From cfoster at oda.state.or.us  Wed Nov  5 12:45:11 2008
From: cfoster at oda.state.or.us (Chris Foster)
Date: Wed, 5 Nov 2008 09:45:11 -0800
Subject: [reportlab-users] even number of pages
Message-ID: <ACACA682-4C5D-49D8-BCB1-90591A347BD6@oda.state.or.us>

I have variable length, Platypus reports that will be printed double- 
sided.  Is there an easy way to be sure I always have an even number  
of pages per report?

From timr at probo.com  Wed Nov  5 13:19:07 2008
From: timr at probo.com (Tim Roberts)
Date: Wed, 05 Nov 2008 10:19:07 -0800
Subject: [reportlab-users] even number of pages
In-Reply-To: <ACACA682-4C5D-49D8-BCB1-90591A347BD6@oda.state.or.us>
References: <ACACA682-4C5D-49D8-BCB1-90591A347BD6@oda.state.or.us>
Message-ID: <4911E39B.9010302@probo.com>

Chris Foster wrote:
> I have variable length, Platypus reports that will be printed
> double-sided.  Is there an easy way to be sure I always have an even
> number of pages per report?

Are you going to be pumping out several reports in a row as a single PDF
file?  If not, your printer driver will handle this for you.

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From cfoster at oda.state.or.us  Wed Nov  5 13:24:49 2008
From: cfoster at oda.state.or.us (Chris Foster)
Date: Wed, 5 Nov 2008 10:24:49 -0800
Subject: [reportlab-users] even number of pages (nevermind)
Message-ID: <379CE716-35E1-40C8-A39D-E47949EA656D@oda.state.or.us>

Sorry, just a brief moment of panic (two days until production).

I forgot I'm also merging variable length letters with pyPDF in post  
processing and the problem is easily handled there.  Obviously, it is  
not a ReportLab issue at all.

From andy at reportlab.com  Thu Nov  6 02:28:08 2008
From: andy at reportlab.com (Andy Robinson)
Date: Thu, 6 Nov 2008 07:28:08 +0000
Subject: [reportlab-users] even number of pages (nevermind)
In-Reply-To: <379CE716-35E1-40C8-A39D-E47949EA656D@oda.state.or.us>
References: <379CE716-35E1-40C8-A39D-E47949EA656D@oda.state.or.us>
Message-ID: <956003ae0811052328q72535cb2lb8b719768efd837e@mail.gmail.com>

2008/11/5 Chris Foster <cfoster at oda.state.or.us>:
> Sorry, just a brief moment of panic (two days until production).
>
> I forgot I'm also merging variable length letters with pyPDF in post
> processing and the problem is easily handled there.  Obviously, it is not a
> ReportLab issue at all.

Funnily enough we have added some (as yet undocumented) flowables to
help this kind of thing - in our case making sure certain things
happen in the right places on 4-page spreads.  These new flowables can
include content or not depending on the current page, frame or
position.  If curious, look at the bottom of
reportlab/platypus/flowables.py on trunk.

But PyPDF is definitely the easiest way if you just need to
conditionally add a blank page.

- Andy

From zhangbinjr at gmail.com  Fri Nov  7 02:13:12 2008
From: zhangbinjr at gmail.com (zhang bin)
Date: Fri, 7 Nov 2008 15:13:12 +0800
Subject: [reportlab-users] how to make text-image searchable
Message-ID: <932c22540811062313l342e33c5h1c150ee6ec4cad4e@mail.gmail.com>

hello,guys,
       Have you ever noticed that some old pdf files's character glyphs are
not as clear as the standard font.Besides,you can search some characters by
the adobe reader. But some can not.
I thought this type of pdf is just an image from the scanner drawing on the
pdf canvas.The producer can not ensure  the high accuracy of the text OCR.So
they just provide extract text information.In this way ,the pdf file's text
is right when reading,but reduce the search text accuracy.
       My question is  how can they relate the image's information with the
text information,which makes adobe reader can search the text. can reportlab
do this job? I have tried reportlab simply paint the text image to pdf which
can not search the characters.And I have a "stupid" solution, paint the text
image to the canvas,the show the character in a transparent way in the exact
position as the text image.But any better ways? Thanks a lot.
-- 
best regards,
zhangbin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/reportlab-users/attachments/20081107/5af65714/attachment.html>

From andy at reportlab.com  Fri Nov  7 02:33:44 2008
From: andy at reportlab.com (Andy Robinson)
Date: Fri, 7 Nov 2008 07:33:44 +0000
Subject: [reportlab-users] how to make text-image searchable
In-Reply-To: <932c22540811062313l342e33c5h1c150ee6ec4cad4e@mail.gmail.com>
References: <932c22540811062313l342e33c5h1c150ee6ec4cad4e@mail.gmail.com>
Message-ID: <956003ae0811062333v4abd3c92uad1c34d6f59bee9d@mail.gmail.com>

2008/11/7 zhang bin <zhangbinjr at gmail.com>:
> hello,guys,
>        My question is  how can they relate the image's information with the
> text information,which makes adobe reader can search the text. can reportlab
> do this job?

No, it can't.  I think there was a recent extension to PDF to allow
'tagged text' - basically a separate hidden text stream withing the
file somewhere - which is also used to reflow text if you are reading
on a PDA or some other device with a small screen.  But we have
not implemented this and not even studied it....

Sorry,


-- 
Andy Robinson
CEO/Chief Architect
ReportLab Europe Ltd.
Media House, 3 Palmerston Road, Wimbledon, London SW19 1PG, UK
Tel +44-20-8545-1570

From darren at ontrenet.com  Fri Nov  7 09:39:13 2008
From: darren at ontrenet.com (Darren Govoni)
Date: Fri, 07 Nov 2008 09:39:13 -0500
Subject: [reportlab-users] Embed comments in PDF?
Message-ID: <1226068754.6947.4.camel@nuraku>

Hi,
  Is it possible to read in an existing PDF, write comments into yet (so
they appear as comments to Adobe Reader) and write it back out with
reportlab? Maybe the "comments" are just some metadata, but I couldn't
see in the API how this might be done.

thanks.

Darren

PS. I also tried to contact the PyPDF author, but no response after 3
tries.


From andy at reportlab.com  Fri Nov  7 09:45:01 2008
From: andy at reportlab.com (Andy Robinson)
Date: Fri, 7 Nov 2008 14:45:01 +0000
Subject: [reportlab-users] Embed comments in PDF?
In-Reply-To: <1226068754.6947.4.camel@nuraku>
References: <1226068754.6947.4.camel@nuraku>
Message-ID: <956003ae0811070645m4a1fe6dat9354256b7581b938@mail.gmail.com>

2008/11/7 Darren Govoni <darren at ontrenet.com>:
> Hi,
>  Is it possible to read in an existing PDF, write comments into yet (so
> they appear as comments to Adobe Reader) and write it back out with
> reportlab? Maybe the "comments" are just some metadata, but I couldn't
> see in the API how this might be done.

No it isn't.  Our commercial package doesn't do this either.  Sorry!

> PS. I also tried to contact the PyPDF author, but no response after 3
> tries.

In the open source world people do tend to move onto new projects,
sadly.  Except for us - we mostly remain stationary on the same old
project  ;-)


-- 
Andy Robinson
CEO/Chief Architect
ReportLab Europe Ltd.
Media House, 3 Palmerston Road, Wimbledon, London SW19 1PG, UK
Tel +44-20-8545-1570

From darren at ontrenet.com  Fri Nov  7 09:53:05 2008
From: darren at ontrenet.com (Darren Govoni)
Date: Fri, 07 Nov 2008 09:53:05 -0500
Subject: [reportlab-users] Embed comments in PDF?
In-Reply-To: <956003ae0811070645m4a1fe6dat9354256b7581b938@mail.gmail.com>
References: <1226068754.6947.4.camel@nuraku>
	<956003ae0811070645m4a1fe6dat9354256b7581b938@mail.gmail.com>
Message-ID: <1226069585.6947.8.camel@nuraku>

Hi Andy,
   Well darn. Hehe. Since I'm a one-track-mind I will maybe have to
write my own mod to do this and give it back to the community. 

Now to hunt down the PDF file format spec. Oh joy!

Darren


On Fri, 2008-11-07 at 14:45 +0000, Andy Robinson wrote:
> 2008/11/7 Darren Govoni <darren at ontrenet.com>:
> > Hi,
> >  Is it possible to read in an existing PDF, write comments into yet (so
> > they appear as comments to Adobe Reader) and write it back out with
> > reportlab? Maybe the "comments" are just some metadata, but I couldn't
> > see in the API how this might be done.
> 
> No it isn't.  Our commercial package doesn't do this either.  Sorry!
> 
> > PS. I also tried to contact the PyPDF author, but no response after 3
> > tries.
> 
> In the open source world people do tend to move onto new projects,
> sadly.  Except for us - we mostly remain stationary on the same old
> project  ;-)
> 
> 


From yroman-reportlab at altalang.com  Fri Nov  7 09:57:58 2008
From: yroman-reportlab at altalang.com (Yoann Roman)
Date: Fri, 7 Nov 2008 09:57:58 -0500
Subject: [reportlab-users] Embed comments in PDF?
In-Reply-To: <1226069585.6947.8.camel@nuraku>
References: <1226068754.6947.4.camel@nuraku><956003ae0811070645m4a1fe6dat9354256b7581b938@mail.gmail.com>
	<1226069585.6947.8.camel@nuraku>
Message-ID: <002501c940e9$38d26540$0502000a@ad.altalang.com>

> Hi Andy,
>    Well darn. Hehe. Since I'm a one-track-mind I will maybe have to
> write my own mod to do this and give it back to the community. 
> 
> Now to hunt down the PDF file format spec. Oh joy!

I believe ReportLab is based on 1.3:
http://www.adobe.com/devnet/pdf/pdfs/PDFReference13.pdf

Not sure if Acrobat comments were part of that version or not, though.

-- 
Yoann Roman


From darren at ontrenet.com  Fri Nov  7 10:23:31 2008
From: darren at ontrenet.com (Darren Govoni)
Date: Fri, 07 Nov 2008 10:23:31 -0500
Subject: [reportlab-users] Embed comments in PDF?
In-Reply-To: <002501c940e9$38d26540$0502000a@ad.altalang.com>
References: <1226068754.6947.4.camel@nuraku>
	<956003ae0811070645m4a1fe6dat9354256b7581b938@mail.gmail.com>
	<1226069585.6947.8.camel@nuraku>
	<002501c940e9$38d26540$0502000a@ad.altalang.com>
Message-ID: <1226071411.6947.9.camel@nuraku>

It has metadata API to PDF, so perhaps comments are structured as
metadata fields (I'm a PDF format noob). He released a version in
September. I will have to dig into it deeper to find out I think.

On Fri, 2008-11-07 at 09:57 -0500, Yoann Roman wrote:
> > Hi Andy,
> >    Well darn. Hehe. Since I'm a one-track-mind I will maybe have to
> > write my own mod to do this and give it back to the community. 
> > 
> > Now to hunt down the PDF file format spec. Oh joy!
> 
> I believe ReportLab is based on 1.3:
> http://www.adobe.com/devnet/pdf/pdfs/PDFReference13.pdf
> 
> Not sure if Acrobat comments were part of that version or not, though.
> 


From andy at reportlab.com  Fri Nov  7 10:53:39 2008
From: andy at reportlab.com (Andy Robinson)
Date: Fri, 7 Nov 2008 15:53:39 +0000
Subject: [reportlab-users] Embed comments in PDF?
In-Reply-To: <1226069585.6947.8.camel@nuraku>
References: <1226068754.6947.4.camel@nuraku>
	<956003ae0811070645m4a1fe6dat9354256b7581b938@mail.gmail.com>
	<1226069585.6947.8.camel@nuraku>
Message-ID: <956003ae0811070753s1858cb9ch63e01d596cc25aa6@mail.gmail.com>

2008/11/7 Darren Govoni <darren at ontrenet.com>:
> Hi Andy,
>   Well darn. Hehe. Since I'm a one-track-mind I will maybe have to
> write my own mod to do this and give it back to the community.
> Now to hunt down the PDF file format spec. Oh joy!

Hunting it down isn't the problem  - bottom link here, ignore the
exhortations to pay ISO for a copy....

    http://www.adobe.com/devnet/pdf/pdf_reference.html

You begin to taste the 'joy' when you've spent a night or two reading it,
but the true sensual pleasure only really kicks in when you
try to implement some feature and need a few dozen tries...

Py2PDf is probably a better starting point since it is inherently
read/write - our own library is designed to be read-only.  It's
also worth looking at Bruno Lowagie's iText library in Java
in case he's solved it...

But we'd be very happy to accept any patches if you get something
going.

-- 
Andy

From henry at itaapy.com  Fri Nov  7 13:29:27 2008
From: henry at itaapy.com (Obein Henry)
Date: Fri, 07 Nov 2008 19:29:27 +0100
Subject: [reportlab-users] Indenter inside a table cell
Message-ID: <49148907.7020907@itaapy.com>

Hello,

I try to add an Indenter flowable inside a table cell but
the table object does not handle the FrameActionFlowable object.
I mix Indenter and Paragraph flowable to simulate a list of items
like ul/ol tag in HTML.
It works inside a frame but not inside a table.
Is there another way (maybe easier) to do this ?

Regards,
Henry

From gabor at nekomancer.net  Mon Nov 10 05:39:33 2008
From: gabor at nekomancer.net (=?ISO-8859-1?Q?G=E1bor_Farkas?=)
Date: Mon, 10 Nov 2008 11:39:33 +0100
Subject: [reportlab-users] ttf "fallback".. is it supported somehow?
Message-ID: <2c173a350811100239k4ebed996v84cc99bc92638fde@mail.gmail.com>

hi,

is it somehow possible, to specify multiple ttf files for a paragraph,
and reportlab would use by default the first ttf, and if some glyph is
not found in it,
then it would try to find those glyphs in the second font etc? or some
other way?

my problem is that i do not have one single TTF font which covers all
the characters i need to
enter into the PDF file. but i have a collection of TTF files which
could together cover it.

basically i'd like to do in a PDF what  web-browsers do when
displaying a html-page.

thanks,
gabor

From robin at reportlab.com  Mon Nov 10 06:03:39 2008
From: robin at reportlab.com (Robin Becker)
Date: Mon, 10 Nov 2008 11:03:39 +0000
Subject: [reportlab-users] ttf "fallback".. is it supported somehow?
In-Reply-To: <2c173a350811100239k4ebed996v84cc99bc92638fde@mail.gmail.com>
References: <2c173a350811100239k4ebed996v84cc99bc92638fde@mail.gmail.com>
Message-ID: <4918150B.3000408@chamonix.reportlab.co.uk>

G?bor Farkas wrote:
> hi,
> 
> is it somehow possible, to specify multiple ttf files for a paragraph,
> and reportlab would use by default the first ttf, and if some glyph is
> not found in it,
> then it would try to find those glyphs in the second font etc? or some
> other way?
> 
> my problem is that i do not have one single TTF font which covers all
> the characters i need to
> enter into the PDF file. but i have a collection of TTF files which
> could together cover it.
> 
> basically i'd like to do in a PDF what  web-browsers do when
> displaying a html-page.
> 
> thanks,
> gabor

art present we only have fallbacks for the T1 fonts. The truetype fonts have a 
very different layout model which involves subsetting etc etc. I suppose we 
could add an extra layer above that and create a multi-ttf font type which would 
do as you suggest, but not right now.
-- 
Robin Becker

From gherman at darwin.in-berlin.de  Mon Nov 10 06:20:40 2008
From: gherman at darwin.in-berlin.de (Dinu Gherman)
Date: Mon, 10 Nov 2008 12:20:40 +0100
Subject: [reportlab-users] ANN: svglib 0.6.1
Message-ID: <008E05C8-0AD0-49C0-B3BF-6B6F4B024D04@darwin.in-berlin.de>

Hi,

I'm glad to release svglib 0.6.1, a library for reading
SVG and converting it to ReportLab Graphics objects:

   http://pypi.python.org/pypi/svglib

This release fixes a few buglets and uses a new set of 199
SVG flags accessed online from Wikipedia.org [1]. These
make great examples because many of them are very simple,
while a few others are quite complex.

When you run the testsuite these SVG samples will be down-
loaded and converted to PDF. As you will see there are
some buglets for a few of these flags. If this should be
the case for the country you live in you're invited to
help debug the code to fix the respective flag!

Also, I've added conversion tests using a tool named uni-
converter (if present) which has generated recently some
enthusiasm on this list. It's interesting to compare it
to svglib by running the respective tests (after uncommen-
ting them). I had to uncomment these because they take
very long and generally produce more errors and inferior
results, compared to svglib, but YMMV.

Regards,

Dinu

[1] http://en.wikipedia.org/wiki/Gallery_of_sovereign_state_flags


From chris.kratz at vistashare.com  Wed Nov 12 09:09:06 2008
From: chris.kratz at vistashare.com (Chris Kratz)
Date: Wed, 12 Nov 2008 09:09:06 -0500
Subject: [reportlab-users] Dual pass numbered canvas causes images to
	disappear
Message-ID: <3642025c0811120609y49b77afeg9ed3e142364f9a4e@mail.gmail.com>

Hello all,

When generating a pdf, if I use the normal canvas, my image shows up just
fine in the generated pdf.
doc.build(story, onFirstPage=self.firstPage, onLaterPages=self.laterPages)

But, I need the ability to print page x of y in the footers.  So, I used a
slightly modified version of the NumberedCanvas class found here:
http://code.activestate.com/recipes/546511/

So, when I call doc.build with the NumberedCanvas, I get the Page x of y in
the footer, but images just seem to not show.  No errors or any other
indications of problems, just no images.  The document size seems to
indicate the image is included in the pdf, even if it isn't displaying.

doc.build(story, onFirstPage=self.firstPage,
onLaterPages=self.laterPages, canvasmaker=NumberedCanvas)

class NumberedCanvas(canvas.Canvas):

    def __init__(self, *args, **kwargs):
        canvas.Canvas.__init__(self, *args, **kwargs)
        self._codes = []

    def showPage(self):
        self._codes.append({'code': self._code, 'stack': self._codeStack})
        self._startPage()

    def save(self):
        """add page info to each page (page x of y)"""
        # reset page counter
        self._pageNumber = 0
        for code in self._codes:
            # recall saved page
            self._code = code['code']
            self._codeStack = code['stack']
            # handle any headers like page numbers that need to updated
after complete rendering
            self.setFont("Times-Roman",9)
            for x,y,hdrStr,func in self.liveHdrs:
                func(x,y,LIVE_HDR_KEYS[hdrStr] % { 'pageNumber' :
self._pageNumber+1,
                                                   'pageTotal' :
len(self._codes)})

            canvas.Canvas.showPage(self)
        canvas.Canvas.save(self)

* the call to func is a handle to either drawString, drawCentredString or
drawRightString depending on where the user asked for the footer to be
placed.

Anyone have any ideas why NumberedCanvas might stop images from showing?  Is
there anything I can do to debug this further or a better way to do "Page x
of y"?

Thanks

-Chris
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/reportlab-users/attachments/20081112/422fce01/attachment.html>

From gabor at nekomancer.net  Wed Nov 12 09:48:27 2008
From: gabor at nekomancer.net (=?ISO-8859-1?Q?G=E1bor_Farkas?=)
Date: Wed, 12 Nov 2008 15:48:27 +0100
Subject: [reportlab-users] aalways using wordWrapping="CJK",
	any disadvantages?
Message-ID: <2c173a350811120648o37c78428p1432a2cf8b2c5b87@mail.gmail.com>

hi,

i'm working on a pdf document, that sometimes might contain japanese
or chinese text,
so i'd like it to wrap such text correctly.

i know that saying wordWrapping="CJK"..

but i cannot see any disadvantages when using it for non-cjk-text, so
i was thinking
about simply turning it on for every possible language.

are there are problems with such approach?

in other words, are there any situations, when it's not a good idea to
have wordWrappign="CJK" ?


thanks,
gabor

From gabor at nekomancer.net  Wed Nov 12 09:58:45 2008
From: gabor at nekomancer.net (=?ISO-8859-1?Q?G=E1bor_Farkas?=)
Date: Wed, 12 Nov 2008 15:58:45 +0100
Subject: [reportlab-users] aalways using wordWrapping="CJK",
	any disadvantages?
In-Reply-To: <2c173a350811120648o37c78428p1432a2cf8b2c5b87@mail.gmail.com>
References: <2c173a350811120648o37c78428p1432a2cf8b2c5b87@mail.gmail.com>
Message-ID: <2c173a350811120658o6ee4d16ep9e976cee1a29681a@mail.gmail.com>

On Wed, Nov 12, 2008 at 3:48 PM, G?bor Farkas <gabor at nekomancer.net> wrote:
>
> in other words, are there any situations, when it's not a good idea to
> have wordWrappign="CJK" ?

ok, please ignore this question, i've found the problematic situation
already :-)

thanks,
gabor

From timr at probo.com  Wed Nov 12 17:09:31 2008
From: timr at probo.com (Tim Roberts)
Date: Wed, 12 Nov 2008 14:09:31 -0800
Subject: [reportlab-users] Dual pass numbered canvas causes images to
 disappear
In-Reply-To: <3642025c0811120609y49b77afeg9ed3e142364f9a4e@mail.gmail.com>
References: <3642025c0811120609y49b77afeg9ed3e142364f9a4e@mail.gmail.com>
Message-ID: <491B541B.80700@probo.com>

Chris Kratz wrote:
>
> When generating a pdf, if I use the normal canvas, my image shows up
> just fine in the generated pdf.
>
> doc.build(story, onFirstPage=self.firstPage, onLaterPages=self.laterPages)
>
> But, I need the ability to print page x of y in the footers.  So, I
> used a slightly modified version of the NumberedCanvas class found
> here: http://code.activestate.com/recipes/546511/
>
> So, when I call doc.build with the NumberedCanvas, I get the Page x of
> y in the footer, but images just seem to not show.  No errors or any
> other indications of problems, just no images.  The document size
> seems to indicate the image is included in the pdf, even if it isn't
> displaying.
> ...
> Anyone have any ideas why NumberedCanvas might stop images from showing?

Yes.  It's because that recipe relies on the undocumented internal
structure of ReportLab, and the structure has changed since it was
written.  The standard Canvas showPage procedure sets a flag in the
queued up page that says whether the page contains images or not.  When
the page does not contain images, pdfbase\pdfdoc.py skips the
image-handling Postscript code, which is somewhat lengthy.  In your
code, if the LAST page does not contain images, then NONE of the pages
will get the image-handling Postscript.

For now, I believe you can short-circuit this by adding
    self._currentPageHasImages = 1
before calling "canvas.Canvas.showPage(self)".


> Is there anything I can do to debug this further or a better way to do
> "Page x of y"?

The recommended way is to use doc.multibuild to do a two-pass build: one
to compute the total number of pages, then another to insert that in the
footer.  There's also a way to insert a "placeholder" that can be filled
in later at the end of the build, but I don't remember what that's called.

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From dirk.holtwick at gmail.com  Thu Nov 13 08:50:49 2008
From: dirk.holtwick at gmail.com (Dirk Holtwick)
Date: Thu, 13 Nov 2008 14:50:49 +0100
Subject: [reportlab-users] Visual Diff Tool
Message-ID: <491C30B9.70006@gmail.com>

Hi,

while working on xhtml2pdf <http://www.xhtml2pdf.com/> I have the 
general problem that writing test cases is quite difficult. The normal 
way is to create some test PDF and then have a look at them if they are 
looking fine. But that is a source of human errors. Therefore I am 
looking for a automatic solution.

My idea is to automatically do a visual compare of a reference PDF that 
was already rendered fine once and the recent PDF. Technically I am 
found out that it would work fine if I use ImageMagic to first create 
images from PDF and if they differ I can show them in TortoiseIDiff.

I started writing such a visual diff tool, but maybe there already 
exists something similar and I can use this instead of wasting my time 
on writing the reinvention of the wheel ;)

So my question: has anyone seen a tool like this around? If so, please 
let me know where to find it. Otherwise I would write the visual diff 
tool and maybe Reportlab also likes to use visual test cases for their 
toolkit?

Cheers
Dirk

From andy at reportlab.com  Thu Nov 13 09:10:33 2008
From: andy at reportlab.com (Andy Robinson)
Date: Thu, 13 Nov 2008 14:10:33 +0000
Subject: [reportlab-users] Visual Diff Tool
In-Reply-To: <491C30B9.70006@gmail.com>
References: <491C30B9.70006@gmail.com>
Message-ID: <956003ae0811130610i7f1c5bdbte1231075a902cb71@mail.gmail.com>

2008/11/13 Dirk Holtwick <dirk.holtwick at gmail.com>:
> Hi,
>
> I started writing such a visual diff tool, but maybe there already exists
> something similar and I can use this instead of wasting my time on writing
> the reinvention of the wheel ;)

We had good results internally with a utility called "pdftoppm", which
I think is part of the Xpdf distro.
        http://www.foolabs.com/xpdf/download.html

To be honest we are not using it for test cases.  I know we should be ;-)
 We have used it in various "tracing tools" to take measurements off other
peoples' PDFs, and to get thumbnails for page previews in a CMS.

> So my question: has anyone seen a tool like this around? If so, please let
> me know where to find it. Otherwise I would write the visual diff tool and
> maybe Reportlab also likes to use visual test cases for their toolkit?

Definitely!

- Andy

From andy at reportlab.com  Thu Nov 13 09:21:34 2008
From: andy at reportlab.com (Andy Robinson)
Date: Thu, 13 Nov 2008 14:21:34 +0000
Subject: [reportlab-users] Subversion checkin rights available
Message-ID: <956003ae0811130621p7b453548tbf3ab1737b2a87e0@mail.gmail.com>

Hi everyone,

We've set up an area in our SVN repo with suitable permissions, and a
copy of the present trunk for experimental purposes.  Anyone who would
like checkin rights should email Robin Becker (robin at
you-know-where) and myself (andy at you-know-where) with....

(a) their preferred username  (BTW we have no internal employees named
Dirk, Dinu or Jerome so these are available to the obvious people...)

(b) the hash for your password generated by the htpasswd utility (see below)

(c) if you are not a regular poster to this list, a brief statement of
what you'd like to work on.


You can generate the hash we need without needing a password file by
doing the following and sending us the last line as follows.

andy at kenya:~/Desktop$ htpasswd -n andy
New password:
Re-type new password:
andy:10kgkGlaXfgx.




Best Regards,



-- 
Andy Robinson
CEO/Chief Architect
ReportLab Europe Ltd.
Media House, 3 Palmerston Road, Wimbledon, London SW19 1PG, UK
Tel +44-20-8545-1570

From dirk.holtwick at gmail.com  Thu Nov 13 10:07:36 2008
From: dirk.holtwick at gmail.com (Dirk Holtwick)
Date: Thu, 13 Nov 2008 16:07:36 +0100
Subject: [reportlab-users] Visual Diff Tool
In-Reply-To: <956003ae0811130610i7f1c5bdbte1231075a902cb71@mail.gmail.com>
References: <491C30B9.70006@gmail.com>
	<956003ae0811130610i7f1c5bdbte1231075a902cb71@mail.gmail.com>
Message-ID: <491C42B8.4050206@gmail.com>

> We had good results internally with a utility called "pdftoppm", which
> I think is part of the Xpdf distro.
>         http://www.foolabs.com/xpdf/download.html
> 
> To be honest we are not using it for test cases.  I know we should be ;-)
>  We have used it in various "tracing tools" to take measurements off other
> peoples' PDFs, and to get thumbnails for page previews in a CMS.

I tested it and it works fine. But I still tend to use ImageMagic because:

- It has an easies handling of fonts
- It creates PNG, JPG, TIF etc. files which can be compared using 
TortoiseIDiff (PPM is not supported)

>> So my question: has anyone seen a tool like this around? If so, please let
>> me know where to find it. Otherwise I would write the visual diff tool and
>> maybe Reportlab also likes to use visual test cases for their toolkit?
> 
> Definitely!

Fine. I'll keep you updated.

Dirk

From gherman at darwin.in-berlin.de  Thu Nov 13 11:03:35 2008
From: gherman at darwin.in-berlin.de (Dinu Gherman)
Date: Thu, 13 Nov 2008 17:03:35 +0100
Subject: [reportlab-users] Visual Diff Tool
In-Reply-To: <491C30B9.70006@gmail.com>
References: <491C30B9.70006@gmail.com>
Message-ID: <C88C4808-E6D7-4100-93EF-97D1A0B15E7B@darwin.in-berlin.de>

Dirk Holtwick:

> So my question: has anyone seen a tool like this around? If so,  
> please let me know where to find it. Otherwise I would write the  
> visual diff tool and maybe Reportlab also likes to use visual test  
> cases for their toolkit?

I think, people just write this kind of stuff themselves,
if they really need it, as I did a while ago [1]. If I re-
member correctly, Stefan Richter also did that for z3c.rml,
maybe you want to look into that code...

Regards,

Dinu

[1] http://two.pairlist.net/pipermail/reportlab-users/2003-September/002094.html

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/reportlab-users/attachments/20081113/ff585852/attachment.html>

From andy at reportlab.com  Thu Nov 13 11:43:54 2008
From: andy at reportlab.com (Andy Robinson)
Date: Thu, 13 Nov 2008 16:43:54 +0000
Subject: [reportlab-users] Visual Diff Tool
In-Reply-To: <491C42B8.4050206@gmail.com>
References: <491C30B9.70006@gmail.com>
	<956003ae0811130610i7f1c5bdbte1231075a902cb71@mail.gmail.com>
	<491C42B8.4050206@gmail.com>
Message-ID: <956003ae0811130843y3227db00i7f174c2b77663b7@mail.gmail.com>

2008/11/13 Dirk Holtwick <dirk.holtwick at gmail.com>:
>> We had good results internally with a utility called "pdftoppm", which
> I tested it and it works fine. But I still tend to use ImageMagic because:
>
> - It has an easies handling of fonts
> - It creates PNG, JPG, TIF etc. files which can be compared using
> TortoiseIDiff (PPM is not supported)

pdftoppm is an unfortunate name - it produces most formats and we
always asked it to give us .png output.

I need to check out ImageMagick and see how well it supports PDFs with
embedded form objects, asian fonts, different font models and so on.
For our own test suite it really needs to support all the common stuff
in PDF.

- Andy

From andy at reportlab.com  Thu Nov 13 11:57:01 2008
From: andy at reportlab.com (Andy Robinson)
Date: Thu, 13 Nov 2008 16:57:01 +0000
Subject: [reportlab-users] Visual Diff Tool
In-Reply-To: <956003ae0811130843y3227db00i7f174c2b77663b7@mail.gmail.com>
References: <491C30B9.70006@gmail.com>
	<956003ae0811130610i7f1c5bdbte1231075a902cb71@mail.gmail.com>
	<491C42B8.4050206@gmail.com>
	<956003ae0811130843y3227db00i7f174c2b77663b7@mail.gmail.com>
Message-ID: <956003ae0811130857s9b737few1a95335a588f2246@mail.gmail.com>

2008/11/13 Andy Robinson <andy at reportlab.com>:
> I need to check out ImageMagick and see how well it supports PDFs with
> embedded form objects, asian fonts, different font models and so on.


It seems to work great on several of our test files but will not
handle the output of, for example, tests/test_multibyte_jpn.py / .pdf.
 I now recall this is why we went with Xpdf; Ghostscript and Quartz
used to be bad at Asian text (although it's probably fixed now
and as people move over to TrueType it will be less of an issue)

I'm still impressed by its 'normal' output though - certainly good
enough for diffing, and I could imagine writing tests which use this
to assert that certain parts of the page end up coloured.  Best of all
the binaries are available on all platforms.

Thanks for the tip!

- Andy

From robin at reportlab.com  Thu Nov 13 12:01:51 2008
From: robin at reportlab.com (Robin Becker)
Date: Thu, 13 Nov 2008 17:01:51 +0000
Subject: [reportlab-users] Visual Diff Tool
In-Reply-To: <491C30B9.70006@gmail.com>
References: <491C30B9.70006@gmail.com>
Message-ID: <491C5D7F.1040702@chamonix.reportlab.co.uk>

Dirk Holtwick wrote:
> Hi,
> 
> while working on xhtml2pdf <http://www.xhtml2pdf.com/> I have the 
> general problem that writing test cases is quite difficult. The normal 
> way is to create some test PDF and then have a look at them if they are 
> looking fine. But that is a source of human errors. Therefore I am 
> looking for a automatic solution.
> 
> My idea is to automatically do a visual compare of a reference PDF that 
> was already rendered fine once and the recent PDF. Technically I am 
> found out that it would work fine if I use ImageMagic to first create 
> images from PDF and if they differ I can show them in TortoiseIDiff.
> 
> I started writing such a visual diff tool, but maybe there already 
> exists something similar and I can use this instead of wasting my time 
> on writing the reinvention of the wheel ;)
> 
> So my question: has anyone seen a tool like this around? If so, please 
> let me know where to find it. Otherwise I would write the visual diff 
> tool and maybe Reportlab also likes to use visual test cases for their 
> toolkit?
> 
......
We have been using ghostscript in one of our server based tools. I tried the 
pdftoppm route, but couldn't get a working version from source on freebsd.

Ghostscript works pretty well, we use commands like

gs -q -dSAFER -dNOPAUSE -dBATCH \
     -sOutputFile=rl_hello-page%04d.jpg \
     -sDEVICE=jpeg -r144x144 -f rl_hello.pdf


-- 
Robin Becker

From dirk.holtwick at gmail.com  Thu Nov 13 12:34:39 2008
From: dirk.holtwick at gmail.com (Dirk Holtwick)
Date: Thu, 13 Nov 2008 18:34:39 +0100
Subject: [reportlab-users] Visual Diff Tool
In-Reply-To: <956003ae0811130857s9b737few1a95335a588f2246@mail.gmail.com>
References: <491C30B9.70006@gmail.com>	<956003ae0811130610i7f1c5bdbte1231075a902cb71@mail.gmail.com>	<491C42B8.4050206@gmail.com>	<956003ae0811130843y3227db00i7f174c2b77663b7@mail.gmail.com>
	<956003ae0811130857s9b737few1a95335a588f2246@mail.gmail.com>
Message-ID: <491C652F.60902@gmail.com>

Andy Robinson schrieb:
> Thanks for the tip!

You are welcome. I implemented a dirty first version of the visual 
differ that seems to work on my Windows machine. If you like to have a 
look it's here:

http://code.google.com/p/xhtml2pdf/source/browse/branches/3.0/tests/test_samples.py

Dirk

From gherman at darwin.in-berlin.de  Fri Nov 14 04:16:38 2008
From: gherman at darwin.in-berlin.de (Dinu Gherman)
Date: Fri, 14 Nov 2008 10:16:38 +0100
Subject: [reportlab-users] Tracker
In-Reply-To: <956003ae0810141622j12c8af74n86c2b2c4b6b38ea0@mail.gmail.com>
References: <956003ae0810141622j12c8af74n86c2b2c4b6b38ea0@mail.gmail.com>
Message-ID: <0556A503-B113-4417-A9B6-C45E0422C9D8@darwin.in-berlin.de>

Andy Robinson:

> There is a basic public tracker up now:
>
>   https://admin.reportlab.com/mantis/project/core/tracker/
>
> For now it's just a brief list of known issues.  (It's part of a  
> bigger
> 2-year-old app which we need to upgrade to Django 1.0+ before we
> can hack on it easily...).
>
> If anyone has reported issues or sent patches we've missed, please
> repost them and we'll ensure they are logged and dealt with.

I cannot see the two buglets I reported on this list on
october 26 and 30. Shall I repost them?

Also, it would be tremendeously useful to be able to sort
the bug tracking page above by date.

Regards,

Dinu


From andy at reportlab.com  Fri Nov 14 05:10:50 2008
From: andy at reportlab.com (Andy Robinson)
Date: Fri, 14 Nov 2008 10:10:50 +0000
Subject: [reportlab-users] Tracker
In-Reply-To: <0556A503-B113-4417-A9B6-C45E0422C9D8@darwin.in-berlin.de>
References: <956003ae0810141622j12c8af74n86c2b2c4b6b38ea0@mail.gmail.com>
	<0556A503-B113-4417-A9B6-C45E0422C9D8@darwin.in-berlin.de>
Message-ID: <956003ae0811140210y4790bd4fu31bb2e7a829e610c@mail.gmail.com>

2008/11/14 Dinu Gherman <gherman at darwin.in-berlin.de>:
> I cannot see the two buglets I reported on this list on
> october 26 and 30. Shall I repost them?
>


Aha, I wasn't around then.  I will berate our support-person-on-duty
and ensure they are aware open source bugs should be at least
logged ;-)


- Andy

From ralsina at netmanagers.com.ar  Fri Nov 14 08:18:37 2008
From: ralsina at netmanagers.com.ar (Roberto Alsina)
Date: Fri, 14 Nov 2008 11:18:37 -0200
Subject: [reportlab-users] Doubts about split()
Message-ID: <200811141118.43700.ralsina@netmanagers.com.ar>

I am not sure I understand split() for platypus flowables.

The docs say if a flowable can split:

  Clever flowables should split themselves and return a list of flowables. It
  is up to the client code to ensure that repeated attempts to split are
  avoided.

This is because split can return [self] right? So I must check if the flowable 
has already been "splitted" and in that case what should I return?

  If the space is sufficient the split method should return [self]. Otherwise
  the flowable should rearrange itself and return a list [f0,...] of flowables
  which will be considered in order.

Suppose I have a box B containing paragraphs P1 P2 P3. The suppose P1 would 
fit in the available space.

I should return [Box(P1),Box(P2,P3)] , right?

Sorry about the pseudo code, I mean two boxes, one containing P1, the other P2 
and P3 ;-)

What should I return when there is no way to fit anything in the available 
space and everything should be moved to the next page?

-- 
 ("\''/").__..-''"`-. .         Roberto Alsina
 `9_ 9  )   `-. (    ).`-._.`)  KDE Developer (MFCH)
 (_Y_.)' ._   ) `._`.  " -.-'  http://lateral.netmanagers.com.ar 
  _..`-'_..-_/ /-'_.'     The 6,855th most popular site of Slovenia   
(l)-'' ((i).' ((!.'       according to alexa.com (27/5/2007) 
                    
Debugging is twice as hard as writing the code in the first place. 
Therefore, if you write the code as cleverly as possible, you are, 
by definition, not smart enough to debug it. --Brian W. Kernighan

From dirk.holtwick at gmail.com  Fri Nov 14 08:23:59 2008
From: dirk.holtwick at gmail.com (Dirk Holtwick)
Date: Fri, 14 Nov 2008 14:23:59 +0100
Subject: [reportlab-users] Leading in Paragraph
Message-ID: <491D7BEF.9060302@gmail.com>

Hi,

I try to optimize the paragraph stuff for my needs and one issue is the 
handling of leading.

If I understand right Reportlab in Paragraph draws  a line and then 
starts the next line with a regular distance of "leading". In HTML 
documents this seems to be handle a little bit different for paragraphs 
with paddings or borders. There a line is rendered like: Draw text on 
(Y-Position + Leading / 2). This is nice because in a bordered paragraph 
the distances from top and bottom are equal.

My question. Is there a way in Paragraph to get the same results?

Dirk

From andy at reportlab.com  Fri Nov 14 08:33:28 2008
From: andy at reportlab.com (Andy Robinson)
Date: Fri, 14 Nov 2008 13:33:28 +0000
Subject: [reportlab-users] Doubts about split()
In-Reply-To: <200811141118.43700.ralsina@netmanagers.com.ar>
References: <200811141118.43700.ralsina@netmanagers.com.ar>
Message-ID: <956003ae0811140533r1e79d367p6aea27508986eabf@mail.gmail.com>

2008/11/14 Roberto Alsina <ralsina at netmanagers.com.ar>:
> I am not sure I understand split() for platypus flowables.
>
> I should return [Box(P1),Box(P2,P3)] , right?

That should work.  I don't think we ever tried a list of more than 2
items; it probably also works if you just return this, as I would hope
anything not fitting is added to the list for the next page.
  [Box(P1),  Box(P1),  Box(P1)]

> What should I return when there is no way to fit anything in the available
> space and everything should be moved to the next page?

Return [].  This is the default implementation for a Flowable anyway.
I just slightly expanded the docstring in Flowable.split to mention this.

- Andy

From gherman at darwin.in-berlin.de  Fri Nov 14 08:38:13 2008
From: gherman at darwin.in-berlin.de (Dinu Gherman)
Date: Fri, 14 Nov 2008 14:38:13 +0100
Subject: [reportlab-users] Doubts about split()
In-Reply-To: <200811141118.43700.ralsina@netmanagers.com.ar>
References: <200811141118.43700.ralsina@netmanagers.com.ar>
Message-ID: <8FA0C229-9F7E-4323-A566-6168FD4BB42C@darwin.in-berlin.de>

Roberto Alsina:

> I am not sure I understand split() for platypus flowables.

Welcome to the club! ;-)

> The docs say if a flowable can split: [...]

The docs are not very helpful with this topic and I've had
a hard time myself splitting flowables on my head!

I suggest you look into my alterparagraphs code which might
solve some questions:

   http://pypi.python.org/pypi/alterparagraphs

Regards,

Dinu

From jonas at reportlab.com  Fri Nov 14 10:20:02 2008
From: jonas at reportlab.com (jonas wustrack)
Date: Fri, 14 Nov 2008 15:20:02 +0000
Subject: [reportlab-users] Tracker
In-Reply-To: <0556A503-B113-4417-A9B6-C45E0422C9D8@darwin.in-berlin.de>
References: <956003ae0810141622j12c8af74n86c2b2c4b6b38ea0@mail.gmail.com>
	<0556A503-B113-4417-A9B6-C45E0422C9D8@darwin.in-berlin.de>
Message-ID: <be0cc4270811140720tbbd3ca3y46071dfbab614add@mail.gmail.com>

On Fri, Nov 14, 2008 at 9:16 AM, Dinu Gherman
<gherman at darwin.in-berlin.de> wrote:
> Andy Robinson:
>
>> There is a basic public tracker up now:
>>
>>  https://admin.reportlab.com/mantis/project/core/tracker/
>>
>> For now it's just a brief list of known issues.  (It's part of a bigger
>> 2-year-old app which we need to upgrade to Django 1.0+ before we
>> can hack on it easily...).
>>
>> If anyone has reported issues or sent patches we've missed, please
>> repost them and we'll ensure they are logged and dealt with.
>
> I cannot see the two buglets I reported on this list on
> october 26 and 30. Shall I repost them?
>
> Also, it would be tremendeously useful to be able to sort
> the bug tracking page above by date.

You can, now ;-)

Regards,
-- 
Jonas

From webinconcept at gmail.com  Fri Nov 14 11:09:38 2008
From: webinconcept at gmail.com (Webin Concept)
Date: Fri, 14 Nov 2008 17:09:38 +0100
Subject: [reportlab-users] which version of adobe acrobat reader (minimal
	requirements)
Message-ID: <9b4634000811140809o6ab74923vf1457b8de272c4ea@mail.gmail.com>

Can you tell me wich version of Acrobat reader is required to read the
pdf output of reportlab?

I'm asking for the *minimum* requirements here: 5? 6?

Kind regards,

J
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/reportlab-users/attachments/20081114/4053c507/attachment.html>

From andy at reportlab.com  Fri Nov 14 12:01:39 2008
From: andy at reportlab.com (Andy Robinson)
Date: Fri, 14 Nov 2008 17:01:39 +0000
Subject: [reportlab-users] which version of adobe acrobat reader
	(minimal requirements)
In-Reply-To: <9b4634000811140809o6ab74923vf1457b8de272c4ea@mail.gmail.com>
References: <9b4634000811140809o6ab74923vf1457b8de272c4ea@mail.gmail.com>
Message-ID: <956003ae0811140901y59774a63la0a1ed7381102045@mail.gmail.com>

2008/11/14 Webin Concept <webinconcept at gmail.com>:
> Can you tell me wich version of Acrobat reader is required to read the
> pdf output of reportlab?
>
> I'm asking for the *minimum* requirements here: 5? 6?

4.0 should be safe.  With 3.0 there were some occasional
problems which we decided not to bother to support.


Best Regards,

-- 
Andy Robinson
CEO/Chief Architect
ReportLab Europe Ltd.
Media House, 3 Palmerston Road, Wimbledon, London SW19 1PG, UK
Tel +44-20-8545-1570

From mlacunza at gmail.com  Fri Nov 14 12:24:10 2008
From: mlacunza at gmail.com (Mario Lacunza)
Date: Fri, 14 Nov 2008 12:24:10 -0500
Subject: [reportlab-users] Visual Designer
Message-ID: <771741b20811140924r4e50172cy2101f5d59a1c6231@mail.gmail.com>

Hello,

There are plans to create a Visual Report Designer for Reportlab? Something
like Crystal Report or other similar tool?

Thanks!

-- 
Saludos / Best regards

Mario Lacunza
Software Architect - Webmaster

Email: mlacunza [AT] gmail [DOT] com
http://www.lacunza.info
Lima - Peru
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/reportlab-users/attachments/20081114/22163c95/attachment.htm>

From andy at reportlab.com  Fri Nov 14 12:28:19 2008
From: andy at reportlab.com (Andy Robinson)
Date: Fri, 14 Nov 2008 17:28:19 +0000
Subject: [reportlab-users] Visual Designer
In-Reply-To: <771741b20811140924r4e50172cy2101f5d59a1c6231@mail.gmail.com>
References: <771741b20811140924r4e50172cy2101f5d59a1c6231@mail.gmail.com>
Message-ID: <956003ae0811140928i46b2233cma11ae606f45798c6@mail.gmail.com>

2008/11/14 Mario Lacunza <mlacunza at gmail.com>:
> Hello,
>
> There are plans to create a Visual Report Designer for Reportlab? Something
> like Crystal Report or other similar tool?
>

No, sorry (at least not within ReportLab the company).  Building such a tool
well would probably take years of effort.

We do have a few ideas to save people some time in coding, but
I'll keep quiet on them for now as we have lots of more short term
issues to fix first...

Best Regards,


-- 
Andy Robinson
CEO/Chief Architect
ReportLab Europe Ltd.
Media House, 3 Palmerston Road, Wimbledon, London SW19 1PG, UK
Tel +44-20-8545-1570

From mlacunza at gmail.com  Fri Nov 14 12:42:25 2008
From: mlacunza at gmail.com (Mario Lacunza)
Date: Fri, 14 Nov 2008 12:42:25 -0500
Subject: [reportlab-users] Visual Designer
In-Reply-To: <956003ae0811140928i46b2233cma11ae606f45798c6@mail.gmail.com>
References: <771741b20811140924r4e50172cy2101f5d59a1c6231@mail.gmail.com>
	<956003ae0811140928i46b2233cma11ae606f45798c6@mail.gmail.com>
Message-ID: <771741b20811140942k6252ad5flf13b3b86cc0071d6@mail.gmail.com>

2008/11/14 Andy Robinson <andy at reportlab.com>

> 2008/11/14 Mario Lacunza <mlacunza at gmail.com>:
> > Hello,
> >
> > There are plans to create a Visual Report Designer for Reportlab?
> Something
> > like Crystal Report or other similar tool?
> >
>
> No, sorry (at least not within ReportLab the company).  Building such a
> tool
> well would probably take years of effort.
>
> We do have a few ideas to save people some time in coding, but
> I'll keep quiet on them for now as we have lots of more short term
> issues to fix first...
>
>
Yes, I understand...  but maybe an Opensource separate project? I think:
others programmers need a graphic tool for design reports like me, if
anyboby in this list have interest in this idea please contact me to create
it with GPL, I think could be a great contribution.

-- 
Saludos / Best regards

Mario Lacunza
Software Architect - Webmaster

Email: mlacunza [AT] gmail [DOT] com
http://www.lacunza.info
Lima - Peru
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/reportlab-users/attachments/20081114/26b3f401/attachment.html>

From webinconcept at gmail.com  Fri Nov 14 13:27:34 2008
From: webinconcept at gmail.com (Webin Concept)
Date: Fri, 14 Nov 2008 19:27:34 +0100
Subject: [reportlab-users] which version of adobe acrobat reader
	(minimal requirements)
In-Reply-To: <956003ae0811140901y59774a63la0a1ed7381102045@mail.gmail.com>
References: <9b4634000811140809o6ab74923vf1457b8de272c4ea@mail.gmail.com>
	<956003ae0811140901y59774a63la0a1ed7381102045@mail.gmail.com>
Message-ID: <9b4634000811141027p9e758e3q3532113e1c38b7bf@mail.gmail.com>

Thank you Andy!

On Fri, Nov 14, 2008 at 6:01 PM, Andy Robinson <andy at reportlab.com> wrote:

> 2008/11/14 Webin Concept <webinconcept at gmail.com>:
> > Can you tell me wich version of Acrobat reader is required to read the
> > pdf output of reportlab?
> >
> > I'm asking for the *minimum* requirements here: 5? 6?
>
> 4.0 should be safe.  With 3.0 there were some occasional
> problems which we decided not to bother to support.
>
>
> Best Regards,
>
> --
> Andy Robinson
> CEO/Chief Architect
> ReportLab Europe Ltd.
> Media House, 3 Palmerston Road, Wimbledon, London SW19 1PG, UK
> Tel +44-20-8545-1570
> _______________________________________________
> reportlab-users mailing list
> reportlab-users at reportlab.com
> http://two.pairlist.net/mailman/listinfo/reportlab-users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/reportlab-users/attachments/20081114/4a2d27e4/attachment.html>

From gherman at darwin.in-berlin.de  Mon Nov 17 08:35:42 2008
From: gherman at darwin.in-berlin.de (Dinu Gherman)
Date: Mon, 17 Nov 2008 14:35:42 +0100
Subject: [reportlab-users] PDF test pages for printers?
Message-ID: <1D630AF4-8B0C-41C5-9048-C447473CCB39@darwin.in-berlin.de>

Hi,

I wonder if anybody here has created sample PDF pages to be
tested with different printers before deciding which one to
buy?

Since all printers have different demo pages, it is hard to
compare them. Features to be tested might include line den-
sity, grey levels, color, resolution, non-printable margins,
etc. I found this, but it lacks "printing on the edge":

   http://cream.sourceforge.net/ColorCard.pdf

Well, maybe someone wants to test his ReportLab knowledge
to create something like this or better for the community?
Just an idea...

Regards,

Dinu

.....................................................................
Heute Utopist werden: http://www.utopia.de - Deutschlands gr??te On-
line-Community f?r strategischen Konsum und nachhaltigen Lebensstil.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/reportlab-users/attachments/20081117/2f5aa342/attachment.htm>

From alet at librelogiciel.com  Mon Nov 17 08:41:59 2008
From: alet at librelogiciel.com (alet at librelogiciel.com)
Date: Mon, 17 Nov 2008 14:41:59 +0100
Subject: [reportlab-users] PDF test pages for printers?
In-Reply-To: <1D630AF4-8B0C-41C5-9048-C447473CCB39@darwin.in-berlin.de>
References: <1D630AF4-8B0C-41C5-9048-C447473CCB39@darwin.in-berlin.de>
Message-ID: <20081117134159.GD27934@vazy.pykota.com>

On Mon, Nov 17, 2008 at 02:35:42PM +0100, Dinu Gherman wrote:
>
> I wonder if anybody here has created sample PDF pages to be
> tested with different printers before deciding which one to
> buy?
>
> Since all printers have different demo pages, it is hard to
> compare them. Features to be tested might include line den-
> sity, grey levels, color, resolution, non-printable margins,
> etc. I found this, but it lacks "printing on the edge":
>
>   http://cream.sourceforge.net/ColorCard.pdf
>
> Well, maybe someone wants to test his ReportLab knowledge
> to create something like this or better for the community?
> Just an idea...

I've created a generator for the attached PDF file, which I use not to test
printers but to test my pkpgcounter software. Maybe this can help or
serve as a basis for more complex test documents.

The source is available at :

http://trac.pykota.com/browser/pkpgcounter/trunk/tests/runcolors.py

(or http://svn.pykota.com/pkpgcounter/trunk/tests/runcolors.py)

hth

Jerome Alet
-------------- next part --------------
A non-text attachment was scrubbed...
Name: colors.pdf
Type: application/pdf
Size: 9022 bytes
Desc: not available
Url : <http://two.pairlist.net/pipermail/reportlab-users/attachments/20081117/a93a7493/attachment.pdf>

From timr at probo.com  Mon Nov 17 14:01:28 2008
From: timr at probo.com (Tim Roberts)
Date: Mon, 17 Nov 2008 11:01:28 -0800
Subject: [reportlab-users] PDF test pages for printers?
In-Reply-To: <1D630AF4-8B0C-41C5-9048-C447473CCB39@darwin.in-berlin.de>
References: <1D630AF4-8B0C-41C5-9048-C447473CCB39@darwin.in-berlin.de>
Message-ID: <4921BF88.2030301@probo.com>

Dinu Gherman wrote:
> Hi,
>
> I wonder if anybody here has created sample PDF pages to be
> tested with different printers before deciding which one to
> buy? 
>
> Since all printers have different demo pages, it is hard to 
> compare them. Features to be tested might include line den-
> sity, grey levels, color, resolution, non-printable margins, 
> etc. I found this, but it lacks "printing on the edge":

Full bleed printing is rather rare in laser printers.  If that's a
requirement for you, then it probably narrows the field down more than
any of these other criteria.

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From tante at the-gay-bar.com  Mon Nov 17 15:39:52 2008
From: tante at the-gay-bar.com (=?ISO-8859-1?Q?J=FCrgen?= Geuter)
Date: Mon, 17 Nov 2008 21:39:52 +0100
Subject: [reportlab-users] Size of Barcodes in Reportlab
Message-ID: <1226954392.7672.16.camel@localhost>

Hello.

I am trying to generate barcodes with reportlab and it works quite well:
---- example code -----
from reportlab.pdfgen import canvas
from reportlab.lib.pagesizes import A4,landscape
from reportlab.graphics.barcode import code39
from reportlab.lib.units import mm

c = canvas.Canvas("/tmp/barcode.pdf", pagesize=landscape(A4))
barcode = code39.Extended39("123456")
barcode.drawOn(c,100*mm,100*mm)

c.showpage()
c.save()
---------------------------

This draws a correct barcode to the position I wanted it printed, but
it's too small. The barcode's size is not influenced by the font or the
current line width, I tried setting the "height" and "width" attributes
of my barcode before calling "drawOn". Nothing would change the
barcode's size.
I just need it bigger so our scanners can properly read it (there is
some dirt involved sometimes and the scanners are not always perfect so
in this case, size _does_matter :)

Thanks in Advance

J?rgen Geuter

-- 
ICQ #81510866  -  http://the-gay-bar.com  -  jabber tante at jabber.org
   Occam's Razor: 
       -"Entia non sunt multiplicanda praeter necessitatem."-
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: This is a digitally signed message part
Url : <http://two.pairlist.net/pipermail/reportlab-users/attachments/20081117/aabc3064/attachment.pgp>

From timr at probo.com  Mon Nov 17 16:10:53 2008
From: timr at probo.com (Tim Roberts)
Date: Mon, 17 Nov 2008 13:10:53 -0800
Subject: [reportlab-users] Size of Barcodes in Reportlab
In-Reply-To: <1226954392.7672.16.camel@localhost>
References: <1226954392.7672.16.camel@localhost>
Message-ID: <4921DDDD.4080304@probo.com>

J?rgen Geuter wrote:
> Hello.
>
> I am trying to generate barcodes with reportlab and it works quite well:
> ---- example code -----
> from reportlab.pdfgen import canvas
> from reportlab.lib.pagesizes import A4,landscape
> from reportlab.graphics.barcode import code39
> from reportlab.lib.units import mm
>
> c = canvas.Canvas("/tmp/barcode.pdf", pagesize=landscape(A4))
> barcode = code39.Extended39("123456")
> barcode.drawOn(c,100*mm,100*mm)
>
> c.showpage()
> c.save()
> ---------------------------
>
> This draws a correct barcode to the position I wanted it printed, but
> it's too small. The barcode's size is not influenced by the font or the
> current line width, I tried setting the "height" and "width" attributes
> of my barcode before calling "drawOn". Nothing would change the
> barcode's size

Use the source, Luke.  ;)

The Extended39 constructor accepts a "barWidth" parameter that specifies
the base width of the smallest element.  It defaults to .0075 inch.

-- 
Tim Roberts, timr at probo.com
Providenza & Boekelheide, Inc.


From haraldarminmassa at gmail.com  Tue Nov 18 09:40:39 2008
From: haraldarminmassa at gmail.com (Harald Armin Massa)
Date: Tue, 18 Nov 2008 15:40:39 +0100
Subject: [reportlab-users] Wordaxe 0.3.0, ReportLab 2.2,
	MultiBuild and splitting error
Message-ID: <7be3f35d0811180640h71517e8dq48403e112296ccbb@mail.gmail.com>

Hello,

on updating some projekt to RL 2.2; and wordaxe 0.3.0, I ran into some
obscure wrapping / splitting bug.

As it turned out:

the SplitRoutine within NewParagraph get's problems when doc.multibuild is used.

I tracked it down to:

        if not hasattr(self, "_lines"):
            # This can only happen when split has been called
            # without a previous wrap for this Paragraph.
            # From looking at doctemplate.py and frames.py,
            # I assume this is only the case if the free space
            # in the frame is not even enough for getSpaceBefore.
            # Thus we can safely return []
            #~ print "split without previous wrap"
            return []

NewParagraph of Wordaxe relying on self._lines NOT being present when
there was no "wrap" in the actual build pass, to return "I cannot
split"

BUT...

in a multibuild, in the second build pass that self._lines is present
from the first build pass ... , and _unused is an empty list (as some
wraps from the build pass before have used up everything)

So MY fix was in doctemplate multibuild:

            #clean up so multi-build does not go wrong - the frame
            #packer might have tacked an attribute onto some flowables
            for elem in story:
                if hasattr(elem, '_postponed'):
                    del elem._postponed
                if hasattr(elem, '_lines'):
                    del elem._lines

to extend the "cleanup" part to also delete the _lines attribute of a
NewParagraph; together with the _postponed attribut.

So everything works fine ... but I am not sure if this is really the
best approach, and the NewParagraph split/wrap is quite to challenging
to really be sure.

That's why I am posting this fix for others to enjoy and for more
enlightened ones to check:)

Harald

-- 
GHUM Harald Massa
persuadere et programmare
Harald Armin Massa
Spielberger Stra?e 49
70435 Stuttgart
0173/9409607
no fx, no carrier pigeon
-
EuroPython 2009 will take place in Birmingham - Stay tuned!

From robin at reportlab.com  Tue Nov 18 10:01:45 2008
From: robin at reportlab.com (Robin Becker)
Date: Tue, 18 Nov 2008 15:01:45 +0000
Subject: [reportlab-users] Wordaxe 0.3.0, ReportLab 2.2,
 MultiBuild and splitting error
In-Reply-To: <7be3f35d0811180640h71517e8dq48403e112296ccbb@mail.gmail.com>
References: <7be3f35d0811180640h71517e8dq48403e112296ccbb@mail.gmail.com>
Message-ID: <4922D8D9.9000000@chamonix.reportlab.co.uk>

Harald Armin Massa wrote:
> Hello,
........ some
> wraps from the build pass before have used up everything)
> 
> So MY fix was in doctemplate multibuild:
> 
>             #clean up so multi-build does not go wrong - the frame
>             #packer might have tacked an attribute onto some flowables
>             for elem in story:
>                 if hasattr(elem, '_postponed'):
>                     del elem._postponed
>                 if hasattr(elem, '_lines'):
>                     del elem._lines
> 
> to extend the "cleanup" part to also delete the _lines attribute of a
> NewParagraph; together with the _postponed attribut.
> 
> So everything works fine ... but I am not sure if this is really the
> best approach, and the NewParagraph split/wrap is quite to challenging
> to really be sure.
> 
> That's why I am posting this fix for others to enjoy and for more
> enlightened ones to check:)
> 
> Harald
> 
great, we have a sort of combined approach now in the trunk with something 
called doc._multiBuildEdits which appends to a list of things that need to be 
cleaned up. Currently that's keepWithNext stuff and _postponed. That mechanism 
requires us to store (operation, object, attributename) tuples at the point 
where we know the edit is required to be unwound. This just avoids having to 
iterate through the whole story to clean stuff up; the clean up code looks like

#work through any edits
while mbe:
	e = mbe.pop(0)
	e[0](*e[1:])

It's not clear whether this approach will work for your _lines attribute. In 
trunk code at least it's possible for a flowable to get access to the document 
context during some phases (using self.doctemplateAttr). If this works you could 
do something like

	mbe=self._doctemplateAttr('_multiBuildEdits')
	mbe((self,delattr,'_lines'))

where self._lines is created.
-- 
Robin Becker

From H.vonBargen at t-p.com  Tue Nov 18 12:39:09 2008
From: H.vonBargen at t-p.com (Henning von Bargen)
Date: Tue, 18 Nov 2008 18:39:09 +0100
Subject: [reportlab-users] Wordaxe 0.3.0, ReportLab 2.2,
	MultiBuild and splitting error
In-Reply-To: <mailman.25.1227027633.92157.reportlab-users@reportlab.com>
Message-ID: <77A9E75DA75C564D82BDC4B18A7D8525305666@ROHRPOSTIX.tup.ads>

Thanks for your bug report, Harald.
I added it as item 2310279 to the wordaxe bug tracker at
http://sourceforge.net/tracker2/?func=detail&aid=2310279&group_id=105867&atid=642478
Can you perhaps create a simple test case?

Henning

> From: "Harald Armin Massa" <haraldarminmassa at gmail.com>
> Subject: Wordaxe 0.3.0, ReportLab 2.2,MultiBuild and splitting error
> 
> Hello,
> 
> on updating some projekt to RL 2.2; and wordaxe 0.3.0, I ran into some
> obscure wrapping / splitting bug.
> 
> As it turned out:
> 
> the SplitRoutine within NewParagraph get's problems when 
> doc.multibuild is used.
> 
> I tracked it down to:
> 
>         if not hasattr(self, "_lines"):
>             # This can only happen when split has been called
>             # without a previous wrap for this Paragraph.
>             # From looking at doctemplate.py and frames.py,
>             # I assume this is only the case if the free space
>             # in the frame is not even enough for getSpaceBefore.
>             # Thus we can safely return []
>             #~ print "split without previous wrap"
>             return []
> 
> NewParagraph of Wordaxe relying on self._lines NOT being present when
> there was no "wrap" in the actual build pass, to return "I cannot
> split"
> 
> BUT...
> 
> in a multibuild, in the second build pass that self._lines is present
> from the first build pass ... , and _unused is an empty list (as some
> wraps from the build pass before have used up everything)
> 
> So MY fix was in doctemplate multibuild:
> 
>             #clean up so multi-build does not go wrong - the frame
>             #packer might have tacked an attribute onto some flowables
>             for elem in story:
>                 if hasattr(elem, '_postponed'):
>                     del elem._postponed
>                 if hasattr(elem, '_lines'):
>                     del elem._lines
> 
> to extend the "cleanup" part to also delete the _lines attribute of a
> NewParagraph; together with the _postponed attribut.
> 
> So everything works fine ... but I am not sure if this is really the
> best approach, and the NewParagraph split/wrap is quite to challenging
> to really be sure.
> 
> That's why I am posting this fix for others to enjoy and for more
> enlightened ones to check:)
> 
> Harald

From matt at tplus1.com  Tue Nov 18 16:27:19 2008
From: matt at tplus1.com (Matthew Wilson)
Date: Tue, 18 Nov 2008 16:27:19 -0500
Subject: [reportlab-users] Can anyone share some code to draw employee
	schedules?
Message-ID: <f6f643a20811181327r457e0be7q784a958e78d20abf@mail.gmail.com>

I want to print a grid where the rows are employees and the columns
are dates.  In each cell, I'll leave it blank or draw a big X that
means this particular employee works on this date.

This is such a vanilla report that I thought I'd check if anyone
already has something like this that they can share with me.

More generally, is there any kind of script repository for reportlab
where I can see how people did what they did?

Matt

-- 
Matthew Wilson
matt at tplus1.com
http://tplus1.com

From gherman at darwin.in-berlin.de  Wed Nov 19 05:50:48 2008
From: gherman at darwin.in-berlin.de (Dinu Gherman)
Date: Wed, 19 Nov 2008 11:50:48 +0100
Subject: [reportlab-users] Future of py2pdf
Message-ID: <3A314D7C-5CBD-4606-94C6-5308CACF7F28@darwin.in-berlin.de>

Hi,

I've just looked again at a tool named py2pdf, that I wrote
and that was included in the ReportLab distribution in 2001
I think (with reportlab 0.92 or so!). AFAIK, the Python Wing
IDE used this module and maybe still does.

Apart from being pretty old code (that still seems to work,
though, and for which I got maybe two little buglets repor-
ted in all these years) I wonder if I should put it again
on my list of things to refurbish. Ideally, I'd like to
distribute it as a self contained package, which should
demand less repackaging and reportlab knowledge from users
to get it running right off the shelf.

I understand that I can have access to the RL SVN repository
to work on it, but I wonder if that would change the instal-
lation issues? Ideally, py2pdf would only need the right
entries in the setup.py script of the RL distro, but the RL
release cycle is counterproductive for a quick propagation
of updates.

So I think I will demand from ReportLab to remove py2pdf
from its distribution starting with the next release, and
distribute it myself as a seperate package, including as
one of these insanely ubiquitous "eggs".

Therefore, I'd like to know if anybody here would suffer
from such a plan? If this would be the case for you, please
let me know.

Regards,

Dinu

......................................................................
Heute Utopist werden: http://www.utopia.de - Deutschlands gr??te On-
line-Community f?r strategischen Konsum und nachhaltigen Lebensstil.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/reportlab-users/attachments/20081119/d8568159/attachment.html>

From andy at reportlab.com  Wed Nov 19 06:37:53 2008
From: andy at reportlab.com (Andy Robinson)
Date: Wed, 19 Nov 2008 11:37:53 +0000
Subject: [reportlab-users] Future of py2pdf
In-Reply-To: <3A314D7C-5CBD-4606-94C6-5308CACF7F28@darwin.in-berlin.de>
References: <3A314D7C-5CBD-4606-94C6-5308CACF7F28@darwin.in-berlin.de>
Message-ID: <956003ae0811190337h4c01cf1cn87f0cef70610fb12@mail.gmail.com>

2008/11/19 Dinu Gherman <gherman at darwin.in-berlin.de>:
> I understand that I can have access to the RL SVN repository
> to work on it, but I wonder if that would change the instal-
> lation issues? Ideally, py2pdf would only need the right
> entries in the setup.py script of the RL distro, but the RL
> release cycle is counterproductive for a quick propagation
> of updates.

We should stop referring to it as a cycle ;-)  Think of it as an
unexpected 'event', like a white Christmas or sunny August
day (for a Brit).

Take this with a pinch of salt, but our next target for a release
is December.  Project work is getting quieter and we have a
commercial need to get a better-packaged release out with
updated docs, which will require an open source release too.

> So I think I will demand from ReportLab to remove py2pdf
> from its distribution starting with the next release, and
> distribute it myself as a seperate package, including as
> one of these insanely ubiquitous "eggs".

It's your call.  Just let us know.     We'll be setting up a page of
links to related packages in December anyway.  It would be
good to decide for sure this month, though.

> Therefore, I'd like to know if anybody here would suffer
> from such a plan? If this would be the case for you, please
> let me know.

-- 
Andy

From haraldarminmassa at gmail.com  Wed Nov 19 06:42:35 2008
From: haraldarminmassa at gmail.com (Harald Armin Massa)
Date: Wed, 19 Nov 2008 12:42:35 +0100
Subject: [reportlab-users] Future of py2pdf
In-Reply-To: <3A314D7C-5CBD-4606-94C6-5308CACF7F28@darwin.in-berlin.de>
References: <3A314D7C-5CBD-4606-94C6-5308CACF7F28@darwin.in-berlin.de>
Message-ID: <7be3f35d0811190342w10d31e41w3ac2de19d5102a4@mail.gmail.com>

Dinu,

> So I think I will demand from ReportLab to remove py2pdf
> from its distribution starting with the next release, and
> distribute it myself as a seperate package, including as
> one of these insanely ubiquitous "eggs".

as the integrated version "works", my recommendation is to leave it
where it is, just to update it with some DOC-String like

"this may be not the hottest and greatest version of py2pdf; if you
need something fresher, please go to <yoururl>"

That would be a free marketing for the module, it's better to have a
not so active one than none, and it frees the cycles.

Harald



-- 
GHUM Harald Massa
persuadere et programmare
Harald Armin Massa
Spielberger Stra?e 49
70435 Stuttgart
0173/9409607
no fx, no carrier pigeon
-
EuroPython 2009 will take place in Birmingham - Stay tuned!

From andy at reportlab.com  Wed Nov 19 06:53:06 2008
From: andy at reportlab.com (Andy Robinson)
Date: Wed, 19 Nov 2008 11:53:06 +0000
Subject: [reportlab-users] Future of py2pdf
In-Reply-To: <7be3f35d0811190342w10d31e41w3ac2de19d5102a4@mail.gmail.com>
References: <3A314D7C-5CBD-4606-94C6-5308CACF7F28@darwin.in-berlin.de>
	<7be3f35d0811190342w10d31e41w3ac2de19d5102a4@mail.gmail.com>
Message-ID: <956003ae0811190353t309624fatcba628492907be7c@mail.gmail.com>

2008/11/19 Harald Armin Massa <haraldarminmassa at gmail.com>:
> Dinu,
>
>> So I think I will demand from ReportLab to remove py2pdf
>> from its distribution starting with the next release, and
>> distribute it myself as a seperate package, including as
>> one of these insanely ubiquitous "eggs".
>
> as the integrated version "works", my recommendation is to leave it
> where it is, just to update it with some DOC-String like
>
> "this may be not the hottest and greatest version of py2pdf; if you
> need something fresher, please go to <yoururl>"

I would prefer that we either
(a) got rid of the demos directory altogether, or

(b) ensured things in it help to form relevant, properly commented
teaching examples (e.g. the long-awaited "long document
with a table of contents" or "print labels") and are maintained.

It's a real problem getting even the 80% of ReportLab staff
familiar with our framework up to speed, let alone people outside.
Removing things which we don't really need is probably a good step.

IMHO a clear link from a reportlab.org page listing related packages
will market the package better than being in a 'demos' directory which
will probably get excluded by some of the packaging mechanisms
anyway.    We will be rearranging the open source part of the site
in December and are happy to promote anyone's work there.


- Andy

From gabor at nekomancer.net  Wed Nov 19 08:31:58 2008
From: gabor at nekomancer.net (=?ISO-8859-1?Q?G=E1bor_Farkas?=)
Date: Wed, 19 Nov 2008 14:31:58 +0100
Subject: [reportlab-users] cjk-word-wrapping and xml-character-entities
Message-ID: <2c173a350811190531v32f71fu90243b75e1b3b464@mail.gmail.com>

hi,

the following script fails:

"""
from reportlab.platypus import SimpleDocTemplate, Paragraph
from reportlab.lib.styles import ParagraphStyle

doc = SimpleDocTemplate("test.pdf")

p = Paragraph(u'aaa&#x61;',ParagraphStyle('s',wordWrap='CJK'))
story = [p]
doc.build(story)
"""

please note, that "&#x61;" is simply "a", so the text in the paragraph
is simply "aaaa".
the problem is a combination of 2 issues:

1. the CJK algorithm only handles paragraps with 1 ParaFrags
2. the ParaParser, when parses the text, creates 2 frags because of
the char-entity

is there any way to work around this problem?

of course, my real problem is with chinese text. chinese text with
"&amp;" in it,
to be precise. this problem makes it impossible for me to have that
text word-wrapped
correctly.
or, maybe, is there a way to represent "&" in a paragraph in a different way?
in one that does not cause the ParaParser to create new ParaFrags?

thanks,
gabor

From andy at reportlab.com  Wed Nov 19 08:58:43 2008
From: andy at reportlab.com (Andy Robinson)
Date: Wed, 19 Nov 2008 13:58:43 +0000
Subject: [reportlab-users] cjk-word-wrapping and xml-character-entities
In-Reply-To: <2c173a350811190531v32f71fu90243b75e1b3b464@mail.gmail.com>
References: <2c173a350811190531v32f71fu90243b75e1b3b464@mail.gmail.com>
Message-ID: <956003ae0811190558m9667e3esb1c24421d0d81f4d@mail.gmail.com>

2008/11/19 G?bor Farkas <gabor at nekomancer.net>:
> 1. the CJK algorithm only handles paragraps with 1 ParaFrags
> 2. the ParaParser, when parses the text, creates 2 frags because of
> the char-entity

Thanks for such an easy-to-reproduce example!

We fixed this several months ago - I believe before the last release 2.2.
Your script runs fine for me against SVN trunk.  What version are you using?

Best Regards,


Andy Robinson

From matt at tplus1.com  Wed Nov 19 11:36:09 2008
From: matt at tplus1.com (Matthew Wilson)
Date: Wed, 19 Nov 2008 11:36:09 -0500
Subject: [reportlab-users] How do I left-align my table?
Message-ID: <f6f643a20811190836i3d1ee793wa6a21c56033af25f@mail.gmail.com>

I have a small table and it is displayed in the center of the page.
How can I have it line up on the left margin like my paragraphs?  I
understand how to mess with stuff inside the various table cells.  I
don't care about that.  Instead, I want to push the whole table box to
the left.


Here's the code I'm using:

def tfun(title, dt, data):

    t = Table(data)

    t.setStyle(TableStyle([
        ('GRID', (0, 0), (-1, -1), 0.5, colors.grey),
        ('BOX', (0, 0), (-1, -1), 2, colors.black),
    ]))

    styles = getSampleStyleSheet()

    def go():
        doc = SimpleDocTemplate("foodservice.pdf")
        style = styles["Normal"]

        Story = []
        Story.append(Paragraph(title, styles['Heading1']))
        Story.append(Paragraph(dt.strftime('%A, %B %d, %Y'),
styles['Heading2']))

        # This draws the table on to the page.
        Story.append(t)

        # This renders the doc.
        doc.build(Story)

    go()

if __name__ == '__main__':

    employees = ('Matt Wilson', 'Matt Wilson', 'Matt Wilson')
    data = [(emp, '6:00 AM') for emp in employees]

    tfun('Food Service Schedule', datetime.now().date(),
        data)


-- 
Matthew Wilson
matt at tplus1.com
http://tplus1.com

From paul.barrass at safeonlinebilling.com  Wed Nov 19 11:44:09 2008
From: paul.barrass at safeonlinebilling.com (Paul Barrass)
Date: Wed, 19 Nov 2008 16:44:09 +0000
Subject: [reportlab-users] How do I left-align my table?
In-Reply-To: <f6f643a20811190836i3d1ee793wa6a21c56033af25f@mail.gmail.com>
References: <f6f643a20811190836i3d1ee793wa6a21c56033af25f@mail.gmail.com>
Message-ID: <49244259.5050807@safeonlinebilling.com>

Matthew Wilson wrote:
> I have a small table and it is displayed in the center of the page.
>
> Here's the code I'm using:
>
> def tfun(title, dt, data):
>
>     t = Table(data)
>
> <snip>
>   

t.hAlign='LEFT' is what you want.

Paul Barrass.

From matt at tplus1.com  Wed Nov 19 11:48:29 2008
From: matt at tplus1.com (Matthew Wilson)
Date: Wed, 19 Nov 2008 11:48:29 -0500
Subject: [reportlab-users] How do I left-align my table?
In-Reply-To: <49244259.5050807@safeonlinebilling.com>
References: <f6f643a20811190836i3d1ee793wa6a21c56033af25f@mail.gmail.com>
	<49244259.5050807@safeonlinebilling.com>
Message-ID: <f6f643a20811190848w27ec0d3bo39e2ad9473df1550@mail.gmail.com>

Thanks Paul!  That was exactly right.

On Wed, Nov 19, 2008 at 11:44 AM, Paul Barrass
<paul.barrass at safeonlinebilling.com> wrote:
> Matthew Wilson wrote:
>>
>> I have a small table and it is displayed in the center of the page.
>>
>> Here's the code I'm using:
>>
>> def tfun(title, dt, data):
>>
>>    t = Table(data)
>>
>> <snip>
>>
>
> t.hAlign='LEFT' is what you want.
>
> Paul Barrass.
> _______________________________________________
> reportlab-users mailing list
> reportlab-users at reportlab.com
> http://two.pairlist.net/mailman/listinfo/reportlab-users
>



-- 
Matthew Wilson
matt at tplus1.com
http://tplus1.com

From Samuele.Kaplun at cern.ch  Wed Nov 19 12:29:10 2008
From: Samuele.Kaplun at cern.ch (Samuele Kaplun)
Date: Wed, 19 Nov 2008 18:29:10 +0100
Subject: [reportlab-users] pyRXP sources
Message-ID: <200811191829.10494.Samuele.Kaplun@cern.ch>

Dear ReportLab team,
I'm interested in your pyRXP project but I wasn't able to retrieve it from 
your website. Around I found inconsistent release numbers (1.12, 1.09...) 
while on your website you refer to version 0.9. The svn tree seems to not 
contain it anymore.
Is the project abandoned? Is there a way to retrieve the last valid version 
ever?

Best regards,
	Samuele
-- 
.O.
..O
OOO

From andy at reportlab.com  Wed Nov 19 12:36:13 2008
From: andy at reportlab.com (Andy Robinson)
Date: Wed, 19 Nov 2008 17:36:13 +0000
Subject: [reportlab-users] pyRXP sources
In-Reply-To: <200811191829.10494.Samuele.Kaplun@cern.ch>
References: <200811191829.10494.Samuele.Kaplun@cern.ch>
Message-ID: <956003ae0811190936l35bb8f64tf5dead2166194cdc@mail.gmail.com>

2008/11/19 Samuele Kaplun <Samuele.Kaplun at cern.ch>:
> Dear ReportLab team,
> I'm interested in your pyRXP project but I wasn't able to retrieve it from
> your website. Around I found inconsistent release numbers (1.12, 1.09...)
> while on your website you refer to version 0.9. The svn tree seems to not
> contain it anymore.
> Is the project abandoned? Is there a way to retrieve the last valid version
> ever?

We use it daily in production in many applications, but we haven't updated
its web page properly in recent months.

It's mature with no changes in a long time, and definitely production quality.

The latest code is at
     http://svn.reportlab.com/svn/public/pyRXP/trunk/

Best Regards

-- 
Andy Robinson
CEO/Chief Architect
ReportLab Europe Ltd.
Media House, 3 Palmerston Road, Wimbledon, London SW19 1PG, UK
Tel +44-20-8545-1570

From gabor at nekomancer.net  Thu Nov 20 02:56:12 2008
From: gabor at nekomancer.net (=?ISO-8859-1?Q?G=E1bor_Farkas?=)
Date: Thu, 20 Nov 2008 08:56:12 +0100
Subject: [reportlab-users] cjk-word-wrapping and xml-character-entities
In-Reply-To: <956003ae0811190558m9667e3esb1c24421d0d81f4d@mail.gmail.com>
References: <2c173a350811190531v32f71fu90243b75e1b3b464@mail.gmail.com>
	<956003ae0811190558m9667e3esb1c24421d0d81f4d@mail.gmail.com>
Message-ID: <2c173a350811192356k5066c933y1f0950d1a5fd75ae@mail.gmail.com>

On Wed, Nov 19, 2008 at 2:58 PM, Andy Robinson <andy at reportlab.com> wrote:
> 2008/11/19 G?bor Farkas <gabor at nekomancer.net>:
>> 1. the CJK algorithm only handles paragraps with 1 ParaFrags
>> 2. the ParaParser, when parses the text, creates 2 frags because of
>> the char-entity
>
> Thanks for such an easy-to-reproduce example!
>
> We fixed this several months ago - I believe before the last release 2.2.
> Your script runs fine for me against SVN trunk.  What version are you using?

thanks..... yes, it works in 2.2 (i was using 2.1)

gabor

From stephane.travostino at gmail.com  Thu Nov 20 09:29:43 2008
From: stephane.travostino at gmail.com (=?UTF-8?Q?St=C3=A9phane_Travostino?=)
Date: Thu, 20 Nov 2008 15:29:43 +0100
Subject: [reportlab-users] Bug when drawing human readable barcodes
Message-ID: <a4f9daf00811200629i5ca077f1q5853fc6bf96e3139@mail.gmail.com>

Hello,
I spotted a bug in the barcode addon, and specifically when trying to draw a
human readable barcode.
Consider this simple example:

from reportlab.pdfgen import canvas
from reportlab.lib.pagesizes import A4
from reportlab.lib.units import mm
from reportlab.graphics.barcode import code39

TEXT = "TEST01L4"

c=canvas.Canvas("barcode.pdf",pagesize=A4)
barcode=code39.Standard39(TEXT, barWidth=0.423*mm, barHeight=10*mm,
ratio=2.75, humanReadable=True)
barcode.drawOn(c,50*mm,50*mm)
c.showPage()
c.save()

When run on Ubuntu (python-reportlab 2.1 from intrepid repos and 2.2
imported from debian sid), I get this backtrace:
...
  File "/usr/lib/python2.5/site-packages/reportlab/pdfbase/pdfmetrics.py",
line 399, in _py_stringWidth
    if not isinstance(text,unicode): text = text.decode(encoding)
AttributeError: 'float' object has no attribute 'decode'


It seems that somewhere the "text" variable is substituted with a float
value. Here is the patch for this issue (diffed from SVN
https://svn.reportlab.com/svn/public/reportlab/trunk/src/reportlab/graphics/barcode/common.py
):
$ diff -ruN common.py
/usr/lib/python2.5/site-packages/reportlab/graphics/barcode/common.py
--- common.py    2007-04-13 14:32:39.000000000 +0200
+++ /usr/lib/python2.5/site-packages/reportlab/graphics/barcode/common.py
2008-11-20 15:19:38.000000000 +0100
@@ -173,7 +173,7 @@
         if anchor=='middle': func = 'drawCentredString'
         elif anchor=='end': func = 'drawRightString'
         else: func = 'drawString'
-        getattr(canv,func)(text,x,y)
+        getattr(canv,func)(x,y,text)
         canv.restoreState()

 class MultiWidthBarcode(Barcode):

St?phane.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/reportlab-users/attachments/20081120/6d6a59e8/attachment.html>

From stephane.travostino at gmail.com  Thu Nov 20 09:59:02 2008
From: stephane.travostino at gmail.com (=?UTF-8?Q?St=C3=A9phane_Travostino?=)
Date: Thu, 20 Nov 2008 15:59:02 +0100
Subject: [reportlab-users] Barcode checksum printed in human readable format
Message-ID: <a4f9daf00811200659he4dcd90w298609bb6f61e9ea@mail.gmail.com>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello,
sorry for pasting the same example as in the previous mail.
When a Code39 barcode is created with the humanReadable variable set, the
text is printed including the checksum.
The expected behaviour is to see the text without the checksum.

from reportlab.pdfgen import canvas
from reportlab.lib.pagesizes import A4
from reportlab.lib.units import mm
from reportlab.graphics.barcode import code39

TEXT = "TEST01L4"

c=canvas.Canvas("barcode.pdf",pagesize=A4)
barcode=code39.Standard39(TEXT, barWidth=0.423*mm, barHeight=10*mm,
ratio=2.75, humanReadable=True)
barcode.drawOn(c,50*mm,50*mm)
c.showPage()
c.save()

You'll see that the checksum is printed with the text.
Here is the patch to solve this issue, diffed from SVN
http://svn.reportlab.com/svn/public/reportlab/community/trunk/src/reportlab/graphics/barcode/code39.py
:
- --- code39.py    2006-05-04 12:48:36.000000000 +0200
+++
/usr/lib/python2.5/site-packages/reportlab/graphics/barcode/code39.py
2008-11-20 15:51:12.000000000 +0100
@@ -138,7 +138,8 @@
         return self.decomposed

     def _humanText(self):
- -        return self.stop and
self.encoded[1:-1] or self.encoded
+        text = self.stop and
self.encoded[1:-1] or self.encoded
+        return self.checksum and
text[:-1] or text

 class Standard39(_Code39Base):
     """

Stephane
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/reportlab-users/attachments/20081120/3417e37f/attachment.html>

From matt at tplus1.com  Thu Nov 20 21:54:40 2008
From: matt at tplus1.com (Matthew Wilson)
Date: Thu, 20 Nov 2008 21:54:40 -0500
Subject: [reportlab-users] How can I set the width of a table?
Message-ID: <f6f643a20811201854t149c61f9yaae4530753fd7613@mail.gmail.com>

I'm drawing several tables and their width seems to be dynamic based
on the contents.  It looks goofy.  How can I require that every table
draws with a fixed width?

Matt

-- 
Matthew Wilson
matt at tplus1.com
http://tplus1.com

From paul.barrass at safeonlinebilling.com  Thu Nov 20 22:00:04 2008
From: paul.barrass at safeonlinebilling.com (Paul Barrass)
Date: Fri, 21 Nov 2008 03:00:04 +0000
Subject: [reportlab-users] How can I set the width of a table?
In-Reply-To: <f6f643a20811201854t149c61f9yaae4530753fd7613@mail.gmail.com>
References: <f6f643a20811201854t149c61f9yaae4530753fd7613@mail.gmail.com>
Message-ID: <49262434.9040100@safeonlinebilling.com>

Matthew Wilson wrote:
> I'm drawing several tables and their width seems to be dynamic based
> on the contents.  It looks goofy.  How can I require that every table
> draws with a fixed width?
You can pass a colWidths parameter (a list of widths which must be the 
same length as the number of columns) when you create the table if you 
want  fix the columns widths for each table.  I'm not sure it's possible 
to specify a fixed total width, but with dynamically assigned widths of 
the individual columns, though someone may know different.

Paul Barrass.

From Samuele.Kaplun at cern.ch  Fri Nov 21 08:41:29 2008
From: Samuele.Kaplun at cern.ch (Samuele Kaplun)
Date: Fri, 21 Nov 2008 14:41:29 +0100
Subject: [reportlab-users] pyRXP sources and question about potential
	memory leak?
In-Reply-To: <956003ae0811190936l35bb8f64tf5dead2166194cdc@mail.gmail.com>
References: <200811191829.10494.Samuele.Kaplun@cern.ch>
	<956003ae0811190936l35bb8f64tf5dead2166194cdc@mail.gmail.com>
Message-ID: <200811211441.30729.Samuele.Kaplun@cern.ch>

Dear Andy,

Il Wednesday 19 November 2008 18:36:13 Andy Robinson ha scritto:
> We use it daily in production in many applications, but we haven't updated
> its web page properly in recent months.
>
> It's mature with no changes in a long time, and definitely production
> quality.
>
> The latest code is at
>      http://svn.reportlab.com/svn/public/pyRXP/trunk/

thanks a lot for the quick answer and for the URL. Glad to discover that pyRXP 
has not disappeared :-)

BTW in the project I'm involved in we use it as a part of one of our core 
library, and we're experiencing a strange behaviour: as part of our tools we 
instantiate an instance of the parser, and we parse thousands of snippets of 
XML, one at a time via the parse method of this same pyRXP.Parser instance. 
We keep the same instance in order to avoid latency in the initialization. 
However, this procedure is getting slower and slower each time a snipped is 
parsed. Could it be that pyRXP has a memory leak or is keeping permanent 
information although these will not be used in the next snippet parsing?

Best regards,
	Samuele

-- 
.O.
..O
OOO

From andy at reportlab.com  Fri Nov 21 08:49:54 2008
From: andy at reportlab.com (Andy Robinson)
Date: Fri, 21 Nov 2008 13:49:54 +0000
Subject: [reportlab-users] pyRXP sources and question about potential
	memory leak?
In-Reply-To: <200811211441.30729.Samuele.Kaplun@cern.ch>
References: <200811191829.10494.Samuele.Kaplun@cern.ch>
	<956003ae0811190936l35bb8f64tf5dead2166194cdc@mail.gmail.com>
	<200811211441.30729.Samuele.Kaplun@cern.ch>
Message-ID: <956003ae0811210549k18ecc35dy573a454cccb04a5d@mail.gmail.com>

2008/11/21 Samuele Kaplun <Samuele.Kaplun at cern.ch>:
> parsed. Could it be that pyRXP has a memory leak or is keeping permanent
> information although these will not be used in the next snippet parsing?
>

I did find one memory leak a couple of years ago.  This was in the
callback to locate and parse a DTD; it only happened when parsing XML
files with a DTD reference at the top, and we fixed it.  Have you
tried with the latest SVN code?

If you still see the problem,  can you send us a script which shows
the leak (e.g. parse your XML document a few thousand times) and we
will look at it quickly? ... any bugs might bite us one day as well!


Best Regards,

-- 
Andy Robinson
CEO/Chief Architect
ReportLab Europe Ltd.
Media House, 3 Palmerston Road, Wimbledon, London SW19 1PG, UK
Tel +44-20-8545-1570

From Samuele.Kaplun at cern.ch  Fri Nov 21 10:33:00 2008
From: Samuele.Kaplun at cern.ch (Samuele Kaplun)
Date: Fri, 21 Nov 2008 16:33:00 +0100
Subject: [reportlab-users] pyRXP sources and question about potential
	memory leak?
In-Reply-To: <956003ae0811210549k18ecc35dy573a454cccb04a5d@mail.gmail.com>
References: <200811191829.10494.Samuele.Kaplun@cern.ch>
	<200811211441.30729.Samuele.Kaplun@cern.ch>
	<956003ae0811210549k18ecc35dy573a454cccb04a5d@mail.gmail.com>
Message-ID: <200811211633.00889.Samuele.Kaplun@cern.ch>

Dear Andy,

Il Friday 21 November 2008 14:49:54 Andy Robinson ha scritto:
> I did find one memory leak a couple of years ago.  This was in the
> callback to locate and parse a DTD; it only happened when parsing XML
> files with a DTD reference at the top, and we fixed it.  Have you
> tried with the latest SVN code?
>
> If you still see the problem,  can you send us a script which shows
> the leak (e.g. parse your XML document a few thousand times) and we
> will look at it quickly? ... any bugs might bite us one day as well!

great! It was this!! On our production server we were still using version 
pyRXP-1.12 (from 2006-11-04) so you were even perfectly right with your guess 
of a couple of years ago! I've tested by removing the DTD line and it doesn't 
take a linear amount of time (WRT the number of parsing operation) anymore!!

Thanks a lot! 
We definitively need to upgrade our installation...

Best regards,
	Sam

Ps infact I wasn't able to reproduce this on my testing machine, since I was 
using the latest SVN version...


-- 
.O.
..O
OOO

From dirk.holtwick at gmail.com  Fri Nov 21 11:14:30 2008
From: dirk.holtwick at gmail.com (Dirk Holtwick)
Date: Fri, 21 Nov 2008 17:14:30 +0100
Subject: [reportlab-users] ANNOUNCE pisa 3.0.28
Message-ID: <4926DE66.3040501@gmail.com>

Hi,

I just published a new version of Pisa/XHTML2PDF. You can get it here:

	http://pypi.python.org/pypi/pisa/
	http://www.xhtml2pdf.com/

So this could also be interesting for the Reportlab community since it 
includes a modified Paragraph implementation that is able to handle 
inline background colors.

Another interesting feature for your PDF development could be the visual 
testing feature that uses ImageMagick and TortoiseIDiff to show you if 
two PDF differ from each other. This could be increasing the quality of 
your code if used in your unit tests.

Enjoy it.

Dirk

From b.aldemir at e-evolution.de  Mon Nov 24 05:58:33 2008
From: b.aldemir at e-evolution.de (=?ISO-8859-1?Q?B=FClent_Aldemir?=)
Date: Mon, 24 Nov 2008 11:58:33 +0100
Subject: [reportlab-users] Using EPS Files / PDF Images in canvas
Message-ID: <c036d95e0811240258p6e61879dt97aaa63693069ecc@mail.gmail.com>

Hi,

I have googled but I found no solution for embedding EPS files or PDF
images in a canvas.

Is this possible?

Greetings
B?lent

From andy at reportlab.com  Mon Nov 24 06:03:40 2008
From: andy at reportlab.com (Andy Robinson)
Date: Mon, 24 Nov 2008 11:03:40 +0000
Subject: [reportlab-users] Using EPS Files / PDF Images in canvas
In-Reply-To: <c036d95e0811240258p6e61879dt97aaa63693069ecc@mail.gmail.com>
References: <c036d95e0811240258p6e61879dt97aaa63693069ecc@mail.gmail.com>
Message-ID: <956003ae0811240303q72fcae43rae3aef2ffd21a2b9@mail.gmail.com>

2008/11/24 B?lent Aldemir <b.aldemir at e-evolution.de>:
> Hi,
>
> I have googled but I found no solution for embedding EPS files or PDF
> images in a canvas.
>
> Is this possible?


It needs our commercial products - Report Markup Language can
basically treat PDF files as images and position them in many ways.
The open source code lacks a PDF or EPS parser and can't do this.

If you'd like a quotation for the commercial libraries, please email
me directly and let me know the intended use and rough volumes of PDFs
you would be producing.

Best Regards,


-- 
Andy Robinson
CEO/Chief Architect
ReportLab Europe Ltd.
Media House, 3 Palmerston Road, Wimbledon, London SW19 1PG, UK
Tel +44-20-8545-1570

From ralsina at netmanagers.com.ar  Mon Nov 24 06:22:38 2008
From: ralsina at netmanagers.com.ar (Roberto Alsina)
Date: Mon, 24 Nov 2008 09:22:38 -0200
Subject: [reportlab-users] Using EPS Files / PDF Images in canvas
In-Reply-To: <c036d95e0811240258p6e61879dt97aaa63693069ecc@mail.gmail.com>
References: <c036d95e0811240258p6e61879dt97aaa63693069ecc@mail.gmail.com>
Message-ID: <200811240922.48816.ralsina@netmanagers.com.ar>

On Monday 24 November 2008 08:58:33 B?lent Aldemir wrote:
> Hi,
>
> I have googled but I found no solution for embedding EPS files or PDF
> images in a canvas.
>
> Is this possible?

You can convert EPS to SVG and then use Dinu's svglib, or you can see if 
uniconvertor can handle the specific file and use my SVGImage flowable (don't 
be fooled by the name, it can handle some EPS files)

My flowable is just these two files (and uniconvertor)

http://code.google.com/p/rst2pdf/source/browse/trunk/rst2pdf/svgimage.py
http://code.google.com/p/rst2pdf/source/browse/trunk/rst2pdf/uniconvsaver.py

-- 
 ("\''/").__..-''"`-. .         Roberto Alsina
 `9_ 9  )   `-. (    ).`-._.`)  KDE Developer (MFCH)
 (_Y_.)' ._   ) `._`.  " -.-'  http://lateral.netmanagers.com.ar 
  _..`-'_..-_/ /-'_.'     The 6,855th most popular site of Slovenia   
(l)-'' ((i).' ((!.'       according to alexa.com (27/5/2007) 
                    
Debugging is twice as hard as writing the code in the first place. 
Therefore, if you write the code as cleverly as possible, you are, 
by definition, not smart enough to debug it. --Brian W. Kernighan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/reportlab-users/attachments/20081124/7db8aa2a/attachment.html>

From b.aldemir at e-evolution.de  Mon Nov 24 06:28:22 2008
From: b.aldemir at e-evolution.de (=?ISO-8859-1?Q?B=FClent_Aldemir?=)
Date: Mon, 24 Nov 2008 12:28:22 +0100
Subject: [reportlab-users] Using EPS Files / PDF Images in canvas
In-Reply-To: <200811240922.48816.ralsina@netmanagers.com.ar>
References: <c036d95e0811240258p6e61879dt97aaa63693069ecc@mail.gmail.com>
	<200811240922.48816.ralsina@netmanagers.com.ar>
Message-ID: <c036d95e0811240328r612bd7fftd0273735c32d5086@mail.gmail.com>

Thanks Roberto. I will give it a try! If it's so simple, I think it will
work for me ;-))

Greetings B?lent

2008/11/24 Roberto Alsina <ralsina at netmanagers.com.ar>

>  On Monday 24 November 2008 08:58:33 B?lent Aldemir wrote:
>
> > Hi,
>
> >
>
> > I have googled but I found no solution for embedding EPS files or PDF
>
> > images in a canvas.
>
> >
>
> > Is this possible?
>
> You can convert EPS to SVG and then use Dinu's svglib, or you can see if
> uniconvertor can handle the specific file and use my SVGImage flowable
> (don't be fooled by the name, it can handle some EPS files)
>
> My flowable is just these two files (and uniconvertor)
>
> http://code.google.com/p/rst2pdf/source/browse/trunk/rst2pdf/svgimage.py
>
>
> http://code.google.com/p/rst2pdf/source/browse/trunk/rst2pdf/uniconvsaver.py
>
> --
>
> ("\''/").__..-''"`-. . Roberto Alsina
>
> `9_ 9 ) `-. ( ).`-._.`) KDE Developer (MFCH)
>
> (_Y_.)' ._ ) `._`. " -.-' http://lateral.netmanagers.com.ar
>
> _..`-'_..-_/ /-'_.' The 6,855th most popular site of Slovenia
>
> (l)-'' ((i).' ((!.' according to alexa.com (27/5/2007)
>
> Debugging is twice as hard as writing the code in the first place.
>
> Therefore, if you write the code as cleverly as possible, you are,
>
> by definition, not smart enough to debug it. --Brian W. Kernighan
>
> _______________________________________________
> reportlab-users mailing list
> reportlab-users at reportlab.com
> http://two.pairlist.net/mailman/listinfo/reportlab-users
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/reportlab-users/attachments/20081124/69d65d68/attachment.html>

From gherman at darwin.in-berlin.de  Mon Nov 24 06:51:56 2008
From: gherman at darwin.in-berlin.de (Dinu Gherman)
Date: Mon, 24 Nov 2008 12:51:56 +0100
Subject: [reportlab-users] Using EPS Files / PDF Images in canvas
In-Reply-To: <c036d95e0811240328r612bd7fftd0273735c32d5086@mail.gmail.com>
References: <c036d95e0811240258p6e61879dt97aaa63693069ecc@mail.gmail.com>
	<200811240922.48816.ralsina@netmanagers.com.ar>
	<c036d95e0811240328r612bd7fftd0273735c32d5086@mail.gmail.com>
Message-ID: <A1C61B69-6519-4B8F-9F1F-996C1937C643@darwin.in-berlin.de>

B?lent Aldemir:

> Thanks Roberto. I will give it a try! If it's so simple,
> I think it will work for me ;-))

As Roberto said, if you have SVG you can use my free svglib
[1] which has a few hundred test cases. I found it to con-
sistenty outperform uniconvertor for SVG import.

I also have a direct PDF import tool in the making, but
it's not available yet, unless you want to make financial
contributions to boost its development. I posted some ex-
amples here that you can find in the archives on this list.

Regards,

Dinu

[1] http://pypi.python.org/pypi/svglib


From b.aldemir at e-evolution.de  Mon Nov 24 07:22:24 2008
From: b.aldemir at e-evolution.de (=?ISO-8859-1?Q?B=FClent_Aldemir?=)
Date: Mon, 24 Nov 2008 13:22:24 +0100
Subject: [reportlab-users] Using EPS Files / PDF Images in canvas
In-Reply-To: <A1C61B69-6519-4B8F-9F1F-996C1937C643@darwin.in-berlin.de>
References: <c036d95e0811240258p6e61879dt97aaa63693069ecc@mail.gmail.com>
	<200811240922.48816.ralsina@netmanagers.com.ar>
	<c036d95e0811240328r612bd7fftd0273735c32d5086@mail.gmail.com>
	<A1C61B69-6519-4B8F-9F1F-996C1937C643@darwin.in-berlin.de>
Message-ID: <c036d95e0811240422jca5987bgdea8be46ea1aa8b7@mail.gmail.com>

> As Roberto said, if you have SVG you can use my free svglib
> [1] which has a few hundred test cases. I found it to con-
> sistenty outperform uniconvertor for SVG import.
>
 Actually I was reading your presentation [2]. Are the links [3-4] on your
website outdated too?

I will have a look on the test cases

:-))) B?lent

[1] http://pypi.python.org/pypi/svglib
 [2] http://python.net/~gherman/presentations/rlgsvg-epc2002.pdf
[3] http://python.net/~gherman/projects/svglib/<http://python.net/~gherman/svglib.html>
[4] http://python.net/~gherman/svglib.html
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/reportlab-users/attachments/20081124/5342b161/attachment-0001.htm>

From gherman at darwin.in-berlin.de  Mon Nov 24 07:27:00 2008
From: gherman at darwin.in-berlin.de (Dinu Gherman)
Date: Mon, 24 Nov 2008 13:27:00 +0100
Subject: [reportlab-users] Using EPS Files / PDF Images in canvas
In-Reply-To: <c036d95e0811240422jca5987bgdea8be46ea1aa8b7@mail.gmail.com>
References: <c036d95e0811240258p6e61879dt97aaa63693069ecc@mail.gmail.com>
	<200811240922.48816.ralsina@netmanagers.com.ar>
	<c036d95e0811240328r612bd7fftd0273735c32d5086@mail.gmail.com>
	<A1C61B69-6519-4B8F-9F1F-996C1937C643@darwin.in-berlin.de>
	<c036d95e0811240422jca5987bgdea8be46ea1aa8b7@mail.gmail.com>
Message-ID: <10DA915F-BAC3-4161-B221-7696D06777A3@darwin.in-berlin.de>

B?lent Aldemir:

> Actually I was reading your presentation [2]. Are the links [3-4] on  
> your website outdated too?


You'll find all my new stuff on www.python.org/pypi or [1].
Svglib has been updated to 0.6.2.

Regards,

Dinu

[1] http://www.dinu-gherman.net/tmp

From gherman at darwin.in-berlin.de  Mon Nov 24 09:30:30 2008
From: gherman at darwin.in-berlin.de (Dinu Gherman)
Date: Mon, 24 Nov 2008 15:30:30 +0100
Subject: [reportlab-users] Future of py2pdf
In-Reply-To: <956003ae0811190337h4c01cf1cn87f0cef70610fb12@mail.gmail.com>
References: <3A314D7C-5CBD-4606-94C6-5308CACF7F28@darwin.in-berlin.de>
	<956003ae0811190337h4c01cf1cn87f0cef70610fb12@mail.gmail.com>
Message-ID: <3081AA8C-F87D-4B62-8E20-1BA0DE4108AE@darwin.in-berlin.de>

Andy Robinson:

> It's your call.  Just let us know.     We'll be setting up a page of
> links to related packages in December anyway.  It would be
> good to decide for sure this month, though.

Ok, then please remove py2pdf from the distribution! I'll
provide later a link to the seperate replacement that you
can point to on the reportlab.org website.

Thanks,

Dinu


From andy at reportlab.com  Mon Nov 24 10:31:34 2008
From: andy at reportlab.com (Andy Robinson)
Date: Mon, 24 Nov 2008 15:31:34 +0000
Subject: [reportlab-users] Future of py2pdf
In-Reply-To: <3081AA8C-F87D-4B62-8E20-1BA0DE4108AE@darwin.in-berlin.de>
References: <3A314D7C-5CBD-4606-94C6-5308CACF7F28@darwin.in-berlin.de>
	<956003ae0811190337h4c01cf1cn87f0cef70610fb12@mail.gmail.com>
	<3081AA8C-F87D-4B62-8E20-1BA0DE4108AE@darwin.in-berlin.de>
Message-ID: <956003ae0811240731t3aafe92s42ee35fd2a6bd3a8@mail.gmail.com>

2008/11/24 Dinu Gherman <gherman at darwin.in-berlin.de>:
> Andy Robinson:
>
>> It's your call.  Just let us know.     We'll be setting up a page of
>> links to related packages in December anyway.  It would be
>> good to decide for sure this month, though.
>
> Ok, then please remove py2pdf from the distribution! I'll
> provide later a link to the seperate replacement that you
> can point to on the reportlab.org website.
>

Committed now.    We'll probably move the other things in
'tools' shortly - Pythonpoint can go to 'demos', and the other
bits are small enough to go into 'lib'.  Less top level directories
makes everything easier to understand.

- Andy

From mpeters42 at gmail.com  Mon Nov 24 13:09:12 2008
From: mpeters42 at gmail.com (Mark Peters)
Date: Mon, 24 Nov 2008 13:09:12 -0500
Subject: [reportlab-users] Using EPS Files / PDF Images in canvas
In-Reply-To: <c036d95e0811240258p6e61879dt97aaa63693069ecc@mail.gmail.com>
References: <c036d95e0811240258p6e61879dt97aaa63693069ecc@mail.gmail.com>
Message-ID: <25a37b830811241009y695f009ctbfd93d685a12e1ac@mail.gmail.com>

I have recently resurrected an older project, initially started as a
proposal for the first Google Summer of Code project.  In the last few
months, I've refactored most of the code and have something I'd call alpha
functionality.  Right now, it handles most EPS clipart I've thrown at it,
but I still have some work to do getting embedded fonts and subsets passed
along to the PDF.

See my original proposal here: http://ivanhouse.com/gsc/proposal.pdf

What kind of EPS files do you want to include?  If you send me an example
via email, I can run it through what I have now and let you know if there
are any problems.

Once I get some more time to thow at this project and polish it up some
more, my plan is to submit a patch.

Mark Peters

On Mon, Nov 24, 2008 at 5:58 AM, B?lent Aldemir <b.aldemir at e-evolution.de>wrote:

> Hi,
>
> I have googled but I found no solution for embedding EPS files or PDF
> images in a canvas.
>
> Is this possible?
>
> Greetings
> B?lent
> _______________________________________________
> reportlab-users mailing list
> reportlab-users at reportlab.com
> http://two.pairlist.net/mailman/listinfo/reportlab-users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/reportlab-users/attachments/20081124/6c2b8d9f/attachment.html>

From matt at tplus1.com  Mon Nov 24 18:33:21 2008
From: matt at tplus1.com (Matthew Wilson)
Date: Mon, 24 Nov 2008 18:33:21 -0500
Subject: [reportlab-users] Why isn't my really wide table splitting?
Message-ID: <f6f643a20811241533i6a377f09k5d4b26db81b277b9@mail.gmail.com>

I want my really wide table to wrap, but instead, it just runs off the
edge of the PDF.

This is how I'm building it:

from reportlab.platypus import Table, SimpleDocTemplate
row = range(1, 101)
data = [row, row, row]
doc = SimpleDocTemplate('x.pdf')
doc.build([Table(data)])

And when I look at the PDF, it doesn't wrap.  Instead, I just see the middle.

What am I doing wrong?

-- 
Matthew Wilson
matt at tplus1.com
http://tplus1.com

From bfreeman at appropriatesolutions.com  Mon Nov 24 19:23:04 2008
From: bfreeman at appropriatesolutions.com (Bill Freeman)
Date: Mon, 24 Nov 2008 19:23:04 -0500
Subject: [reportlab-users] Why isn't my really wide table splitting?
In-Reply-To: <f6f643a20811241533i6a377f09k5d4b26db81b277b9@mail.gmail.com>
References: <f6f643a20811241533i6a377f09k5d4b26db81b277b9@mail.gmail.com>
Message-ID: <492B4568.5080608@appropriatesolutions.com>

Matthew Wilson wrote:
> I want my really wide table to wrap, but instead, it just runs off the
> edge of the PDF.
>
> This is how I'm building it:
>
> from reportlab.platypus import Table, SimpleDocTemplate
> row = range(1, 101)
> data = [row, row, row]
> doc = SimpleDocTemplate('x.pdf')
> doc.build([Table(data)])
>
> And when I look at the PDF, it doesn't wrap.  Instead, I just see the middle.
>
> What am I doing wrong?
>
>   
Last time I looked at the code (maybe 3 months ago) there was no code 
for splitting horizontally.
(A nice variable to say what you preferred, but only vertical splitting 
code.)

Bill


From andy at reportlab.com  Tue Nov 25 02:24:59 2008
From: andy at reportlab.com (Andy Robinson)
Date: Tue, 25 Nov 2008 07:24:59 +0000
Subject: [reportlab-users] Why isn't my really wide table splitting?
In-Reply-To: <492B4568.5080608@appropriatesolutions.com>
References: <f6f643a20811241533i6a377f09k5d4b26db81b277b9@mail.gmail.com>
	<492B4568.5080608@appropriatesolutions.com>
Message-ID: <956003ae0811242324v6acf2e9s9eccae7d394f9030@mail.gmail.com>

2008/11/25 Bill Freeman <bfreeman at appropriatesolutions.com>:
> Matthew Wilson wrote:
>> I want my really wide table to wrap, but instead, it just runs off the
>> edge of the PDF.
> Last time I looked at the code (maybe 3 months ago) there was no code for
> splitting horizontally.
Bill's right.  Although there is some old and probably unmaintained
code from years back to help a programmer explicitly split tables
horizontally, the main Platypus algorithm assumes all your content
fits within the width of your frame, and asks them to split
vertically.

IMHO it's much easier for programmers to cut up their data into chunks
in Python and generate a sequence of smaller tables than it is for us
to split tables in 2 dimensions and correctly take care of all the box
and border styling rules.

Sorry,

-- 
Andy Robinson
CEO/Chief Architect
ReportLab Europe Ltd.
Media House, 3 Palmerston Road, Wimbledon, London SW19 1PG, UK
Tel +44-20-8545-1570

From matt at tplus1.com  Tue Nov 25 07:13:15 2008
From: matt at tplus1.com (Matthew Wilson)
Date: Tue, 25 Nov 2008 07:13:15 -0500
Subject: [reportlab-users] Why isn't my really wide table splitting?
In-Reply-To: <956003ae0811242324v6acf2e9s9eccae7d394f9030@mail.gmail.com>
References: <f6f643a20811241533i6a377f09k5d4b26db81b277b9@mail.gmail.com>
	<492B4568.5080608@appropriatesolutions.com>
	<956003ae0811242324v6acf2e9s9eccae7d394f9030@mail.gmail.com>
Message-ID: <f6f643a20811250413obcaf45fqbf4b81925fa4ff59@mail.gmail.com>

That's cool -- but I would really like to request that the
documentation make that clearer.

Can I submit a patch with a change to the docs?

Matt

On Tue, Nov 25, 2008 at 2:24 AM, Andy Robinson <andy at reportlab.com> wrote:
> 2008/11/25 Bill Freeman <bfreeman at appropriatesolutions.com>:
>> Matthew Wilson wrote:
>>> I want my really wide table to wrap, but instead, it just runs off the
>>> edge of the PDF.
>> Last time I looked at the code (maybe 3 months ago) there was no code for
>> splitting horizontally.
> Bill's right.  Although there is some old and probably unmaintained
> code from years back to help a programmer explicitly split tables
> horizontally, the main Platypus algorithm assumes all your content
> fits within the width of your frame, and asks them to split
> vertically.
>
> IMHO it's much easier for programmers to cut up their data into chunks
> in Python and generate a sequence of smaller tables than it is for us
> to split tables in 2 dimensions and correctly take care of all the box
> and border styling rules.
>
> Sorry,
>
> --
> Andy Robinson
> CEO/Chief Architect
> ReportLab Europe Ltd.
> Media House, 3 Palmerston Road, Wimbledon, London SW19 1PG, UK
> Tel +44-20-8545-1570
> _______________________________________________
> reportlab-users mailing list
> reportlab-users at reportlab.com
> http://two.pairlist.net/mailman/listinfo/reportlab-users
>



-- 
Matthew Wilson
matt at tplus1.com
http://tplus1.com

From andy at reportlab.com  Tue Nov 25 07:44:28 2008
From: andy at reportlab.com (Andy Robinson)
Date: Tue, 25 Nov 2008 12:44:28 +0000
Subject: [reportlab-users] Why isn't my really wide table splitting?
In-Reply-To: <f6f643a20811250413obcaf45fqbf4b81925fa4ff59@mail.gmail.com>
References: <f6f643a20811241533i6a377f09k5d4b26db81b277b9@mail.gmail.com>
	<492B4568.5080608@appropriatesolutions.com>
	<956003ae0811242324v6acf2e9s9eccae7d394f9030@mail.gmail.com>
	<f6f643a20811250413obcaf45fqbf4b81925fa4ff59@mail.gmail.com>
Message-ID: <956003ae0811250444l4f4cec4ane78afa4676efe167@mail.gmail.com>

2008/11/25 Matthew Wilson <matt at tplus1.com>:
> That's cool -- but I would really like to request that the
> documentation make that clearer.
>
> Can I submit a patch with a change to the docs?
>

Yes please!  We love such patches and will commit it very quickly
if it looks reasonable ;-)

- Andy

From dirk.holtwick at gmail.com  Wed Nov 26 12:41:10 2008
From: dirk.holtwick at gmail.com (Dirk Holtwick)
Date: Wed, 26 Nov 2008 18:41:10 +0100
Subject: [reportlab-users] BUGFIX: Re:  &nbsp; in paragraph
In-Reply-To: <956003ae0810060450m39d20d74pb6e15b9e92e97ff3@mail.gmail.com>
References: <48E778F6.3010601@gmail.com>
	<956003ae0810060450m39d20d74pb6e15b9e92e97ff3@mail.gmail.com>
Message-ID: <492D8A36.3070302@gmail.com>

Hi,

to fix the described error please modify the following function in 
"paragraph.py":

-----------------8<---------------[cut here]
#on UTF8 branch, split and strip must be unicode-safe!
def split(text, delim=None):
     if type(text) is str: text = text.decode('utf8')
     if type(delim) is str: delim = delim.decode('utf8')
     # This fixes &nbsp; issue and multiple linebraks on splitted page part
     if delim is None and text == u'\xa0':
         delim = ' '
     return [uword.encode('utf8') for uword in text.split(delim)]
-----------------8<---------------[cut here]

Cheers
Dirk

Andy Robinson schrieb:
> 2008/10/4 Dirk Holtwick <dirk.holtwick at gmail.com>:
>> Hi,
>>
>> for my project pisa <http://www.htmltopdf.org> I create the paragraph
>> fragments directly. Until version 2.1 of RL I used a non breaking space
>> placeholder like u"\u00a0" and it worked fine. Since 2.2 this one seems to
>> be considered a normal whitespace and therefore removed. Any hints how I can
>> define a fragment with just nothing in it and the length of a space
>> character?
> 
> This sounds like a bug.  I have logged it in our tracker (issue 679,
> and hopefully the tracker will be visible by midweek); Robin's back
> from vacation now and will hopefully be able to look at it this week
> before we do 2.2.1.
> 
> 
> - Andy
> _______________________________________________
> reportlab-users mailing list
> reportlab-users at reportlab.com
> http://two.pairlist.net/mailman/listinfo/reportlab-users

From simon.king at motorola.com  Thu Nov 27 12:24:58 2008
From: simon.king at motorola.com (King Simon-NFHD78)
Date: Thu, 27 Nov 2008 17:24:58 -0000
Subject: [reportlab-users] PDF Transparency
Message-ID: <8D20BBB55F590E42AADF592A815E861703BF1ACB@zuk35exm65.ds.mot.com>

Hi,

Does anyone have any plans to support the PDF Transparency model in
ReportLab?

I had a need for basic support (just the stroking and non-stroking alpha
constants), which I've managed to hack in to reportlab. However, I don't
know if the method is suitable for general consumption. In particular, I
don't know if it would make it difficult to support the more general
transparency model

The basic idea was:

 * add an attribute to the canvas that tracks ExtGStates

 * add set[Non]StrokeAlphaConstant methods that create and activate new
ExtGStates

 * in canvas.showPage, pass the ExtGStates on to the page

 * add strokeOpacity and fillOpacity to some of the shapes in
reportlab.graphics.shapes

 * update _PDFRenderer.applyStateChanges to pas those properties on to
the canvas.

I also had to update PDFHeader to PDF version 1.4, because that is the
first version that supported transparency.

This is the first time I've looked at both the PDF spec and the
reportlab code, so I've probably done something very silly. Does this
approach sound reasonable?

One flaw at the moment is that a new graphics state gets created every
time you call set[Non]StrokeAlphaConstant, even if you pass the same
parameter. It would be much better to reuse states if possible.

I can post the patch if anyone is interested.

Thanks for listening,

Simon King

From robin at reportlab.com  Thu Nov 27 12:45:31 2008
From: robin at reportlab.com (Robin Becker)
Date: Thu, 27 Nov 2008 17:45:31 +0000
Subject: [reportlab-users] PDF Transparency
In-Reply-To: <8D20BBB55F590E42AADF592A815E861703BF1ACB@zuk35exm65.ds.mot.com>
References: <8D20BBB55F590E42AADF592A815E861703BF1ACB@zuk35exm65.ds.mot.com>
Message-ID: <492EDCBB.2000903@chamonix.reportlab.co.uk>

King Simon-NFHD78 wrote:
> Hi,
> 
> Does anyone have any plans to support the PDF Transparency model in
> ReportLab?
..........
> 
> The basic idea was:
> 
>  * add an attribute to the canvas that tracks ExtGStates
> 
>  * add set[Non]StrokeAlphaConstant methods that create and activate new
> ExtGStates
> 
>  * in canvas.showPage, pass the ExtGStates on to the page
> 
>  * add strokeOpacity and fillOpacity to some of the shapes in
> reportlab.graphics.shapes
> 
>  * update _PDFRenderer.applyStateChanges to pas those properties on to
> the canvas.
> 
> I also had to update PDFHeader to PDF version 1.4, because that is the
> first version that supported transparency.
> 
> This is the first time I've looked at both the PDF spec and the
> reportlab code, so I've probably done something very silly. Does this
> approach sound reasonable?
> 
> One flaw at the moment is that a new graphics state gets created every
> time you call set[Non]StrokeAlphaConstant, even if you pass the same
> parameter. It would be much better to reuse states if possible.
> 
> I can post the patch if anyone is interested.
> 
> Thanks for listening,
> 
> Simon King
.......
by all means post the patch. I guess the main problem is that we'd probably like 
to expose 1.4 capabilities in a controlled way. Perhaps we should have an 
attribute that specified which level of PDF is being aimed at. I guess that 
should probably be on the PDFDoc, but perhaps it should be settable on the 
canvas (as that needs support as well).
-- 
Robin Becker

From simon.king at motorola.com  Thu Nov 27 13:09:37 2008
From: simon.king at motorola.com (King Simon-NFHD78)
Date: Thu, 27 Nov 2008 18:09:37 -0000
Subject: [reportlab-users] PDF Transparency
In-Reply-To: <492EDCBB.2000903@chamonix.reportlab.co.uk>
Message-ID: <8D20BBB55F590E42AADF592A815E861703BF1AD5@zuk35exm65.ds.mot.com>

> -----Original Message-----
> From: reportlab-users-bounces at reportlab.com 
> [mailto:reportlab-users-bounces at reportlab.com] On Behalf Of 
> Robin Becker
> Sent: 27 November 2008 17:46
> To: Support list for users of Reportlab software
> Subject: Re: [reportlab-users] PDF Transparency
> 
> King Simon-NFHD78 wrote:
> > Hi,
> > 
> > Does anyone have any plans to support the PDF Transparency model in
> > ReportLab?
> ..........
> > 
> > I can post the patch if anyone is interested.
> > 
> > Thanks for listening,
> > 
> > Simon King
> .......
> by all means post the patch. I guess the main problem is that 
> we'd probably like 
> to expose 1.4 capabilities in a controlled way. Perhaps we 
> should have an 
> attribute that specified which level of PDF is being aimed 
> at. I guess that 
> should probably be on the PDFDoc, but perhaps it should be 
> settable on the 
> canvas (as that needs support as well).
> -- 
> Robin Becker

OK, here's the patch - it's a little ugly at the moment.

I was thinking a little more about managing the ExtGState dictionaries.
Do you think it is reasonable to create a state for each 'extended
attribute' value? For example, setting both the stroke alpha and fill
alpha results in the creation of two states. This keeps the canvas
interface nice and simple, possibly at the expense of efficiency.

I suppose you could expose add_extgstate method and activate_extgstate
as well, for users who want to manage them manually.

Thanks again,

Simon
-------------- next part --------------
A non-text attachment was scrubbed...
Name: reportlab_opacity_patch.patch
Type: application/octet-stream
Size: 5865 bytes
Desc: reportlab_opacity_patch.patch
Url : <http://two.pairlist.net/pipermail/reportlab-users/attachments/20081127/c8c5ab87/attachment.obj>

From simon.king at motorola.com  Fri Nov 28 09:13:59 2008
From: simon.king at motorola.com (King Simon-NFHD78)
Date: Fri, 28 Nov 2008 14:13:59 -0000
Subject: [reportlab-users] [PATCH] PDF Transparency
In-Reply-To: <8D20BBB55F590E42AADF592A815E861703BF1AD5@zuk35exm65.ds.mot.com>
Message-ID: <8D20BBB55F590E42AADF592A815E861703BF1C76@zuk35exm65.ds.mot.com>

> -----Original Message-----
> From: reportlab-users-bounces at reportlab.com 
> [mailto:reportlab-users-bounces at reportlab.com] On Behalf Of 
> King Simon-NFHD78
> Sent: 27 November 2008 18:10
> To: Support list for users of Reportlab software
> Subject: Re: [reportlab-users] PDF Transparency
> 
> OK, here's the patch - it's a little ugly at the moment.
> 
> I was thinking a little more about managing the ExtGState 
> dictionaries.
> Do you think it is reasonable to create a state for each 'extended
> attribute' value? For example, setting both the stroke alpha and fill
> alpha results in the creation of two states. This keeps the canvas
> interface nice and simple, possibly at the expense of efficiency.
> 
> I suppose you could expose add_extgstate method and activate_extgstate
> as well, for users who want to manage them manually.
> 
> Thanks again,
> 
> Simon
> 

Here's an updated version of the patch, and an example PDF with the
results.

With this new patch, reportlab should generate exactly the same PDFs
that it has always done (including the version, 1.3) if you don't ever
call canvas.setStrokeAlphaConstant or canvas.setFillAlphaConstant. If
you call one of those functions, a version 1.4 PDF will be generated
instead.

It now caches the graphics states, so there will only be a single state
for each transparency level used.

I think it would be fairly easy to add different blend modes, but I
haven't done it. I also haven't done anything about transparency mask
shapes - I've no idea if they would be easy to implement.

I hope this can be considered for inclusion in ReportLab, but I am happy
to continue to use my patched version if it is not acceptable. I think
it does everything I need at this point so I probably won't be putting
any more work into it for the moment.

Thanks,

Simon King
-------------- next part --------------
A non-text attachment was scrubbed...
Name: transparency.pdf
Type: application/octet-stream
Size: 3025 bytes
Desc: transparency.pdf
Url : <http://two.pairlist.net/pipermail/reportlab-users/attachments/20081128/c9308681/attachment-0002.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: reportlab_opacity_patch2.patch
Type: application/octet-stream
Size: 11984 bytes
Desc: reportlab_opacity_patch2.patch
Url : <http://two.pairlist.net/pipermail/reportlab-users/attachments/20081128/c9308681/attachment-0003.obj>

From gherman at darwin.in-berlin.de  Fri Nov 28 10:50:18 2008
From: gherman at darwin.in-berlin.de (Dinu Gherman)
Date: Fri, 28 Nov 2008 16:50:18 +0100
Subject: [reportlab-users] Adding word/hyphenations to wordaxe
Message-ID: <B194BF6E-B517-4CDB-B822-B850CF8D23E3@darwin.in-berlin.de>

Hi,

does anybody here use wordaxe (I'm using 0.3.0) and knows
how to pass it a list of additional words with allowed
hyphenation points?

Thanks,

Dinu