.70.camel@ida.dylanreinhardt.com> <m2he8or7a1.fsf@orbtech.com> <1051155688.22265.120.camel@ida.dylanreinhardt.com> <16039.64072.657758.943971@montanaro.dyndns.org> <m24r4mel88.fsf@orbtech.com> <20030425135844.GA11633@panix.com>
Message-ID: <20030425192029.GH21581@logilab.fr>

On Fri, Apr 25, 2003 at 09:58:44AM -0400, Aahz wrote:
> [cc'ing doc-sig in case anyone is interested]
> 
> On Fri, Apr 25, 2003, Patrick K. O'Brien wrote:
> >
> > When I gave my presentation to the ACM I just created an html page,
> > increased the font size in the browser, and scrolled through the
> > items.  I've done my share of PowerPoint slides, but I'd rather have a
> > way to generate slides using reST.  Kevin Altis said he's toying
> > around with something to do this.
> 
> If you'd like, you can have my front-end to PythonPoint.  (I created a
> dead-simple half-reSTed syntax to avoid the drudgery of writing XML.)
> Once Kevin does his work, it should be simple to modify my front-end to
> write out reST.

I want it! I almost wrote one six month ago and stopped halfway due to lack of time.

-- 
Nicolas Chauvat

http://www.logilab.com - "Mais oł est donc Ornicar ?" - LOGILAB, Paris (France)


From aahz@pythoncraft.com  Sat Apr 26 15:05:55 2003
From: aahz@pythoncraft.com (Aahz)
Date: Sat, 26 Apr 2003 10:05:55 -0400
Subject: [Doc-SIG] Re: PythonPoint front-end
In-Reply-To: <KJEOLDOPMIDKCMJDCNDPMEIFDFAA.altis@semi-retired.com>
References: <20030425152415.GC13187@panix.com> <KJEOLDOPMIDKCMJDCNDPMEIFDFAA.altis@semi-retired.com>
Message-ID: <20030426140554.GA21156@panix.com>

--AhhlLboLdkugWU4S
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline

On Fri, Apr 25, 2003, Kevin Altis wrote:
>
> Hey Aahz,
> I couldn't get the script to run.
> 
> ImportError: No module named style

Oops.  I've only run this as part of the PythonPoint toolchain, so I
forgot about that.  I've long since had to run PythonPoint only with a
style file, so my stx2xml partly relies on that.

I'm attaching the style.py that I used.

> What are the module/package requirements? Do you have a URL to the PDF
> output so I can just look at that?

Real Soon Now.  ;-)  (It's part of my full PyCon package.)

> The thing I was thinking of doing would be pretty darn simple, so it would
> be interesting to get your requirements for the presentation tool/output
> since I haven't done many presentations. I'm sort of leaning towards wiki
> pages with some special CSS style sheets.

Well, I don't like web pages much, so that doesn't work for me.  I chose
this route because I needed the fine-grained control.  (Just like
PowerPoint.)
-- 
Aahz (aahz@pythoncraft.com)           <*>         http://www.pythoncraft.com/

"In many ways, it's a dull language, borrowing solid old concepts from
many other languages & styles:  boring syntax, unsurprising semantics,
few automatic coercions, etc etc.  But that's one of the things I like
about it."  --Tim Peters on Python, 16 Sep 93

--AhhlLboLdkugWU4S
Content-Type: text/plain; charset=us-ascii
Content-Disposition: attachment; filename="style.py"

#from reportlab.platypus import TableStyle
from reportlab.platypus.tables import TableStyle
from reportlab.lib import styles
from reportlab.lib.enums import TA_LEFT, TA_CENTER, TA_RIGHT, TA_JUSTIFY
from reportlab.lib import colors

def Styles():
    stylesheet = {}
    ParagraphStyle = styles.ParagraphStyle

    para = ParagraphStyle('Normal', None)   #the ancestor of all
    para.fontName = 'Times-Roman'
    para.fontSize = 30
    para.leading = para.fontSize * 1.2
    stylesheet['Normal'] = para
    stylesheet['normal'] = para

    #This one is spaced out a bit...
    para = ParagraphStyle('BodyText', stylesheet['Normal'])
    para.spaceBefore = 12
    stylesheet['BodyText'] = para

    para = ParagraphStyle('Centered', stylesheet['Normal'])
    para.alignment = TA_CENTER
    stylesheet['Centered'] = para

    para = ParagraphStyle('BigCentered', stylesheet['Normal'])
    para.spaceBefore = 12
    para.alignment = TA_CENTER
    stylesheet['BigCentered'] = para

    para = ParagraphStyle('Italic', stylesheet['BodyText'])
    para.fontName = 'Times-Italic'
    stylesheet['Italic'] = para

    para = ParagraphStyle('Footer', stylesheet['Normal'])
    para.fontSize = 18
    para.leading = 0
    para.alignment = TA_RIGHT
    stylesheet['Footer'] = para

    para = ParagraphStyle('Title', stylesheet['Normal'])
    para.fontName = 'Times-Roman'
    para.fontSize = para.fontSize * 3.0
    para.leading = para.fontSize * 1.0
    para.spaceAfter = para.fontSize * 0.2
    para.alignment = TA_CENTER
    stylesheet['Title'] = para
    stylesheet['title'] = para

    para = ParagraphStyle('SmallTitle', stylesheet['Normal'])
    para.fontName = 'Times-Roman'
    para.fontSize = stylesheet['Title'].fontSize * 0.5
    para.leading = para.fontSize * 1.0
    para.alignment = TA_CENTER
    stylesheet['SmallTitle'] = para
    stylesheet['smalltitle'] = para

    para = ParagraphStyle('Heading1', stylesheet['Normal'])
    para.fontName = 'Times-Bold'
    para.fontSize = para.fontSize * 2.25
    para.leading = para.fontSize * 1.1
    para.spaceAfter = para.fontSize * 0.15
    para.alignment = TA_CENTER
    stylesheet['Heading1'] = para
    stylesheet['head'] = para

    para = ParagraphStyle('Heading2', stylesheet['Normal'])
    para.fontName = 'Times-Bold'
    para.fontSize = para.fontSize * 1.6
    para.leading = para.fontSize * 1.2
    para.spaceBefore = para.fontSize * 0.8
    stylesheet['Heading2'] = para

    para = ParagraphStyle('Heading3', stylesheet['Normal'])
    para.fontName = 'Times-BoldItalic'
    para.spaceBefore = 24
    stylesheet['Heading3'] = para

    para = ParagraphStyle('Heading4', stylesheet['Normal'])
    para.fontName = 'Times-BoldItalic'
    para.spaceBefore = 6
    stylesheet['Heading4'] = para

    para = ParagraphStyle('Bullet', stylesheet['Normal'])
    para.firstLineIndent = 10
    para.leftIndent = 16
    para.fontSize = para.fontSize * 1.5
    para.leading = para.fontSize * 1.1
    para.spaceBefore = para.fontSize * 0.4
    para.bulletFontName = 'Symbol'
    para.bulletFontSize = para.fontSize * 0.7
    para.bulletIndent = 20
    stylesheet['Bullet'] = para
    stylesheet['bullet'] = para         # doesn't work

    #Indented, for lists
    para = ParagraphStyle('Indent', stylesheet['Normal'])
    para.fontSize = stylesheet['Bullet'].fontSize * 0.75
    para.leftIndent = para.fontSize * 2.5
    para.leading = para.fontSize * 1.1
    para.spaceBefore = para.fontSize * 0.2
    stylesheet['Indent'] = para
    stylesheet['indent'] = para

    para = ParagraphStyle('Definition', stylesheet['Normal'])
    #use this for definition lists
    para.firstLineIndent = 72
    para.leftIndent = 72
    para.bulletIndent = 0
    para.spaceBefore = 12
    para.bulletFontName = 'Helvetica-BoldOblique'
    para.bulletFontSize = 24
    stylesheet['Definition'] = para

    para = ParagraphStyle('Code', stylesheet['Normal'])
    para.fontName = 'Courier'
    #para.fontSize = para.fontSize * 0.75
    para.fontSize = para.fontSize * 0.8
    para.leading = para.fontSize + 1
    para.leftIndent = 36
    stylesheet['Code'] = para
    stylesheet['code'] = para

    para = ParagraphStyle('CodeHeader', stylesheet['Normal'])
    para.fontSize = stylesheet['Code'].fontSize * 1.5
    para.spaceAfter = para.fontSize * 0.2
    para.leftIndent = 36
    para.firstLineIndent = para.leftIndent
    stylesheet['CodeHeader'] = para

    para = ParagraphStyle('NumList', stylesheet['Normal'])
    para.fontSize = 24
    para.leading = para.fontSize * 1.1
    stylesheet['NumList'] = para

    para = ParagraphStyle('URL', stylesheet['Normal'])
    para.fontName = 'Courier'
    para.fontSize = 24
    para.leading = para.fontSize + 2
    para.leftIndent = 10
    stylesheet['url'] = para

    para = ParagraphStyle('Small', stylesheet['Normal'])
    para.fontSize = 12
    para.leading = 14
    stylesheet['Small'] = para

    #now for a table
    ts = TableStyle([
         ('FONT', (0,0), (-1,-1), 'Times-Roman', 24),
         ('LINEABOVE', (0,0), (-1,0), 2, colors.green),
         ('LINEABOVE', (0,1), (-1,-1), 0.25, colors.black),
         ('LINEBELOW', (0,-1), (-1,-1), 2, colors.green),
         ('LINEBEFORE', (-1,0), (-1,-1), 2, colors.black),
         ('ALIGN', (1,1), (-1,-1), 'RIGHT'),   #all numeric cells right aligned
         ('TEXTCOLOR', (0,1), (0,-1), colors.red),
         ('BACKGROUND', (0,0), (-1,0), colors.Color(0,0.7,0.7))
         ])
    stylesheet['table1'] = ts

    para = stylesheet['Code']
    ts = TableStyle([
        ('FONT', (0,0), (-1,0), 'Times-Roman', para.fontSize*1.5 ),
        ('FONT', (0,1), (-1,-1), 'Courier', para.fontSize),
        ])
    stylesheet['2code'] = ts


    return stylesheet

--AhhlLboLdkugWU4S--