From jonas at reportlab.com  Mon Feb  2 06:19:32 2009
From: jonas at reportlab.com (jonas wustrack)
Date: Mon, 2 Feb 2009 11:19:32 +0000
Subject: [reportlab-users] Release 2.3 approaching...
In-Reply-To: <59828.213.10.61.26.1233224979.squirrel@marcans.nl>
References: <956003ae0901290106k2c5b55c5v33c6f3fee2c4996d@mail.gmail.com>
	<59828.213.10.61.26.1233224979.squirrel@marcans.nl>
Message-ID: <be0cc4270902020319y546cc7aej4e261eba197ba0cb@mail.gmail.com>

Hi Hans,

Sorry for not having replied earlier. We somehow must have missed your
initial post. We now checked your patch and it looks fine. It has been
applied to the subversion trunk and will be included in the 2.3
release.

Thanks,
Jonas

On Thu, Jan 29, 2009 at 10:29 AM, Hans Brand <hans at marcans.nl> wrote:
>
>> We have fixed a lot of broken and obsolete documentation and dealt
>> with all the recent patches and bug reports.
>>
>>
>
> Hi Andy,
>
> Recently I mentioned a problem in doctemplate.py
> (http://two.pairlist.net/pipermail/reportlab-users/2009-January/007946.html)
> and also posted a possible solution.
>
> I never got any response but the change I suggested in doctemplate.py is
> functioning in my system perfectly fine.
>
> Is it possible for you to include this bugfix in Release 2.3?
>
> Gr, Hans
>
>
>
> _______________________________________________
> reportlab-users mailing list
> reportlab-users at reportlab.com
> http://two.pairlist.net/mailman/listinfo/reportlab-users
>



-- 
Jonas

From farid.adigamov at gmail.com  Mon Feb  2 14:17:21 2009
From: farid.adigamov at gmail.com (Farid Adigamov)
Date: Mon, 2 Feb 2009 22:17:21 +0300
Subject: [reportlab-users] Fitting the area ( table cells for example ) with
	text
Message-ID: <84c131010902021117i2b798ea7hf5a39808bee090ab@mail.gmail.com>

Hello!
I'm fitting the area ( table cells for example ) with text.
How could it be done? Is it possible?
For example I have the text:

a = "HAVE A NICE DAY"

I have to put it to the cell of table, but don't know how long is the
string.
What steps are possible in this situation?

Thanks. Have a nice day ;)

-- 
,,, ^ .. ^ ,,,
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/reportlab-users/attachments/20090202/60e83a51/attachment.htm>

From paul.barrass at safeonlinebilling.com  Mon Feb  2 14:23:11 2009
From: paul.barrass at safeonlinebilling.com (Paul Barrass)
Date: Mon, 02 Feb 2009 19:23:11 +0000
Subject: [reportlab-users] Fitting the area ( table cells for example )
 with text
In-Reply-To: <84c131010902021117i2b798ea7hf5a39808bee090ab@mail.gmail.com>
References: <84c131010902021117i2b798ea7hf5a39808bee090ab@mail.gmail.com>
Message-ID: <4987481F.7040902@safeonlinebilling.com>

Farid Adigamov wrote:
> Hello!
> I'm fitting the area ( table cells for example ) with text.
> How could it be done? Is it possible?
> For example I have the text:
>
> a = "HAVE A NICE DAY"
>
> I have to put it to the cell of table, but don't know how long is the 
> string.
> What steps are possible in this situation?
You could either:

[for tables only] not specify the column widths, so the column will 
expand to the size required to fix the text (which can lead to tables 
wider than the page)
Place the text into a Paragraph() to enable word-wrapping.
Wrap the text (or a Paragraph() containing the text) into a 
KeepInFrame() which can either (by default) shrink or crop it's contents 
to fit.

Hope one of the above helps you out.

Paul Barrass.

From IStevens at globeandmail.com  Mon Feb  2 17:09:06 2009
From: IStevens at globeandmail.com (Stevens, Ian)
Date: Mon, 2 Feb 2009 17:09:06 -0500
Subject: [reportlab-users] Patch for inclusion of CMYK images as CMYK,
	not converted to RGB
Message-ID: <DBA10110A4D98644A505089C23131871040A76CB@GMTOMB01.globeandmail.net>

The following patch to pdfimage includes unconverted CMYK images in the
PDF to preserve colours. We have clients who pass their PDFs through a
preflight tool before sending it to press and want all content to be
CMYK. The patch uses PIL's Image.mode to decide whether to convert the
image. We've been using this on our production servers since 2.1 without
problem.
 
Would it be possible to include this in the upcoming release? 
 
Thanks,
Ian.
 
--- reportlab-trunk/src/reportlab/pdfgen/pdfimages.py   (revision 3413)
+++ reportlab-trunk/src/reportlab/pdfgen/pdfimages.py   (working copy)
@@ -95,17 +95,27 @@
         self.source = 'PIL'
         zlib = import_zlib()
         if not zlib: return
-        myimage = image.convert('RGB')
+
+        # Use the colorspace in the image
+        if image.mode == 'CMYK':
+            myimage = image
+            colorspace = 'DeviceCMYK'
+        else:
+            myimage = image.convert('RGB')
+            colorspace = 'RGB'
+
         imgwidth, imgheight = myimage.size
 
         # this describes what is in the image itself
         # *NB* according to the spec you can only use the short form in
inline images
         #imagedata=['BI /Width %d /Height /BitsPerComponent 8
/ColorSpace /%s /Filter [/Filter [ /ASCII85Decode /FlateDecode] ID]' %
(imgwidth, imgheight,'RGB')]
-        imagedata=['BI /W %d /H %d /BPC 8 /CS /RGB /F [/A85 /Fl] ID' %
(imgwidth, imgheight)]
+        imagedata=['BI /W %d /H %d /BPC 8 /CS /%s /F [/A85 /Fl] ID' %
(imgwidth, imgheight, colorspace)]
 
         #use a flate filter and Ascii Base 85 to compress
         raw = myimage.tostring()
-        assert len(raw) == imgwidth*imgheight*3, "Wrong amount of data
for image"
+
+        bpp = image.mode == 'CMYK' and 4 or 3
+        assert len(raw) == imgwidth*imgheight*bpp, "Wrong amount of
data for image"
         compressed = zlib.compress(raw)   #this bit is very fast...
         encoded = pdfutils._AsciiBase85Encode(compressed) #...sadly
this may not be
         #append in blocks of 60 characters

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/reportlab-users/attachments/20090202/9eba37f2/attachment.htm>

From timr at probo.com  Mon Feb  2 17:35:47 2009
From: timr at probo.com (Tim Roberts)
Date: Mon, 02 Feb 2009 14:35:47 -0800
Subject: [reportlab-users] Patch for inclusion of CMYK images as CMYK,
 not converted to RGB
In-Reply-To: <DBA10110A4D98644A505089C23131871040A76CB@GMTOMB01.globeandmail.net>
References: <DBA10110A4D98644A505089C23131871040A76CB@GMTOMB01.globeandmail.net>
Message-ID: <49877543.60702@probo.com>

Stevens, Ian wrote:
> The following patch to pdfimage includes unconverted CMYK images in
> the PDF to preserve colours. We have clients who pass their PDFs
> through a preflight tool before sending it to press and want all
> content to be CMYK. The patch uses PIL's Image.mode to decide whether
> to convert the image. We've been using this on our production servers
> since 2.1 without problem.

For what it's worth, this is a VERY useful thing to have.  I've been
through that same professional preflight process, and you really do get
substantially better results if your PDF's JPEGs remain in CMYK.

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


From robin at reportlab.com  Tue Feb  3 05:45:51 2009
From: robin at reportlab.com (Robin Becker)
Date: Tue, 03 Feb 2009 10:45:51 +0000
Subject: [reportlab-users] Patch for inclusion of CMYK images as CMYK,
 not converted to RGB
In-Reply-To: <DBA10110A4D98644A505089C23131871040A76CB@GMTOMB01.globeandmail.net>
References: <DBA10110A4D98644A505089C23131871040A76CB@GMTOMB01.globeandmail.net>
Message-ID: <4988205F.9040907@chamonix.reportlab.co.uk>

Stevens, Ian wrote:
> The following patch to pdfimage includes unconverted CMYK images in the
> PDF to preserve colours. We have clients who pass their PDFs through a
> preflight tool before sending it to press and want all content to be
> CMYK. The patch uses PIL's Image.mode to decide whether to convert the
> image. We've been using this on our production servers since 2.1 without
> problem.
>  
> Would it be possible to include this in the upcoming release? 
>

Yes, I see no obvious problem here. I think it's more sensible to move the bpp 
setting into the first part of the patch though.

We should probably think more about greyscale and other common image formats, 
but not today.

As for CMYK in general we have had patches in the past related to CMYK in jpegs. 
I notice in the code where we load a CMYK image from jpeg that we set a decode 
array parameter which seems to reverse the sense of the colour values so instead 
of running 0-1 they run 1-0 instead. Anyone clever here know why we have to do 
that. I seem to remember when we started doing these patches that the tool (name 
forgotten) which was creating the CMYK images required this transform (otherwise 
all the colours were messed up).

> Thanks,
> Ian.

--
Robin Becker

From IStevens at globeandmail.com  Tue Feb  3 13:16:05 2009
From: IStevens at globeandmail.com (Stevens, Ian)
Date: Tue, 3 Feb 2009 13:16:05 -0500
Subject: [reportlab-users] Patch for inclusion of CMYK images as CMYK,
	not converted to RGB
In-Reply-To: <4988205F.9040907@chamonix.reportlab.co.uk>
References: <DBA10110A4D98644A505089C23131871040A76CB@GMTOMB01.globeandmail.net>
	<4988205F.9040907@chamonix.reportlab.co.uk>
Message-ID: <DBA10110A4D98644A505089C23131871040A7743@GMTOMB01.globeandmail.net>

Thanks for including the patch. There's a typo, though:

===================================================================
--- src/reportlab/pdfgen/pdfimages.py   (revision 3416)
+++ src/reportlab/pdfgen/pdfimages.py   (working copy)
@@ -100,11 +100,11 @@
         # Use the colorspace in the image
         if image.mode == 'CMYK':
             myimage = image
-            colorspace = 'DeviceCMYK'
+            colorSpace = 'DeviceCMYK'
             bpp = 4
         else:
             myimage = image.convert('RGB')
-            colorspace = 'RGB'
+            colorSpace = 'RGB'
             bpp = 3
         imgwidth, imgheight = myimage.size

I've noticed the reverse color on JPEG images before but since went with
TIFF. I'll see what I can come up with.

Ian. 

> -----Original Message-----
> From: reportlab-users-bounces at reportlab.com 
> [mailto:reportlab-users-bounces at reportlab.com] On Behalf Of 
> Robin Becker
> Sent: February 3, 2009 5:46 AM
> To: Support list for users of Reportlab software
> Subject: Re: [reportlab-users] Patch for inclusion of CMYK 
> images as CMYK, not converted to RGB
> 
> Stevens, Ian wrote:
> > The following patch to pdfimage includes unconverted CMYK images in 
> > the PDF to preserve colours. We have clients who pass their PDFs 
> > through a preflight tool before sending it to press and want all 
> > content to be CMYK. The patch uses PIL's Image.mode to 
> decide whether 
> > to convert the image. We've been using this on our 
> production servers 
> > since 2.1 without problem.
> >  
> > Would it be possible to include this in the upcoming release? 
> >
> 
> Yes, I see no obvious problem here. I think it's more 
> sensible to move the bpp setting into the first part of the 
> patch though.
> 
> We should probably think more about greyscale and other 
> common image formats, but not today.
> 
> As for CMYK in general we have had patches in the past 
> related to CMYK in jpegs. 
> I notice in the code where we load a CMYK image from jpeg 
> that we set a decode array parameter which seems to reverse 
> the sense of the colour values so instead of running 0-1 they 
> run 1-0 instead. Anyone clever here know why we have to do 
> that. I seem to remember when we started doing these patches 
> that the tool (name
> forgotten) which was creating the CMYK images required this 
> transform (otherwise all the colours were messed up).
> 
> > Thanks,
> > Ian.
> 
> --
> Robin Becker
> _______________________________________________
> reportlab-users mailing list
> reportlab-users at reportlab.com
> http://two.pairlist.net/mailman/listinfo/reportlab-users
> 

From IStevens at globeandmail.com  Tue Feb  3 13:32:19 2009
From: IStevens at globeandmail.com (Stevens, Ian)
Date: Tue, 3 Feb 2009 13:32:19 -0500
Subject: [reportlab-users] Patch for inclusion of CMYK images as CMYK,
	not converted to RGB
In-Reply-To: <DBA10110A4D98644A505089C23131871040A7743@GMTOMB01.globeandmail.net>
References: <DBA10110A4D98644A505089C23131871040A76CB@GMTOMB01.globeandmail.net><4988205F.9040907@chamonix.reportlab.co.uk>
	<DBA10110A4D98644A505089C23131871040A7743@GMTOMB01.globeandmail.net>
Message-ID: <DBA10110A4D98644A505089C23131871040A7747@GMTOMB01.globeandmail.net>

Also, you'll want to remove the first call to image.convert('RGB') on
line 98.

Ian.

> -----Original Message-----
> From: reportlab-users-bounces at reportlab.com 
> [mailto:reportlab-users-bounces at reportlab.com] On Behalf Of 
> Stevens, Ian
> Sent: February 3, 2009 1:16 PM
> To: Support list for users of Reportlab software
> Subject: Re: [reportlab-users] Patch for inclusion of CMYK 
> images as CMYK,not converted to RGB
> 
> Thanks for including the patch. There's a typo, though:
> 
> ===================================================================
> --- src/reportlab/pdfgen/pdfimages.py   (revision 3416)
> +++ src/reportlab/pdfgen/pdfimages.py   (working copy)
> @@ -100,11 +100,11 @@
>          # Use the colorspace in the image
>          if image.mode == 'CMYK':
>              myimage = image
> -            colorspace = 'DeviceCMYK'
> +            colorSpace = 'DeviceCMYK'
>              bpp = 4
>          else:
>              myimage = image.convert('RGB')
> -            colorspace = 'RGB'
> +            colorSpace = 'RGB'
>              bpp = 3
>          imgwidth, imgheight = myimage.size
> 
> I've noticed the reverse color on JPEG images before but 
> since went with TIFF. I'll see what I can come up with.
> 
> Ian. 
> 
> > -----Original Message-----
> > From: reportlab-users-bounces at reportlab.com
> > [mailto:reportlab-users-bounces at reportlab.com] On Behalf Of Robin 
> > Becker
> > Sent: February 3, 2009 5:46 AM
> > To: Support list for users of Reportlab software
> > Subject: Re: [reportlab-users] Patch for inclusion of CMYK 
> images as 
> > CMYK, not converted to RGB
> > 
> > Stevens, Ian wrote:
> > > The following patch to pdfimage includes unconverted CMYK 
> images in 
> > > the PDF to preserve colours. We have clients who pass their PDFs 
> > > through a preflight tool before sending it to press and want all 
> > > content to be CMYK. The patch uses PIL's Image.mode to
> > decide whether
> > > to convert the image. We've been using this on our
> > production servers
> > > since 2.1 without problem.
> > >  
> > > Would it be possible to include this in the upcoming release? 
> > >
> > 
> > Yes, I see no obvious problem here. I think it's more 
> sensible to move 
> > the bpp setting into the first part of the patch though.
> > 
> > We should probably think more about greyscale and other 
> common image 
> > formats, but not today.
> > 
> > As for CMYK in general we have had patches in the past 
> related to CMYK 
> > in jpegs.
> > I notice in the code where we load a CMYK image from jpeg 
> that we set 
> > a decode array parameter which seems to reverse the sense of the 
> > colour values so instead of running 0-1 they run 1-0 
> instead. Anyone 
> > clever here know why we have to do that. I seem to remember when we 
> > started doing these patches that the tool (name
> > forgotten) which was creating the CMYK images required this 
> transform 
> > (otherwise all the colours were messed up).
> > 
> > > Thanks,
> > > Ian.
> > 
> > --
> > Robin Becker
> > _______________________________________________
> > reportlab-users mailing list
> > reportlab-users at reportlab.com
> > http://two.pairlist.net/mailman/listinfo/reportlab-users
> > 
> _______________________________________________
> reportlab-users mailing list
> reportlab-users at reportlab.com
> http://two.pairlist.net/mailman/listinfo/reportlab-users
> 

From robin at reportlab.com  Tue Feb  3 16:28:49 2009
From: robin at reportlab.com (Robin Becker)
Date: Tue, 03 Feb 2009 21:28:49 +0000
Subject: [reportlab-users] Patch for inclusion of CMYK images as CMYK,
 not converted to RGB
In-Reply-To: <DBA10110A4D98644A505089C23131871040A7747@GMTOMB01.globeandmail.net>
References: <DBA10110A4D98644A505089C23131871040A76CB@GMTOMB01.globeandmail.net><4988205F.9040907@chamonix.reportlab.co.uk>	<DBA10110A4D98644A505089C23131871040A7743@GMTOMB01.globeandmail.net>
	<DBA10110A4D98644A505089C23131871040A7747@GMTOMB01.globeandmail.net>
Message-ID: <4988B711.3050103@jessikat.plus.net>

Stevens, Ian wrote:
> Also, you'll want to remove the first call to image.convert('RGB') on
> line 98.
> 
.....
OK redone, can you check when time permits? Also is there an image which 
we could use to see that stuff is right? Doesn't have to be anything 
pretty four stripes would be enough.
-- 
Robin Becker

From IStevens at globeandmail.com  Tue Feb  3 18:07:46 2009
From: IStevens at globeandmail.com (Stevens, Ian)
Date: Tue, 3 Feb 2009 18:07:46 -0500
Subject: [reportlab-users] Patch for inclusion of CMYK images as CMYK,
	not converted to RGB
In-Reply-To: <4988B711.3050103@jessikat.plus.net>
References: <DBA10110A4D98644A505089C23131871040A76CB@GMTOMB01.globeandmail.net><4988205F.9040907@chamonix.reportlab.co.uk>	<DBA10110A4D98644A505089C23131871040A7743@GMTOMB01.globeandmail.net><DBA10110A4D98644A505089C23131871040A7747@GMTOMB01.globeandmail.net>
	<4988B711.3050103@jessikat.plus.net>
Message-ID: <DBA10110A4D98644A505089C23131871040A778A@GMTOMB01.globeandmail.net>

Looks good. I use ImageMagick for most of my images. If you specify a
-colorspace CMYK parameter, it will output a CMYK image:

> convert logo: -colorspace CMYK logo.jpg
> identify -verbose logo.jpg | head -13
Image: logo.jpg
  Format: JPEG (Joint Photographic Experts Group JFIF format)
  Class: DirectClass
  Geometry: 641x481+0+0
  Type: ColorSeparation
  Endianess: Undefined
  Colorspace: CMYK
  Depth: 8-bit
  Channel depth:
    Cyan: 1-bit
    Magenta: 8-bit
    Yellow: 8-bit
    Black: 8-bit

Also found this simple image online:
http://labs.teppefall.com/cmyk_test2.jpg

Sorry, I don't have anything more definitive. Someone else might be able
to help out with that.

Ian.

> -----Original Message-----
> From: reportlab-users-bounces at reportlab.com 
> [mailto:reportlab-users-bounces at reportlab.com] On Behalf Of 
> Robin Becker
> Sent: February 3, 2009 4:29 PM
> To: Support list for users of Reportlab software
> Subject: Re: [reportlab-users] Patch for inclusion of CMYK 
> images as CMYK, not converted to RGB
> 
> Stevens, Ian wrote:
> > Also, you'll want to remove the first call to 
> image.convert('RGB') on 
> > line 98.
> > 
> .....
> OK redone, can you check when time permits? Also is there an 
> image which we could use to see that stuff is right? Doesn't 
> have to be anything pretty four stripes would be enough.
> --
> Robin Becker
> _______________________________________________
> reportlab-users mailing list
> reportlab-users at reportlab.com
> http://two.pairlist.net/mailman/listinfo/reportlab-users
> 

From rvalyi at gmail.com  Tue Feb  3 20:41:29 2009
From: rvalyi at gmail.com (=?ISO-8859-1?Q?Rapha=EBl_Valyi?=)
Date: Wed, 4 Feb 2009 02:41:29 +0100
Subject: [reportlab-users] patch to make Reportlab work again on Jython
In-Reply-To: <956003ae0901300824v39358922p7f93d8452375c893@mail.gmail.com>
References: <806313f00901300507o7c427687ic3900621755d79b7@mail.gmail.com>
	<956003ae0901300816v604f3c79kb7106e9ca21278dd@mail.gmail.com>
	<956003ae0901300824v39358922p7f93d8452375c893@mail.gmail.com>
Message-ID: <806313f00902031741x53693945odf4bdf7644750fe5@mail.gmail.com>

Hi Andy and all,

yes, Jython knows the  '__all__' variable, so may be you can use it
for the patch.
Also, I didn't try that, but I think my patch could differ less from
the original file. Indeed, I wrapped almost every large list
instantiation of pdfbase/_fontdata.py in a private method, but I guess
that wrapping only a few of them will already make the containing
method small enough for the JVM to handle it.

Any way, as I said, it's more about applying that dummy patch that has
no impact except it's not beautiful (but not so ugly either). Flag it
with a big #FIXME remove that when Jython folks fix the issue, and
indeed remove the patch in one or two years when Jython doesn't has
the issue anymore. As I said, it's not a top priority bug for Jython
(hard to fix but rarely occurs). But on the contrary, having softwares
like Reportlab or OpenERP running on Jython would also certainly help
Sun putting more resources on that, finally helping reportlab too.

And to answer your other question, I'm running my test without any
accelerator for the moment (happy to know there were one). Having an
accelerator is not a top priority for OpenERP on Jython, we should
first focus on compatibility, and even if Reportlab is expected to be
100% compatible after my patch is applied, there are a few other
issues to port OpenERP on Jython. And since it's still a spare time
challenge...

Finally, be sure I can run your test suite and provide some feedback
(including speed) if Reportlab lets me run it on Jython. I doubt
Reportlab will be faster on Jython that soon, but it might be for the
end of the year. Bare in mind that JRuby is 2 to 5 times faster than C
Ruby (http://shootout.alioth.debian.org/u64q/benchmark.php?test=all&lang=jruby&lang2=ruby
  more to come...) . JRuby got much more momentum at Sun to go there,
plus they had outstanding rock stars like Charles Nutter in their
team. Jython got less momentum, but reborn after JRuby proved amazing
things could be done on the top of the JVM. After specialists like
Charles Nutter or Jim Baker, Jython will go there, probably by the end
of the year already, they can reuse a lot what has been done for JRuby
too. A 99% 2.5 Python compliant Jython is expected by February.

An other thing to consider if you are interested in Jython is Jython
multitreading abilities (unlike Python which is sticking to a GIL).
For multithreading enable applications, that means more performance
and much less memory use (thanks to the shared runtime that can
completely compensate the JVM extra memory usage) under huge
concurrent loads. Other benefit are portability.

If one of you is interested in the OpenERP+reportlab on Jython topic,
I started a blog post here:
http://rvalyi.blogspot.com/2009/02/openerp-running-on-java-jython-round-1.html


Hope to see something done for Jython compatibility. Best regards,

Rapha?l Valyi.


On Fri, Jan 30, 2009 at 5:24 PM, Andy Robinson <andy at reportlab.com> wrote:
> Rapha?l, does Jython pay attention to '__all__' ?    If so,
> perhaps we could do something less hacky
> and just list the bits we want in an '__all__' variable.
>
> Sorry but I don't have time to experiment on Jython itself.
>
> - Andy
>

From andy at reportlab.com  Wed Feb  4 11:59:27 2009
From: andy at reportlab.com (Andy Robinson)
Date: Wed, 4 Feb 2009 16:59:27 +0000
Subject: [reportlab-users] ReportLab 2.3 is out!
Message-ID: <956003ae0902040859r70828034rc8118150f737cdaa@mail.gmail.com>

Version 2.3 of the ReportLab library has been released today.  New
features are listed here:

      http://www.reportlab.org/whatsnew_2_3.html

It's also available on PyPI.

Thanks to everyone who contributed patches, bug fixes and other feedback.

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 andy at reportlab.com  Wed Feb  4 12:05:24 2009
From: andy at reportlab.com (Andy Robinson)
Date: Wed, 4 Feb 2009 17:05:24 +0000
Subject: [reportlab-users] patch to make Reportlab work again on Jython
In-Reply-To: <806313f00902031741x53693945odf4bdf7644750fe5@mail.gmail.com>
References: <806313f00901300507o7c427687ic3900621755d79b7@mail.gmail.com>
	<956003ae0901300816v604f3c79kb7106e9ca21278dd@mail.gmail.com>
	<956003ae0901300824v39358922p7f93d8452375c893@mail.gmail.com>
	<806313f00902031741x53693945odf4bdf7644750fe5@mail.gmail.com>
Message-ID: <956003ae0902040905j57bc01bbt76291d3dc4dadc6f@mail.gmail.com>

2009/2/4 Rapha?l Valyi <rvalyi at gmail.com>:
> Hi Andy and all,
>
> yes, Jython knows the  '__all__' variable, so may be you can use it
> for the patch.

I regret to say we had to leave the 'Jython support' out of today's
release; we had a deadline to ship the release, and need a bit longer
to study this problem and play with Jython ourselves before we pick
a solution.    We might even try to update our jar-file deployment from
5 years ago.



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 james.yoo at gmail.com  Wed Feb  4 12:51:27 2009
From: james.yoo at gmail.com (James Yoo)
Date: Wed, 4 Feb 2009 11:51:27 -0600
Subject: [reportlab-users] small bug in setting tablestyle ???
Message-ID: <487323a00902040951q61ad00ayb7f0d377df0ac019@mail.gmail.com>

Hello list,

disclaimer: I'm new to both python AND reportlab

I /think/* I may have found a small bug in the tables.py... I couldn't find
any other information relevant to my particular scenario, so I'm inclined to
believe that mine is an edge case, and perhaps a bit contrived since the
table style documentation seems to assume that one will always be setting a
style in a table on a range of cells, or aways referencing the last row or
column with -1

Nevertheless, here's the scenario:
1) I am dynamically creating a list of lists to feed to platypus.Table and
also conditionally setting table styles on a cell by cell basis...

len(data) = 3
len(data[0]) = 2

so, 3 rows, 2 cols.

2) if you happen to be setting style for a single cell in the last column
data[0][1]

eg: TableStyle.add('FONT', (1,0),(1,0), 'Times-Bold')

tables._addCommand does this test to calculate the cell coords for the
style:

    873         else:
    874             (op, (sc, sr), (ec, er)), values = cmd[:3] , cmd[3:]
    875             if sc < 0: sc = sc + self._ncols
    876             if ec < 0: ec = ec + self._ncols
    877             if sr < 0: sr = sr + self._nrows
    878             if er < 0: er = er + self._nrows
*    879             for i in range(sr, er+1):
*    *880*                * for j in range(sc, ec+1):*
    881                     _setCellStyle(self._cellStyles, i, j, op,
values)

you'll get a list index out of range error because of the the range() in
line 880

It's simply not checking that your "end column" is actually the index for
the last column and NOT -1

Of course, I can work around this in my own code, but it seems like a
relatively minor fix to the code above or a minor adjustment to the
documentation.


- James


* maybe it is a feature...
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/reportlab-users/attachments/20090204/0ac293fc/attachment.html>

From Michael at Hipp.com  Wed Feb  4 13:12:02 2009
From: Michael at Hipp.com (Michael Hipp)
Date: Wed, 04 Feb 2009 12:12:02 -0600
Subject: [reportlab-users] ReportLab 2.3 is out!
In-Reply-To: <956003ae0902040859r70828034rc8118150f737cdaa@mail.gmail.com>
References: <956003ae0902040859r70828034rc8118150f737cdaa@mail.gmail.com>
Message-ID: <4989DA72.4020903@Hipp.com>

Andy Robinson wrote:
> Version 2.3 of the ReportLab library has been released today.  New
> features are listed here:
> 
>       http://www.reportlab.org/whatsnew_2_3.html
> 
> It's also available on PyPI.
> 
> Thanks to everyone who contributed patches, bug fixes and other feedback.

Thanks.

I don't see anything about full Python 2.6 compatibility. Did it have to be 
postponed?

Thanks,
Michael

From robin at reportlab.com  Wed Feb  4 13:13:38 2009
From: robin at reportlab.com (Robin Becker)
Date: Wed, 04 Feb 2009 18:13:38 +0000
Subject: [reportlab-users] small bug in setting tablestyle ???
In-Reply-To: <487323a00902040951q61ad00ayb7f0d377df0ac019@mail.gmail.com>
References: <487323a00902040951q61ad00ayb7f0d377df0ac019@mail.gmail.com>
Message-ID: <4989DAD2.9050301@chamonix.reportlab.co.uk>

James Yoo wrote:
> Hello list,
> 
.........
> 2) if you happen to be setting style for a single cell in the last column
> data[0][1]
> 
> eg: TableStyle.add('FONT', (1,0),(1,0), 'Times-Bold')
> 
> tables._addCommand does this test to calculate the cell coords for the
> style:
> 
>     873         else:
>     874             (op, (sc, sr), (ec, er)), values = cmd[:3] , cmd[3:]
>     875             if sc < 0: sc = sc + self._ncols
>     876             if ec < 0: ec = ec + self._ncols
>     877             if sr < 0: sr = sr + self._nrows
>     878             if er < 0: er = er + self._nrows
> *    879             for i in range(sr, er+1):
> *    *880*                * for j in range(sc, ec+1):*
>     881                     _setCellStyle(self._cellStyles, i, j, op,
> values)
> 
> you'll get a list index out of range error because of the the range() in
> line 880
> 
> It's simply not checking that your "end column" is actually the index for
> the last column and NOT -1
> 
> Of course, I can work around this in my own code, but it seems like a
> relatively minor fix to the code above or a minor adjustment to the
> documentation.

If it's a bug then it's good to know, but here's what happens when I simulate 
your test scenario by hand.

 >>> cmd=('FONT', (1,0),(1,0), 'Times-Bold')
 >>> (op, (sc, sr), (ec, er)), values = cmd[:3] , cmd[3:]
 >>> for i in range(sr, er+1):
... 	for j in range(sc, ec+1):
... 		print i,j
...
0 1
 >>>

so we ought to be addressing only i=0 and j=1 which we claim is there providing 
we've added one style for each cell which we attempt to do in the init method 
where we set

cellrows = []
for i in xrange(nrows):
	cellcols = []
	for j in xrange(ncols):
		cellcols.append(CellStyle(`(i,j)`))
	cellrows.append(cellcols)
self._cellStyles = cellrows


For what it's worth I see your code at line 988 so our versions are out of sync; 
therefore it might be that something has been fixed.

Can you illustrate your cell/style addition with a tiny example? I/we have used 
essentially this technique before so I believe it ought to work.

It is of course easy to break this simplistic scheme by just adding in styles 
for cells which don't exist. So logically the sr, er, sc & ec values ought to be 
checked against the actual table dimensions.
-- 
Robin Becker

From robin at reportlab.com  Wed Feb  4 13:16:14 2009
From: robin at reportlab.com (Robin Becker)
Date: Wed, 04 Feb 2009 18:16:14 +0000
Subject: [reportlab-users] ReportLab 2.3 is out!
In-Reply-To: <4989DA72.4020903@Hipp.com>
References: <956003ae0902040859r70828034rc8118150f737cdaa@mail.gmail.com>
	<4989DA72.4020903@Hipp.com>
Message-ID: <4989DB6E.5080009@chamonix.reportlab.co.uk>

Michael Hipp wrote:
> Andy Robinson wrote:
>> Version 2.3 of the ReportLab library has been released today.  New
>> features are listed here:
>>
>>       http://www.reportlab.org/whatsnew_2_3.html
>>
>> It's also available on PyPI.
>>
>> Thanks to everyone who contributed patches, bug fixes and other feedback.
> 
> Thanks.
> 
> I don't see anything about full Python 2.6 compatibility. Did it have to 
> be postponed?
> 
> Thanks,
> Michael
......

We have tried to make it compatible with 2.6. Certainly all the tests run fine 
and I believe we got rid of some annoying warnings.
-- 
Robin Becker

From Michael at Hipp.com  Wed Feb  4 13:38:55 2009
From: Michael at Hipp.com (Michael Hipp)
Date: Wed, 04 Feb 2009 12:38:55 -0600
Subject: [reportlab-users] ReportLab 2.3 is out!
In-Reply-To: <4989DB6E.5080009@chamonix.reportlab.co.uk>
References: <956003ae0902040859r70828034rc8118150f737cdaa@mail.gmail.com>
	<4989DA72.4020903@Hipp.com>
	<4989DB6E.5080009@chamonix.reportlab.co.uk>
Message-ID: <4989E0BF.7090906@Hipp.com>

Robin Becker wrote:
> Michael Hipp wrote:
>> I don't see anything about full Python 2.6 compatibility. Did it have 
>> to be postponed?
> 
> We have tried to make it compatible with 2.6. Certainly all the tests 
> run fine and I believe we got rid of some annoying warnings.

Great.

But I still don't see a binary installer (for 2.6) listed here:

   http://www.reportlab.org/downloads.html

Thanks,
Michael

From timr at probo.com  Wed Feb  4 14:12:19 2009
From: timr at probo.com (Tim Roberts)
Date: Wed, 04 Feb 2009 11:12:19 -0800
Subject: [reportlab-users] small bug in setting tablestyle ???
In-Reply-To: <487323a00902040951q61ad00ayb7f0d377df0ac019@mail.gmail.com>
References: <487323a00902040951q61ad00ayb7f0d377df0ac019@mail.gmail.com>
Message-ID: <4989E893.1080307@probo.com>

James Yoo wrote:
>
> disclaimer: I'm new to both python AND reportlab
> ...
> Nevertheless, here's the scenario:
> 1) I am dynamically creating a list of lists to feed to platypus.Table
> and also conditionally setting table styles on a cell by cell basis...
> ...
> 2) if you happen to be setting style for a single cell in the last
> column data[0][1]
>
> eg: TableStyle.add('FONT', (1,0),(1,0), 'Times-Bold')
>
> tables._addCommand does this test to calculate the cell coords for the
> style:
>
>     873         else:
>     874             (op, (sc, sr), (ec, er)), values = cmd[:3] , cmd[3:]
>     875             if sc < 0: sc = sc + self._ncols
>     876             if ec < 0: ec = ec + self._ncols
>     877             if sr < 0: sr = sr + self._nrows
>     878             if er < 0: er = er + self._nrows
> *    879             for i in range(sr, er+1):
> *    *880*                * for j in range(sc, ec+1):*
>     881                     _setCellStyle(self._cellStyles, i, j, op,
> values)
>
> you'll get a list index out of range error because of the the range()
> in line 880

Does it?  Remember that a Python range is exclusive of the endpoint. 
That is, range(0,1) returns one thing: 0.

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


From james.yoo at gmail.com  Wed Feb  4 14:59:30 2009
From: james.yoo at gmail.com (James Yoo)
Date: Wed, 4 Feb 2009 13:59:30 -0600
Subject: [reportlab-users] small bug in setting tablestyle ???
In-Reply-To: <4989E893.1080307@probo.com>
References: <487323a00902040951q61ad00ayb7f0d377df0ac019@mail.gmail.com>
	<4989E893.1080307@probo.com>
Message-ID: <487323a00902041159h43d5b5bcmb5c5cd2fe53618fa@mail.gmail.com>

Robin, Tim, honorable list members....

After I realized the fundamental error I was making in my claim
(amature!)...

I retraced my steps.

Turns out, I was stepping in it, in a big way, and walking all over the
carpet to boot....

I had a STATIC var set to a particular TableStyle in my module....
and was using it within my function thinking that I was getting a new copy
with each invocation...

so I was constantly adding new cellstyles between invocations to the same
TableStyle obj...

thanks for pointing me out...

- "that guy"



On Wed, Feb 4, 2009 at 1:12 PM, Tim Roberts <timr at probo.com> wrote:

> James Yoo wrote:
> >
> > disclaimer: I'm new to both python AND reportlab
> > ...
> > Nevertheless, here's the scenario:
> > 1) I am dynamically creating a list of lists to feed to platypus.Table
> > and also conditionally setting table styles on a cell by cell basis...
> > ...
> > 2) if you happen to be setting style for a single cell in the last
> > column data[0][1]
> >
> > eg: TableStyle.add('FONT', (1,0),(1,0), 'Times-Bold')
> >
> > tables._addCommand does this test to calculate the cell coords for the
> > style:
> >
> >     873         else:
> >     874             (op, (sc, sr), (ec, er)), values = cmd[:3] , cmd[3:]
> >     875             if sc < 0: sc = sc + self._ncols
> >     876             if ec < 0: ec = ec + self._ncols
> >     877             if sr < 0: sr = sr + self._nrows
> >     878             if er < 0: er = er + self._nrows
> > *    879             for i in range(sr, er+1):
> > *    *880*                * for j in range(sc, ec+1):*
> >     881                     _setCellStyle(self._cellStyles, i, j, op,
> > values)
> >
> > you'll get a list index out of range error because of the the range()
> > in line 880
>
> Does it?  Remember that a Python range is exclusive of the endpoint.
> That is, range(0,1) returns one thing: 0.
>
> --
> Tim Roberts, timr at probo.com
> Providenza & Boekelheide, Inc.
>
> _______________________________________________
> 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/20090204/ed8c439c/attachment.html>

From jconner at EmpireDI.com  Wed Feb  4 15:38:41 2009
From: jconner at EmpireDI.com (John Conner)
Date: Wed, 04 Feb 2009 13:38:41 -0700
Subject: [reportlab-users] ReportLab 2.3 - still getting errors with 2.6
Message-ID: <4989FCD1.3030306@EmpireDI.com>

Hi,

Not sure if it is a problem with report lab or maybe python2.6 or maybe 
freetype. I have been seeing the
following error for a while now.

Running 'python setup.py tests-preinstall' I get the following:

...................................................................................................................Traceback 
(most recent call last):
  File "genreference.py", line 31, in <module>
    run()
  File "genreference.py", line 17, in run
    yaml2pdf.run('reference.yml','reportlab-reference.pdf')
  File 
"G:\downloads\Python\ReportLab\ReportLab_2_3\ReportLab_2_3\tools\docco\yaml2pdf.py", 
line 54, in run
    story.append(Paragraph(text, style, bulletText=bulletText))
  File 
"G:\downloads\Python\ReportLab\ReportLab_2_3\ReportLab_2_3\src\reportlab\platypus\paragraph.py", 
line 798, in __init__
    self._setup(text, style, bulletText, frags, cleanBlockQuotedText)
  File 
"G:\downloads\Python\ReportLab\ReportLab_2_3\ReportLab_2_3\src\reportlab\platypus\paragraph.py", 
line 816, in _setup
    % (_parser.errors[0],text[:min(30,len(text))]))
ValueError: xml parser error (garbage in start tag) in paragraph beginning
'Very simple and fast XML parse'
Failed to import renderPM.
F..................................................................
======================================================================
FAIL: Test if all manuals buildable from source.
----------------------------------------------------------------------
Traceback (most recent call last):
  File 
"G:\downloads\Python\ReportLab\ReportLab_2_3\ReportLab_2_3\tests\test_docs_build.py", 
line 32, in test0
    assert os.path.isfile('reportlab-reference.pdf'), 'genAll.py failed 
to generate reportlab-reference.pdf!'
AssertionError: genAll.py failed to generate reportlab-reference.pdf!

----------------------------------------------------------------------
Ran 182 tests in 71.657s

FAILED (failures=1)


running 'python setup.py install' gives:

################################################
#Attempting install of _rl_accel, sgmlop & pyHnj
#extensions from 
'G:\\downloads\\Python\\ReportLab\\ReportLab_2_3\\ReportLab_2_3\\src\\rl_addons\\rl_accel'
################################################
################################################
#Attempting install of _renderPM
#extensions from 
'G:\\downloads\\Python\\ReportLab\\ReportLab_2_3\\ReportLab_2_3\\src\\rl_addons\\renderPM'
# installing with win32 freetype 'freetype6.dll'
################################################
running install
running build
running build_py
copying src\reportlab\lib\hyphen.mashed -> build\lib.win32-2.6\reportlab\lib
running build_clib
building '_renderPM_libart' library
error: None


Finally running 'python setup.py tests' results in:
 
...................................................................................................................Traceback 
(most recent call last):
  File "genreference.py", line 31, in <module>
    run()
  File "genreference.py", line 17, in run
    yaml2pdf.run('reference.yml','reportlab-reference.pdf')
  File 
"G:\downloads\Python\ReportLab\ReportLab_2_3\ReportLab_2_3\tools\docco\yaml2pdf.py", 
line 54, in run
    story.append(Paragraph(text, style, bulletText=bulletText))
  File 
"G:\downloads\Python\ReportLab\ReportLab_2_3\ReportLab_2_3\src\reportlab\platypus\paragraph.py", 
line 798, in __init__
    self._setup(text, style, bulletText, frags, cleanBlockQuotedText)
  File 
"G:\downloads\Python\ReportLab\ReportLab_2_3\ReportLab_2_3\src\reportlab\platypus\paragraph.py", 
line 816, in _setup
    % (_parser.errors[0],text[:min(30,len(text))]))
ValueError: xml parser error (garbage in start tag) in paragraph beginning
'Very simple and fast XML parse'
Failed to import renderPM.
F..................................................................
======================================================================
FAIL: Test if all manuals buildable from source.
----------------------------------------------------------------------
Traceback (most recent call last):
  File 
"G:\downloads\Python\ReportLab\ReportLab_2_3\ReportLab_2_3\tests\test_docs_build.py", 
line 32, in test0
    assert os.path.isfile('reportlab-reference.pdf'), 'genAll.py failed 
to generate reportlab-reference.pdf!'
AssertionError: genAll.py failed to generate reportlab-reference.pdf!

----------------------------------------------------------------------
Ran 182 tests in 72.141s

FAILED (failures=1)


Any ideas what we've got messed up????

John Conner
_____________________
Empire Digital Instruments
Colorado Springs, Colorado



From andy at reportlab.com  Wed Feb  4 15:53:57 2009
From: andy at reportlab.com (Andy Robinson)
Date: Wed, 4 Feb 2009 20:53:57 +0000
Subject: [reportlab-users] ReportLab 2.3 - still getting errors with 2.6
In-Reply-To: <4989FCD1.3030306@EmpireDI.com>
References: <4989FCD1.3030306@EmpireDI.com>
Message-ID: <956003ae0902041253q88b48f4ye07902a02e9c7f52@mail.gmail.com>

2009/2/4 John Conner <jconner at empiredi.com>:
> Hi,
>
> Not sure if it is a problem with report lab or maybe python2.6 or maybe
> freetype. I have been seeing the
> following error for a while now.

I see you are compiling the C extensions on Windows.  Is this because
you want to, or because we didn't make clear that Windows packages are
available?  ;-)   It's very likely there is some difference between
your compiler setup and ours.

We ran tests of the pre-built Windows installers on a bunch of VMs,
from Python 2.3 up to Python 2.6, until they all came up clean.  Could
you try the installer below?

   http://www.reportlab.org/ftp/reportlab-2.3.win32-py2.6.exe

This will install the library and DLLs in site-packages, but none of
the docs and tests. That seems to be how the default exe-builder in
distutils does things but you could then separately run the test
directory you have.

Alternatively, just grab the built DLLs from here and put them on your
path.  The test suite will then hopefully run to completion.

     http://www.reportlab.org/ftp/win32-dlls/2.6/

Our windows test environments are in the office so we'll investigate
further in the (UK) morning..


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 jconner at EmpireDI.com  Wed Feb  4 16:49:38 2009
From: jconner at EmpireDI.com (John Conner)
Date: Wed, 04 Feb 2009 14:49:38 -0700
Subject: [reportlab-users] ReportLab 2.3 - still getting errors with 2.6
In-Reply-To: <956003ae0902041253q88b48f4ye07902a02e9c7f52@mail.gmail.com>
References: <4989FCD1.3030306@EmpireDI.com>
	<956003ae0902041253q88b48f4ye07902a02e9c7f52@mail.gmail.com>
Message-ID: <498A0D72.7000302@EmpireDI.com>

Andy,

I figured it was some confusion on my part. I've been away from Python 
for several
years - Mostly doing Fortran.

The installer worked fine.  Ran the tests and 193 passed with an OK. 

Next time my buddy is in London (a consultant for some part of what used 
to be
BR) I'll have him by you a pint.


Thanks,

John Conner
_____________________
Empire Digital Instruments
Colorado Springs, Colorado



Andy Robinson wrote:
> 2009/2/4 John Conner <jconner at empiredi.com>:
>   
>> Hi,
>>
>> Not sure if it is a problem with report lab or maybe python2.6 or maybe
>> freetype. I have been seeing the
>> following error for a while now.
>>     
>
> I see you are compiling the C extensions on Windows.  Is this because
> you want to, or because we didn't make clear that Windows packages are
> available?  ;-)   It's very likely there is some difference between
> your compiler setup and ours.
>
> We ran tests of the pre-built Windows installers on a bunch of VMs,
> from Python 2.3 up to Python 2.6, until they all came up clean.  Could
> you try the installer below?
>
>    http://www.reportlab.org/ftp/reportlab-2.3.win32-py2.6.exe
>
> This will install the library and DLLs in site-packages, but none of
> the docs and tests. That seems to be how the default exe-builder in
> distutils does things but you could then separately run the test
> directory you have.
>
> Alternatively, just grab the built DLLs from here and put them on your
> path.  The test suite will then hopefully run to completion.
>
>      http://www.reportlab.org/ftp/win32-dlls/2.6/
>
> Our windows test environments are in the office so we'll investigate
> further in the (UK) morning..
>
>
> Best Regards,
>
>
>   
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/reportlab-users/attachments/20090204/e4f62848/attachment.htm>

From tamonashroy at gmail.com  Thu Feb  5 02:32:24 2009
From: tamonashroy at gmail.com (Tamonash Roy)
Date: Thu, 5 Feb 2009 13:02:24 +0530
Subject: [reportlab-users] Problem with transparent png's
Message-ID: <2d0a01960902042332j1da0122fw404b97db5c41ea77@mail.gmail.com>

Hi !
I've created a pdf document which has a lot of images in it. These images
are transparent png images. In my development machine I have windows with
python 2.5 and reportlab2.2, whereas in the server it is debian linux,
python2.5 and reportlab2.2. The problem is I have a perfect pdf document
while I use my development machine, but when the pdf is generated in the
server and downloaded, the transparent images have a black background. I got
to use these images in the document but dont know how. please help!

Sample code used to generate pdf is as follows:

        f2 = Frame( inch*1.75, inch*7.6,  inch*4.75, inch*2.95,
leftPadding=0, bottomPadding=0,rightPadding=0, topPadding=2, showBoundary=1)
        f3 = Frame( inch*1.25, inch*5.75,  inch*3.85, inch*1.7,
leftPadding=0, bottomPadding=0,rightPadding=0, topPadding=2, showBoundary=1)

mainPage = PageTemplate(frames=[f2, f3])
doc = BaseDocTemplate('report_1.pdf',  pageTemplates=mainPage, leftMargin=72
,title=Title, author=Author, showBoundary=0)

story = []
story.append(Image("a.png"))
story.append(FrameBreak())
story.append(Image("b.png"))

doc.build(story)

Thanks,
Tamonash Roy
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/reportlab-users/attachments/20090205/0cce096a/attachment.html>

From andy at reportlab.com  Thu Feb  5 04:20:46 2009
From: andy at reportlab.com (Andy Robinson)
Date: Thu, 5 Feb 2009 09:20:46 +0000
Subject: [reportlab-users] ReportLab 2.3 - still getting errors with 2.6
In-Reply-To: <498A0D72.7000302@EmpireDI.com>
References: <4989FCD1.3030306@EmpireDI.com>
	<956003ae0902041253q88b48f4ye07902a02e9c7f52@mail.gmail.com>
	<498A0D72.7000302@EmpireDI.com>
Message-ID: <956003ae0902050120u7db21bcfufe29a52d4b502d11@mail.gmail.com>

2009/2/4 John Conner <jconner at empiredi.com>:
> The installer worked fine.  Ran the tests and 193 passed with an OK.

Great news!  I'll amend the changes page to boast more about Python
2.6 compatibility.

- Andy

From robin at reportlab.com  Thu Feb  5 04:57:15 2009
From: robin at reportlab.com (Robin Becker)
Date: Thu, 05 Feb 2009 09:57:15 +0000
Subject: [reportlab-users] ReportLab 2.3 is out!
In-Reply-To: <4989E0BF.7090906@Hipp.com>
References: <956003ae0902040859r70828034rc8118150f737cdaa@mail.gmail.com>	<4989DA72.4020903@Hipp.com>	<4989DB6E.5080009@chamonix.reportlab.co.uk>
	<4989E0BF.7090906@Hipp.com>
Message-ID: <498AB7FB.5060700@chamonix.reportlab.co.uk>

Michael Hipp wrote:
> Robin Becker wrote:
>> Michael Hipp wrote:
>>> I don't see anything about full Python 2.6 compatibility. Did it have 
>>> to be postponed?
>>
>> We have tried to make it compatible with 2.6. Certainly all the tests 
>> run fine and I believe we got rid of some annoying warnings.
> 
> Great.
> 
> But I still don't see a binary installer (for 2.6) listed here:
> 
>   http://www.reportlab.org/downloads.html
> 
> Thanks,
> Michael
......the exe is there at

http://www.reportlab.org/ftp/reportlab-2.3.win32-py2.6.exe

just some one decided to use static version number rather than dynamic. I'll fix 
that page up right away.
-- 
Robin Becker

From robin at reportlab.com  Thu Feb  5 06:10:21 2009
From: robin at reportlab.com (Robin Becker)
Date: Thu, 05 Feb 2009 11:10:21 +0000
Subject: [reportlab-users] patch to make Reportlab work again on Jython
In-Reply-To: <806313f00901300507o7c427687ic3900621755d79b7@mail.gmail.com>
References: <806313f00901300507o7c427687ic3900621755d79b7@mail.gmail.com>
Message-ID: <498AC91D.6020407@chamonix.reportlab.co.uk>

Rapha?l,

we recently introduced some ability for selected people to access and modify at 
least parts of the reportlab repository. Would it not make sense for us to 
create a jython branch based on the current reportlab 2.3 and then allow you 
(and perhaps others) to make patches to that branch?

It seems wrong to mess with the base implementation just because the latest 
jython has some limits. I suppose jython's new model(ie not creating java files) 
has caused these problems with class size, but I'm sure you'll find other areas 
where our code needs some tweaks to make it work under jython.

Even if you don't want to do the commits yourself it seems to make more sense 
than doing everything on the trunk.


Rapha?l Valyi wrote:
> Dear Reportlab developers,
........
> What is your position regarding to this?
>
-- 
Robin Becker

From rvalyi at gmail.com  Thu Feb  5 09:01:47 2009
From: rvalyi at gmail.com (=?ISO-8859-1?Q?Rapha=EBl_Valyi?=)
Date: Thu, 5 Feb 2009 15:01:47 +0100
Subject: [reportlab-users] patch to make Reportlab work again on Jython
In-Reply-To: <498AC91D.6020407@chamonix.reportlab.co.uk>
References: <806313f00901300507o7c427687ic3900621755d79b7@mail.gmail.com>
	<498AC91D.6020407@chamonix.reportlab.co.uk>
Message-ID: <806313f00902050601j1dd7427an24e34477f4c90f4f@mail.gmail.com>

Hi Robin, inlined responses,

On Thu, Feb 5, 2009 at 12:10 PM, Robin Becker <robin at reportlab.com> wrote:
> Rapha?l,
>
> we recently introduced some ability for selected people to access and modify
> at least parts of the reportlab repository. Would it not make sense for us
> to create a jython branch based on the current reportlab 2.3 and then allow
> you (and perhaps others) to make patches to that branch?

Ok, will do that at least for the current patch I had.

>
> It seems wrong to mess with the base implementation just because the latest
> jython has some limits. I suppose jython's new model(ie not creating java
> files) has caused these problems with class size, but I'm sure you'll find
> other areas where our code needs some tweaks to make it work under jython.

Well may be. I mean if you imagine to release a Jython version at
least along with the Python one that would make sense. Moreover, I
think you would have to bother propagating your Reportlab improvements
on the Jython branch because I unfortunately I'm afraid I'll never
assume that. I'm only helping out in my spare time in making the
things I believe are right, but you can only expect me to commit that
pacth and may be simple others as needs occur.
I the mean time if you can't assume to maintain that parallel Jython
distro, I think it's just easier to include the minor tweak in the
regular distro to make it work.
There might be more business around Jython in the near future
(depending a lot on what the Sun team can achieve), but currently I'm
not sure it will generate enough to sustain large efforts.


> Even if you don't want to do the commits yourself it seems to make more
> sense than doing everything on the trunk.

Well you have my view. On the other way, I can make the commit myself
and even see if the tweak could be smaller. In fact I saw that a
method was too big and didn't bother trying/guessing what was the
minimal change, instead I just wrapped every table/list instantiation
here to make it work. May be just one or two wrappings could do the
job. Indeed, if that worked before (as you seem to tell), knowing that
Jython always had this JVM limitation, I guess this is because that
particular file was simply smaller (had no time to confirm though).
That would mean that only a few wrapping would make it work. In such a
case, may be that tweak would be so minor that it could perfectly be
in the regular distro.

Meanwhile, if you Reportlab want to bring the Java Reportlab
accelerator back to life, I think it definitely makes sense to have a
separate branch + distro. Ideally thinks would be modular and clean so
that the difference between the two branches would be minimal and
auto-magically propagated.
In the ideal world that might be much like the JRuby on Rails world.
Rails started ignoring JRuby (while Sun started ignoring Rails). Now
Sun is fare from ignoring Rails (their Kenai dev platform is just
meant to be the regular corporate dev platform while it was a personal
proto of some JRuby employee, now Glassfish, the fame J2EE server is
just to be developed on...  ...Kenai). And now Rails  itself started
distributing the database JDBC adapter in the regular Rails distro...

As I said the same will probably happen with Jython (I see no reason
why not), but it might simply take more time.


As I said, my own goal is to show the world that a killer ERP
(OpenERP) which is using Reportab can be and should be worth running
in a Java servlet containter (Modjy). I won't bother making it happen
alone in my spare time past this stage. But I'm very much implied
professionally with OpenERP integration with increasing demand, so
it's just about bridging the gap sooner or later, if that includes
giving a hand to reportlab, I would be happy to do my part.

Best regards,

Rapha?l Valyi.

>
>
> Rapha?l Valyi wrote:
>>
>> Dear Reportlab developers,
>
> ........
>>
>> What is your position regarding to this?
>>
> --
> Robin Becker
> _______________________________________________
> reportlab-users mailing list
> reportlab-users at reportlab.com
> http://two.pairlist.net/mailman/listinfo/reportlab-users
>

From Michael at Hipp.com  Thu Feb  5 09:22:35 2009
From: Michael at Hipp.com (Michael Hipp)
Date: Thu, 05 Feb 2009 08:22:35 -0600
Subject: [reportlab-users] ReportLab 2.3 is out!
In-Reply-To: <498AB7FB.5060700@chamonix.reportlab.co.uk>
References: <956003ae0902040859r70828034rc8118150f737cdaa@mail.gmail.com>	<4989DA72.4020903@Hipp.com>	<4989DB6E.5080009@chamonix.reportlab.co.uk>
	<4989E0BF.7090906@Hipp.com>
	<498AB7FB.5060700@chamonix.reportlab.co.uk>
Message-ID: <498AF62B.1060803@Hipp.com>

Robin Becker wrote:
> Michael Hipp wrote:
>> But I still don't see a binary installer (for 2.6) listed here:
>>
>>   http://www.reportlab.org/downloads.html
>>
> ......the exe is there at
> 
> http://www.reportlab.org/ftp/reportlab-2.3.win32-py2.6.exe
> 
> just some one decided to use static version number rather than dynamic. 
> I'll fix that page up right away.

Thanks!

Question: On the download page for Win binaries, do I need to get both the 
'Binary Installers' as well as the 'Win32 DLLs and Python Extensions'? Or do I 
just need one of them? It's not clear to me.

http://www.reportlab.org/downloads.html#dlls

Thanks,
Michael

From andy at reportlab.com  Thu Feb  5 09:43:37 2009
From: andy at reportlab.com (Andy Robinson)
Date: Thu, 5 Feb 2009 14:43:37 +0000
Subject: [reportlab-users] ReportLab 2.3 is out!
In-Reply-To: <498AF62B.1060803@Hipp.com>
References: <956003ae0902040859r70828034rc8118150f737cdaa@mail.gmail.com>
	<4989DA72.4020903@Hipp.com>
	<4989DB6E.5080009@chamonix.reportlab.co.uk>
	<4989E0BF.7090906@Hipp.com>
	<498AB7FB.5060700@chamonix.reportlab.co.uk>
	<498AF62B.1060803@Hipp.com>
Message-ID: <956003ae0902050643y4402faa1vde22b468773acf44@mail.gmail.com>

2009/2/5 Michael Hipp <Michael at hipp.com>:
> Question: On the download page for Win binaries, do I need to get both the
> 'Binary Installers' as well as the 'Win32 DLLs and Python Extensions'? Or do
> I just need one of them? It's not clear to me.

Just the binaries.  The DLLs are "backup" for people who like to roll
their own distributions.

We know this page needs a complete rewrite.  We're working on it now!

- Andy

From Michael at Hipp.com  Thu Feb  5 11:49:13 2009
From: Michael at Hipp.com (Michael Hipp)
Date: Thu, 05 Feb 2009 10:49:13 -0600
Subject: [reportlab-users] ReportLab 2.3 is out!
In-Reply-To: <956003ae0902050643y4402faa1vde22b468773acf44@mail.gmail.com>
References: <956003ae0902040859r70828034rc8118150f737cdaa@mail.gmail.com>	<4989DA72.4020903@Hipp.com>	<4989DB6E.5080009@chamonix.reportlab.co.uk>	<4989E0BF.7090906@Hipp.com>	<498AB7FB.5060700@chamonix.reportlab.co.uk>	<498AF62B.1060803@Hipp.com>
	<956003ae0902050643y4402faa1vde22b468773acf44@mail.gmail.com>
Message-ID: <498B1889.8010302@Hipp.com>

Andy Robinson wrote:
> 2009/2/5 Michael Hipp <Michael at hipp.com>:
>> Question: On the download page for Win binaries, do I need to get both the
>> 'Binary Installers' as well as the 'Win32 DLLs and Python Extensions'? Or do
>> I just need one of them? It's not clear to me.
> 
> Just the binaries.  The DLLs are "backup" for people who like to roll
> their own distributions.
> 
> We know this page needs a complete rewrite.  We're working on it now!

Thanks. Some of us simpletons need everything spelled out in baby-talk language :-)

Michael

From james.yoo at gmail.com  Thu Feb  5 17:58:01 2009
From: james.yoo at gmail.com (James Yoo)
Date: Thu, 5 Feb 2009 16:58:01 -0600
Subject: [reportlab-users] colwidths question
Message-ID: <487323a00902051458v1e42669cv51c1fe5e8735915f@mail.gmail.com>

Hi... me again..

There is something I am not understanding about colWidths...
I'm trying to create a Table with colWidths = (28,10,10,10), those are the
widest character lengths of the strings in my columns...

 I'm obviously not understanding about the units for the column widths
because I get a really squished table.

Is there an idiom for setting exact widths for columns that I am missing?

thanks,
James
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/reportlab-users/attachments/20090205/57490da4/attachment.html>

From DJESSUP at usnews.com  Thu Feb  5 21:11:36 2009
From: DJESSUP at usnews.com (Jessup, David)
Date: Thu, 5 Feb 2009 21:11:36 -0500
Subject: [reportlab-users] colwidths question
In-Reply-To: <487323a00902051458v1e42669cv51c1fe5e8735915f@mail.gmail.com>
References: <487323a00902051458v1e42669cv51c1fe5e8735915f@mail.gmail.com>
Message-ID: <97886A555FD8E14E8F7445D75C11D54DD2E133@EXCHANGE.usn.root.ent>

What I tend to do is perform the following import:
 
> from reportlab.lib.units import unit
 
where unit is "inch", "cm", "mm", or "pica".  I can then set colWidths =
(inch, 2.5 * inch, inch, inch) in units I can measure with a ruler.

________________________________

From: reportlab-users-bounces at reportlab.com
[mailto:reportlab-users-bounces at reportlab.com] On Behalf Of James Yoo
Sent: Thursday, February 05, 2009 5:58 PM
To: Support list for users of Reportlab software
Subject: [reportlab-users] colwidths question


Hi... me again..

There is something I am not understanding about colWidths...
I'm trying to create a Table with colWidths = (28,10,10,10), those are
the widest character lengths of the strings in my columns...

 I'm obviously not understanding about the units for the column widths
because I get a really squished table.

Is there an idiom for setting exact widths for columns that I am
missing?

thanks,
James


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/reportlab-users/attachments/20090205/67bdb98f/attachment.html>

From DJESSUP at usnews.com  Thu Feb  5 21:17:36 2009
From: DJESSUP at usnews.com (Jessup, David)
Date: Thu, 5 Feb 2009 21:17:36 -0500
Subject: [reportlab-users] colwidths question
In-Reply-To: <97886A555FD8E14E8F7445D75C11D54DD2E133@EXCHANGE.usn.root.ent>
References: <487323a00902051458v1e42669cv51c1fe5e8735915f@mail.gmail.com>
	<97886A555FD8E14E8F7445D75C11D54DD2E133@EXCHANGE.usn.root.ent>
Message-ID: <97886A555FD8E14E8F7445D75C11D54DD2E134@EXCHANGE.usn.root.ent>

One more thing:  If I recall correctly, if you don't specify the
colWidths, they'll shrink to just fitting the displayed content; you
only need to specify the colWidths if your table width would otherwise
not fit on a page.

________________________________

From: reportlab-users-bounces at reportlab.com
[mailto:reportlab-users-bounces at reportlab.com] On Behalf Of Jessup,
David
Sent: Thursday, February 05, 2009 9:12 PM
To: Support list for users of Reportlab software
Subject: Re: [reportlab-users] colwidths question


What I tend to do is perform the following import:
 
> from reportlab.lib.units import unit
 
where unit is "inch", "cm", "mm", or "pica".  I can then set colWidths =
(inch, 2.5 * inch, inch, inch) in units I can measure with a ruler.

________________________________

From: reportlab-users-bounces at reportlab.com
[mailto:reportlab-users-bounces at reportlab.com] On Behalf Of James Yoo
Sent: Thursday, February 05, 2009 5:58 PM
To: Support list for users of Reportlab software
Subject: [reportlab-users] colwidths question


Hi... me again..

There is something I am not understanding about colWidths...
I'm trying to create a Table with colWidths = (28,10,10,10), those are
the widest character lengths of the strings in my columns...

 I'm obviously not understanding about the units for the column widths
because I get a really squished table.

Is there an idiom for setting exact widths for columns that I am
missing?

thanks,
James


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/reportlab-users/attachments/20090205/c6276517/attachment.htm>

From aklaver at comcast.net  Thu Feb  5 21:24:14 2009
From: aklaver at comcast.net (Adrian Klaver)
Date: Thu, 5 Feb 2009 18:24:14 -0800
Subject: [reportlab-users] colwidths question
In-Reply-To: <487323a00902051458v1e42669cv51c1fe5e8735915f@mail.gmail.com>
References: <487323a00902051458v1e42669cv51c1fe5e8735915f@mail.gmail.com>
Message-ID: <200902051824.14795.aklaver@comcast.net>

On Thursday 05 February 2009 2:58:01 pm James Yoo wrote:
> Hi... me again..
>
> There is something I am not understanding about colWidths...
> I'm trying to create a Table with colWidths = (28,10,10,10), those are the
> widest character lengths of the strings in my columns...
>
>  I'm obviously not understanding about the units for the column widths
> because I get a really squished table.
>
> Is there an idiom for setting exact widths for columns that I am missing?
>
> thanks,
> James

If I remember correctly the units are in picas(1/72 inch).

-- 
Adrian Klaver
aklaver at comcast.net

From dave at bcs.co.nz  Thu Feb  5 22:54:12 2009
From: dave at bcs.co.nz (David Brooks)
Date: Fri, 06 Feb 2009 16:54:12 +1300
Subject: [reportlab-users] ReportLab 2.3 is out!
In-Reply-To: <956003ae0902040859r70828034rc8118150f737cdaa@mail.gmail.com>
References: <956003ae0902040859r70828034rc8118150f737cdaa@mail.gmail.com>
Message-ID: <498BB464.60803@bcs.co.nz>

I've just tried an "easy_install" on a Centos box which has Python 2.5 
and got:

    # easy_install reportlab
    Searching for reportlab
    Reading http://pypi.python.org/simple/reportlab/
    Reading http://www.reportlab.com/
    Reading http://www.reportlab.org/downloads.html
    Best match: reportlab daily-win32
    Downloading http://www.reportlab.org/ftp/reportlab-daily-win32.zip

Where did "reportlab-daily-win32.zip" come from...??


Thanks,

Dave

On 5/2/09 5:59 AM, Andy Robinson wrote:
> Version 2.3 of the ReportLab library has been released today.  New
> features are listed here:
>
>       http://www.reportlab.org/whatsnew_2_3.html
>
> It's also available on PyPI.
>
> Thanks to everyone who contributed patches, bug fixes and other feedback.
>
> Best Regards,
>
>
>   
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/reportlab-users/attachments/20090206/c39f6ea7/attachment.html>

From andy at reportlab.com  Fri Feb  6 04:09:36 2009
From: andy at reportlab.com (Andy Robinson)
Date: Fri, 6 Feb 2009 09:09:36 +0000
Subject: [reportlab-users] colwidths question
In-Reply-To: <487323a00902051458v1e42669cv51c1fe5e8735915f@mail.gmail.com>
References: <487323a00902051458v1e42669cv51c1fe5e8735915f@mail.gmail.com>
Message-ID: <956003ae0902060109p4033dfes753b94e848914db3@mail.gmail.com>

2009/2/5 James Yoo <james.yoo at gmail.com>:
> Hi... me again..
>
> There is something I am not understanding about colWidths...
> I'm trying to create a Table with colWidths = (28,10,10,10), those are the
> widest character lengths of the strings in my columns...

All units are in points (1/72 of an inch).   The table class will
auto-size for you if you omit the colWidths.

- Andy

From andy at reportlab.com  Fri Feb  6 04:14:39 2009
From: andy at reportlab.com (Andy Robinson)
Date: Fri, 6 Feb 2009 09:14:39 +0000
Subject: [reportlab-users] ReportLab 2.3 is out!
In-Reply-To: <498BB464.60803@bcs.co.nz>
References: <956003ae0902040859r70828034rc8118150f737cdaa@mail.gmail.com>
	<498BB464.60803@bcs.co.nz>
Message-ID: <956003ae0902060114y459eaa81s41ba03fad517f80f@mail.gmail.com>

2009/2/6 David Brooks <dave at bcs.co.nz>:
> I've just tried an "easy_install" on a Centos box which has Python 2.5 and
> got:

I haven't a clue.  There is no reference to this file in the
information we loaded onto PyPI, and we offer a classic
distutils-style setup script which works as far as we can tell.

To be clear, we do not support easy_install.    Clearly, it has
decided to (half-) support us in some kind of devious and broken
manner, by telling the world to fetch a random package from our web
site.

Maybe someone who understands pypi can tell us what's wrong with the
following, and how easy_install manages to get from what we shipped to
the unrelated file you mentioned...

     http://pypi.python.org/pypi/reportlab/2.3

The files are available there and on our own web site.

- Andy

From cito at online.de  Fri Feb  6 05:37:13 2009
From: cito at online.de (Christoph Zwerschke)
Date: Fri, 06 Feb 2009 11:37:13 +0100
Subject: [reportlab-users] ReportLab 2.3 is out!
In-Reply-To: <956003ae0902060114y459eaa81s41ba03fad517f80f@mail.gmail.com>
References: <956003ae0902040859r70828034rc8118150f737cdaa@mail.gmail.com>	<498BB464.60803@bcs.co.nz>
	<956003ae0902060114y459eaa81s41ba03fad517f80f@mail.gmail.com>
Message-ID: <498C12D9.7010903@online.de>

Andy Robinson schrieb:
> Maybe someone who understands pypi can tell us what's wrong with the
> following, and how easy_install manages to get from what we shipped to
> the unrelated file you mentioned...
> 
>      http://pypi.python.org/pypi/reportlab/2.3
> 
> The files are available there and on our own web site.

http://pypi.python.org/pypi/reportlab/2.0 is still active which has 
http://www.reportlab.org/downloads.html as download page and the daily 
snapshot is obviously the latest version it can find there.

-- Christoph

From robin at reportlab.com  Fri Feb  6 05:55:05 2009
From: robin at reportlab.com (Robin Becker)
Date: Fri, 06 Feb 2009 10:55:05 +0000
Subject: [reportlab-users] ReportLab 2.3 is out!
In-Reply-To: <498C12D9.7010903@online.de>
References: <956003ae0902040859r70828034rc8118150f737cdaa@mail.gmail.com>	<498BB464.60803@bcs.co.nz>	<956003ae0902060114y459eaa81s41ba03fad517f80f@mail.gmail.com>
	<498C12D9.7010903@online.de>
Message-ID: <498C1709.3030604@chamonix.reportlab.co.uk>

Christoph Zwerschke wrote:
> Andy Robinson schrieb:
>> Maybe someone who understands pypi can tell us what's wrong with the
>> following, and how easy_install manages to get from what we shipped to
>> the unrelated file you mentioned...
>>
>>      http://pypi.python.org/pypi/reportlab/2.3
>>
>> The files are available there and on our own web site.
> 
> http://pypi.python.org/pypi/reportlab/2.0 is still active which has 
> http://www.reportlab.org/downloads.html as download page and the daily 
> snapshot is obviously the latest version it can find there.
> 
> -- Christoph
........

I'm not sure what it means to find a version. Does easy_install attempt to guess 
what files to download from a web page that is intended to be read by humans? If 
so, easy_install must be an advanced program indeed.
-- 
Robin Becker

From andy at reportlab.com  Fri Feb  6 05:56:00 2009
From: andy at reportlab.com (Andy Robinson)
Date: Fri, 6 Feb 2009 10:56:00 +0000
Subject: [reportlab-users] ReportLab 2.3 is out!
In-Reply-To: <498C12D9.7010903@online.de>
References: <956003ae0902040859r70828034rc8118150f737cdaa@mail.gmail.com>
	<498BB464.60803@bcs.co.nz>
	<956003ae0902060114y459eaa81s41ba03fad517f80f@mail.gmail.com>
	<498C12D9.7010903@online.de>
Message-ID: <956003ae0902060256j425203d9x29ba611383a39411@mail.gmail.com>

2009/2/6 Christoph Zwerschke <cito at online.de>:
> http://pypi.python.org/pypi/reportlab/2.0 is still active which has
> http://www.reportlab.org/downloads.html as download page and the daily
> snapshot is obviously the latest version it can find there.


So easy_install parses our HTML and finds the most recent timestamped
file on it? How utterly fragile and moronic.

Their documentation is no use on this, except to suggest that I
can use --allow-hosts=None to stop the world from using easy_install
(arguably a good solution).

I guess I have to read the setuptools code in order to 'not support it' better

- Andy

From cito at online.de  Fri Feb  6 06:30:26 2009
From: cito at online.de (Christoph Zwerschke)
Date: Fri, 06 Feb 2009 12:30:26 +0100
Subject: [reportlab-users] ReportLab 2.3 is out!
In-Reply-To: <956003ae0902060256j425203d9x29ba611383a39411@mail.gmail.com>
References: <956003ae0902040859r70828034rc8118150f737cdaa@mail.gmail.com>	<498BB464.60803@bcs.co.nz>	<956003ae0902060114y459eaa81s41ba03fad517f80f@mail.gmail.com>	<498C12D9.7010903@online.de>
	<956003ae0902060256j425203d9x29ba611383a39411@mail.gmail.com>
Message-ID: <498C1F52.4000209@online.de>

Andy Robinson schrieb:
> 2009/2/6 Christoph Zwerschke <cito at online.de>:
>> http://pypi.python.org/pypi/reportlab/2.0 is still active which has
>> http://www.reportlab.org/downloads.html as download page and the daily
>> snapshot is obviously the latest version it can find there.
> 
> So easy_install parses our HTML and finds the most recent timestamped
> file on it? How utterly fragile and moronic.

Yes, it scans the links in the HTML code. But the selection is done by 
interpretation of the name of the URLs (which can contain version, 
checksum etc.) and introspection of the packages, not by timestamp.

There is some documentation here:
http://peak.telecommunity.com/DevCenter/setuptools#making-your-package-available-for-easyinstall
http://peak.telecommunity.com/DevCenter/EasyInstall#package-index-api

Daily snapshots should be built like this:
http://peak.telecommunity.com/DevCenter/setuptools#egg-info-examples

-- Christoph


From andy at reportlab.com  Fri Feb  6 06:59:59 2009
From: andy at reportlab.com (Andy Robinson)
Date: Fri, 6 Feb 2009 11:59:59 +0000
Subject: [reportlab-users] ReportLab 2.3 is out!
In-Reply-To: <498C1F52.4000209@online.de>
References: <956003ae0902040859r70828034rc8118150f737cdaa@mail.gmail.com>
	<498BB464.60803@bcs.co.nz>
	<956003ae0902060114y459eaa81s41ba03fad517f80f@mail.gmail.com>
	<498C12D9.7010903@online.de>
	<956003ae0902060256j425203d9x29ba611383a39411@mail.gmail.com>
	<498C1F52.4000209@online.de>
Message-ID: <956003ae0902060359w47d2e4aaxc7829d524c8c254d@mail.gmail.com>

2009/2/6 Christoph Zwerschke <cito at online.de>:
> Yes, it scans the links in the HTML code. But the selection is done by
> interpretation of the name of the URLs (which can contain version, checksum
> etc.) and introspection of the packages, not by timestamp.

Thanks for the links, but we already studied those and they don't help.
Everything on PyPI looks correctly configured; 2.0 is marked as 'hidden'
but I see no reason why we should take it down.

I had to trace through the setuptools code as it runs to find the problem.
It parses a simple listing file on python.org with the distros we uploade
(which seems reasonable), and also spiders our whole damned site,
looking for files ending in .tar.gz, .tgz, .zip, .egg etc.  But it
makes assumptions about the way packages are named, and the fact
that we used "ReportLab_2_3.tgz" instead of "reportlab-2.3.tgz" causes it
to skip past and say "unknown version" on almost all of them.

I went for a quick workaround and removed the download URLs from PyPI,
so easy_install at least gives up now rather than getting the wrong package.

I'll try a few more experiments and update people later.  Changing the
package names we used for 7 years is frankly a major pain in the arse,
but maybe we can rename the output packages uploaded to PyPI

- Andy

From andy at reportlab.com  Fri Feb  6 07:38:59 2009
From: andy at reportlab.com (Andy Robinson)
Date: Fri, 6 Feb 2009 12:38:59 +0000
Subject: [reportlab-users] ReportLab 2.3 is out!
In-Reply-To: <956003ae0902060359w47d2e4aaxc7829d524c8c254d@mail.gmail.com>
References: <956003ae0902040859r70828034rc8118150f737cdaa@mail.gmail.com>
	<498BB464.60803@bcs.co.nz>
	<956003ae0902060114y459eaa81s41ba03fad517f80f@mail.gmail.com>
	<498C12D9.7010903@online.de>
	<956003ae0902060256j425203d9x29ba611383a39411@mail.gmail.com>
	<498C1F52.4000209@online.de>
	<956003ae0902060359w47d2e4aaxc7829d524c8c254d@mail.gmail.com>
Message-ID: <956003ae0902060438h5d9fe286t6083c50339c2afe8@mail.gmail.com>

2009/2/6 Andy Robinson <andy at reportlab.com>:
> I'll try a few more experiments and update people later.  Changing the
> package names we used for 7 years is frankly a major pain in the arse,
> but maybe we can rename the output packages uploaded to PyPI

OK, I renamed the packages on PyPI, and it finds them.  We'll probably
change our naming conventions; what he expects seems reasonable
and common...

http://peak.telecommunity.com/DevCenter/setuptools#specifying-your-project-s-version

easy_install now finds the latest package fast, but fails for less trivial
reasons (compiler errors with libart).  However, when I unpack and run
"python setup.py install", it works perfectly.    I suspect the problem
is that our source distro is made with 'svn export' and 'tar', and lacks
some magic stuff 'sdist' does.

This is as far as we can take it today.   If anyone has time to look at
why easy_install can't handle our working (classic) setup script, I'd be
interested, but I have to work on other stuff.  However, we are not
going to switch to a setuptools-based setup script at this time.

- Andy

From timr at probo.com  Fri Feb  6 12:38:43 2009
From: timr at probo.com (Tim Roberts)
Date: Fri, 06 Feb 2009 09:38:43 -0800
Subject: [reportlab-users] colwidths question
In-Reply-To: <200902051824.14795.aklaver@comcast.net>
References: <487323a00902051458v1e42669cv51c1fe5e8735915f@mail.gmail.com>
	<200902051824.14795.aklaver@comcast.net>
Message-ID: <498C75A3.7060606@probo.com>

Adrian Klaver wrote:
> If I remember correctly the units are in picas(1/72 inch).
>   

Just for completeness, Postscript's native units (which is what a PDF
is) are 1/72 inch, but those are points.  A pica is 12 points, or 1/6 inch.

There are other weird units you sometimes encounter.  A traditional
printer's point is actually just over 1/72.27 inch.  A "Didot point",
used in some European typesetting, is about 1/67.55 inch.  It was
defined as 1/72 of the "French royal inch", which was slightly longer
than the English inch.  A cicero is 12 Didot points.

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


From james.yoo at gmail.com  Fri Feb  6 13:07:22 2009
From: james.yoo at gmail.com (James Yoo)
Date: Fri, 6 Feb 2009 12:07:22 -0600
Subject: [reportlab-users] colwidths question
In-Reply-To: <498C75A3.7060606@probo.com>
References: <487323a00902051458v1e42669cv51c1fe5e8735915f@mail.gmail.com>
	<200902051824.14795.aklaver@comcast.net> <498C75A3.7060606@probo.com>
Message-ID: <487323a00902061007o139382e0o14e708694671c226@mail.gmail.com>

thanks everyone... I found a stringWidths function inside pdfmetrics module
and am using that with good results.


On Fri, Feb 6, 2009 at 11:38 AM, Tim Roberts <timr at probo.com> wrote:

> Adrian Klaver wrote:
> > If I remember correctly the units are in picas(1/72 inch).
> >
>
> Just for completeness, Postscript's native units (which is what a PDF
> is) are 1/72 inch, but those are points.  A pica is 12 points, or 1/6 inch.
>
> There are other weird units you sometimes encounter.  A traditional
> printer's point is actually just over 1/72.27 inch.  A "Didot point",
> used in some European typesetting, is about 1/67.55 inch.  It was
> defined as 1/72 of the "French royal inch", which was slightly longer
> than the English inch.  A cicero is 12 Didot points.
>
> --
> Tim Roberts, timr at probo.com
> Providenza & Boekelheide, Inc.
>
> _______________________________________________
> 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/20090206/da0ccf1a/attachment.htm>

From andy at reportlab.com  Fri Feb  6 14:48:50 2009
From: andy at reportlab.com (Andy Robinson)
Date: Fri, 6 Feb 2009 19:48:50 +0000
Subject: [reportlab-users] Download page tidied up a little
Message-ID: <956003ae0902061148ua37bd5t2ec0d275f183fad3@mail.gmail.com>

This page hopefully makes a lot more sense now...

    http://www.reportlab.org/downloads.html

We plan to redo the site completely this quarter, but hopefully this
will help in the meantime.

Best Regards,

-- 
Andy Robinson
CEO/Chief Architect
ReportLab Europe Ltd.

From Michael at Hipp.com  Fri Feb  6 16:26:20 2009
From: Michael at Hipp.com (Michael Hipp)
Date: Fri, 06 Feb 2009 15:26:20 -0600
Subject: [reportlab-users] Download page tidied up a little
In-Reply-To: <956003ae0902061148ua37bd5t2ec0d275f183fad3@mail.gmail.com>
References: <956003ae0902061148ua37bd5t2ec0d275f183fad3@mail.gmail.com>
Message-ID: <498CAAFC.8040602@Hipp.com>

Andy Robinson wrote:
> This page hopefully makes a lot more sense now...
> 
>     http://www.reportlab.org/downloads.html
> 
> We plan to redo the site completely this quarter, but hopefully this
> will help in the meantime.

Looks good; much clearer. Thanks.

Michael

From cfoster at oda.state.or.us  Mon Feb  9 18:52:48 2009
From: cfoster at oda.state.or.us (Chris Foster)
Date: Mon, 9 Feb 2009 15:52:48 -0800
Subject: [reportlab-users] Saving annotations with Platypus
Message-ID: <AE978569-49B6-4F3E-BE67-D39E18BDD4C9@oda.state.or.us>

I'm passing a "subject" to a DocTemplate, but it doesn't seem to  
stick.  Is there something I need to do to explicitly save it to the  
canvas?

From david at davidheath.org  Tue Feb 10 03:30:43 2009
From: david at davidheath.org (David Heath)
Date: Tue, 10 Feb 2009 08:30:43 +0000
Subject: [reportlab-users] Can I build a PDF from a template?
In-Reply-To: <345b57c50902090423n5793058bn38016ea7d5f6ccd5@mail.gmail.com>
References: <345b57c50902090423n5793058bn38016ea7d5f6ccd5@mail.gmail.com>
Message-ID: <345b57c50902100030h683e26cfo6a485f4e2e308f53@mail.gmail.com>

Hi,

I've been using Reportlab for a little while for generating PDFs for a web
application I'm working on. The library is excellent - very well thought
through and documented .. thanks!

I now have a need to allow a graphic designer to independently create and
amend visual designs for the generated PDFs, and have designated areas of
the design filled with dynamic content.

Do you have any suggestions about how to do this? Is there a way, for
example to import a base 'layer' from a PDF file, and then add additional
elements programmatically? Or is there some other vector format (XML:FO, PS
or EPS?) which could be imported and added to using reportlab. I'm not very
familiar with the export capabilities of current editing programs such as
illustrator (I'm a linux-based programmer :-)), so I'm open minded about
possible solutions.

Looking forward to hearing any thoughts or suggestions you can offer.

Best wishes

David Heath
-- 
http://davidheath.org/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/reportlab-users/attachments/20090210/576c95f8/attachment.html>

From gherman at darwin.in-berlin.de  Tue Feb 10 03:47:48 2009
From: gherman at darwin.in-berlin.de (Dinu Gherman)
Date: Tue, 10 Feb 2009 09:47:48 +0100
Subject: [reportlab-users] Can I build a PDF from a template?
In-Reply-To: <345b57c50902100030h683e26cfo6a485f4e2e308f53@mail.gmail.com>
References: <345b57c50902090423n5793058bn38016ea7d5f6ccd5@mail.gmail.com>
	<345b57c50902100030h683e26cfo6a485f4e2e308f53@mail.gmail.com>
Message-ID: <385F600C-4186-47C4-8B5D-1E599E7827B7@darwin.in-berlin.de>

David Heath:

> Looking forward to hearing any thoughts or suggestions you can offer.

There are many responses to this in the archive of this mailing list.

Regards,

Dinu


From cito at online.de  Tue Feb 10 04:58:54 2009
From: cito at online.de (Christoph Zwerschke)
Date: Tue, 10 Feb 2009 10:58:54 +0100
Subject: [reportlab-users] Can I build a PDF from a template?
In-Reply-To: <385F600C-4186-47C4-8B5D-1E599E7827B7@darwin.in-berlin.de>
References: <345b57c50902090423n5793058bn38016ea7d5f6ccd5@mail.gmail.com>	<345b57c50902100030h683e26cfo6a485f4e2e308f53@mail.gmail.com>
	<385F600C-4186-47C4-8B5D-1E599E7827B7@darwin.in-berlin.de>
Message-ID: <49914FDE.2040702@online.de>

Dinu Gherman schrieb:
>> Looking forward to hearing any thoughts or suggestions you can offer.
> 
> There are many responses to this in the archive of this mailing list.

Which indicates that it should be added to the FAQ (e.g. as 2.1.6).

-- Christoph

From wthie at thiengineering.ch  Tue Feb 10 05:11:39 2009
From: wthie at thiengineering.ch (Werner Thie)
Date: Tue, 10 Feb 2009 11:11:39 +0100
Subject: [reportlab-users] Can I build a PDF from a template?
In-Reply-To: <49914FDE.2040702@online.de>
References: <345b57c50902090423n5793058bn38016ea7d5f6ccd5@mail.gmail.com>	<345b57c50902100030h683e26cfo6a485f4e2e308f53@mail.gmail.com>	<385F600C-4186-47C4-8B5D-1E599E7827B7@darwin.in-berlin.de>
	<49914FDE.2040702@online.de>
Message-ID: <499152DB.1010103@thiengineering.ch>

I personally use the z3c.rml implementation and are quite happy with it. 
You get a FOSS reimplementation of RML which does PDF as a template 
thing and a lot more.

HTH, Werner

Christoph Zwerschke wrote:
> Dinu Gherman schrieb:
>>> Looking forward to hearing any thoughts or suggestions you can offer.
>> There are many responses to this in the archive of this mailing list.
> 
> Which indicates that it should be added to the FAQ (e.g. as 2.1.6).
> 
> -- Christoph
> _______________________________________________
> reportlab-users mailing list
> reportlab-users at reportlab.com
> http://two.pairlist.net/mailman/listinfo/reportlab-users

From robin at reportlab.com  Tue Feb 10 05:32:06 2009
From: robin at reportlab.com (Robin Becker)
Date: Tue, 10 Feb 2009 10:32:06 +0000
Subject: [reportlab-users] Saving annotations with Platypus
In-Reply-To: <AE978569-49B6-4F3E-BE67-D39E18BDD4C9@oda.state.or.us>
References: <AE978569-49B6-4F3E-BE67-D39E18BDD4C9@oda.state.or.us>
Message-ID: <499157A6.8050304@chamonix.reportlab.co.uk>

Chris Foster wrote:
> I'm passing a "subject" to a DocTemplate, but it doesn't seem to stick.  
> Is there something I need to do to explicitly save it to the canvas?
> _______________________________________________
......
well I've just tried this in test_platypus_general.py. I added subject='test0' 
so that line 555 in func run lookes like

doc = AndyTemplate(outputfile('test_platypus_general.pdf'),subject='test0')

when I run the test the document  test_platypus_general.pdf does have a subject 
of test0 (as viewed in the document properties). However, this is not a standard 
annotation, but a PDFInfo object which is a special dictionary that's part of 
the PDF trailer.
-- 
Robin Becker

From david at davidheath.org  Tue Feb 10 05:33:14 2009
From: david at davidheath.org (David Heath)
Date: Tue, 10 Feb 2009 10:33:14 +0000
Subject: [reportlab-users] Can I build a PDF from a template?
In-Reply-To: <499152DB.1010103@thiengineering.ch>
References: <345b57c50902090423n5793058bn38016ea7d5f6ccd5@mail.gmail.com>
	<345b57c50902100030h683e26cfo6a485f4e2e308f53@mail.gmail.com>
	<385F600C-4186-47C4-8B5D-1E599E7827B7@darwin.in-berlin.de>
	<49914FDE.2040702@online.de> <499152DB.1010103@thiengineering.ch>
Message-ID: <345b57c50902100233q6761b5e5ga29dc983bd46acfc@mail.gmail.com>

On 2/10/09, Werner Thie <wthie at thiengineering.ch> wrote:
> I personally use the z3c.rml implementation and are quite happy with it.
> You get a FOSS reimplementation of RML which does PDF as a template
> thing and a lot more.

Hi Werner,

thanks that sounds very useful and interesting, I hadn't heard of the
z3c.rml implementation. I had a look at the documentation, but I
couldn't see any obvious way to use a PDF file as a template. Would
you be able to give me a simple example, or point me in the right
direction?

thanks

David

From andy at reportlab.com  Tue Feb 10 06:09:43 2009
From: andy at reportlab.com (Andy Robinson)
Date: Tue, 10 Feb 2009 11:09:43 +0000
Subject: [reportlab-users] Can I build a PDF from a template?
In-Reply-To: <49914FDE.2040702@online.de>
References: <345b57c50902090423n5793058bn38016ea7d5f6ccd5@mail.gmail.com>
	<345b57c50902100030h683e26cfo6a485f4e2e308f53@mail.gmail.com>
	<385F600C-4186-47C4-8B5D-1E599E7827B7@darwin.in-berlin.de>
	<49914FDE.2040702@online.de>
Message-ID: <956003ae0902100309y3db8a0d1oe259489c5f0cd758@mail.gmail.com>

2009/2/10 Christoph Zwerschke <cito at online.de>:
> Dinu Gherman schrieb:
>>> Looking forward to hearing any thoughts or suggestions you can offer.
>>
>> There are many responses to this in the archive of this mailing list.
>
> Which indicates that it should be added to the FAQ (e.g. as 2.1.6).
>

Done.

- Andy

From david at davidheath.org  Tue Feb 10 08:01:38 2009
From: david at davidheath.org (David Heath)
Date: Tue, 10 Feb 2009 13:01:38 +0000
Subject: [reportlab-users] Can I build a PDF from a template?
In-Reply-To: <4991702B.6010907@thiengineering.ch>
References: <345b57c50902090423n5793058bn38016ea7d5f6ccd5@mail.gmail.com>
	<345b57c50902100030h683e26cfo6a485f4e2e308f53@mail.gmail.com>
	<385F600C-4186-47C4-8B5D-1E599E7827B7@darwin.in-berlin.de>
	<49914FDE.2040702@online.de> <499152DB.1010103@thiengineering.ch>
	<345b57c50902100233q6761b5e5ga29dc983bd46acfc@mail.gmail.com>
	<4991702B.6010907@thiengineering.ch>
Message-ID: <345b57c50902100501t5afdf7c7jef02f6b1363527c6@mail.gmail.com>

> some .pdf file called fw2.pdf is merged into the current document and
> then gets something drawn on top. I doubt that there is an out of the
> box solution which pulls up .pdf back to the same level as RML and let
> you change it, but for me the mergePage solved all my templating
> business, as far as graphics or or letterhead or other prefabbed pdf
> stuff goes.

Great, that sounds like it would do the job. I just need a 'decorative
background' above which I'll add additional content.

Thanks for your help.

David

From cito at online.de  Tue Feb 10 09:03:35 2009
From: cito at online.de (Christoph Zwerschke)
Date: Tue, 10 Feb 2009 15:03:35 +0100
Subject: [reportlab-users] Can I build a PDF from a template?
In-Reply-To: <956003ae0902100309y3db8a0d1oe259489c5f0cd758@mail.gmail.com>
References: <345b57c50902090423n5793058bn38016ea7d5f6ccd5@mail.gmail.com>	<345b57c50902100030h683e26cfo6a485f4e2e308f53@mail.gmail.com>	<385F600C-4186-47C4-8B5D-1E599E7827B7@darwin.in-berlin.de>	<49914FDE.2040702@online.de>
	<956003ae0902100309y3db8a0d1oe259489c5f0cd758@mail.gmail.com>
Message-ID: <49918937.7090901@online.de>

Andy Robinson schrieb:
> 2009/2/10 Christoph Zwerschke <cito at online.de>:
>> Dinu Gherman schrieb:
>>> There are many responses to this in the archive of this mailing list.
>> Which indicates that it should be added to the FAQ (e.g. as 2.1.6).
> 
> Done.

Thanks, though I think you answered a different question than the one of
the op ;-)

-- Christoph

From cfoster at oda.state.or.us  Tue Feb 10 18:58:15 2009
From: cfoster at oda.state.or.us (Chris Foster)
Date: Tue, 10 Feb 2009 15:58:15 -0800
Subject: [reportlab-users] Saving annotations with Platypus
In-Reply-To: <499157A6.8050304@chamonix.reportlab.co.uk>
References: <AE978569-49B6-4F3E-BE67-D39E18BDD4C9@oda.state.or.us>
	<499157A6.8050304@chamonix.reportlab.co.uk>
Message-ID: <2F658387-82B7-4FAF-B29D-6E398ED566C1@oda.state.or.us>

Thank you.

Your test code works fine and I re-implemented my DocTemplate subclass  
to closely match the test code and it is now working.  I can see the  
results with the Mac Preview application BTW.  However, I'm still  
missing something subtle.  I had to explicitly save the canvas.  It  
looks like this should have happened by default in _endBuild()?

On Feb 10, 2009, at 2:32 AM, Robin Becker wrote:

> Chris Foster wrote:
>> I'm passing a "subject" to a DocTemplate, but it doesn't seem to  
>> stick.  Is there something I need to do to explicitly save it to  
>> the canvas?
>> _______________________________________________
> ......
> well I've just tried this in test_platypus_general.py. I added  
> subject='test0' so that line 555 in func run lookes like
>
> doc =  
> AndyTemplate(outputfile('test_platypus_general.pdf'),subject='test0')
>
> when I run the test the document  test_platypus_general.pdf does  
> have a subject of test0 (as viewed in the document properties).  
> However, this is not a standard annotation, but a PDFInfo object  
> which is a special dictionary that's part of the PDF trailer.
> -- 
> Robin Becker
> _______________________________________________
> reportlab-users mailing list
> reportlab-users at reportlab.com
> http://two.pairlist.net/mailman/listinfo/reportlab-users


From gogtesuyash at gmail.com  Wed Feb 11 05:50:16 2009
From: gogtesuyash at gmail.com (Suyash Gogte)
Date: Wed, 11 Feb 2009 16:20:16 +0530
Subject: [reportlab-users] How to add two flowables horizontally
Message-ID: <5797af200902110250q9eb1126nd66ecd099ea1638b@mail.gmail.com>

hi frndz...
i m adding one table of say 8 rows n piechart....but i want to add them on
my Doc horizontally ...means want to add them side by side or on same
line...
How can i do this...i m using Story object which has append method which
appends vertically....means
Plz do reply...Its urgent
thnks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/reportlab-users/attachments/20090211/80695970/attachment.html>

From robin at reportlab.com  Wed Feb 11 05:54:27 2009
From: robin at reportlab.com (Robin Becker)
Date: Wed, 11 Feb 2009 10:54:27 +0000
Subject: [reportlab-users] Saving annotations with Platypus
In-Reply-To: <2F658387-82B7-4FAF-B29D-6E398ED566C1@oda.state.or.us>
References: <AE978569-49B6-4F3E-BE67-D39E18BDD4C9@oda.state.or.us>	<499157A6.8050304@chamonix.reportlab.co.uk>
	<2F658387-82B7-4FAF-B29D-6E398ED566C1@oda.state.or.us>
Message-ID: <4992AE63.10403@chamonix.reportlab.co.uk>

Chris Foster wrote:
> Thank you.
> 
> Your test code works fine and I re-implemented my DocTemplate subclass 
> to closely match the test code and it is now working.  I can see the 
> results with the Mac Preview application BTW.  However, I'm still 
> missing something subtle.  I had to explicitly save the canvas.  It 
> looks like this should have happened by default in _endBuild()?
> 
........
Certainly if you are calling all the standard methods eg build/multiBuild then 
_endBuild gets called and providing you don't mess with _doSave the canvas gets 
saved. Since you've got your own class perhaps you've missed out a call to 
something.

Without seeing your class it's pretty hard to figure out what's going wrong.
-- 
Robin Becker

From wietse.j at gmail.com  Wed Feb 11 06:33:21 2009
From: wietse.j at gmail.com (Wietse Jacobs)
Date: Wed, 11 Feb 2009 12:33:21 +0100
Subject: [reportlab-users] How to add two flowables horizontally
In-Reply-To: <5797af200902110250q9eb1126nd66ecd099ea1638b@mail.gmail.com>
References: <5797af200902110250q9eb1126nd66ecd099ea1638b@mail.gmail.com>
Message-ID: <7d9c65800902110333v2d63d289o19070e5b86f03d35@mail.gmail.com>

2009/2/11 Suyash Gogte <gogtesuyash at gmail.com>:
> i m adding one table of say 8 rows n piechart....but i want to add them on
> my Doc horizontally ...means want to add them side by side or on same
> line...
> How can i do this...i m using Story object which has append method which
> appends vertically....means

You can put them both in a table with 1 row:
    story.append(Table([[chart_obj, table_obj]]), style=appropriate_style)

-- 
--Wietse

From gogtesuyash at gmail.com  Wed Feb 11 07:20:22 2009
From: gogtesuyash at gmail.com (Suyash Gogte)
Date: Wed, 11 Feb 2009 17:50:22 +0530
Subject: [reportlab-users] How to align object
Message-ID: <5797af200902110420n45f4d129i858d6b9f75b2adf4@mail.gmail.com>

hi frnds
can anyone tell me how to align boject that is passsed to Story object
e.g
Story.append(table Instance)
i want to align this table instance how to do that
thnks in advance
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/reportlab-users/attachments/20090211/c01b0eb7/attachment.html>

From meyer at mesw.de  Wed Feb 11 07:23:19 2009
From: meyer at mesw.de (Markus Meyer)
Date: Wed, 11 Feb 2009 13:23:19 +0100
Subject: [reportlab-users] EAN-5 barcodes?
Message-ID: <4992C337.8050602@mesw.de>

Hi everyone,

who is responsible for maintaining the barcode functionality in 
ReportLab? I see that ReportLab has support for creating EAN-8 and 
EAN-13 barcodes, but I would like to create an EAN-5 barcode, which is 
often used next to a ISBN barcode to designate the book price.

I would even be willing to pay someone for adding this functionality if 
it is not trivial.

Regards,
Markus


From gogtesuyash at gmail.com  Wed Feb 11 07:34:46 2009
From: gogtesuyash at gmail.com (Suyash Gogte)
Date: Wed, 11 Feb 2009 18:04:46 +0530
Subject: [reportlab-users] How to add two flowables horizontally
In-Reply-To: <7d9c65800902110333v2d63d289o19070e5b86f03d35@mail.gmail.com>
References: <5797af200902110250q9eb1126nd66ecd099ea1638b@mail.gmail.com>
	<7d9c65800902110333v2d63d289o19070e5b86f03d35@mail.gmail.com>
Message-ID: <5797af200902110434k4b1f0f84k42e7135f25bae109@mail.gmail.com>

thnks its working but can u plz tell me......how to align it means
for e.g

Global Data :


On Wed, Feb 11, 2009 at 5:03 PM, Wietse Jacobs <wietse.j at gmail.com> wrote:

> 2009/2/11 Suyash Gogte <gogtesuyash at gmail.com>:
> > i m adding one table of say 8 rows n piechart....but i want to add them
> on
> > my Doc horizontally ...means want to add them side by side or on same
> > line...
> > How can i do this...i m using Story object which has append method which
> > appends vertically....means
>
> You can put them both in a table with 1 row:
>    story.append(Table([[chart_obj, table_obj]]), style=appropriate_style)
>
> --
> --Wietse
> _______________________________________________
> 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/20090211/20ff835b/attachment.htm>

From wietse.j at gmail.com  Wed Feb 11 08:30:20 2009
From: wietse.j at gmail.com (Wietse Jacobs)
Date: Wed, 11 Feb 2009 14:30:20 +0100
Subject: [reportlab-users] How to add two flowables horizontally
In-Reply-To: <5797af200902110434k4b1f0f84k42e7135f25bae109@mail.gmail.com>
References: <5797af200902110250q9eb1126nd66ecd099ea1638b@mail.gmail.com>
	<7d9c65800902110333v2d63d289o19070e5b86f03d35@mail.gmail.com>
	<5797af200902110434k4b1f0f84k42e7135f25bae109@mail.gmail.com>
Message-ID: <7d9c65800902110530q5a3ecb9ex8684e4c1b5cb2570@mail.gmail.com>

2009/2/11 Suyash Gogte <gogtesuyash at gmail.com>:
> thnks its working but can u plz tell me......how to align it means
> for e.g
>
> Global Data :

That's chapter 7 of the User Guide: "Tables and TableStyles".

    story.append(Table([[chart_obj, table_obj]], style=appropriate_style))

with an appropriate appropriate_style, e.g.:

    appropriate_style = [
        ('VALIGN', (0,0), (-1,-1), 'MIDDLE'),
        ('ALIGN', (0,0), (0,-1), 'RIGHT'),
        ('ALIGN', (1,0), (1,-1), 'LEFT'),
        ('GRID', (0,0), (-1,-1), 1, colors.pink),
    ]

-- 
--Wietse

From robin at reportlab.com  Wed Feb 11 08:57:24 2009
From: robin at reportlab.com (Robin Becker)
Date: Wed, 11 Feb 2009 13:57:24 +0000
Subject: [reportlab-users] EAN-5 barcodes?
In-Reply-To: <4992C337.8050602@mesw.de>
References: <4992C337.8050602@mesw.de>
Message-ID: <4992D944.5040108@chamonix.reportlab.co.uk>

Markus Meyer wrote:
> Hi everyone,
> 
> who is responsible for maintaining the barcode functionality in 
> ReportLab? I see that ReportLab has support for creating EAN-8 and 
> EAN-13 barcodes, but I would like to create an EAN-5 barcode, which is 
> often used next to a ISBN barcode to designate the book price.
> 
> I would even be willing to pay someone for adding this functionality if 
> it is not trivial.
> 
> Regards,
> Markus
......

the original barcode coding was contributed by Ty Sarna, but I have been 
maintaining and improving it for a while now. As I understand it ean-5 is an 
extension to ean-13 (as is ean-8), but its structure differs somewhat. I don't 
think this would take too long, but we're flat out at present.
-- 
Robin Becker

From meyer at mesw.de  Wed Feb 11 09:43:25 2009
From: meyer at mesw.de (Markus Meyer)
Date: Wed, 11 Feb 2009 15:43:25 +0100
Subject: [reportlab-users] EAN-5 barcodes?
In-Reply-To: <4992D944.5040108@chamonix.reportlab.co.uk>
References: <4992C337.8050602@mesw.de>
	<4992D944.5040108@chamonix.reportlab.co.uk>
Message-ID: <4992E40D.7040506@mesw.de>

Robin Becker schrieb:
> the original barcode coding was contributed by Ty Sarna, but I have been 
> maintaining and improving it for a while now. As I understand it ean-5 
> is an extension to ean-13 (as is ean-8), but its structure differs 
> somewhat. I don't think this would take too long, but we're flat out at 
> present.

What a pity! What would be a time frame for your availability?

Anyone else?


Markus

From gherman at darwin.in-berlin.de  Wed Feb 11 11:44:14 2009
From: gherman at darwin.in-berlin.de (Dinu Gherman)
Date: Wed, 11 Feb 2009 17:44:14 +0100
Subject: [reportlab-users] Paragraphs as chart titles?
Message-ID: <2BD2B9CF-FA6B-41FD-954F-E23F8F0CC312@darwin.in-berlin.de>

Hi,

since it seems to be hard to guarantee that a paragraph and a
subsequent chart will not be split by Platypus, I'd like to use
multi-line paragraphs (of initially unknown length) as chart
titles, but can find only the String shape to add to drawings
(where I'd need to do the line wrapping myself).

Has anybody found a straightforward way of combining Paragraphs
and charts?

Thanks,

Dinu


From andy at reportlab.com  Wed Feb 11 12:30:56 2009
From: andy at reportlab.com (Andy Robinson)
Date: Wed, 11 Feb 2009 17:30:56 +0000
Subject: [reportlab-users] Paragraphs as chart titles?
In-Reply-To: <2BD2B9CF-FA6B-41FD-954F-E23F8F0CC312@darwin.in-berlin.de>
References: <2BD2B9CF-FA6B-41FD-954F-E23F8F0CC312@darwin.in-berlin.de>
Message-ID: <956003ae0902110930s65f8ba73t89cc75ba06a5a2c2@mail.gmail.com>

2009/2/11 Dinu Gherman <gherman at darwin.in-berlin.de>:
> Has anybody found a straightforward way of combining Paragraphs
> and charts?

We haven't.  With hindsight or time for a rewrite, we would never have
distinguished between charts and Platypus objects, but it's too late.

However I think there is some kind of label class used for data points
which can wrap text if it exceeds the desired width - but without any
intra-paragraph formatting.

- Andy

From gogtesuyash at gmail.com  Wed Feb 11 22:56:42 2009
From: gogtesuyash at gmail.com (Suyash Gogte)
Date: Thu, 12 Feb 2009 09:26:42 +0530
Subject: [reportlab-users] How to add two flowables horizontally
In-Reply-To: <7d9c65800902110530q5a3ecb9ex8684e4c1b5cb2570@mail.gmail.com>
References: <5797af200902110250q9eb1126nd66ecd099ea1638b@mail.gmail.com>
	<7d9c65800902110333v2d63d289o19070e5b86f03d35@mail.gmail.com>
	<5797af200902110434k4b1f0f84k42e7135f25bae109@mail.gmail.com>
	<7d9c65800902110530q5a3ecb9ex8684e4c1b5cb2570@mail.gmail.com>
Message-ID: <5797af200902111956p245eb99cx765b8ac01f575f42@mail.gmail.com>

yes we can use style but its only for data or contents of the table...we can
align content of cell..using above styles...but i want to align table...

On Wed, Feb 11, 2009 at 7:00 PM, Wietse Jacobs <wietse.j at gmail.com> wrote:

> 2009/2/11 Suyash Gogte <gogtesuyash at gmail.com>:
> > thnks its working but can u plz tell me......how to align it means
> > for e.g
> >
> > Global Data :
>
> That's chapter 7 of the User Guide: "Tables and TableStyles".
>
>    story.append(Table([[chart_obj, table_obj]], style=appropriate_style))
>
> with an appropriate appropriate_style, e.g.:
>
>    appropriate_style = [
>        ('VALIGN', (0,0), (-1,-1), 'MIDDLE'),
>        ('ALIGN', (0,0), (0,-1), 'RIGHT'),
>        ('ALIGN', (1,0), (1,-1), 'LEFT'),
>        ('GRID', (0,0), (-1,-1), 1, colors.pink),
>    ]
>
> --
> --Wietse
> _______________________________________________
> 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/20090212/9bfbe5e1/attachment.html>

From david at davidheath.org  Thu Feb 12 08:09:30 2009
From: david at davidheath.org (David Heath)
Date: Thu, 12 Feb 2009 13:09:30 +0000
Subject: [reportlab-users] Can I build a PDF from a template - small
	freelance job?
Message-ID: <345b57c50902120509v4e9b1adcpdb0653fa60fbcbc9@mail.gmail.com>

On 2/10/09, David Heath <david at davidheath.org> wrote:
>> some .pdf file called fw2.pdf is merged into the current document and
>> then gets something drawn on top. I doubt that there is an out of the
>> box solution which pulls up .pdf back to the same level as RML and let
>> you change it, but for me the mergePage solved all my templating
>> business, as far as graphics or or letterhead or other prefabbed pdf
>> stuff goes.
>
> Great, that sounds like it would do the job. I just need a 'decorative
> background' above which I'll add additional content.

Due to time pressures, I don't have time to work on this. Are there
any experienced reportlab hackers on this list who would be willing to
do this as a small freelance job?

The requirements are to produce a python script. The script would
generate a single-page PDF file used as a postal address wrapper.

The script will be given as input a PDF 'template' file plus a plain
text JSON of the following structure:

{
"from": "Gordon Brown, 10 Downing Street, London W1A 1AA, UK",
"locale": "us",
"to": ["Daniel Jackson", "124 Pleasant St", "Portsmouth", "Hants", "UK"]
}

locale is either 'us' or 'uk'. For US a US-letter size page should be
produced, for UK an A4 page should be produced (this may be determined
instead by the input template file).

The script will combine the 'template' file and add to it the from and
to addresses. I can envisage several possible levels of functionality:

1. in the most basic version, the sender and recipient addresses would
be positioned on the page in fixed known positions. This is the least
desirable of the three options, but ok if nothing else is possible.
2. in a more sophisticated version, the positioning, bounding box,
font size (size, weight, typeface) of the sender and recipient
addresses would be supplied as input data in the JSON data structure
3. in the most sophisticated version, the positioning of the sender
and recipient addresses would be determined by specially tagged
elements or placeholders in the PDF template which would indicate the
position, font style and bounding box of the elements.

I don't know if (3) is possible at all with reportlab.

If you're interested please contact me privately to discuss.

Thanks

David Heath
david at davidheath.org

From peter at maubp.freeserve.co.uk  Thu Feb 12 08:55:55 2009
From: peter at maubp.freeserve.co.uk (Peter)
Date: Thu, 12 Feb 2009 13:55:55 +0000
Subject: [reportlab-users] External hyperlinks in SVG output
In-Reply-To: <956003ae0901290838g6393c59dxb36e52626ff6429c@mail.gmail.com>
References: <320fb6e00901290737k1269fb4en92975edbcf87f88a@mail.gmail.com>
	<956003ae0901290838g6393c59dxb36e52626ff6429c@mail.gmail.com>
Message-ID: <320fb6e00902120555i37c3e7c8ibb0332901c81c0b8@mail.gmail.com>

On Thu, Jan 29, 2009 at 4:38 PM, Andy Robinson <andy at reportlab.com> wrote:
> There are two parts to this. ?First, the shapes hierarchy in
> reportlab/graphics would need some extra properties so that
> a shape could have link properties. ?This is straightforward for
> manually constructed shapes, but ...

The list of properties for HTML like output is potentially quite long,
see http://www.w3.org/TR/REC-html40/struct/links.html - and would be
different for SVG, PDF, ... output.  The core two for me are the HREF
or URL, and the TITLE (usually rendered as a tooltip).  Another
interesting one in HTML is the TARGET (useful if you want to have the
link open in another window), although for SVG is appear they prefer
the new SHOW attribute).  There are also things like mouse over, on
click etc.

What can a PDF (or PS) hyperlink do, or what attributes can is have?
There is the link URL itself, but is that it?

If we added a dictionary (or object) to the basic shape objects
(called "link" or "hyperlink" or whatever) then the user could put as
many attributes into the dictionary as they liked.  The different
render backends would each look at a different subset of these
entries.  This would be extendible, but for the end user could require
different properties to be set depending on the intended output file
format.  I also feel this dictionary approach would be a departure
form current ReportLab style.  For example, font properties are not
bundled together but are held as multiple properties (fontface,
fontsize, ...).

Therefore, perhaps we should instead implement just a handful of core
link properties.  e.g. "href_url" as a string, "href_tooltip" as a
string, and perhaps "href_new_window" as a boolean (an HTML image map
back end would use target="_blank", an SVG or XML backend might use
show="new", while for a PDF this has probably has no real meaning and
would be ignored).  This could be extended gradually if other
attributes were deemed sufficiently important.

> Secondly, each renderer would need to do the right thing.
> I can see this would be really easy with SVG.
>
> If you're willing to do a little work on this and contribute
> patches for core and SVG back end, we'd be happy to help out and
> maybe do the same in PDF in step with it.

I've been experimenting with an HTML image map renderer, inspired by
Dirk Datzert's approach (see July 2003 email linked to earlier) but
with additional support for polygons.  It occurs to me that even
strings with links could be dealt with here by generating a link
region from their bounding box.  Other basic ReportLab shapes include
the ellipse which is not support by an HTML image map, so a crude
polygon approximation may be required for completeness.  However, for
the moment, my biggest unsolved problem is dealing with group
transformations - I expect looking more closely at the existing render
backends will help here.

Trying to extend renderSVG to cope with links is on my to do list -
but it would help to have an idea of how you would prefer to store the
extra attributes within the basic shapes (discussed above).

Thanks,

Peter

From cito at online.de  Thu Feb 12 11:38:53 2009
From: cito at online.de (Christoph Zwerschke)
Date: Thu, 12 Feb 2009 17:38:53 +0100
Subject: [reportlab-users] Problem with importing Image from PIL
Message-ID: <4994509D.5040007@online.de>

I noticed that in reportlab.lib.utils, the main "import PIL.Image" is 
done with a fallback to "import Image" if it was installed top-level. 
That's good.

However, there are two more places in the same file which only try 
"import PIL.Image" and do not try "import Image". These places should 
use the same fallback or even simpler just use the Image that has 
already been imported as a global variable.

-- Christoph

From peter at maubp.freeserve.co.uk  Thu Feb 12 18:09:22 2009
From: peter at maubp.freeserve.co.uk (Peter)
Date: Thu, 12 Feb 2009 23:09:22 +0000
Subject: [reportlab-users] External hyperlinks in SVG output
In-Reply-To: <320fb6e00902120555i37c3e7c8ibb0332901c81c0b8@mail.gmail.com>
References: <320fb6e00901290737k1269fb4en92975edbcf87f88a@mail.gmail.com>
	<956003ae0901290838g6393c59dxb36e52626ff6429c@mail.gmail.com>
	<320fb6e00902120555i37c3e7c8ibb0332901c81c0b8@mail.gmail.com>
Message-ID: <320fb6e00902121509n3233235dlf44cc1ba6604423b@mail.gmail.com>

On Thu, Feb 12, 2009 at 1:55 PM, Peter <peter at maubp.freeserve.co.uk> wrote:
>
> Trying to extend renderSVG to cope with links is on my to do list -
> but it would help to have an idea of how you would prefer to store the
> extra attributes within the basic shapes (discussed above).
>

I've got something with SVG links working here, but I had to make a
refinement to this bit of renderSVG.py:

        #these suggested by Tim Roberts
        self.svg.setAttribute("xmlns", "http://www.w3.org/2000/svg")
        self.svg.setAttribute("xmlns:link", "http://www.w3.org/1999/xlink")
        self.svg.setAttribute("version", "1.0")
        self.svg.setAttribute("baseProfile", "full")

I found my links don't work (currently testing on Firefox 1.5 on
Linux) unless I change the "xmlns:link" to read "xmlns:xlink" instead.
 This change is consistent with most online examples, e.g. the
official examples on http://www.w3.org/TR/SVG10/linking.html or
http://www.w3.org/TR/SVG/linking.html but some cases of the "wrong"
form do crop up via Google.  With this fix the
http://validator.w3.org/ is also much happier - but still grumbles.

Apparently "version" and "baseProfile" shouldn't be there for the
currently used doctype. Interestingly renderSVG doesn't specify the
doctype, letting python's xml.dom decide, so it could vary with
version of python (and I am using python 2.4 on this machine).  We
should be able to specify them when creating the DOM...

Peter

From robin at reportlab.com  Fri Feb 13 05:01:08 2009
From: robin at reportlab.com (Robin Becker)
Date: Fri, 13 Feb 2009 10:01:08 +0000
Subject: [reportlab-users] Problem with importing Image from PIL
In-Reply-To: <4994509D.5040007@online.de>
References: <4994509D.5040007@online.de>
Message-ID: <499544E4.5090802@chamonix.reportlab.co.uk>

Christoph Zwerschke wrote:
> I noticed that in reportlab.lib.utils, the main "import PIL.Image" is 
> done with a fallback to "import Image" if it was installed top-level. 
> That's good.
> 
> However, there are two more places in the same file which only try 
> "import PIL.Image" and do not try "import Image". These places should 
> use the same fallback or even simpler just use the Image that has 
> already been imported as a global variable.
> 
> -- Christoph
.......
thanks I'll take a look later.
-- 
Robin Becker

From henning.vonbargen at arcor.de  Sun Feb 15 09:46:28 2009
From: henning.vonbargen at arcor.de (Henning von Bargen)
Date: Sun, 15 Feb 2009 15:46:28 +0100
Subject: [reportlab-users] Am I allowed to copy and distribute fonts and
	images from the RL test suite with wordaxe?
Message-ID: <041946CBB1754F50975E8E0BF38F3B6D@max>

For the wordaxe hyphenation library 
(http://sourceforge.net/projects/deco-cow/),
I have slightly modified some test scripts from the ReportLab test suite in 
order to see
if these also run with hyphenation enabled or with wordaxe's Paragraph class 
from
wordaxe/rl/NewParagraph.py instead of the original 
reportlab/platypus/Paragraph.py.

Some of these tests reference fonts or images that are no longer part of the 
ReportLab
distribution. This causes the tests to fail.See
http://sourceforge.net/tracker/index.php?func=detail&aid=2599961&group_id=105867&atid=642478
http://sourceforge.net/tracker/index.php?func=detail&aid=2599892&group_id=105867&atid=642478

Am I allowed to copy the missing resource files for the LeERC___ font and 
redsquare.png
(and possibly others that were part of the ReportlLab test suite) and 
distribute them with
wordaxe?

Henning



From robin at reportlab.com  Sun Feb 15 14:25:28 2009
From: robin at reportlab.com (Robin Becker)
Date: Sun, 15 Feb 2009 19:25:28 +0000
Subject: [reportlab-users] Am I allowed to copy and distribute fonts and
 images from the RL test suite with wordaxe?
In-Reply-To: <041946CBB1754F50975E8E0BF38F3B6D@max>
References: <041946CBB1754F50975E8E0BF38F3B6D@max>
Message-ID: <49986C28.6080506@jessikat.plus.net>

Henning von Bargen wrote:
> For the wordaxe hyphenation library 
> (http://sourceforge.net/projects/deco-cow/),
> I have slightly modified some test scripts from the ReportLab test suite 
> in order to see
> if these also run with hyphenation enabled or with wordaxe's Paragraph 
> class from
> wordaxe/rl/NewParagraph.py instead of the original 
> reportlab/platypus/Paragraph.py.
> 
> Some of these tests reference fonts or images that are no longer part of 
> the ReportLab
> distribution. This causes the tests to fail.See
> http://sourceforge.net/tracker/index.php?func=detail&aid=2599961&group_id=105867&atid=642478 
> 
> http://sourceforge.net/tracker/index.php?func=detail&aid=2599892&group_id=105867&atid=642478 
> 
........
I'm not sure; these fonts are certainly still in older revsions of the 
repository. They were originally contributed to ReportLab by Just van 
Rossum so perhaps he's the best person to ask. On the other hand is it 
not possible to substitute the latest fonts which we substituted (ie 
DarkGarden)? I believe these were added specifically to assist onward 
distribution at the request of various package maintainers.
-- 
Robin Becker

From Robert at jli.com  Mon Feb 16 00:56:53 2009
From: Robert at jli.com (Robert D. Young)
Date: Sun, 15 Feb 2009 21:56:53 -0800
Subject: [reportlab-users] Platypus
Message-ID: <F47DB73DCE1B4218A9B978CEAB486104@AbilitySys.local>

Sorry for the re-post, if that's what it is. It's my first posting and I
expected to see an ack or copy of the posting and didn't. So I went and
changed my settings and now am reposting to make sure it is getting
through:

---------------------------- Original Message ----------------------------
Subject: Platypus
Date:    Fri, February 13, 2009 5:58 pm
To:      reportlab-users at reportlab.com
--------------------------------------------------------------------------

Newbie to reportlab's lib, so trying samples and fail on import of
SimpleDocTemplate, which doesn' appear to be in the 2.3 version? I'm using
Py 2.6.1 under WinXP and installed the reportlab file 2.3 package for
Win32 Python 2.6.

Can someone point me o a way to add the platypus parts tht appear to be
missong?

- Robert
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/reportlab-users/attachments/20090215/02c37f76/attachment.html>

From cito at online.de  Mon Feb 16 01:55:23 2009
From: cito at online.de (Christoph Zwerschke)
Date: Mon, 16 Feb 2009 07:55:23 +0100
Subject: [reportlab-users] Platypus
In-Reply-To: <F47DB73DCE1B4218A9B978CEAB486104@AbilitySys.local>
References: <F47DB73DCE1B4218A9B978CEAB486104@AbilitySys.local>
Message-ID: <49990DDB.8000700@online.de>

Robert D. Young schrieb:
> Newbie to reportlab's lib, so trying samples and fail on import of
> SimpleDocTemplate, which doesn' appear to be in the 2.3 version?

Did you do the following which is working for me?

from reportlab.platypus import SimpleDocTemplate

-- Christoph

From Robert at jli.com  Mon Feb 16 02:22:08 2009
From: Robert at jli.com (Robert D. Young)
Date: Sun, 15 Feb 2009 23:22:08 -0800
Subject: [reportlab-users] Platypus
In-Reply-To: <49990DDB.8000700@online.de>
References: <F47DB73DCE1B4218A9B978CEAB486104@AbilitySys.local>
	<49990DDB.8000700@online.de>
Message-ID: <9844C827DFB94D309C334A9586764851@AbilitySys.local>


Doggone it - I thought I had, but retried it just now and - well, i must 
have been a typo before because it works now! Sorry for the bother.

Great product, BTW.

- Robert

----- Original Message ----- 
From: "Christoph Zwerschke" <cito at online.de>
To: "Support list for users of Reportlab software" 
<reportlab-users at reportlab.com>
Sent: Sunday, February 15, 2009 10:55 PM
Subject: Re: [reportlab-users] Platypus


> Robert D. Young schrieb:
>> Newbie to reportlab's lib, so trying samples and fail on import of
>> SimpleDocTemplate, which doesn' appear to be in the 2.3 version?
>
> Did you do the following which is working for me?
>
> from reportlab.platypus import SimpleDocTemplate
>
> -- Christoph
> _______________________________________________
> reportlab-users mailing list
> reportlab-users at reportlab.com
> http://two.pairlist.net/mailman/listinfo/reportlab-users
>
>
> 



From gogtesuyash at gmail.com  Mon Feb 16 03:54:49 2009
From: gogtesuyash at gmail.com (Suyash Gogte)
Date: Mon, 16 Feb 2009 14:24:49 +0530
Subject: [reportlab-users] Error
Message-ID: <5797af200902160054q6321956fj53f06e22bb9b4ffa@mail.gmail.com>

hi frnds..
I m getting following error..its run time error at doc.build....
can anybody tell me way to fix it
thnks ...

Traceback (most recent call last):
  File "test1.py", line 58, in <module>
    go()
  File "test1.py", line 56, in go
    doc.build(Story, onFirstPage=myFirstPage, onLaterPages=myLaterPages)
  File
"/usr/local/lib/python2.5/site-packages/reportlab/platypus/doctemplate.py",
line 987, in build
    BaseDocTemplate.build(self,flowables, canvasmaker=canvasmaker)
  File
"/usr/local/lib/python2.5/site-packages/reportlab/platypus/doctemplate.py",
line 756, in build
    self.handle_flowable(flowables)
  File
"/usr/local/lib/python2.5/site-packages/reportlab/platypus/doctemplate.py",
line 630, in handle_flowable
    self.handle_keepWithNext(flowables)
  File
"/usr/local/lib/python2.5/site-packages/reportlab/platypus/doctemplate.py",
line 599, in handle_keepWithNext
    while i<n and flowables[i].getKeepWithNext(): i += 1
*AttributeError: 'str' object has no attribute 'getKeepWithNext'*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/reportlab-users/attachments/20090216/420c8dd3/attachment.htm>

From cito at online.de  Mon Feb 16 04:25:21 2009
From: cito at online.de (Christoph Zwerschke)
Date: Mon, 16 Feb 2009 10:25:21 +0100
Subject: [reportlab-users] Error
In-Reply-To: <5797af200902160054q6321956fj53f06e22bb9b4ffa@mail.gmail.com>
References: <5797af200902160054q6321956fj53f06e22bb9b4ffa@mail.gmail.com>
Message-ID: <49993101.4000709@online.de>

Suyash Gogte schrieb:
> I m getting following error..its run time error at doc.build....
> can anybody tell me way to fix it

Seems you've put a string instead of a Flowable in your story. Wrap the
string in a Paragraph or something.

-- Christoph

From gogtesuyash at gmail.com  Mon Feb 16 04:34:01 2009
From: gogtesuyash at gmail.com (Suyash Gogte)
Date: Mon, 16 Feb 2009 15:04:01 +0530
Subject: [reportlab-users] Error
In-Reply-To: <49993101.4000709@online.de>
References: <5797af200902160054q6321956fj53f06e22bb9b4ffa@mail.gmail.com>
	<49993101.4000709@online.de>
Message-ID: <5797af200902160134t19104e21yfece80a80f1abd98@mail.gmail.com>

hey Christoph thnks a lot...
Now its working by mistake i have used string in flowables...
thnks man......


On Mon, Feb 16, 2009 at 2:55 PM, Christoph Zwerschke <cito at online.de> wrote:

> Suyash Gogte schrieb:
> > I m getting following error..its run time error at doc.build....
> > can anybody tell me way to fix it
>
> Seems you've put a string instead of a Flowable in your story. Wrap the
> string in a Paragraph or something.
>
> -- Christoph
> _______________________________________________
> 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/20090216/feb98668/attachment-0001.html>

From peter at maubp.freeserve.co.uk  Mon Feb 16 07:51:41 2009
From: peter at maubp.freeserve.co.uk (Peter)
Date: Mon, 16 Feb 2009 12:51:41 +0000
Subject: [reportlab-users] External hyperlinks in SVG output
In-Reply-To: <320fb6e00902121509n3233235dlf44cc1ba6604423b@mail.gmail.com>
References: <320fb6e00901290737k1269fb4en92975edbcf87f88a@mail.gmail.com>
	<956003ae0901290838g6393c59dxb36e52626ff6429c@mail.gmail.com>
	<320fb6e00902120555i37c3e7c8ibb0332901c81c0b8@mail.gmail.com>
	<320fb6e00902121509n3233235dlf44cc1ba6604423b@mail.gmail.com>
Message-ID: <320fb6e00902160451m190404c4j9afb3e76ee784f5f@mail.gmail.com>

> Apparently "version" and "baseProfile" shouldn't be there for the
> currently used doctype. Interestingly renderSVG doesn't specify the
> doctype, letting python's xml.dom decide, so it could vary with
> version of python (and I am using python 2.4 on this machine).  We
> should be able to specify them when creating the DOM...

My mistake - the current renderSVG.py actually has the doctype hard
coded in the save method - it does not handle this via the dom python
library.

Anyway, attached is my proof of concept updates to renderSVG.py
(replacement file and as a patch against
https://svn.reportlab.com/svn/public/reportlab/trunk/src/reportlab/graphics/renderSVG.py
which I assume it the latest) which allows the use of external
hyper-references.  All you get to specify are the URL and its title
string (tooltip text on most browsers) via two optional properties of
the drawing objects.

As part of this work I have changed how the doctype is declared, and
made sure the new output does pass the W3 validation (the current
renderSVG output does not).

An example script using this follows below.

Peter

--------------

from reportlab.graphics import renderPS, renderPDF, renderPM #, renderSVG
from reportlab.lib.units import cm
from reportlab.graphics import shapes
from reportlab.lib import colors
import renderSVG_links as renderSVG #my local copy of
reportlab.graphics.renderSVG

width = 10*cm
height = 2*cm

#Create very simple drawing object,
drawing = shapes.Drawing(width, height)
drawing.add(shapes.String(1*cm, 1*cm, "Hello World!",
                          hrefURL = "http://en.wikipedia.org/wiki/Hello_world",
                          hrefTitle = "Why 'Hello World'?",
                          fillColor = colors.darkgreen))
drawing.add(shapes.Rect(4.5*cm, 0.5*cm, 5*cm, 1*cm,
                         hrefURL = "http://en.wikipedia.org/wiki/Rectangle",
                         hrefTitle = "Wikipedia page on rectangles",
                         strokeColor = colors.blue,
                         fillColor = colors.red))
drawing.add(shapes.Ellipse(7*cm, 1*cm, 2*cm, 0.95*cm,
                           hrefURL = "http://en.wikipedia.org/wiki/Ellipse",
                           strokeColor = colors.black,
                           fillColor = colors.yellow))
drawing.add(shapes.Circle(7*cm, 1*cm, 0.9*cm,
                           hrefURL = "http://en.wikipedia.org/wiki/Circle",
                          strokeColor = colors.black,
                          fillColor = colors.brown))
drawing.add(shapes.Ellipse(7*cm, 1*cm, 0.5*cm, 0.9*cm,
                           hrefTitle = "Tooltip with no link? Does nothing!",
                           strokeColor = colors.black,
                           fillColor = colors.black))
drawing.add(shapes.Polygon([4.5*cm, 1.25*cm, 5*cm, 0.1*cm, 4*cm, 0.1*cm],
                           hrefURL = "http://en.wikipedia.org/wiki/Polygon",
                           hrefTitle = "This triangle is a simple polygon.",
                           strokeColor = colors.darkgreen,
                           fillColor = colors.green))

if True :
    print "Rotating the drawing... to check this all works still ;)"
    group = drawing.asGroup()
    group.rotate(90) #rotates about top left corner
    group.translate(0, -2*cm)
    drawing = shapes.Drawing(height, width)
    drawing.add(group)

renderPDF.drawToFile(drawing, "hello.pdf")
renderPS.drawToFile(drawing, "hello.ps")
renderPM.drawToFile(drawing, "hello.png", "PNG")
renderSVG.drawToFile(drawing, "hello.svg")

print "-" * 50
print renderSVG.drawToString(drawing)
print "-" * 50
print "Done"

From robin at reportlab.com  Mon Feb 16 08:09:18 2009
From: robin at reportlab.com (Robin Becker)
Date: Mon, 16 Feb 2009 13:09:18 +0000
Subject: [reportlab-users] External hyperlinks in SVG output
In-Reply-To: <320fb6e00902160451m190404c4j9afb3e76ee784f5f@mail.gmail.com>
References: <320fb6e00901290737k1269fb4en92975edbcf87f88a@mail.gmail.com>	<956003ae0901290838g6393c59dxb36e52626ff6429c@mail.gmail.com>	<320fb6e00902120555i37c3e7c8ibb0332901c81c0b8@mail.gmail.com>	<320fb6e00902121509n3233235dlf44cc1ba6604423b@mail.gmail.com>
	<320fb6e00902160451m190404c4j9afb3e76ee784f5f@mail.gmail.com>
Message-ID: <4999657E.1050506@chamonix.reportlab.co.uk>

Peter wrote:
.............
> Anyway, attached is my proof of concept updates to renderSVG.py
> (replacement file and as a patch against
> https://svn.reportlab.com/svn/public/reportlab/trunk/src/reportlab/graphics/renderSVG.py
> which I assume it the latest) which allows the use of external
> hyper-references.  All you get to specify are the URL and its title
> string (tooltip text on most browsers) via two optional properties of
> the drawing objects.
> 

did you forget to attach your replacement? I did not see it?

> As part of this work I have changed how the doctype is declared, and
> made sure the new output does pass the W3 validation (the current
> renderSVG output does not).
> 
> An example script using this follows below.
> 
.......
-- 
Robin Becker

From peter at maubp.freeserve.co.uk  Mon Feb 16 08:30:12 2009
From: peter at maubp.freeserve.co.uk (Peter)
Date: Mon, 16 Feb 2009 13:30:12 +0000
Subject: [reportlab-users] External hyperlinks in SVG output
In-Reply-To: <4999657E.1050506@chamonix.reportlab.co.uk>
References: <320fb6e00901290737k1269fb4en92975edbcf87f88a@mail.gmail.com>
	<956003ae0901290838g6393c59dxb36e52626ff6429c@mail.gmail.com>
	<320fb6e00902120555i37c3e7c8ibb0332901c81c0b8@mail.gmail.com>
	<320fb6e00902121509n3233235dlf44cc1ba6604423b@mail.gmail.com>
	<320fb6e00902160451m190404c4j9afb3e76ee784f5f@mail.gmail.com>
	<4999657E.1050506@chamonix.reportlab.co.uk>
Message-ID: <320fb6e00902160530y414c939dj2053e54018303cfe@mail.gmail.com>

>
> did you forget to attach your replacement? I did not see it?
>

Yes, sorry :(

Second attempt...

Peter
-------------- next part --------------
A non-text attachment was scrubbed...
Name: svg_links.patch
Type: application/octet-stream
Size: 8373 bytes
Desc: not available
Url : <http://two.pairlist.net/pipermail/reportlab-users/attachments/20090216/822934fa/attachment-0001.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: renderSVG_links.py
Type: text/x-script.phyton
Size: 30955 bytes
Desc: not available
Url : <http://two.pairlist.net/pipermail/reportlab-users/attachments/20090216/822934fa/attachment-0001.bin>

From peter at maubp.freeserve.co.uk  Mon Feb 16 11:11:13 2009
From: peter at maubp.freeserve.co.uk (Peter)
Date: Mon, 16 Feb 2009 16:11:13 +0000
Subject: [reportlab-users] SVG font-size values lack units (invalid CSS)
Message-ID: <320fb6e00902160811jb151383g617c040a9035d712@mail.gmail.com>

I noticed that font sizes were not working consistently across
viewers, and eventually identified the problem.  Firefox 3.0 and Opera
9.20 will ignore a font-size in a style attribute if it has no units
(this is considered invalid CSS), showing text in its default size.
On the other hand, Safari 3.2.1, Adobe Illustrator CS3 and Mac OS X
10.5 preview will tolerate the missing units, and show the images as I
expected (arguably buggy behaviour).  See
https://bugzilla.mozilla.org/show_bug.cgi?id=311569 for a discussion
of this.

Currently renderSVG outputs this kind of invalid CSS in its SVG files,

                <text style="font-size: 10; font-family: Times-Roman;
fill: rgb(0%,39%,0%);" transform="translate(0,56) scale(1,-1)"
x="28.3464566929" y="28.3464566929">
                    Hello World!
                </text>

This should be:

                <text style="font-size: 10pt; font-family:
Times-Roman; fill: rgb(0%,39%,0%);" transform="translate(0,56)
scale(1,-1)" x="28.3464566929" y="28.3464566929">
                    Hello World!
                </text>

Or, instead of using the style attribute we could use the font-size
attribute where the units are optional, e.g.

                <text font-size="10" style="font-family: Times-Roman;
fill: rgb(0%,39%,0%);" transform="translate(0,56) scale(1,-1)"
x="28.3464566929" y="28.3464566929">
                    Hello World!
                </text>

My suggested fix is to update renderSVG.py to replace this:

    def setFont(self, font, fontSize):
        if self._font != font or self._fontSize != fontSize:
            self._font, self._fontSize = (font, fontSize)
            self.style['font-family'] = font
            self.style['font-size'] = fontSize

with this:

    def setFont(self, font, fontSize):
        if self._font != font or self._fontSize != fontSize:
            self._font, self._fontSize = (font, fontSize)
            self.style['font-family'] = font
            #The style attribute contains CSS, so the units are required:
            self.style['font-size'] = "%ipt" % fontSize

The images don't look identical to those from the PNG output from
renderPM (the fonts sizes differ a little), but this does look pretty
close and it now works on all the browsers I have tested.

Peter

From peter at maubp.freeserve.co.uk  Mon Feb 16 12:23:20 2009
From: peter at maubp.freeserve.co.uk (Peter)
Date: Mon, 16 Feb 2009 17:23:20 +0000
Subject: [reportlab-users] External hyperlinks in SVG output
In-Reply-To: <320fb6e00902160530y414c939dj2053e54018303cfe@mail.gmail.com>
References: <320fb6e00901290737k1269fb4en92975edbcf87f88a@mail.gmail.com>
	<956003ae0901290838g6393c59dxb36e52626ff6429c@mail.gmail.com>
	<320fb6e00902120555i37c3e7c8ibb0332901c81c0b8@mail.gmail.com>
	<320fb6e00902121509n3233235dlf44cc1ba6604423b@mail.gmail.com>
	<320fb6e00902160451m190404c4j9afb3e76ee784f5f@mail.gmail.com>
	<4999657E.1050506@chamonix.reportlab.co.uk>
	<320fb6e00902160530y414c939dj2053e54018303cfe@mail.gmail.com>
Message-ID: <320fb6e00902160923o7d609de9s903d9ea99db736a3@mail.gmail.com>

I see this is in your SVN repository now :)
https://svn.reportlab.com/svn/public/reportlab/trunk/src/reportlab/graphics/renderSVG.py

[As an aside, I don't suppose you could turn tweak the website to let
it show the history too?  I find the revision log very helpful in
other projects, especially if it includes tags/labels for official
releases.]

I'm pleased to report that in addition to the previous tests using
Opera, Firefox and Safari on the Mac, I have now also tested this on
Windows XP using IE 6.0 with either Adobe SVG Viewer (ASV) 6.0 beta,
or Examotion's current IE SVG plugin (version 1.1.1).  Clickable links
are also working in IE6 with either of these plugins (except for the
annoying extra "Click to activate and use this control" step from
Microsoft avoiding the Eolas patent) with the new renderSVG.py file :)

http://www.adobe.com/svg/viewer/install/beta.html
http://www.examotion.com

Peter

From haraldarminmassa at gmail.com  Mon Feb 16 12:25:42 2009
From: haraldarminmassa at gmail.com (Harald Armin Massa)
Date: Mon, 16 Feb 2009 18:25:42 +0100
Subject: [reportlab-users] Wordaxe 0.3.0, ReportLab 2.2: Assertion Failure,
	cannot understand 	why this has to be asserted...
Message-ID: <7be3f35d0902160925m56e3c9ebo8be49a838e493f3d@mail.gmail.com>

hello,

while running a document in preparation for updating wordaxe /
reportlab, I get the Assertion failure:

Assertion failure
self.width=429.536000
nFrags=31
printrange=0:31
printwidth=431.760000
0 SW() 0
1 SP( ) 2.224
2 SW(wegen) 23.568
3 SP( ) 2.224
4 SW(Sch?den,) 33.8
5 SP( ) 2.224
6 SW(die) 10.672
7 SP( ) 2.224
8 SW(der) 11.56
9 SP( ) 2.224
10 SW(Versicherungsnehmer,) 80.472
11 SP( ) 2.224
12 SW(ein) 10.672
13 SP( ) 2.224
14 SW(Mitversicherter) 52.448
15 SP( ) 2.224
16 SW(oder) 16.008
17 SP( ) 2.224
18 SW(eine) 15.12
19 SP( ) 2.224
20 SW(von) 12.896
21 SP( ) 2.224
22 SW(ihnen) 19.568
23 SP( ) 2.224
24 SW(bestellte) 29.792
25 SP( ) 2.224
26 SW(oder) 16.008
27 SP( ) 2.224
28 SW(beauftragte) 40.472
29 SP( ) 2.224
30 SW(Person) 25.344

asserted was:

if not abs(self.width - sum(getattr(f,"width",0) for f in
fragments[print_indx_start:print_indx_end])) <= 1e-5:

and I a have no Idea WHY this get's asserted. My "first look"
understanding is that this is checking that floating point errors stay
below some point, but... what is the bigger idea behind this?

best wishes

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  Mon Feb 16 12:50:34 2009
From: robin at reportlab.com (Robin Becker)
Date: Mon, 16 Feb 2009 17:50:34 +0000
Subject: [reportlab-users] Wordaxe 0.3.0,
 ReportLab 2.2: Assertion Failure,
 cannot understand 	why this has to be asserted...
In-Reply-To: <7be3f35d0902160925m56e3c9ebo8be49a838e493f3d@mail.gmail.com>
References: <7be3f35d0902160925m56e3c9ebo8be49a838e493f3d@mail.gmail.com>
Message-ID: <4999A76A.4030701@chamonix.reportlab.co.uk>

Harald Armin Massa wrote:
> hello,
> 
> while running a document in preparation for updating wordaxe /
> reportlab, I get the Assertion failure:
> 
> Assertion failure
> self.width=429.536000
> nFrags=31
> printrange=0:31
> printwidth=431.760000
> 0 SW() 0
> 1 SP( ) 2.224
> 2 SW(wegen) 23.568
> 3 SP( ) 2.224
> 4 SW(Sch?den,) 33.8
> 5 SP( ) 2.224
> 6 SW(die) 10.672
> 7 SP( ) 2.224
> 8 SW(der) 11.56
> 9 SP( ) 2.224
> 10 SW(Versicherungsnehmer,) 80.472
> 11 SP( ) 2.224
> 12 SW(ein) 10.672
> 13 SP( ) 2.224
> 14 SW(Mitversicherter) 52.448
> 15 SP( ) 2.224
> 16 SW(oder) 16.008
> 17 SP( ) 2.224
> 18 SW(eine) 15.12
> 19 SP( ) 2.224
> 20 SW(von) 12.896
> 21 SP( ) 2.224
> 22 SW(ihnen) 19.568
> 23 SP( ) 2.224
> 24 SW(bestellte) 29.792
> 25 SP( ) 2.224
> 26 SW(oder) 16.008
> 27 SP( ) 2.224
> 28 SW(beauftragte) 40.472
> 29 SP( ) 2.224
> 30 SW(Person) 25.344
> 
> asserted was:
> 
> if not abs(self.width - sum(getattr(f,"width",0) for f in
> fragments[print_indx_start:print_indx_end])) <= 1e-5:
> 
> and I a have no Idea WHY this get's asserted. My "first look"
> understanding is that this is checking that floating point errors stay
> below some point, but... what is the bigger idea behind this?
> 
> best wishes
> 
> Harald
> 

I believe there was some awful hackery around reconstructing texts when 
paragraphs get split. Even when the new paragraph was supposed to fit it 
sometimes didn't because of slight differences in accumulation errors etc etc.

The above looks like a sanity check for seeing if the current set of fragments 
actually fits on the line.

However, since that code isn't currently part of our paragraph I can't be 
totally sure.
-- 
Robin Becker

From robin at reportlab.com  Mon Feb 16 12:59:08 2009
From: robin at reportlab.com (Robin Becker)
Date: Mon, 16 Feb 2009 17:59:08 +0000
Subject: [reportlab-users] SVG font-size values lack units (invalid CSS)
In-Reply-To: <320fb6e00902160811jb151383g617c040a9035d712@mail.gmail.com>
References: <320fb6e00902160811jb151383g617c040a9035d712@mail.gmail.com>
Message-ID: <4999A96C.6070502@chamonix.reportlab.co.uk>

Peter wrote:
........
> 
> The images don't look identical to those from the PNG output from
> renderPM (the fonts sizes differ a little), but this does look pretty
> close and it now works on all the browsers I have tested.
> 
well I just looked at the definitions currently being talked about for svg 1.1 
and font-size appears here

http://www.w3.org/TR/SVG11/propidx.html

there the allowed value types for font-size are

  <absolute-size>  |  <relative-size>  | <length>  |  <percentage>  |  inherit


and following the link for length gives this definition


> <length>: A length is a distance measurement. The format of a <length> is a <number>  optionally followed immediately by a unit identifier. (Note that the specification of a <number> is different for property  values than for XML attribute values.)
> If the <length> is expressed as a value without a unit identifier (e.g., 48), then the <length> represents a distance in the current user coordinate system.
> If one of the unit identifiers is provided (e.g., 12mm), then the <length> is processed according to the description in Units.
> Percentage values (e.g., 10%) depend on the particular property or attribute to which the percentage value has been assigned. Two common cases are: (a) when a percentage value represents a percent of the viewport (refer to the section that discusses Units in general), and (b) when a percentage value represents a percent of the bounding box on a given object (refer to the section that describes Object bounding box units).
> Within the SVG DOM, a <length> is represented as an SVGLength or an SVGAnimatedLength.



I take the definition to allow font-size to be allowed not to be followed by a unit.

That said it's not at all obvious how these lengths should be properly interpreted.

For PDF the default coordinate system is certainly defined in point units ie 
1=1pt. For our bitmaps we decided to make 1 pixel == 1point, but I believe you 
can specify a dpi value. For SVG I'm not sure. According to

http://www.w3.org/TR/SVG11/coords.html#InitialCoordinateSystem

the SVG User agent (browser?) is supposed to set up the default view port and 
the default coordinate system consistently so that 1unit=1pixel. That should be 
consistent with our current usage. However, looking at the standard examples 
it's obvious that SVG is not obeying these rules. I believe the default unit 
should be pixels.

Has anyone else got any better ideas?
-- 
Robin Becker

From peter at maubp.freeserve.co.uk  Mon Feb 16 13:10:47 2009
From: peter at maubp.freeserve.co.uk (Peter)
Date: Mon, 16 Feb 2009 18:10:47 +0000
Subject: [reportlab-users] SVG font-size values lack units (invalid CSS)
In-Reply-To: <4999A96C.6070502@chamonix.reportlab.co.uk>
References: <320fb6e00902160811jb151383g617c040a9035d712@mail.gmail.com>
	<4999A96C.6070502@chamonix.reportlab.co.uk>
Message-ID: <320fb6e00902161010g8e0096csd60cc237bcea2375@mail.gmail.com>

On Mon, Feb 16, 2009 at 5:59 PM, Robin Becker <robin at reportlab.com> wrote:
>
> Peter wrote:
> ........
>>
>> The images don't look identical to those from the PNG output from
>> renderPM (the fonts sizes differ a little), but this does look pretty
>> close and it now works on all the browsers I have tested.
>>
> well I just looked at the definitions currently being talked about for svg
> 1.1 and font-size appears here
>
> http://www.w3.org/TR/SVG11/propidx.html
>
> there the allowed value types for font-size are
>
>  <absolute-size>  |  <relative-size>  | <length>  |  <percentage>  |
>  inherit
>
>
> and following the link for length gives this definition ...

I think you are reading the spec for the font-size attribute where I
agree the units are optional, and from me testing all the browsers
cope. e.g. <text font-size="10">Hello World!</text>
(I'm not sure what the units used are - this is another reason for us
to be explicit if we can be.)

The current renderSVG (and my suggested small change) doesn't use the
font-size attribute but instead supplies an bit of CSS to the style
attribute where units *are* required.  e.g <text style="font-size:
10pt;">Hello World!</text>

Peter

From andy at reportlab.com  Tue Feb 17 02:22:22 2009
From: andy at reportlab.com (Andy Robinson)
Date: Tue, 17 Feb 2009 07:22:22 +0000
Subject: [reportlab-users] Am I allowed to copy and distribute fonts and
	images from the RL test suite with wordaxe?
In-Reply-To: <041946CBB1754F50975E8E0BF38F3B6D@max>
References: <041946CBB1754F50975E8E0BF38F3B6D@max>
Message-ID: <956003ae0902162322u1a505b07kefe01ae0ab816e8b@mail.gmail.com>

2009/2/15 Henning von Bargen <henning.vonbargen at arcor.de>
> Am I allowed to copy the missing resource files for the LeERC___ font and
> redsquare.png
> (and possibly others that were part of the ReportlLab test suite) and
> distribute them with
> wordaxe?


We removed the font because it is not GPL; JvR did NOT give people
permission to further modify his font, which is a freedom the Ubuntu
folks wanted to include us in their packages. We wanted a font which
was very obviously "not Helvetica", so we switched to DarkGarden which
fills the same needs for us.


Just van Rossum explicitly gave permission for these to be included
for testing purposes.  If you already wrote a lot of tests based on
the metrics of LettError, am sure he won't mind if you keep it there,
but you should note in your credits that it is his copyright.

I don't know what 'redsquare.png' was (I'm mostly offline this week)
but am willing to bet it was a red square PNG we created, so I hereby
permit you to redistribute this :-)

- Andy

From gherman at darwin.in-berlin.de  Tue Feb 17 04:18:32 2009
From: gherman at darwin.in-berlin.de (Dinu Gherman)
Date: Tue, 17 Feb 2009 10:18:32 +0100
Subject: [reportlab-users] Am I allowed to copy and distribute fonts and
	images from the RL test suite with wordaxe?
In-Reply-To: <956003ae0902162322u1a505b07kefe01ae0ab816e8b@mail.gmail.com>
References: <041946CBB1754F50975E8E0BF38F3B6D@max>
	<956003ae0902162322u1a505b07kefe01ae0ab816e8b@mail.gmail.com>
Message-ID: <07AAEB48-6EA4-4049-8BE3-33332A56C723@darwin.in-berlin.de>

Andy Robinson:

> We removed the font because it is not GPL; JvR did NOT give people
> permission to further modify his font, which is a freedom the Ubuntu
> folks wanted to include us in their packages. We wanted a font which
> was very obviously "not Helvetica", so we switched to DarkGarden which
> fills the same needs for us.

So RL goes GPL, now? I'm holding my breath...

Dinu


From H.vonBargen at t-p.com  Tue Feb 17 04:46:20 2009
From: H.vonBargen at t-p.com (Henning von Bargen)
Date: Tue, 17 Feb 2009 10:46:20 +0100
Subject: [reportlab-users] Wordaxe 0.3.0,
	ReportLab 2.2: Assertion Failure, cannot understand why...
In-Reply-To: <mailman.37.1234862315.77885.reportlab-users@reportlab.com>
Message-ID: <77A9E75DA75C564D82BDC4B18A7D8525305740@ROHRPOSTIX.tup.ads>

Harald Armin Massa wrote:

> while running a document in preparation for updating wordaxe /
> reportlab, I get the Assertion failure:
> 
> Assertion failure
> self.width=429.536000
> nFrags=31
> printrange=0:31
> printwidth=431.760000
> 0 SW() 0
> 1 SP( ) 2.224
> 2 SW(wegen) 23.568
> 3 SP( ) 2.224
> 4 SW(Sch?den,) 33.8
...
> 28 SW(beauftragte) 40.472
> 29 SP( ) 2.224
> 30 SW(Person) 25.344
> 
> asserted was:
> 
> if not abs(self.width - sum(getattr(f,"width",0) for f in
> fragments[print_indx_start:print_indx_end])) <= 1e-5:
> 
> and I a have no Idea WHY this get's asserted. My "first look"
> understanding is that this is checking that floating point errors stay
> below some point, but... what is the bigger idea behind this?

Harald,

the code you are mentioning is part of wordaxe, not ReportLab.
Please file a bug for wordaxe instead of posting to this list next time.

But to answer your question:

There's no bigger idea. The intention is just as you guessed:
When I coded NewParagraph, I assumed that the sum of the fragments' widths
should match self.width - since floats are used, the comparison is a bit fuzzy.
This is to assure that the paragraph doesn't "write across its border" when
finally it will be rendered.

Seems like either there's a bug in the wordaxe code or the assertion is wrong.
You may try to comment the assertion and carefully look at the generated PDF
for this paragraph.

Henning



From gherman at darwin.in-berlin.de  Sat Feb 21 12:44:18 2009
From: gherman at darwin.in-berlin.de (Dinu Gherman)
Date: Sat, 21 Feb 2009 18:44:18 +0100
Subject: [reportlab-users] Sample code for repeating table headers/footers?
Message-ID: <33243C82-19C4-42E2-831E-C5D7511549EC@darwin.in-berlin.de>

Hi,

I cannot find any sample code for repeating header/footer rows
in tables? I would be very glad if somebody could point me to
such existing code or even post something like that here, since
neither the RL User Guide nor the test suite seem to contain
such a code samples.

Thanks,

Dinu


From aklaver at comcast.net  Sat Feb 21 13:00:07 2009
From: aklaver at comcast.net (Adrian Klaver)
Date: Sat, 21 Feb 2009 10:00:07 -0800
Subject: [reportlab-users] Sample code for repeating table
	headers/footers?
In-Reply-To: <33243C82-19C4-42E2-831E-C5D7511549EC@darwin.in-berlin.de>
References: <33243C82-19C4-42E2-831E-C5D7511549EC@darwin.in-berlin.de>
Message-ID: <200902211000.08414.aklaver@comcast.net>

On Saturday 21 February 2009 9:44:18 am Dinu Gherman wrote:
> Hi,
>
> I cannot find any sample code for repeating header/footer rows
> in tables? I would be very glad if somebody could point me to
> such existing code or even post something like that here, since
> neither the RL User Guide nor the test suite seem to contain
> such a code samples.
>
> Thanks,
>
> Dinu
>
> _______________________________________________
> reportlab-users mailing list
> reportlab-users at reportlab.com
> http://two.pairlist.net/mailman/listinfo/reportlab-users

For a header row, straight from the User Guide. Repeats the first row passed to 
the Table(), in this case the data in dList. 
tbl = Table(dList,style=TBL_STYLE,colWidths=COL_WIDTHS,repeatRows=1)

I have yet to find a footer equivalent for a table.

-- 
Adrian Klaver
aklaver at comcast.net

From paul.barrass at safeonlinebilling.com  Sun Feb 22 09:04:54 2009
From: paul.barrass at safeonlinebilling.com (Paul Barrass)
Date: Sun, 22 Feb 2009 14:04:54 +0000
Subject: [reportlab-users] Sample code for repeating table
	headers/footers?
In-Reply-To: <200902211000.08414.aklaver@comcast.net>
References: <33243C82-19C4-42E2-831E-C5D7511549EC@darwin.in-berlin.de>
	<200902211000.08414.aklaver@comcast.net>
Message-ID: <49A15B86.9000200@safeonlinebilling.com>

Adrian Klaver wrote:
> On Saturday 21 February 2009 9:44:18 am Dinu Gherman wrote:
>   
>> Hi,
>>
>> I cannot find any sample code for repeating header/footer rows
>> in tables?
> For a header row, straight from the User Guide. Repeats the first row passed to 
> the Table(), in this case the data in dList. 
> tbl = Table(dList,style=TBL_STYLE,colWidths=COL_WIDTHS,repeatRows=1)
>
> I have yet to find a footer equivalent for a table
As far as I'm aware, there is no per-page Table footer - however you 
should be able to adjust the workaround I posted for long table 
performance a while back ( 
http://two.pairlist.net/pipermail/reportlab-users/2008-October/007594.html 
) to provide a footer as well as a header - though I've never checked to 
see if this actually works.  There would be no need to split the data 
into multiple Tables unless you needed the performance boost, and that 
should remove some of the constraints.

That suggestion is really an evil gratuitous hack, and you'd probably be 
a lot better off altering Table to repeat footer rows, but that would 
probably involve a lot of work updating the styling etc.

Paul Barrass.

From chris.hoy at hoyc.fsnet.co.uk  Tue Feb 24 03:06:13 2009
From: chris.hoy at hoyc.fsnet.co.uk (Chris Hoy)
Date: Tue, 24 Feb 2009 08:06:13 -0000
Subject: [reportlab-users] Problem install 2.3 via easy_install on Unbunto
Message-ID: <39A6FE6D0D79457FBDDFCF4190A11759@Yellowstone>

Hi 

 

I?m getting the following error when trying to install ReportLab 2.3 via
easy_install, I suspect that I haven?t install all the necessary packages
for the complier any help would be appreciated 

 

root at prmax01:~# easy_install -U ReportLab

Searching for ReportLab

Reading http://pypi.python.org/simple/ReportLab/

Reading http://www.reportlab.com/

Best match: reportLab 2.3

Downloading
http://pypi.python.org/packages/source/r/reportlab/reportLab-2.3.zip#md5=7d9
8b26fa287a9e4be4d35d682ce64ac

Processing reportLab-2.3.zip

Running ReportLab_2_3/setup.py -q bdist_egg --dist-dir
/tmp/easy_install-QsosE7/ReportLab_2_3/egg-dist-tmp-xJI7Vq

################################################

#Attempting install of _rl_accel, sgmlop & pyHnj

#extensions from
'/tmp/easy_install-QsosE7/ReportLab_2_3/src/rl_addons/rl_accel'

################################################

################################################

#Attempting install of _renderPM

#extensions from
'/tmp/easy_install-QsosE7/ReportLab_2_3/src/rl_addons/renderPM'

# installing without freetype no ttf, sorry!

################################################

/tmp/easy_install-QsosE7/ReportLab_2_3/src/rl_addons/rl_accel/_rl_accel.c:
In function 'hex32':

/tmp/easy_install-QsosE7/ReportLab_2_3/src/rl_addons/rl_accel/_rl_accel.c:79
3: warning: format '%8.8X' expects type 'unsigned int', but argument 3 has
type 'long unsigned int'

/tmp/easy_install-QsosE7/ReportLab_2_3/src/rl_addons/rl_accel/_rl_accel.c:
In function '_instanceStringWidthU':

/tmp/easy_install-QsosE7/ReportLab_2_3/src/rl_addons/rl_accel/_rl_accel.c:12
00: warning: pointer targets in assignment differ in signedness

/tmp/easy_install-QsosE7/ReportLab_2_3/src/rl_addons/rl_accel/_rl_accel.c:11
23: warning: 'f' may be used uninitialized in this function

/tmp/easy_install-QsosE7/ReportLab_2_3/src/rl_addons/rl_accel/_rl_accel.c:11
23: warning: 't' may be used uninitialized in this function

/tmp/easy_install-QsosE7/ReportLab_2_3/src/rl_addons/rl_accel/_rl_accel.c:11
23: warning: 'L' may be used uninitialized in this function

/usr/bin/ld: cannot find -l_renderPM_libart

collect2: ld returned 1 exit status

error: Setup script exited with error: command 'gcc' failed with exit status
1

 

 

Chris

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/reportlab-users/attachments/20090224/0aabdf12/attachment.html>

From andy at reportlab.com  Tue Feb 24 03:56:22 2009
From: andy at reportlab.com (Andy Robinson)
Date: Tue, 24 Feb 2009 08:56:22 +0000
Subject: [reportlab-users] Problem install 2.3 via easy_install on
	Unbunto
In-Reply-To: <39A6FE6D0D79457FBDDFCF4190A11759@Yellowstone>
References: <39A6FE6D0D79457FBDDFCF4190A11759@Yellowstone>
Message-ID: <956003ae0902240056q78a3d6c7xd1f12bda69065c08@mail.gmail.com>

2009/2/24 Chris Hoy <chris.hoy at hoyc.fsnet.co.uk>:
> Hi

Hi.  Awesome performance in Beijing, by the way!


> I?m getting the following error when trying to install ReportLab 2.3 via
> easy_install, I suspect that I haven?t install all the necessary packages
> for the complier any help would be appreciated

We don't currently support easy_install.  (Unfortunately, easy_install attempts
to support us, but that's the author's problem!).  Just download the package and
run 'python setup.py install', and all should work.

Why is a long story but we laid out our directories, version numbers
and build processes long before easy_install was invented and each time
we rearrange these things it causes us some pain and takes up a lot of time
(for no functionality in return).  We might get there on our next
major version.

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 peter at maubp.freeserve.co.uk  Tue Feb 24 11:41:30 2009
From: peter at maubp.freeserve.co.uk (Peter)
Date: Tue, 24 Feb 2009 16:41:30 +0000
Subject: [reportlab-users] External hyperlinks in SVG output
In-Reply-To: <320fb6e00902160923o7d609de9s903d9ea99db736a3@mail.gmail.com>
References: <320fb6e00901290737k1269fb4en92975edbcf87f88a@mail.gmail.com>
	<956003ae0901290838g6393c59dxb36e52626ff6429c@mail.gmail.com>
	<320fb6e00902120555i37c3e7c8ibb0332901c81c0b8@mail.gmail.com>
	<320fb6e00902121509n3233235dlf44cc1ba6604423b@mail.gmail.com>
	<320fb6e00902160451m190404c4j9afb3e76ee784f5f@mail.gmail.com>
	<4999657E.1050506@chamonix.reportlab.co.uk>
	<320fb6e00902160530y414c939dj2053e54018303cfe@mail.gmail.com>
	<320fb6e00902160923o7d609de9s903d9ea99db736a3@mail.gmail.com>
Message-ID: <320fb6e00902240841h71869ed8p2bf397e82622dc83@mail.gmail.com>

On Mon, Feb 16, 2009 at 5:23 PM, Peter <peter at maubp.freeserve.co.uk> wrote:
> I see this is in your SVN repository now :)

I've attached a patch against the current SVN version of renderSVG.py
to support making links on closed SVG paths.  Without this only the
simplest of vector shapes had working links (rectangles, polygons,
circles, etc), but by supporting closed paths this lets the user
create any complicated shape - and have it linkable :)

At the end of the email is an example of how this might get used.

Peter

--------------
from reportlab.graphics import renderPS, renderPDF, renderSVG, renderPM
from reportlab.lib.units import cm
from reportlab.graphics import shapes
from reportlab.lib import colors

width=10*cm
height=2*cm

#Create fairly simple drawing object,
drawing=shapes.Drawing(width, height)

p=shapes.ArcPath(strokeColor=colors.darkgreen,
                   fillColor=colors.green,
                   hrefURL="http://en.wikipedia.org/wiki/Vector_path",
                   hrefTitle="This big letter C is actually a closed
vector path.",
                   strokewidth=0)
p.addArc(1*cm, 1*cm, 0.8*cm, 20, 340, moveTo=True)
p.addArc(1*cm, 1*cm, 0.9*cm, 20, 340, reverse=True)
p.closePath()
drawing.add(p)

drawing.add(shapes.Rect(2.25*cm, 0.1*cm, 1.5*cm, 0.8*cm, rx=0.25*cm, ry=0.25*cm,

hrefURL="http://en.wikipedia.org/wiki/Rounded_rectangle",
                        hrefTitle="Rounded Rectangle",
                        strokeColor=colors.red,
                        fillColor=colors.yellow))

drawing.add(shapes.String(1*cm, 1*cm, "Hello World!",
                          hrefURL="http://en.wikipedia.org/wiki/Hello_world",
                          hrefTitle="Why 'Hello World'?",
                          fillColor=colors.darkgreen))
drawing.add(shapes.Rect(4.5*cm, 0.5*cm, 5*cm, 1*cm,
                         hrefURL="http://en.wikipedia.org/wiki/Rectangle",
                         hrefTitle="Wikipedia page on rectangles",
                         strokeColor=colors.blue,
                         fillColor=colors.red))
drawing.add(shapes.Ellipse(7*cm, 1*cm, 2*cm, 0.95*cm,
                           hrefURL="http://en.wikipedia.org/wiki/Ellipse",
                           strokeColor=colors.black,
                           fillColor=colors.yellow))
drawing.add(shapes.Circle(7*cm, 1*cm, 0.9*cm,
                           hrefURL="http://en.wikipedia.org/wiki/Circle",
                          strokeColor=colors.black,
                          fillColor=colors.brown))
drawing.add(shapes.Ellipse(7*cm, 1*cm, 0.5*cm, 0.9*cm,
                           hrefTitle="Tooltip with no link?",
                           strokeColor=colors.black,
                           fillColor=colors.black))
drawing.add(shapes.Polygon([4.5*cm, 1.25*cm, 5*cm, 0.1*cm, 4*cm, 0.1*cm],
                           hrefURL="http://en.wikipedia.org/wiki/Polygon",
                           hrefTitle="This triangle is a simple polygon.",
                           strokeColor=colors.darkgreen,
                           fillColor=colors.green))

renderPDF.drawToFile(drawing, "hello.pdf")
renderPS.drawToFile(drawing, "hello.ps")
renderSVG.drawToFile(drawing, "hello.svg")
renderPM.drawToFile(drawing, "hello.png", "PNG")

print "-" * 50
print renderSVG.drawToString(drawing)
print "-" * 50
print "Done"
-------------- next part --------------
A non-text attachment was scrubbed...
Name: svg_link_closed_paths.patch
Type: application/octet-stream
Size: 1425 bytes
Desc: not available
Url : <http://two.pairlist.net/pipermail/reportlab-users/attachments/20090224/2a4db1bf/attachment.obj>

From robin at reportlab.com  Tue Feb 24 12:07:24 2009
From: robin at reportlab.com (Robin Becker)
Date: Tue, 24 Feb 2009 17:07:24 +0000
Subject: [reportlab-users] jython25 branch
Message-ID: <49A4294C.8080001@chamonix.reportlab.co.uk>

Hi,

I've created a jython25 branch in the public repository at

http://svn.reportlab.com/svn/public/reportlab/branches/jython25/src

I've tested this with jython2.5b1 and it is mostly working.

Clearly there are some issues with image handling etc etc as we don't have PIL, 
but there are only a few test failures. I'm glad to say that ttf fonts can now 
be used as in ordinary python based ReportLab.
-- 
Robin Becker

From peter at maubp.freeserve.co.uk  Tue Feb 24 12:42:28 2009
From: peter at maubp.freeserve.co.uk (Peter)
Date: Tue, 24 Feb 2009 17:42:28 +0000
Subject: [reportlab-users] External hyperlinks in SVG output
In-Reply-To: <956003ae0901290838g6393c59dxb36e52626ff6429c@mail.gmail.com>
References: <320fb6e00901290737k1269fb4en92975edbcf87f88a@mail.gmail.com>
	<956003ae0901290838g6393c59dxb36e52626ff6429c@mail.gmail.com>
Message-ID: <320fb6e00902240942g264848bfta23eb143b0c6331e@mail.gmail.com>

On Thu, Jan 29, 2009 at 4:38 PM, Andy Robinson <andy at reportlab.com> wrote:
>
> 2009/1/29 Peter <peter at maubp.freeserve.co.uk>:
>> I imagine there are parallels here with PDF output, which can in
>> theory also include externals URLs. ?Is any of this currently possible
>> in ReportLab? ?I've been looking over the documentation and the source
>> code, but have concluded that it is not yet supported.
>
> No. ? A few people have asked us about charts with click-throughs
> and drill-downs, but we never did it.
>
> There are two parts to this. ?First, the shapes hierarchy in
> reportlab/graphics would need some extra properties so that
> a shape could have link properties. ?This is straightforward for
> manually constructed shapes, but presumably for charts one
> would need to allow some subclass or callback mechanism so the caller
> can define what URLs a data point or bar would link to.

What's in SVN now defines two optional attributes on the shape
objects, hrefURL and hrefTitle, which do nothing unless the render
code is aware of them.

> Secondly, each renderer would need to do the right thing.
> I can see this would be really easy with SVG.

Yes, this does now seem to be working pretty well for SVG.

> If you're willing to do a little work on this and contribute
> patches for core and SVG back end, we'd be happy to help out and
> maybe do the same in PDF in step with it.

I've been looking at ReportLab's PDF code, in particular the PDF
canvas object (which lives in reportlab/pdfgen/canvas.py rather than
reportlab/graphics/renderPDF.py) and see there is some code for "PDF
Actions" (a rather general feature which includes linking to an
external URL, or URI as Adobe seem to prefer).  In the PDF canvas
linkURL method a "PDF Action" is associated with an invisible "PDF
Annotation", which is a rectangular region.  If am I reading the code
right, this "link region" has to be a rectangle - it can't be a
rotated rectangle where the edges are not parallel to the page edges.
Can a "PDF Action" be attached to an ordinary shape, like a visible
(rotated) rectangle, or any closed path?

Peter

From robin at reportlab.com  Tue Feb 24 13:05:16 2009
From: robin at reportlab.com (Robin Becker)
Date: Tue, 24 Feb 2009 18:05:16 +0000
Subject: [reportlab-users] External hyperlinks in SVG output
In-Reply-To: <320fb6e00902240942g264848bfta23eb143b0c6331e@mail.gmail.com>
References: <320fb6e00901290737k1269fb4en92975edbcf87f88a@mail.gmail.com>	<956003ae0901290838g6393c59dxb36e52626ff6429c@mail.gmail.com>
	<320fb6e00902240942g264848bfta23eb143b0c6331e@mail.gmail.com>
Message-ID: <49A436DC.10902@chamonix.reportlab.co.uk>

.......
> 
> I've been looking at ReportLab's PDF code, in particular the PDF
> canvas object (which lives in reportlab/pdfgen/canvas.py rather than
> reportlab/graphics/renderPDF.py) and see there is some code for "PDF
> Actions" (a rather general feature which includes linking to an
> external URL, or URI as Adobe seem to prefer).  In the PDF canvas
> linkURL method a "PDF Action" is associated with an invisible "PDF
> Annotation", which is a rectangular region.  If am I reading the code
> right, this "link region" has to be a rectangle - it can't be a
> rotated rectangle where the edges are not parallel to the page edges.
> Can a "PDF Action" be attached to an ordinary shape, like a visible
> (rotated) rectangle, or any closed path?
> 
> Peter
........

my reading of the latest link annotation dictionary definition seems to indicate 
that it can be a rectangle /Rect entry or


> QuadPoints
> array
> (Optional; PDF 1.6) An array of 8 ? n numbers specifying the coordinates of nquadrilaterals in default user space that comprise the region in which the link should be activated. The coordinates for each quadrilateral are given in the order
> x1 y1 x2 y2 x3 y3 x4 y4
> specifying the four vertices of the quadrilateral in counterclockwise order. For orientation purposes, such as when applying an underline border style, the bottom of a quadrilateral is the line formed by (x1 , y1) and (x2 , y2).
> If this entry is not present or the viewer application does not recognize it, the region specified by the Rect entry should be used. QuadPoints should be ignored
> if any coordinate in the array lies outside the region specified by Rect.

it doesn't look like you can get a generic path region as the clickable.
-- 
Robin Becker

From peter at maubp.freeserve.co.uk  Tue Feb 24 13:22:03 2009
From: peter at maubp.freeserve.co.uk (Peter)
Date: Tue, 24 Feb 2009 18:22:03 +0000
Subject: [reportlab-users] External hyperlinks in SVG output
In-Reply-To: <49A436DC.10902@chamonix.reportlab.co.uk>
References: <320fb6e00901290737k1269fb4en92975edbcf87f88a@mail.gmail.com>
	<956003ae0901290838g6393c59dxb36e52626ff6429c@mail.gmail.com>
	<320fb6e00902240942g264848bfta23eb143b0c6331e@mail.gmail.com>
	<49A436DC.10902@chamonix.reportlab.co.uk>
Message-ID: <320fb6e00902241022h7136fd32j2382cdd39de4834f@mail.gmail.com>

On Tue, Feb 24, 2009 at 6:05 PM, Robin Becker <robin at reportlab.com> wrote:
>
> my reading of the latest link annotation dictionary definition seems to
> indicate that it can be a rectangle /Rect entry or
>
>> QuadPoints
>> array
>> (Optional; PDF 1.6) An array of 8 ? n numbers specifying the coordinates
>> of nquadrilaterals in default user space that comprise the region in which
>> the link should be activated. The coordinates for each quadrilateral are
>> given in the order x1 y1 x2 y2 x3 y3 x4 y4 specifying the four vertices of
>> the quadrilateral in counterclockwise order. For orientation purposes,
>> such as when applying an underline border style, the bottom of a
>> quadrilateral is the line formed by (x1 , y1) and (x2 , y2).
>> If this entry is not present or the viewer application does not recognize it,
>> the region specified by the Rect entry should be used. QuadPoints should
>> be ignored if any coordinate in the array lies outside the region specified
>> by Rect.
>
> it doesn't look like you can get a generic path region as the clickable.

Thanks for the information - that is at least better than just
non-rotated rectangular regions.  What version of PDF does ReportLab
normally generate?  If we can use the QuadPoints feature that would
support the ReportLab shapes.Rect nicely (at any rotation or other
translation), and polygons with four points.  Triangles (polygons with
three points) should also be easy.  Other shapes would have to be
approximated (ugly), given as the bounding rectangle (simple), or not
supported at all...

Does any of this apply to PostScript?  I know you can embed some PDF
stuff into PostScript, but I haven't looked at the details of
renderPS.py to see how ReportLab does this.

Thanks,

Peter

From chris.hoy at hoyc.fsnet.co.uk  Wed Feb 25 03:35:00 2009
From: chris.hoy at hoyc.fsnet.co.uk (Chris Hoy)
Date: Wed, 25 Feb 2009 08:35:00 -0000
Subject: [reportlab-users] Problem install 2.3 via easy_install on
	Unbunto
In-Reply-To: <956003ae0902240056q78a3d6c7xd1f12bda69065c08@mail.gmail.com>
References: <39A6FE6D0D79457FBDDFCF4190A11759@Yellowstone>
	<956003ae0902240056q78a3d6c7xd1f12bda69065c08@mail.gmail.com>
Message-ID: <578C90B806B14DF5825849A06877FF4D@Yellowstone>

Hi Andy

Thank, installed and working fine

Chris

>> Hi.  Awesome performance in Beijing, by the way!
>>
>>
>>  I?m getting the following error when trying to install ReportLab 2.3 via
>>  easy_install, I suspect that I haven?t install all the necessary
packages
>>  for the complier any help would be appreciated

>> We don't currently support easy_install.  (Unfortunately, easy_install >>
attempts
>> to support us, but that's the author's problem!).  Just download the
package and
>> run 'python setup.py install', and all should work.

>> Why is a long story but we laid out our directories, version numbers
>> and build processes long before easy_install was invented and each time
>> we rearrange these things it causes us some pain and takes up a lot of >>
time
>> (for no functionality in return).  We might get there on our next
>> major version.

>> 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

No virus found in this outgoing message.
Checked by AVG - www.avg.com
Version: 8.0.237 / Virus Database: 270.11.3/1970 - Release Date: 02/24/09
13:35:00





From gherman at darwin.in-berlin.de  Thu Feb 26 06:02:27 2009
From: gherman at darwin.in-berlin.de (Dinu Gherman)
Date: Thu, 26 Feb 2009 12:02:27 +0100
Subject: [reportlab-users] Tons of table questions
Message-ID: <27F3B42F-D09E-451F-B323-475C2AEDBC18@darwin.in-berlin.de>

Hi,

I guess this is the main (FAQ one) to start with... How can I make
tables render more quickly?

 From my experiments it seems that tables are rendered roughly in
O(n^2) with n being the number of rows (in my current example 1000
rows take a minute, but 10 x 100 take 6 seconds).

Does it have to be like that? What can be done against it? Do plain
strings instead of Paragraphs help? Does it help to preset all col-
umn widths with fixed values? Is the LongTable class the solution?

Also, is there any useful sample code / test / documentation for
the splitlast/splitfirst features in tables that one can understand
without digging into mail archives, only to find that it likely
is not working as it should?

And finally, is there a workaround to speedup table rendering by
splitting it into little chunks and concatenating them? In order
for that to work one would need a table header that is shown only
at the beginning of a new frame. Is that realistic?

Regards,

Dinu


From robin at reportlab.com  Thu Feb 26 06:29:29 2009
From: robin at reportlab.com (Robin Becker)
Date: Thu, 26 Feb 2009 11:29:29 +0000
Subject: [reportlab-users] Tons of table questions
In-Reply-To: <27F3B42F-D09E-451F-B323-475C2AEDBC18@darwin.in-berlin.de>
References: <27F3B42F-D09E-451F-B323-475C2AEDBC18@darwin.in-berlin.de>
Message-ID: <49A67D19.5080806@chamonix.reportlab.co.uk>

Dinu Gherman wrote:
> Hi,
> 
> I guess this is the main (FAQ one) to start with... How can I make
> tables render more quickly?
> 
>  From my experiments it seems that tables are rendered roughly in
> O(n^2) with n being the number of rows (in my current example 1000
> rows take a minute, but 10 x 100 take 6 seconds).
> 
> Does it have to be like that? What can be done against it? Do plain
> strings instead of Paragraphs help? Does it help to preset all col-
> umn widths with fixed values? Is the LongTable class the solution?
> 
> Also, is there any useful sample code / test / documentation for
> the splitlast/splitfirst features in tables that one can understand
> without digging into mail archives, only to find that it likely
> is not working as it should?
> 
> And finally, is there a workaround to speedup table rendering by
> splitting it into little chunks and concatenating them? In order
> for that to work one would need a table header that is shown only
> at the beginning of a new frame. Is that realistic?
> 
> Regards,
> 
> Dinu

Dinu there was a long discussion about table times a few years ago. In 
particular there was a longTable speedup proposed and contributed by Henning von 
Bargen. That was about 5 years ago. In rl_config.py there's a variable to turn 
this optimisation on and off for the default Table class. If you set it to one 
in your local_rl_config then you should get rid of the n**2 behaviour.
-- 
Robin Becker

From paul.barrass at safeonlinebilling.com  Thu Feb 26 06:32:13 2009
From: paul.barrass at safeonlinebilling.com (Paul Barrass)
Date: Thu, 26 Feb 2009 11:32:13 +0000
Subject: [reportlab-users] Tons of table questions
In-Reply-To: <27F3B42F-D09E-451F-B323-475C2AEDBC18@darwin.in-berlin.de>
References: <27F3B42F-D09E-451F-B323-475C2AEDBC18@darwin.in-berlin.de>
Message-ID: <49A67DBD.8060406@safeonlinebilling.com>

Dinu Gherman wrote:
> Hi,
>
> I guess this is the main (FAQ one) to start with... How can I make
> tables render more quickly?
I refer you again to my earlier post 
http://two.pairlist.net/pipermail/reportlab-users/2008-October/007594.html 
[Potential workaround for performance issues on VERY long tables.], and 
the thread cited in that post 
http://two.pairlist.net/pipermail/reportlab-users/2004-May/003019.html 
[Large tables - exponentially increasing run times] which discusses the 
O(n^2) behaviour.  Both posts suggest that the idea is worth making a 
note of in the documentation.

>
> Do plain strings instead of Paragraphs help? 
I've found that plain text rather than Paragraphs leads to an 
approximate doubling of rendering speed, I've also found that the extra 
flexibility obtained by using a Paragraph (most notably word-wrapping, 
but also more control over styling etc.) is worth the extra time, but YMMV

> Does it help to preset all column widths with fixed values? Is the 
> LongTable class the solution?
I've always used fixed-width tables, so can't really comment from 
experience, though I think the LongTable class works to improve 
performance on the width-setting stage of Table generation.

> And finally, is there a workaround to speedup table rendering by
> splitting it into little chunks and concatenating them? In order
> for that to work one would need a table header that is shown only
> at the beginning of a new frame. Is that realistic?
This is, or at least was in 2004, the way to do it.  I'm not sure if any 
improvements to the Table splitting code have been made since then.  My 
workaround of wrapping the fragments in a PTOContainer allows table 
headers (and I guess footers) to be used when the Table is split between 
pages.

Paul Barrass.

From gherman at darwin.in-berlin.de  Thu Feb 26 06:41:24 2009
From: gherman at darwin.in-berlin.de (Dinu Gherman)
Date: Thu, 26 Feb 2009 12:41:24 +0100
Subject: [reportlab-users] Tons of table questions
In-Reply-To: <27F3B42F-D09E-451F-B323-475C2AEDBC18@darwin.in-berlin.de>
References: <27F3B42F-D09E-451F-B323-475C2AEDBC18@darwin.in-berlin.de>
Message-ID: <AAD9FC2C-6ADC-4242-B146-E686CD8B5686@darwin.in-berlin.de>

I wrote:

> Is the LongTable class the solution?


Thanks for the answers so far! I also found that LongTable saves
me about an order of magnitude of time, but otherwise still be-
haves more like O(n^2) than like O(n)...

I'll try using less Paragraphs and more fixed column widths...

Regards,

Dinu


From paul.barrass at safeonlinebilling.com  Thu Feb 26 06:48:14 2009
From: paul.barrass at safeonlinebilling.com (Paul Barrass)
Date: Thu, 26 Feb 2009 11:48:14 +0000
Subject: [reportlab-users] Tons of table questions
In-Reply-To: <49A67D19.5080806@chamonix.reportlab.co.uk>
References: <27F3B42F-D09E-451F-B323-475C2AEDBC18@darwin.in-berlin.de>
	<49A67D19.5080806@chamonix.reportlab.co.uk>
Message-ID: <49A6817E.5030407@safeonlinebilling.com>

Robin Becker wrote:
> Dinu there was a long discussion about table times a few years ago. In 
> particular there was a longTable speedup proposed and contributed by 
> Henning von Bargen. That was about 5 years ago. In rl_config.py 
> there's a variable to turn this optimisation on and off for the 
> default Table class. If you set it to one in your local_rl_config then 
> you should get rid of the n**2 behaviour.
In my experience with long tables, I've found the non-linear behaviour 
both with and without LongTable() - I think the LongTable optimisation 
optimises the calculation of /where/ to split (which can improve 
performance where cell widths/heights are variable) but the non-linear 
behaviour is occurring during the actual splitting, where LongTable is 
identical to Table.

Paul Barrass.

From robin at reportlab.com  Thu Feb 26 07:14:59 2009
From: robin at reportlab.com (Robin Becker)
Date: Thu, 26 Feb 2009 12:14:59 +0000
Subject: [reportlab-users] Tons of table questions
In-Reply-To: <49A6817E.5030407@safeonlinebilling.com>
References: <27F3B42F-D09E-451F-B323-475C2AEDBC18@darwin.in-berlin.de>	<49A67D19.5080806@chamonix.reportlab.co.uk>
	<49A6817E.5030407@safeonlinebilling.com>
Message-ID: <49A687C3.2000202@chamonix.reportlab.co.uk>

Paul Barrass wrote:
> Robin Becker wrote:
>> Dinu there was a long discussion about table times a few years ago. In 
>> particular there was a longTable speedup proposed and contributed by 
>> Henning von Bargen. That was about 5 years ago. In rl_config.py 
>> there's a variable to turn this optimisation on and off for the 
>> default Table class. If you set it to one in your local_rl_config then 
>> you should get rid of the n**2 behaviour.
> In my experience with long tables, I've found the non-linear behaviour 
> both with and without LongTable() - I think the LongTable optimisation 
> optimises the calculation of /where/ to split (which can improve 
> performance where cell widths/heights are variable) but the non-linear 
> behaviour is occurring during the actual splitting, where LongTable is 
> identical to Table.
> 
......
yes there's bound to be an element of n**2 since the split has to create the new 
table from the old by copying.

We currently don't have the concept of a view into the table data.

The original problem which the longTable optimisation addressed was that we 
scanned the whole table each time again rather than splitting early.
-- 
Robin Becker

From robin at reportlab.com  Thu Feb 26 07:20:10 2009
From: robin at reportlab.com (Robin Becker)
Date: Thu, 26 Feb 2009 12:20:10 +0000
Subject: [reportlab-users] Tons of table questions
In-Reply-To: <49A6817E.5030407@safeonlinebilling.com>
References: <27F3B42F-D09E-451F-B323-475C2AEDBC18@darwin.in-berlin.de>	<49A67D19.5080806@chamonix.reportlab.co.uk>
	<49A6817E.5030407@safeonlinebilling.com>
Message-ID: <49A688FA.8040309@chamonix.reportlab.co.uk>

I guess we could imagine a two pass algorithm where the first pass O(n) gathers 
  overall width height data and accumulates stuff for a second pass which then 
lays out the data using a slice of the original data that's left to be rendered 
and does no copying at all of data or styles.
-- 
Robin Becker

From H.vonBargen at t-p.com  Thu Feb 26 07:53:22 2009
From: H.vonBargen at t-p.com (Henning von Bargen)
Date: Thu, 26 Feb 2009 13:53:22 +0100
Subject: [reportlab-users] Tons of table questions
In-Reply-To: <mailman.15.1235650822.17883.reportlab-users@reportlab.com>
Message-ID: <77A9E75DA75C564D82BDC4B18A7D8525305765@ROHRPOSTIX.tup.ads>

> From: Robin Becker <robin at reportlab.com>
> Subject: Re: [reportlab-users] Tons of table questions
> 
> I guess we could imagine a two pass algorithm where the first 
> pass O(n) gathers 
>   overall width height data and accumulates stuff for a 
> second pass which then 
> lays out the data using a slice of the original data that's 
> left to be rendered 
> and does no copying at all of data or styles.
> -- 
> Robin Becker

I guess that further optimization requires additional conditions:
1) The first and all the following frames have to be of the same width.
2) Possibly:  ... the same height.

But for long tables, at least condition 1) should be no problem
for real-world applications.

Henning

From gherman at darwin.in-berlin.de  Thu Feb 26 08:12:55 2009
From: gherman at darwin.in-berlin.de (Dinu Gherman)
Date: Thu, 26 Feb 2009 14:12:55 +0100
Subject: [reportlab-users] Tons of table questions
In-Reply-To: <77A9E75DA75C564D82BDC4B18A7D8525305765@ROHRPOSTIX.tup.ads>
References: <77A9E75DA75C564D82BDC4B18A7D8525305765@ROHRPOSTIX.tup.ads>
Message-ID: <F15CBB83-574F-4DBA-B8D3-EDB8B3E7F1CE@darwin.in-berlin.de>

Henning von Bargen:

> I guess that further optimization requires additional conditions:
> 1) The first and all the following frames have to be of the same  
> width.
> 2) Possibly:  ... the same height.
>
> But for long tables, at least condition 1) should be no problem
> for real-world applications.

I think both conditions are very realistic in practice.
But even without them I cannot see any non-linearity re-
lated to the problem itself. So, maybe an implementation
using generators might solve this?

Dinu


From robin at reportlab.com  Thu Feb 26 09:01:14 2009
From: robin at reportlab.com (Robin Becker)
Date: Thu, 26 Feb 2009 14:01:14 +0000
Subject: [reportlab-users] Tons of table questions
In-Reply-To: <F15CBB83-574F-4DBA-B8D3-EDB8B3E7F1CE@darwin.in-berlin.de>
References: <77A9E75DA75C564D82BDC4B18A7D8525305765@ROHRPOSTIX.tup.ads>
	<F15CBB83-574F-4DBA-B8D3-EDB8B3E7F1CE@darwin.in-berlin.de>
Message-ID: <49A6A0AA.5000805@chamonix.reportlab.co.uk>

Dinu Gherman wrote:
> Henning von Bargen:
> 
>> I guess that further optimization requires additional conditions:
>> 1) The first and all the following frames have to be of the same width.
>> 2) Possibly:  ... the same height.
>>
>> But for long tables, at least condition 1) should be no problem
>> for real-world applications.
> 
> I think both conditions are very realistic in practice.
> But even without them I cannot see any non-linearity re-
> lated to the problem itself. So, maybe an implementation
> using generators might solve this?
> 
> Dinu
.......
I don't think it much matters whether you use generators or something else to 
hold the actual data. What matters is to reduce or eliminate the copying.

The longTable n**2 problem relates to copying the data which is of order n-m 
where m is the number of data rows already dealt with. Since you need to do that 
operation some fraction of n times you get a term in (n-p) + n-2p + n-3p ..... 
which sums to O(n**2).

By putting the tail of the data into a generator (+a lookahead buffer for the 
bits discarded by the current table) the n-m term is indeed reduced, but you 
still have to copy the data around. So the split would look like

T0(B=[],G)

    -->      T1(B=[1...p],None) + T2(B=[p+1,..],G)

where B represent s a buffer of rows popped from the generator, but not yet 
rendered and G represents the tail of the data. Clearly you have to copy the 
[p+1,..p+x] rows that are unused from T1. Hopefully this copying is small (say 
only one overflow row).

If you use a data array or simulation thereof you can represent the split as

T0(1,len(A),A)

    -->     T1(1,p,A) + T2(p+1,len(A),A)

so no copying is required, but we change the view indices as we go forward.
-- 
Robin Becker

From gherman at darwin.in-berlin.de  Thu Feb 26 09:56:32 2009
From: gherman at darwin.in-berlin.de (Dinu Gherman)
Date: Thu, 26 Feb 2009 15:56:32 +0100
Subject: [reportlab-users] Tons of table questions
In-Reply-To: <49A6A0AA.5000805@chamonix.reportlab.co.uk>
References: <77A9E75DA75C564D82BDC4B18A7D8525305765@ROHRPOSTIX.tup.ads>
	<F15CBB83-574F-4DBA-B8D3-EDB8B3E7F1CE@darwin.in-berlin.de>
	<49A6A0AA.5000805@chamonix.reportlab.co.uk>
Message-ID: <7038B056-81E3-48E5-831B-473DE3F66B43@darwin.in-berlin.de>

Robin Becker:

> [...] so no copying is required, but we change the view
> indices as we go forward.

Thanks for the analysis! In essence, yes, the solution would
be to use the appropriate datastructure. Could you add this to
your bug tracker, as a design change request or as whatever is
appropriate to RL?

Thanks,

Dinu



From andy at reportlab.com  Thu Feb 26 11:46:25 2009
From: andy at reportlab.com (Andy Robinson)
Date: Thu, 26 Feb 2009 16:46:25 +0000
Subject: [reportlab-users] Tons of table questions
In-Reply-To: <F15CBB83-574F-4DBA-B8D3-EDB8B3E7F1CE@darwin.in-berlin.de>
References: <77A9E75DA75C564D82BDC4B18A7D8525305765@ROHRPOSTIX.tup.ads>
	<F15CBB83-574F-4DBA-B8D3-EDB8B3E7F1CE@darwin.in-berlin.de>
Message-ID: <956003ae0902260846r383a076bl13278d70cc2de72e@mail.gmail.com>

2009/2/26 Dinu Gherman <gherman at darwin.in-berlin.de>:
> I think both conditions are very realistic in practice.
> But even without them I cannot see any non-linearity re-
> lated to the problem itself. So, maybe an implementation
> using generators might solve this?

The non-linearity is as follows.  If you make a big table with,
say, 1000 rows x 10 cols, and space for 40 rows on each page,
and specify  no widths to help us, then we have to size every
table cell - that's 10,000 objects. We then work out the optimal
column widths for the present page.  We then say "OK, I can
only fit about 40 rows on this page",  and we split it into a 40-
row table and a 960-row table.

Then, on the next page, we start again, this time sizing 9600
cells.  But this page might have a different frame width, so we
just dumbly start again and recalculate all those table cell sizes.

I don't know the maths but the scaling is probably the same as
the song "ten green bottles, hanging on the wall" or the
"twelve days of Christmas".

AFAIR the LongTable optimisation works by just looking at
the first few rows to work out the column widths.

Smarter implementations are certainly possible but I would
urge people to just create a bunch of smaller tables in the
first place.  If you are outputting thousands of rows, you
probably have some kind of grouping mechanism or want
headers and footers anyway, and it's easiest to style things
AND faster to render if you use a separate table for each visually
similar slab of content.

- Andy

From gherman at darwin.in-berlin.de  Thu Feb 26 15:44:19 2009
From: gherman at darwin.in-berlin.de (Dinu Gherman)
Date: Thu, 26 Feb 2009 21:44:19 +0100
Subject: [reportlab-users] Tons of table questions
In-Reply-To: <956003ae0902260846r383a076bl13278d70cc2de72e@mail.gmail.com>
References: <77A9E75DA75C564D82BDC4B18A7D8525305765@ROHRPOSTIX.tup.ads>
	<F15CBB83-574F-4DBA-B8D3-EDB8B3E7F1CE@darwin.in-berlin.de>
	<956003ae0902260846r383a076bl13278d70cc2de72e@mail.gmail.com>
Message-ID: <DD04DCD3-6C3E-4C19-B836-4241AA1F8E0A@darwin.in-berlin.de>

Andy Robinson:

> The non-linearity is as follows.  If you make a big table with,
> say, 1000 rows x 10 cols, and space for 40 rows on each page,
> and specify  no widths to help us, then we have to size every
> table cell - that's 10,000 objects. We then work out the optimal
> column widths for the present page.  We then say "OK, I can
> only fit about 40 rows on this page",  and we split it into a 40-
> row table and a 960-row table.
>
> Then, on the next page, we start again, this time sizing 9600
> cells.  But this page might have a different frame width, so we
> just dumbly start again and recalculate all those table cell sizes.

I see. Is there a reason to continue sizing cells in vain after
reaching a frame split?

> AFAIR the LongTable optimisation works by just looking at
> the first few rows to work out the column widths.

That seems to me the right thing to do, ideally until the
first split occurs.

> Smarter implementations are certainly possible but I would
> urge people to just create a bunch of smaller tables in the
> first place.  If you are outputting thousands of rows, you
> probably have some kind of grouping mechanism or want
> headers and footers anyway, and it's easiest to style things
> AND faster to render if you use a separate table for each visually
> similar slab of content.

Agreed, and I do use repeating a header (repeating footers do
not exist, if I'm not mistaken). But grouping sometimes is just
not an option.

Concatenated tables, as I described before, might be a work-
around, but then one would need to draw a header (if you need
one) only after a split, but not at the beginning of the table,
which is also not possible.

Dinu


From andy at reportlab.com  Thu Feb 26 16:04:58 2009
From: andy at reportlab.com (Andy Robinson)
Date: Thu, 26 Feb 2009 21:04:58 +0000
Subject: [reportlab-users] Tons of table questions
In-Reply-To: <DD04DCD3-6C3E-4C19-B836-4241AA1F8E0A@darwin.in-berlin.de>
References: <77A9E75DA75C564D82BDC4B18A7D8525305765@ROHRPOSTIX.tup.ads>
	<F15CBB83-574F-4DBA-B8D3-EDB8B3E7F1CE@darwin.in-berlin.de>
	<956003ae0902260846r383a076bl13278d70cc2de72e@mail.gmail.com>
	<DD04DCD3-6C3E-4C19-B836-4241AA1F8E0A@darwin.in-berlin.de>
Message-ID: <956003ae0902261304g7101d870t1f4db00543bb469e@mail.gmail.com>

2009/2/26 Dinu Gherman <gherman at darwin.in-berlin.de>:
> Andy Robinson:
>
> I see. Is there a reason to continue sizing cells in vain after
> reaching a frame split?

I just reviewed the code and that's exactly what the longTable option does.

In my opinion this should be 'on' by default (in fact should have been
years ago).  Can anyone here see problems if we change this for the
next release?

A better layout algorithm is possible, but would take a week's
work by someone to do it safely, and it's not an internal priority
for us right now.  Patches are welcome though ;-)

- Andy

From andy at reportlab.com  Thu Feb 26 16:11:02 2009
From: andy at reportlab.com (Andy Robinson)
Date: Thu, 26 Feb 2009 21:11:02 +0000
Subject: [reportlab-users] Tons of table questions
In-Reply-To: <DD04DCD3-6C3E-4C19-B836-4241AA1F8E0A@darwin.in-berlin.de>
References: <77A9E75DA75C564D82BDC4B18A7D8525305765@ROHRPOSTIX.tup.ads>
	<F15CBB83-574F-4DBA-B8D3-EDB8B3E7F1CE@darwin.in-berlin.de>
	<956003ae0902260846r383a076bl13278d70cc2de72e@mail.gmail.com>
	<DD04DCD3-6C3E-4C19-B836-4241AA1F8E0A@darwin.in-berlin.de>
Message-ID: <956003ae0902261311v612c0c49qfaa34202e2ead8a1@mail.gmail.com>

2009/2/26 Dinu Gherman <gherman at darwin.in-berlin.de>:
> Concatenated tables, as I described before, might be a work-
> around, but then one would need to draw a header (if you need
> one) only after a split, but not at the beginning of the table,
> which is also not possible.

That's exactly what the "PTO" mechanism is intended for.  But it's way
more flexible than simply repeating a row - you can specify tables,
other paragraphs or any other chunk of content to be injected at the
top or bottom of a page.

See test_025_pto on this page...the PDF and RML show this, and the
flowables work as you would expect from looking at the RML...
    http://developer.reportlab.com/examples.html


-- 
Andy

From haraldarminmassa at gmail.com  Fri Feb 27 06:10:41 2009
From: haraldarminmassa at gmail.com (Harald Armin Massa[legacy])
Date: Fri, 27 Feb 2009 12:10:41 +0100
Subject: [reportlab-users] WordCount in reportlab paragraph
Message-ID: <e3e180dc0902270310j16fdfef8y2534a6051e67467@mail.gmail.com>

Hello,

I am creating long insurance documents with reportlab. I use a
2-column-layout on DIN-A4, so the lines are rather shortish (<9cm)

And, as it is Liability Insurance AND German language, it is rather layers
talk with VERY long words.

To add insult, those are expected to be justified. Many times it works. BUT
... then I had lines with only 4 words in them.

        'Abgrenzungen und Erweiterungen des'

and they "crossed the border", that is: they were drawn outside the Column.


Tracking it down, I guessed the culprit withing:  _justifyDrawParaLineX
ofparagraph.py

def _justifyDrawParaLineX( tx, offset, line, last=0):
    setXPos(tx,offset)
    extraSpace = line.extraSpace
    #~ nSpaces = line.wordCount - 1
    nSpaces = line.wordCount # This is my fix
    if last or not nSpaces or abs(extraSpace)<=1e-8 or line.lineBreak:
        _putFragLine(offset, tx, line)  #no space modification
    else:
        #~ print "-"*20
        #~ print "wordspace",line.extraSpace,nSpaces ,extraSpace /
float(nSpaces)
        #~ print "'"*20
        #~ print line
        tx.setWordSpace(extraSpace / float(nSpaces))
        _putFragLine(offset, tx, line)
        tx.setWordSpace(0)
    setXPos(tx,-offset)

which gave me the following debug output:

FragLine('lineBreak'  False
[yadda yadda yadda]
      'text'
        'Abgrenzungen und Erweiterungen des'
[yadda yadda yadda]
  'extraSpace'
    76.078425196850333
  'wordCount'
    3
) #FragLine


As of documentation within the code:

class FragLine(ABag):
    """
    class FragLine contains a styled line (ie a line with more than one
style)::

        extraSpace  unused space for justification only
*        wordCount   1+spaces in line for justification purposes*

    """

Now I was curious:         'Abgrenzungen und Erweiterungen des'
are clearly FOUR words, and 3 Spaces. So 1+3 should be 4; and wordCount 3
(as there is no such concept as a "negative word" which could be further
down the FragLine)


Checking out some more FragLines, I discovered that wordCount allways was
"one less then counted by Harald"

So I removed that "-1" from _justifyDrawParaLineX and everything works
great. (or better, acceptable, as "4 words justified in a line" allways look
cruel)


My fix works; and using Hennings Wordaxe, splitting those words in Fragmens
it even looks fair.

BUT: is my fix correct? If yes, can it please be incorporated in RL 2.4 and
SVN Trunk?

best wishes,

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!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://two.pairlist.net/pipermail/reportlab-users/attachments/20090227/8aad2bf8/attachment.html>

From robin at reportlab.com  Fri Feb 27 06:53:51 2009
From: robin at reportlab.com (Robin Becker)
Date: Fri, 27 Feb 2009 11:53:51 +0000
Subject: [reportlab-users] WordCount in reportlab paragraph
In-Reply-To: <e3e180dc0902270310j16fdfef8y2534a6051e67467@mail.gmail.com>
References: <e3e180dc0902270310j16fdfef8y2534a6051e67467@mail.gmail.com>
Message-ID: <49A7D44F.2090109@chamonix.reportlab.co.uk>

Harald Armin Massa[legacy] wrote:
> Hello,
> 
> I am creating long insurance documents with reportlab. I use a
> 2-column-layout on DIN-A4, so the lines are rather shortish (<9cm)
> 
> And, as it is Liability Insurance AND German language, it is rather layers
> talk with VERY long words.
> 
> To add insult, those are expected to be justified. Many times it works. BUT
> ... then I had lines with only 4 words in them.
> 
>         'Abgrenzungen und Erweiterungen des'
> 
> and they "crossed the border", that is: they were drawn outside the Column.
> 
> 
> Tracking it down, I guessed the culprit withing:  _justifyDrawParaLineX
> ofparagraph.py
> 
> def _justifyDrawParaLineX( tx, offset, line, last=0):
>     setXPos(tx,offset)
>     extraSpace = line.extraSpace
>     #~ nSpaces = line.wordCount - 1
>     nSpaces = line.wordCount # This is my fix
........-

Harald, thanks for the report. Unfortunately I think the problem lies elsewhere

.......

When I run this script

from reportlab.platypus.doctemplate import PageTemplate, BaseDocTemplate, 
Indenter, SimpleDocTemplate
doc = SimpleDocTemplate('test_platypus_just.pdf')
doc.rightMargin=3*72
styleSheet = getSampleStyleSheet()
bt = styleSheet['BodyText']
btj = ParagraphStyle('bodyText1j',parent=bt,alignment=TA_JUSTIFY)
story=[Paragraph("""<a name='top'/>Subsequent pages test pageBreakBefore, 
frameBreakBefore and
keepTogether attributes.  Generated at 1111. The number in brackets
at the end of each paragraph is its position in the story. 
llllllllllllllllllllllllll bbbbbbbbbbbbbbbbbbbbbb ccccccccccccccccccccccc 
ddddddddddddddddddddd eeee""",btj)]
doc.build(story)


I see this output if I put a print line.wordCount in _justifyDrawParaLineX
C:\Tmp>tparajust.py
5
10
12
2
2


And the split then looks like

Subsequent pages test pageBreakBefore, frameBreakBefore and           (6)

keepTogether attributes. Generated at 1111. The number in brackets at  (10)

the end of each paragraph is its position in the story.
llllllllllllllllllllllllll (12)

bbbbbbbbbbbbbbbbbbbbbb ccccccccccccccccccccccc (2)
ddddddddddddddddddddd eeee (2)

so all the counts are right except the first. I suspect something in the break 
is improperly incrementing the count (or not decrementing it). I'll take a look 
today.
-- 
Robin Becker

From andy at reportlab.com  Fri Feb 27 08:29:24 2009
From: andy at reportlab.com (Andy Robinson)
Date: Fri, 27 Feb 2009 13:29:24 +0000
Subject: [reportlab-users] WordCount in reportlab paragraph
In-Reply-To: <49A7D44F.2090109@chamonix.reportlab.co.uk>
References: <e3e180dc0902270310j16fdfef8y2534a6051e67467@mail.gmail.com>
	<49A7D44F.2090109@chamonix.reportlab.co.uk>
Message-ID: <956003ae0902270529o5f901c8ao723d7a41862440e7@mail.gmail.com>

2009/2/27 Robin Becker <robin at reportlab.com>:
> Harald, thanks for the report. Unfortunately I think the problem lies
> elsewhere
...

This kind of problem is eminently suited to unit tests, when
you have them.  Harald, could you spare us a paragraph
of super-long German insurance words for a test case without
violating client confidentiality?  It would be more fun!

- Andy

From gherman at darwin.in-berlin.de  Fri Feb 27 10:17:44 2009
From: gherman at darwin.in-berlin.de (Dinu Gherman)
Date: Fri, 27 Feb 2009 16:17:44 +0100
Subject: [reportlab-users] Chart axes without lines?
Message-ID: <A53936C7-9206-4EE9-B826-F9549B3BBFCE@darwin.in-berlin.de>

Hi,

I'm trying to create more Tufte[1]-like charts, i.e. I'd like
to use axes showing labels, but no ticks and especially no
axis line. I'm finding it impossible to switch only the axis
line off, but leave the labels. Is that right?

Regards,

Dinu

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


From robin at reportlab.com  Fri Feb 27 10:22:59 2009
From: robin at reportlab.com (Robin Becker)
Date: Fri, 27 Feb 2009 15:22:59 +0000
Subject: [reportlab-users] WordCount in reportlab paragraph
In-Reply-To: <956003ae0902270529o5f901c8ao723d7a41862440e7@mail.gmail.com>
References: <e3e180dc0902270310j16fdfef8y2534a6051e67467@mail.gmail.com>	<49A7D44F.2090109@chamonix.reportlab.co.uk>
	<956003ae0902270529o5f901c8ao723d7a41862440e7@mail.gmail.com>
Message-ID: <49A80553.3030905@chamonix.reportlab.co.uk>

Andy Robinson wrote:
> 2009/2/27 Robin Becker <robin at reportlab.com>:
>> Harald, thanks for the report. Unfortunately I think the problem lies
>> elsewhere
> ...

Harald, I think the(at least my) problem can be fixed with this patch; can you 
try it out?

Index: /code/reportlab/platypus/paragraph.py
===================================================================
--- /code/reportlab/platypus/paragraph.py       (revision 3441)
+++ /code/reportlab/platypus/paragraph.py       (working copy)
@@ -1118,9 +1118,11 @@
                          if nText!='' and nText[0]!=' ':
                              g.text += ' ' + nText

+                    ni = 0
                      for i in w[2:]:
                          g = i[0].clone()
                          g.text=i[1]
+                        if g.text: ni = 1
                          words.append(g)
                          fontSize = g.fontSize
                          if calcBounds:
@@ -1134,6 +1136,7 @@
                          maxSize = max(maxSize,fontSize)
                          maxAscent = max(maxAscent,ascent)
                          minDescent = min(minDescent,descent)
+                    if not nText and ni: n+=1

                      currentWidth = newWidth
                  else:  #either it won't fit, or it's a lineBreak tag


> 
> This kind of problem is eminently suited to unit tests, when

when the unit test framework can look at our pdfs and see that a word is not 
properly aligned then I'll give in. We could try saving images of these problems 
(say using ghost script) and then compare images.

> you have them.  Harald, could you spare us a paragraph
> of super-long German insurance words for a test case without
> violating client confidentiality?  It would be more fun!
> 
> - Andy
> _______________________________________________
> reportlab-users mailing list
> reportlab-users at reportlab.com
> http://two.pairlist.net/mailman/listinfo/reportlab-users
> 


-- 
Robin Becker

From robin at reportlab.com  Fri Feb 27 10:25:57 2009
From: robin at reportlab.com (Robin Becker)
Date: Fri, 27 Feb 2009 15:25:57 +0000
Subject: [reportlab-users] Chart axes without lines?
In-Reply-To: <A53936C7-9206-4EE9-B826-F9549B3BBFCE@darwin.in-berlin.de>
References: <A53936C7-9206-4EE9-B826-F9549B3BBFCE@darwin.in-berlin.de>
Message-ID: <49A80605.4040606@chamonix.reportlab.co.uk>

Dinu Gherman wrote:
> Hi,
> 
> I'm trying to create more Tufte[1]-like charts, i.e. I'd like
> to use axes showing labels, but no ticks and especially no
> axis line. I'm finding it impossible to switch only the axis
> line off, but leave the labels. Is that right?
> 
> Regards,
> 
> Dinu
> 
> [1] http://en.wikipedia.org/wiki/Edward_Tufte

I think visibleAxis = 0 on the particular axis might be what you seek.

-- 
Robin Becker

From gherman at darwin.in-berlin.de  Fri Feb 27 10:36:35 2009
From: gherman at darwin.in-berlin.de (Dinu Gherman)
Date: Fri, 27 Feb 2009 16:36:35 +0100
Subject: [reportlab-users] Chart axes without lines?
In-Reply-To: <49A80605.4040606@chamonix.reportlab.co.uk>
References: <A53936C7-9206-4EE9-B826-F9549B3BBFCE@darwin.in-berlin.de>
	<49A80605.4040606@chamonix.reportlab.co.uk>
Message-ID: <F3CB78AF-04FD-4451-BB78-F4AD0CABF627@darwin.in-berlin.de>

Robin Becker:

> I think visibleAxis = 0 on the particular axis might be what you seek.

Oops, indeed! Had not seen that (in the sea of attributes).

Thanks,

Dinu


From andy at reportlab.com  Fri Feb 27 10:37:43 2009
From: andy at reportlab.com (Andy Robinson)
Date: Fri, 27 Feb 2009 15:37:43 +0000
Subject: [reportlab-users] WordCount in reportlab paragraph
In-Reply-To: <49A80553.3030905@chamonix.reportlab.co.uk>
References: <e3e180dc0902270310j16fdfef8y2534a6051e67467@mail.gmail.com>
	<49A7D44F.2090109@chamonix.reportlab.co.uk>
	<956003ae0902270529o5f901c8ao723d7a41862440e7@mail.gmail.com>
	<49A80553.3030905@chamonix.reportlab.co.uk>
Message-ID: <956003ae0902270737k666b18dcga266b5ea22b22e20@mail.gmail.com>

2009/2/27 Robin Becker <robin at reportlab.com>:
>> This kind of problem is eminently suited to unit tests, when
>
> when the unit test framework can look at our pdfs and see that a word is not
> properly aligned then I'll give in. We could try saving images of these
> problems (say using ghost script) and then compare images.

Surely we can assume a built-in font (Courier, always 0.6em wide),
construct a paragraph, call wrap(), and then assert that certain
text appears in line 1, line 2 and so on, and that the amount of
extra space on each line is 'as expected'?

- Andy

From robin at reportlab.com  Fri Feb 27 10:56:25 2009
From: robin at reportlab.com (Robin Becker)
Date: Fri, 27 Feb 2009 15:56:25 +0000
Subject: [reportlab-users] WordCount in reportlab paragraph
In-Reply-To: <956003ae0902270737k666b18dcga266b5ea22b22e20@mail.gmail.com>
References: <e3e180dc0902270310j16fdfef8y2534a6051e67467@mail.gmail.com>	<49A7D44F.2090109@chamonix.reportlab.co.uk>	<956003ae0902270529o5f901c8ao723d7a41862440e7@mail.gmail.com>	<49A80553.3030905@chamonix.reportlab.co.uk>
	<956003ae0902270737k666b18dcga266b5ea22b22e20@mail.gmail.com>
Message-ID: <49A80D29.6030900@chamonix.reportlab.co.uk>

Andy Robinson wrote:
> 2009/2/27 Robin Becker <robin at reportlab.com>:
>>> This kind of problem is eminently suited to unit tests, when
>> when the unit test framework can look at our pdfs and see that a word is not
>> properly aligned then I'll give in. We could try saving images of these
>> problems (say using ghost script) and then compare images.
> 
> Surely we can assume a built-in font (Courier, always 0.6em wide),
> construct a paragraph, call wrap(), and then assert that certain
> text appears in line 1, line 2 and so on, and that the amount of
> extra space on each line is 'as expected'?
> 
......
the problem here was not in the breaking, but in some attributes that were 
collected alongside the break and that then affected the layout. A test could be 
devised that required the broken paragraph to have a specific structure and then 
compare all the properties against future outputs.

Of course that won't protect against changes in the later layout layer and makes 
changing the overall algorithm harder since each change requires measuring the 
properties again etc etc. Perhaps we could just record the expected output of 
the paragraph rendering rather than converting the pdf to pixels.
-- 
Robin Becker