From torresj@apb.es  Fri Jun  1 08:59:59 2001
From: torresj@apb.es (Joan Torres)
Date: Fri, 1 Jun 2001 09:59:59 +0200 (MET DST)
Subject: [Image-SIG] PIL 1.1.2 install problems
Message-ID: <200106010759.JAA19489@osona.apb.es>

 Hi everyone,

 I am working on a Sun (solaris 2.8) workstation. I have python2.1 build
 from sources with thinter module, and I try to install Imaging-1.1.2 following the steps :
 1. Build the Imaging core library (libImaging.a)
    cd Imaging-1.1.2/libImaging
    ./configure
    make
    make check (seems to be OK)
 2. Build a dynamically loaded module
    cd ..
    cp Setup.in Setup
    Edit de Setup in order to correct the paths for JPEG, ZLIB libraries
    make -f Makefile.pre.in boot
    make
    During the make process the following errors raise:

gcc -shared  ./_imaging.o ./decode.o ./encode.o ./map.o ./display.o ./outline.o ./path.o  libImaging/libImaging.a -L/opt/sfw/lib -ljpeg -L/usr/lib -lz  -o ./_imaging.so
Text relocation remains                         referenced
    against symbol                  offset      in file
<unknown>                           0x148       libImaging/libImaging.a(Unpack.o)
<unknown>                           0x14c       libImaging/libImaging.a(Unpack.o)
<unknown>                           0x150       libImaging/libImaging.a(Unpack.o)
<unknown>                           0x154       libImaging/libImaging.a(Unpack.o)
<unknown>                           0x158       libImaging/libImaging.a(Unpack.o)
<unknown>                           0x15c       libImaging/libImaging.a(Unpack.o)
<unknown>                           0x160       libImaging/libImaging.a(Unpack.o)
<unknown>                           0x2ac       libImaging/libImaging.a(Unpack.o)
............................
deflateInit2_                       0x1a0       libImaging/libImaging.a(ZipEncode.o)
deflate                             0x6a0       libImaging/libImaging.a(ZipEncode.o)
deflate                             0x760       libImaging/libImaging.a(ZipEncode.o)
ld: fatal: relocations remain against allocatable but non-writable sections
collect2: ld returned 1 exit status
*** Error code 1
make: Fatal error: Command failed for target `_imaging.so'


 Does anybody know what I am doing wrong !!!!

Thanks in advance,

Joan Torres
Autoritat Portuaria de Barcelona


From info@visualspace.nl  Tue Jun  5 19:51:15 2001
From: info@visualspace.nl (VisualSpace)
Date: Tue, 05 Jun 2001 20:51:15 +0200
Subject: [Image-SIG] t1python
Message-ID: <3.0.5.32.20010605205115.00a33400@10.0.0.1>

Dear Fred,

I'am searching for t1python and its not in t1lib-1.1.1. as mentioned on the
t1python - A Type1 Font Renderer for Python page:

..."The source code is distributed with t1lib, so no separate distribution
is required. Build t1lib, and then switch to the t1python/ directory
provided, and follow the build instructions included in the README file
there."...

I realy need the t1python, because I have to complete a Zope extension
drawing text realtime. Could you please help me to locate the thing or
provide me with an older version. I would be very grateful.

Best regards,
Kees van Drongelen
||||||||||||||||||||||||||||||||||||
VisualSpace
grafische en interactieve vormgeving

Krelagestraat 34
2012 CT Haarlem
T 023 532 33 66
F 023 532 33 04
http://www.visualspace.nl
||||||||||||||||||||||||||||||||||||


From klimek@grc.nasa.gov  Tue Jun  5 19:51:49 2001
From: klimek@grc.nasa.gov (Bob Klimek)
Date: Tue, 05 Jun 2001 14:51:49 -0400
Subject: [Image-SIG] new TIFF error
Message-ID: <4.2.2.20010605143123.00a9d688@parrot.grc.nasa.gov>

I'm getting an error when reading some TIFF files. I'm using PIL 1.1.1. The 
error is coming from TiffImagePlugging.py (line 165 ) in function getscalar().

The error reads:  ValueError: not a scalar.

This is a new problem as I used to read these .tif files all the time. It 
has something to do with color because 8-bit grayscale images are read in 
fine, but the error occurs only on some color .tif images (saved by one 
package). However other color tif images read into PIL without a problem.

Any ideas or solutions?

Bob



From fdrake@acm.org  Tue Jun  5 20:16:09 2001
From: fdrake@acm.org (Fred L. Drake, Jr.)
Date: Tue, 5 Jun 2001 15:16:09 -0400 (EDT)
Subject: [Image-SIG] Re: t1python
In-Reply-To: <3.0.5.32.20010605205115.00a33400@10.0.0.1>
References: <3.0.5.32.20010605205115.00a33400@10.0.0.1>
Message-ID: <15133.12281.449736.428925@cj42289-a.reston1.va.home.com>

VisualSpace writes:
 > I'am searching for t1python and its not in t1lib-1.1.1. as mentioned on the
 > t1python - A Type1 Font Renderer for Python page:

  I've not had time to keep up with it for a while.  You can find what
is available at:

	http://sourceforge.net/projects/t1python/

  If you want to bring it up to date or work on it, I'd be happy to
see any patches you'd like to send!


  -Fred

-- 
Fred L. Drake, Jr.  <fdrake at acm.org>
PythonLabs at Digital Creations



From etiennel@geospace.co.za  Wed Jun  6 16:22:57 2001
From: etiennel@geospace.co.za (Etienne Labuschagne)
Date: Wed, 6 Jun 2001 17:22:57 +0200
Subject: [Image-SIG] Please help - Overlaying image with transparent color over anothe
 r image
Message-ID: <6CBE5E8F222DCA4C910079182C5C93E48DC7CC@terraserver>

Hi all

I am trying to overlay two images on top of each other.  None of them have
transparency (normally they are RGB images) but I know which RGB color on
the top image I want to make transparent.   What I have done that works most
of the time, is convert the top image to P (paletted), then search in the
palette for the color matching my RGB value, change that to black in the
palette and all the other colors in the palette to white, thus making a mask
image.  Then using paste to paste the top image over the bottom image, using
my mask.  This works perfect IF the top image has a valid palette.

Now sometimes the top image does not have a palette (it is equal to None).
Ho do I create a palette for it?  I don't want to step through each and
every pixel in the image and build a palette from that as it is very slow.
After consulting the documentation,  I have tried using histogram, but I
can't seem to get it right. It seems to me histogram gives you a list saying
color number 1 is used 10 times, color number 2 is used 13 times, etc.  But
there is not even a palette, so what color is color number 1!!??.  I have
tried using point, but was even more confused as to what value of the pixel
it (point) works on (it seemed to me it also works on the color number,
rather than the RGB value)!

Please, any help would be greatly appreciated!

Regards
Etienne


From image-sig@lee-morgan.net  Thu Jun  7 06:17:40 2001
From: image-sig@lee-morgan.net (image-sig@lee-morgan.net)
Date: 07 Jun 2001 15:17:40 +1000
Subject: [Image-SIG] PIL 1.1.2 win32 IOError saving BMP's
Message-ID: <wkg0dchnrf.fsf@lee-morgan.net>

Hi,

Anyone know why I get this?

I can save from jpg to png ok, its just saving to a bitmap that fails.

Cheers

Lee


Traceback (most recent call last):
  File "bmptopng.py", line 74, in ?
    img.convert("RGB").save( open(name + '.bmp', 'wb'), "BMP" )
  File "c:\python20\pil\Image.py", line 723, in save
    SAVE[string.upper(format)](self, fp, filename)
  File "c:\python20\pil\BmpImagePlugin.py", line 199, in _save
    ImageFile._save(im, fp, [("raw", (0,0)+im.size, 0, (rawmode, stride, -1))])
  File "c:\python20\pil\ImageFile.py", line 374, in _save
    s = e.encode_to_file(fh, bufsize)
IOError: (0, 'Error')



script:

if __name__ == "__main__":

    path = <filename>
    img = Image.open( path )
    (name,ext) = os.path.splitext( path)

    img.save( name + ".bmp" )

    #img.convert("RGB").save( open(name + '.bmp', 'wb'), "BMP" )


-- 
image-sig



From lbates@swamisoft.com  Wed Jun 13 23:27:02 2001
From: lbates@swamisoft.com (Larry Bates)
Date: Wed, 13 Jun 2001 17:27:02 -0500
Subject: [Image-SIG] New to PIL routines
Message-ID: <00bc01c0f457$f8500620$4700a8c0@tsclsa1.al.home.com>

I haven't been using the PIL routines for very long, but I have been
successful in getting most functions to work.  I tried to use new functions
today to combine two .JPG files (background and foreground information)

I used ActiveState's built-in Python Package Manager and downloaded and
installed Python Imaging Library and the PILGraph package.  Some features of
the Python Imaging Library seem to work properly (open, resize, rotate,
thumbnails, etc.).  I've used them extensively in other programs.  Today I
needed to use new features (blend, composite, merge) and they are not
working properly.  The errors don't appear to be path or other easily
diagnosed problems.

Sample programs/output

import Image
#
# Blend two images together to form a composite image
#
file1="c:\\temp\\DollarBill.jpg"
file2="c:\\temp\\revenue001.jpg"
im=Image.blend(file1,file2,0.5)
im.save("c:\\temp\\blend.jpg")

gives following error

Traceback (most recent call last):
  File "c:\python21\pythonwin\pywin\framework\scriptutils.py", line 301, in
RunScript
    exec codeObject in __main__.__dict__
  File "F:\Swami\Projects\EBSCO\2000-2001\python\ALSDE_blend.py", line 12,
in ?
    im=Image.blend(file1,file2,0.5)
  File "C:\PYTHON21\PIL\Image.py", line 914, in blend
    im1.load()
AttributeError: load

and

import Image
#
# Blend two images together to form a composite image
#
file1="c:\\temp\\DollarBill.jpg"
file2="c:\\temp\\revenue001.jpg"
im=Image.composite(file1,file2,"L")
im.save("c:\\temp\\blend.jpg")

gives the following error:

Traceback (most recent call last):
  File "c:\python21\pythonwin\pywin\framework\scriptutils.py", line 301, in
RunScript
    exec codeObject in __main__.__dict__
  File "F:\Swami\Projects\EBSCO\2000-2001\python\ALSDE_blend.py", line 13,
in ?
    im=Image.composite(file1,file2,"L")
  File "C:\PYTHON21\PIL\Image.py", line 921, in composite
    image = image2.copy()
AttributeError: copy


Thanks in advance for any assistance.





From fredrik@pythonware.com  Thu Jun 14 12:35:12 2001
From: fredrik@pythonware.com (Fredrik Lundh)
Date: Thu, 14 Jun 2001 13:35:12 +0200
Subject: [Image-SIG] New to PIL routines
References: <00bc01c0f457$f8500620$4700a8c0@tsclsa1.al.home.com>
Message-ID: <029101c0f4c6$1699e420$0900a8c0@spiff>

larry wrote:
> file1="c:\\temp\\DollarBill.jpg"
> file2="c:\\temp\\revenue001.jpg"
> im=Image.blend(file1,file2,0.5)
> im.save("c:\\temp\\blend.jpg")

blend expects image objects, not strings.  use Image.open
to open the images, and it will work as expected.

hope this helps!

Cheers /F




From tw@engInsight.com  Mon Jun 18 21:22:10 2001
From: tw@engInsight.com (Thomas J. Wielenga)
Date: Mon, 18 Jun 2001 16:22:10 -0400
Subject: [Image-SIG] Has anyone made a TWAIN interface?
Message-ID: <NDBBIABBJMPIBAKELDGJIEOMCBAA.tw@engInsight.com>

I have an application in which I would like to scan in some images from a
sheet fed scanner and then manipulate them, make thumbnails etc.  I use
python extensively now, but have not as yet used PIL.  I need to control the
scanner and was wondering if anyone had made a TWAIN interface (for Win32).
Is this in the works or did I miss it in the PIL documentation?  It seems
like a natural step, since you can acquire all kinds of images through a
TWAIN interface.

Thomas J. Wielenga, Ph.D., P.E.
President, Engineering Insight LLC
email:  tw@EngInsight.com
phone:  (734) 913-8520




From jleach@mail.ocis.net  Wed Jun 20 17:46:51 2001
From: jleach@mail.ocis.net (Jason C. Leach)
Date: Wed, 20 Jun 2001 09:46:51 -0700 (PDT)
Subject: [Image-SIG] Chop up a 750MB TIFF.
Message-ID: <Pine.LNX.4.21.0106200944220.25358-100000@praha>

hi,

I have a 750MB TIFF file I can not do anything with.  I need to chop it up
into quarters. Can python help with this? Can anyone help with the code?

Thanks,
j.

......................
..... Jason C. Leach
.. 



From kent@springfed.com  Wed Jun 20 19:46:46 2001
From: kent@springfed.com (Kent Tenney)
Date: Wed, 20 Jun 2001 13:46:46 -0500
Subject: [Image-SIG] EXIF information
Message-ID: <200106201846.LAA16738@svc1.netwk-innov.net>

Howdy,

Does anyone have pointers on how to
read EXIF in a jpg from a digital camera
using Python?

TIA,

Kent



From jleach@mail.ocis.net  Wed Jun 20 19:52:08 2001
From: jleach@mail.ocis.net (Jason C. Leach)
Date: Wed, 20 Jun 2001 11:52:08 -0700 (PDT)
Subject: [Image-SIG] ImageEnhance
In-Reply-To: <Pine.LNX.4.21.0106200944220.25358-100000@praha>
Message-ID: <Pine.LNX.4.21.0106201149490.25358-100000@praha>

hi,

How do we get images out of ImageEnhance? For example if I:
enh = ImageEnhance.Contrast(im)
does it change 'im', or a copy of 'im'? What I am trying to do is save
'im' after the contrast change.

Thanks,
j.


......................
..... Jason C. Leach
.. 




From Kevin.Cazabon@gretag.com  Wed Jun 20 20:26:04 2001
From: Kevin.Cazabon@gretag.com (Kevin.Cazabon@gretag.com)
Date: Wed, 20 Jun 2001 13:26:04 -0600
Subject: [Image-SIG] ImageEnhance
Message-ID: <OF08383956.7E94EE2C-ON87256A71.006AAD28@fotoprint.com>

Fred helped me with a similar issue I had with huge images (1.5GB+!) a year
or so ago, using the im.tile attribute.

See his respose, it should help:
http://mail.python.org/pipermail/image-sig/1999-April/000713.html



Kevin Cazabon
Product Manager, Software Applications and Systems
Gretag, Lab Systems Group

kevin.cazabon@gretag.com
303-754-0200 x289



                                                                                                                   
                    "Jason C.                                                                                      
                    Leach"               To:     image-sig@python.org                                              
                    <jleach@mail.        cc:     (bcc: Kevin Cazabon/sienna/gig)                                   
                    ocis.net>            Subject:     [Image-SIG] ImageEnhance                                     
                                                                                                                   
                    06/20/2001                                                                                     
                    12:52 PM                                                                                       
                                                                                                                   
                                                                                                                   






hi,

How do we get images out of ImageEnhance? For example if I:
enh = ImageEnhance.Contrast(im)
does it change 'im', or a copy of 'im'? What I am trying to do is save
'im' after the contrast change.

Thanks,
j.


......................
..... Jason C. Leach
..



_______________________________________________
Image-SIG maillist  -  Image-SIG@python.org
http://mail.python.org/mailman/listinfo/image-sig





From Kevin.Cazabon@gretag.com  Wed Jun 20 20:31:16 2001
From: Kevin.Cazabon@gretag.com (Kevin.Cazabon@gretag.com)
Date: Wed, 20 Jun 2001 13:31:16 -0600
Subject: [Image-SIG] ImageEnhance
Message-ID: <OFD85C0596.5FA45E73-ON87256A71.006B2771@fotoprint.com>

And of course, I replied to the wrong one of your two questions... lol...
the suggestion I had was for your other question about breaking images up
into chunks, sorry!


Kevin Cazabon




                                                                                                                   
                    "Jason C.                                                                                      
                    Leach"               To:     image-sig@python.org                                              
                    <jleach@mail.        cc:     (bcc: Kevin Cazabon/sienna/gig)                                   
                    ocis.net>            Subject:     [Image-SIG] ImageEnhance                                     
                                                                                                                   
                    06/20/2001                                                                                     
                    12:52 PM                                                                                       
                                                                                                                   
                                                                                                                   






hi,

How do we get images out of ImageEnhance? For example if I:
enh = ImageEnhance.Contrast(im)
does it change 'im', or a copy of 'im'? What I am trying to do is save
'im' after the contrast change.

Thanks,
j.


......................
..... Jason C. Leach
..



_______________________________________________
Image-SIG maillist  -  Image-SIG@python.org
http://mail.python.org/mailman/listinfo/image-sig





From fredrik@pythonware.com  Wed Jun 20 21:25:27 2001
From: fredrik@pythonware.com (Fredrik Lundh)
Date: Wed, 20 Jun 2001 22:25:27 +0200
Subject: [Image-SIG] ImageEnhance
References: <Pine.LNX.4.21.0106201149490.25358-100000@praha>
Message-ID: <009d01c0f9c7$63f91410$4ffa42d5@hagrid>

Jason C. Leach wrote:
> How do we get images out of ImageEnhance? For example if I:
> enh = ImageEnhance.Contrast(im)
> does it change 'im', or a copy of 'im'? What I am trying to do is save
> 'im' after the contrast change.

http://www.pythonware.com/library/pil/handbook/imageenhance.htm

    All enhancement classes implement a common interface,
    containing a single method:

    enhance(factor). Returns an enhanced image. The factor is a
    floating point value controlling the enhancement. Factor 1.0
    always returns a copy of the original image, lower factors
    means less colour (brightness, contrast, etc), and higher values
    more. There are no restrictions on this value.

in other words, something like this should work:

    imOut = enh.enhance(someValue)
    imOut.save(someFileName)

</F>



From jleach@mail.ocis.net  Wed Jun 20 22:12:06 2001
From: jleach@mail.ocis.net (Jason C. Leach)
Date: Wed, 20 Jun 2001 14:12:06 -0700 (PDT)
Subject: [Image-SIG] ImageEnhance
In-Reply-To: <009d01c0f9c7$63f91410$4ffa42d5@hagrid>
Message-ID: <Pine.LNX.4.21.0106201408550.25627-100000@praha>

hi,

Ok. I got that working ok.  How can I do multiple enhancments?
Right now I:

enh = ImageEnhance.Contrast(im)
enh.enhance(1.15)

To increase the contrast by 15%, but I also need to modify the brightnes
and saturation.  In the code above I pass the obj to a Contrast
constructor; I'm not sure how to pass enh to a Brightnes() constructor or
get enh back into my 'im' (original image obj).

Thanks,
j.


......................
..... Jason C. Leach
.. 

On Wed, 20 Jun 2001, Fredrik Lundh wrote:

> 
>     All enhancement classes implement a common interface,
>     containing a single method:
> 
>     enhance(factor). Returns an enhanced image. The factor is a
>     floating point value controlling the enhancement. Factor 1.0
>     always returns a copy of the original image, lower factors
>     means less colour (brightness, contrast, etc), and higher values
>     more. There are no restrictions on this value.
> 
> in other words, something like this should work:
> 
>     imOut = enh.enhance(someValue)
>     imOut.save(someFileName)
> 
> </F>
> 
> 



From jleach@mail.ocis.net  Wed Jun 20 23:36:50 2001
From: jleach@mail.ocis.net (Jason C. Leach)
Date: Wed, 20 Jun 2001 15:36:50 -0700 (PDT)
Subject: [Image-SIG] PIL can't ID a TIFF.
In-Reply-To: <009d01c0f9c7$63f91410$4ffa42d5@hagrid>
Message-ID: <Pine.LNX.4.21.0106201510300.25687-100000@praha>

hi,

I am trying to open a TIFF with PIL 1.1.2 for Python 1.5.2.  It says it
can not correctly ID the file. Can I do this by manually; if not can I do
anything to correct the problem with the file?

The error is:
Traceback (innermost last):
  File "./bright.py", line 8, in ?
    im = Image.open('tfl15_utm.tif')
  File "./PIL/Image.py", line 960, in open
    raise IOError, "cannot identify image file"
IOError: cannot identify image file



Thanks,
j.

......................
..... Jason C. Leach
.. 




From emmanuel.viennet@lipn.univ-paris13.fr  Thu Jun 21 07:16:46 2001
From: emmanuel.viennet@lipn.univ-paris13.fr (Emmanuel Viennet)
Date: Thu, 21 Jun 2001 08:16:46 +0200
Subject: [Image-SIG] EXIF information
References: <200106201846.LAA16738@svc1.netwk-innov.net>
Message-ID: <3B31914E.C34C7A2F@lipn.univ-paris13.fr>

 I use the exifdump.py script, written by 
Thierry Bousch. 
 You can find it at:
   http://topo.math.u-psud.fr/~bousch/exifdump.py

Emmanuel

Kent Tenney wrote:
> 
> Howdy,
> 
> Does anyone have pointers on how to
> read EXIF in a jpg from a digital camera
> using Python?
> 
> TIA,
> 
> Kent
>


From kent@springfed.com  Thu Jun 21 13:51:45 2001
From: kent@springfed.com (Kent Tenney)
Date: Thu, 21 Jun 2001 07:51:45 -0500
Subject: [Image-SIG] EXIF information
In-Reply-To: <3B31914E.C34C7A2F@lipn.univ-paris13.fr>
Message-ID: <200106211251.FAA18967@svc1.netwk-innov.net>

Thanks, however this script seems to read tiff files
only, I need to read jpg.


On Thu, 21 Jun 2001 08:16:46 +0200, Emmanuel Viennet wrote:
 I use the exifdump.py script, written by 
Thierry Bousch. 
 You can find it at:
   http://topo.math.u-psud.fr/~bousch/exifdump.py

Emmanuel

Kent Tenney wrote:
> 
> Howdy,
> 
> Does anyone have pointers on how to
> read EXIF in a jpg from a digital camera
> using Python?
> 
> TIA,
> 
> Kent
>

_______________________________________________
Image-SIG maillist  -  Image-SIG@python.org
http://mail.python.org/mailman/listinfo/image-sig



From kent@springfed.com  Thu Jun 21 14:14:45 2001
From: kent@springfed.com (Kent Tenney)
Date: Thu, 21 Jun 2001 08:14:45 -0500
Subject: [Image-SIG] EXIF information .. ignore my previous message
In-Reply-To: <3B31914E.C34C7A2F@lipn.univ-paris13.fr>
Message-ID: <200106211314.GAA19675@svc1.netwk-innov.net>

I had tested the script on a bad jpg file.

It indeed does what I need.

Thanks a lot.

Kent


On Thu, 21 Jun 2001 08:16:46 +0200, Emmanuel Viennet wrote:
 I use the exifdump.py script, written by 
Thierry Bousch. 
 You can find it at:
   http://topo.math.u-psud.fr/~bousch/exifdump.py

Emmanuel

Kent Tenney wrote:
> 
> Howdy,
> 
> Does anyone have pointers on how to
> read EXIF in a jpg from a digital camera
> using Python?
> 
> TIA,
> 
> Kent
>

_______________________________________________
Image-SIG maillist  -  Image-SIG@python.org
http://mail.python.org/mailman/listinfo/image-sig



From speno@isc.upenn.edu  Thu Jun 21 17:43:35 2001
From: speno@isc.upenn.edu (John P Speno)
Date: Thu, 21 Jun 2001 12:43:35 -0400
Subject: [Image-SIG] jpeg comments
Message-ID: <20010621124335.A83441@isc.upenn.edu>

I was interested in using PIL 1.1.2 for reading and writing comments from
jpeg files. That feature isn't included with PIL, but I was able to easily
enable the reading of comments with very small changes to
JpegImagePlugin.py. I choose to put the comments into the Image object as
info['comments']. Diffs available by request.

The harder part for me is using PIL to save a jpeg which preserves the
comments. Does anyone have pointers for doing this or patches to PIL which
can do this?

I thought about using SWIG to create a python module out of the
jpeg-6b/wrjpgcom.c, which reads and writes jpeg comments, but I'd rather
have it all in PIL.

Secondly, is there a list of new features planned for PIL 1.2?

Thanks.


From fred@scd.ucar.edu  Mon Jun 25 19:34:00 2001
From: fred@scd.ucar.edu (fred)
Date: Mon, 25 Jun 2001 12:34:00 -0600 (MDT)
Subject: [Image-SIG] drawing lines with Imaging-1.1.2
Message-ID: <200106251834.MAA23262@niwot.scd.ucar.edu>

When I run the following five line code:

import Image
import ImageDraw
image = Image.new("RGB",(512,256),(255,0,0))
draw  = ImageDraw.Draw(image)
draw.line([(0,128),(512,128)],fill=(0,0,0))

using Imaging-1.1.2 and Python-2.1 on an SGI running 
IRIX 6.5, I get:

  Bus error (core dumped)

Drawing text seems to work fine.

When I run the above code on a Sun running SunOS 5.7
using Imaging-1.1.1 and Python-2.0 it works fine.

Any ideas would be appreciated.

  - Fred Clare 

---------------------------------------------
Fred Clare     fred@ucar.edu     303-497-1284
Visualization & Enabling Technologies Section
NCAR Scientific Computing Division
---------------------------------------------


From edcjones@erols.com  Wed Jun 27 03:57:14 2001
From: edcjones@erols.com (Edward C. Jones)
Date: Tue, 26 Jun 2001 22:57:14 -0400
Subject: [Image-SIG] undefined symbol: Py_InitModule4
Message-ID: <3B394B8A.1090608@erols.com>

I have compiled a copy of Python 2.1 with the --with-pydebug 
switch. This is on a RedHat Linux PC. When I compile 
Imaging-1.1.2c1, everything appears to work. But  "import 
_imaging" fails with the message:

ImportError: ./_imaging.so: undefined symbol: Py_InitModule4

What's going on here?

Thanks,
Ed Jones



From edcjones@erols.com  Wed Jun 27 19:58:16 2001
From: edcjones@erols.com (Edward C. Jones)
Date: Wed, 27 Jun 2001 14:58:16 -0400
Subject: [Image-SIG] undefined symbol: Py_InitModule4: more information
Message-ID: <3B3A2CC8.8080109@erols.com>

I have 3 copies of Python on my system: the Python 1.5 that 
RedHat uses in the /usr tree, Python 2.1, also in the /usr tree, 
and Python 2.1 compiled with the "--with-pydebug" switch. This 
latter copy is in the /usr/local directory tree. If I type 
"python", I get /usr/bin/python because of a "ln -s".

I am trying to install PIL with the Python debugging stuff 
included. What am I doing wrong?

The make process in libImaging does not appear to depend on 
Python in any way.

The first try at "make -f Makefile.pre.in boot" failed because it 
couldn't find tcl8.0. A change in Setup to "8.3" fixed that.

I made the following assumption: The default settings for 
installing PIL assume Python is in the /usr/local/lib tree. 
Therefore the "--with-pydebug" version of the include files, etc. 
will be found. A second attempt at "make -f Makefile.pre.in boot" 
and "make" worked. I did "PYTHONPATH=.:./PIL ; export PYTHONPATH" 
and then "/usr/local/bin/python2.1". An attempt to "import 
_imaging" gave an error message. Here is the session:

      Python 2.1 (#8, Jun 26 2001, 16:22:28)
      [GCC 2.96 20000731 (Red Hat Linux 7.1 2.96-81)] on linux2
      Type "copyright", "credits" or "license" for more information.
      >>> import _imaging
      Traceback (most recent call last):
        File "<stdin>", line 1, in ?
      ImportError: ./_imaging.so: undefined symbol: Py_InitModule4
      [6981 refs]
      >>>
      [6981 refs]
      [2374 refs]
      Print left references? [ny] n

If I modify /usr/local/src/Imaging-1.1.2/Makefile.pre.in to say

# Installed python binary (used only by boot target)
PYTHON=         /usr/local/bin/python2.1

then I get the error:

      make[1]: Entering directory `/usr/local/src/Imaging-1.1.2'
      make[1]: *** No rule to make target `Print'.  Stop.
      make[1]: Leaving directory `/usr/local/src/Imaging-1.1.2'
      make: *** [boot] Error 2



From kevnardi@hotmail.com  Thu Jun 28 01:30:26 2001
From: kevnardi@hotmail.com (Kevin Nardi)
Date: Wed, 27 Jun 2001 17:30:26 -0700
Subject: [Image-SIG] Animated/Transparent GIFs
Message-ID: <F127utSgcOrBLSAQV6u00004179@hotmail.com>

I assume that since PIL only saves version 87a GIFs that neither animation 
nor transparency is supported.  Are there any python libraries out there 
that will encode 89a GIFs, or would I have to encode the file format myself?

Thanks.

-Kevin
_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com



From aclover@1value.com  Thu Jun 28 16:22:22 2001
From: aclover@1value.com (Clover Andrew)
Date: Thu, 28 Jun 2001 17:22:22 +0200
Subject: [Image-SIG] Re: Animated/Transparent GIFs
Message-ID: <D58B0195B58937489E89124469E57CA20D5037@EX1.1value.com>

Kevin Nardi <kevnardi@hotmail.com> wrote:

> Are there any python libraries out there that will encode 89a
> GIFs, or would I have to encode the file format myself?

It's not too difficult to monkey up an encoder given the fairly simple
GIF and LZW specifications. Totally hypothetically of course, one could
have made a slow pure-Python (PIL-independent) module for one's own use
that supported transparency and interlacing, and optimal compression
(it's surprising how many commercial encoders use unneccessarily high
initial bits-per-codeword settings). This theoretical module would be
fairly untested (as it would only have been written for one use) and
wouldn't support animation, but could probably be altered relatively
easily to do so.

Of course, having theoretically written it, the aggravating legal
situation with LZW wouldn't allow one to distribute it publically.

Just FYI, there.

--=20
Andrew Clover
Technical Consultant
1VALUE.com AG


From returnco@mandrakesoft.com  Fri Jun 29 00:45:34 2001
From: returnco@mandrakesoft.com (Mandrakesoft team)
Date: Thu, 28 Jun 2001 18:45:34 -0500
Subject: [Image-SIG] Mandrakesoft contributor database
Message-ID: <200106282345.SAA22405@mandrakesoft.mandrakesoft.com>

Hello,

We are in the process of updating our database of contributors so that we may 
reward our main contributors.

You are considered a Linux-Mandrake contributor because you are listed as 
having contributed code in the "python-imaging" packages that are included in our 
product.

If you would like to participate in our Contributors Program, please provide 
us with your mailing address by completing this form:
http://mandrake.com/cgi-bin/regcon.pl?email=image-sig@python.org&ctrl=ai

Thank you for your attention; if you are not interested in this program 
please disregard this email.

Thank you!

Alix Guillard <alix@mandrakesoft.com>