From tapenet at cox.net  Fri Apr  1 06:14:15 2005
From: tapenet at cox.net (Martin Aherron)
Date: Fri Apr  1 06:14:19 2005
Subject: [Image-SIG] PIL 1.1.5 - seek(0) on StringIO file object
Message-ID: <20050401041415.IBVV6521.lakermmtao02.cox.net@smtp.east.cox.net>

This is in the online docs for the PIL
--------------
Reading from a string 

import StringIO

im = Image.open(StringIO.StringIO(buffer))

Note that the library rewinds the file (using seek(0)) before reading the image header.
-------------


I just downloaded Version 1.1.5 and found that the library is not doing the seek before reading from the StringIO object.

Adding a seek(0) just prior to an Image.open fixed the problem.


From rsgalloway at yahoo.com  Sun Apr  3 01:06:06 2005
From: rsgalloway at yahoo.com (Ryan Galloway)
Date: Sun Apr  3 01:06:09 2005
Subject: [Image-SIG] PIL support: scaling hires images
Message-ID: <20050402230606.85309.qmail@web51510.mail.yahoo.com>


Greetings,

I am experiencing problems trying to scale down or create thumbnails from
images which are high resolution, say greater than 2,048 x resolution. Is
there a way around this?

Regards,
Ryan



		
__________________________________ 
Yahoo! Messenger 
Show us what our next emoticon should look like. Join the fun. 
http://www.advision.webevents.yahoo.com/emoticontest
From jwt at OnJapan.net  Sun Apr  3 03:06:22 2005
From: jwt at OnJapan.net (Jim Tittsler)
Date: Sun Apr  3 03:06:27 2005
Subject: [Image-SIG] PIL support: scaling hires images
In-Reply-To: <20050402230606.85309.qmail@web51510.mail.yahoo.com>
References: <20050402230606.85309.qmail@web51510.mail.yahoo.com>
Message-ID: <20050403010622.GB14439@server.onjapan.net>

On Sat, Apr 02, 2005 at 03:06:06PM -0800, Ryan Galloway wrote:
> I am experiencing problems trying to scale down or create thumbnails from
> images which are high resolution, say greater than 2,048 x resolution. Is
> there a way around this?

What sort of problems?  (What format source image?)

FWIW, I'm able to scale large TIFF images without problem.  (It
does take about 15 seconds on my rather old machine.)
>>> import Image
>>> im = Image.open("/home/jwt/sat1/temp200101.tif")
>>> print im.format, im.size, im.mode
TIFF (8192, 4096) RGB
>>> im.thumbnail((128,64))
>>> im.save("temp200101.jpg", "JPEG")


-- 
Jim Tittsler             http://www.OnJapan.net/      GPG: 0x01159DB6
Python Starship          http://Starship.Python.net/
Ringo MUG Tokyo          http://www.ringo.net/rss.html

From postmaster at quost.com  Tue Apr  5 17:18:16 2005
From: postmaster at quost.com (dnsrequest.com PostMaster)
Date: Tue Apr  5 17:18:13 2005
Subject: [Image-SIG] Error sending message [1112714293573.1444.pegasus] from
	[dnsrequest.com]
Message-ID: <20050405151811.E457F1E4004@bag.python.org>

[<00>] XMail bounce: Rcpt=[dan@steelkey.com.au];Error=[550 Mailbox unavailable <dan@steelkey.com.au>]


[<01>] Error sending message [1112714293573.1444.pegasus] from [dnsrequest.com].

ID:        <S63BDC>
Mail From: <image-sig@python.org>
Rcpt To:   <dan@steelkey.com.au>
Server:    <mail3.dnsrequest.com> [12.96.165.198]


[<02>] The reason of the delivery failure was:

550 Mailbox unavailable <dan@steelkey.com.au>


[<05>] Here is listed the initial part of the message:

Received: from python.org (84.204.14.21:2348)
	by mail2.dnsrequest.com with [XMail 1.17 (Win32/Ix86) ESMTP Server]
	id <S63BDC> for <dan@steelkey.com.au> from <image-sig@python.org>;
	Tue, 5 Apr 2005 15:17:47 -0000
From: image-sig@python.org
To: dan@steelkey.com.au
Subject: hi
Date: Thu, 5 May 2005 19:21:03 +0400
MIME-Version: 1.0
Content-Type: multipart/mixed;
	boundary="----=_NextPart_000_0010_58B76FF2.7B88B5C9"
X-Priority: 3
X-MSMail-Priority: Normal

From meesters at uni-mainz.de  Wed Apr  6 13:02:36 2005
From: meesters at uni-mainz.de (Christian Meesters)
Date: Wed Apr  6 13:02:40 2005
Subject: [Image-SIG] ImageFont on OSX
Message-ID: <00a23da8fe023a09eb1debf6b385b2f4@uni-mainz.de>

Hi

I'm using PIL 1.1.4 on OSX (still standard Python 2.3) and would like 
to paste text into images. All I get is some rather small standard font 
text written into my images, because calling ImageFont.load() fails, 
since I cannot specify any font.
Could somebody tell me which are valid fonts and where to obtain such 
font files? The passage in the handbook ("PIL uses it's own font file 
format to store bitmap fonts. You can use the pilfont utility to 
convert BDF and PCF font descriptors (X window font formats) to this 
format. ") tells me nothing: Where to get those font descriptors? And 
all I have are standard font files of Apple ('Font Suitcase').

Any ideas?

Thanks a lot in advance,
Christian

From dvarrazzo at virgilio.it  Wed Apr  6 17:44:23 2005
From: dvarrazzo at virgilio.it (dvarrazzo@virgilio.it)
Date: Wed Apr  6 17:44:26 2005
Subject: [Image-SIG] Help with TIFF images
Message-ID: <4200066600125C5F@ims1d.cp.tin.it>

Hi everybody,

i want to open a 16 bit grayscale TIFF file with photometric interpretation
0 (white is 0). Such files are a common densitometers product.

The corrispondenting format key is (0, 1, 1, (16,), ()) in PIL 1.1.5 (it
used to be (0, 1, (16,), ()) in previous releases. This key is missing from
TIFF OPEN_INFO dict, resulting in an unsupported format.

I addd the same value as for the key (1, 1, 1, (16,), ()), i.e. ("I;16",
"I;16"). The file is open, of course, in negative form. Anyway this is not
a big problem, because i don't have to display the image: i perform image
processing in numarray, passing the data array to it with a fromstring()/tostring()
pair.

What is the correct value to read such images?

When i come to save the image after editing, i pass it back to PIL via fromstring()
with mode="I;16". This way the file is stored "black is 0", so i hack the
interpretation bit "by hand" (some control tools only open this very combination).

Can i set mode so that the file is stored "white is 0"?

Thank you in advance

Daniele



From kevin at cazabon.com  Wed Apr  6 23:15:20 2005
From: kevin at cazabon.com (kevin@cazabon.com)
Date: Wed Apr  6 23:15:55 2005
Subject: [Image-SIG] Saturation function?
References: <4200066600125C5F@ims1d.cp.tin.it>
Message-ID: <060a01c53aed$bd5ac5b0$640aa8c0@duallie>

Does anyone have a simple saturation adjustment function for PIL?

I haven't been able to find anything through Google, other than a VB script
for converting between HSL and RGB.  That could be hacked into a saturation
function, but would be VERY slow...

ANy suggestions would be most welcome, thanks!

Kevin.


From meyer at mesw.de  Wed Apr  6 23:44:01 2005
From: meyer at mesw.de (meyer@mesw.de)
Date: Wed Apr  6 23:46:05 2005
Subject: =?iso-8859-1?Q?Re:_[Image-SIG]_Saturation_function??=
Message-ID: <27414465$1112823759425457cff18a26.10335469@config3.schlund.de>


Can't you use the ImageEnhance module?

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

I'm no expert in these things, but I believe that the "Color" value
should be what is generally called "Saturation".


Markus

kevin@cazabon.com schrieb am 06.04.2005, 23:15:20:
> Does anyone have a simple saturation adjustment function for PIL?
> 
> I haven't been able to find anything through Google, other than a VB script
> for converting between HSL and RGB.  That could be hacked into a saturation
> function, but would be VERY slow...
> 
> ANy suggestions would be most welcome, thanks!
> 
> Kevin.
> 
> 
> _______________________________________________
> Image-SIG maillist  -  Image-SIG@python.org
> http://mail.python.org/mailman/listinfo/image-sig
From kevin at cazabon.com  Thu Apr  7 00:00:57 2005
From: kevin at cazabon.com (kevin@cazabon.com)
Date: Thu Apr  7 00:01:34 2005
Subject: [Image-SIG] Saturation function?
References: <27414465$1112823759425457cff18a26.10335469@config3.schlund.de>
Message-ID: <001901c53af4$1d0d0170$640aa8c0@duallie>

Thanks Markus - I skipped over that one due to the description... but you're
correct that it's what I want.

(the description says it's for adjusting the color balance of an image...
not really an accurate description)

Kevin.
----- Original Message ----- 
From: <meyer@mesw.de>
To: <kevin@cazabon.com>
Cc: <image-sig@python.org>
Sent: Wednesday, April 06, 2005 11:44 PM
Subject: Re: [Image-SIG] Saturation function?


>
> Can't you use the ImageEnhance module?
>
> http://www.pythonware.com/library/pil/handbook/imageenhance.htm
>
> I'm no expert in these things, but I believe that the "Color" value
> should be what is generally called "Saturation".
>
>
> Markus
>
> kevin@cazabon.com schrieb am 06.04.2005, 23:15:20:
> > Does anyone have a simple saturation adjustment function for PIL?
> >
> > I haven't been able to find anything through Google, other than a VB
script
> > for converting between HSL and RGB.  That could be hacked into a
saturation
> > function, but would be VERY slow...
> >
> > ANy suggestions would be most welcome, thanks!
> >
> > Kevin.
> >
> >
> > _______________________________________________
> > Image-SIG maillist  -  Image-SIG@python.org
> > http://mail.python.org/mailman/listinfo/image-sig
>
>
>


From gwidion at mpc.com.br  Thu Apr  7 02:11:08 2005
From: gwidion at mpc.com.br (Joao S. O. Bueno Calligaris)
Date: Thu Apr  7 02:11:15 2005
Subject: [Image-SIG] Saturation function?
In-Reply-To: <27414465$1112823759425457cff18a26.10335469@config3.schlund.de>
References: <27414465$1112823759425457cff18a26.10335469@config3.schlund.de>
Message-ID: <200504062111.08848.gwidion@mpc.com.br>

On Wednesday 06 April 2005 18:44, meyer@mesw.de wrote:
> Can't you use the ImageEnhance module?
>
> http://www.pythonware.com/library/pil/handbook/imageenhance.htm
>
> I'm no expert in these things, but I believe that the "Color" value
> should be what is generally called "Saturation".
>
>

If that doesn't work, RGB <-> HSV conversion is more or less simple, 
and I had worked it in python for a GIMP-Plugin. 
My conversion is for working with palettes so it is a slow 
implementation - you will be better if the above module works. 
Otherwise, I can e -mail you my script. - a NumArray version of it 
should be usable for raster data.

	JS
	-><-




> Markus
>
> kevin@cazabon.com schrieb am 06.04.2005, 23:15:20:
> > Does anyone have a simple saturation adjustment function for PIL?
> >
> > I haven't been able to find anything through Google, other than a
> > VB script for converting between HSL and RGB.  That could be
> > hacked into a saturation function, but would be VERY slow...
> >
> > ANy suggestions would be most welcome, thanks!
> >
> > Kevin.
> >
> >
> > _______________________________________________
> > Image-SIG maillist  -  Image-SIG@python.org
> > http://mail.python.org/mailman/listinfo/image-sig
>
> _______________________________________________
> Image-SIG maillist  -  Image-SIG@python.org
> http://mail.python.org/mailman/listinfo/image-sig
From pascor22234 at yahoo.com  Fri Apr  8 17:03:08 2005
From: pascor22234 at yahoo.com (Ray Pasco)
Date: Fri Apr  8 17:03:11 2005
Subject: [Image-SIG] Saving to JPEG in CMYK
Message-ID: <20050408150308.69551.qmail@web52703.mail.yahoo.com>

When saving an RGB image to JPEG CMYK results in a
totally black image. I'm using all default parameters.
Is there some other parameters that are necessary ? 

BTW, saving to TIFF CMYK works fine.

Thanks,
Ray Pasco


		
__________________________________ 
Do you Yahoo!? 
Read only the mail you want - Yahoo! Mail SpamGuard. 
http://promotions.yahoo.com/new_mail 
From pascor at hotpop.com  Fri Apr  1 18:59:52 2005
From: pascor at hotpop.com (Ray Pasco)
Date: Sat Apr  9 10:54:27 2005
Subject: [Image-SIG] Faster way to examine alpha ?
Message-ID: <424D7E08.1080705@hotpop.com>

I have a routine that examines the alpha values of an RGBA image to 
determine
whether whether there are any "active"alpha (value < 255), and
whether the any active alpha are single-valued and if only a single color
is associted with this alpha. Each pixel must be examined one-by-one
which is rather slow.

Is there way to examine pixel values akin to the .point(lambda) method ?
I just want to examine, not change the valuse.

Thanks,
Ray Pasco

From pascor at hotpop.com  Thu Apr  7 16:31:58 2005
From: pascor at hotpop.com (Ray Pasco)
Date: Sat Apr  9 10:54:28 2005
Subject: [Image-SIG] HSL adjustment, and others
Message-ID: <4255445E.1010707@hotpop.com>

I've been playing around with just this. This program converts to/from 
several different colorspaces
and changes the alternate colorspace's "ramp" slope and offset. The 
program only does linear adjustments.
I've been trying to merge PIL and wxPython and since have come up with a 
better way to display the results,
but I think you want only the processing algorithms, anyway. I don't 
really can't swear to the accuracy
of the colorspace conversions other than they convert from  RGB and then 
back again perfectly.  I was able
to correct several apparent bugs in the algorithms. I don't know where 
to find a "definitive" source for the algorithms,
much less an explanation for how they work.

I'd appreciate any comments/suggestions you may have.  I intend to 
rewrite the app's GUI and
would also appreciate finding a more useful mapping function than a 
linear-ramp to linear-ramp function.

Ray Pasco
-------------- next part --------------

import wx

#----------------------------

def Pil2Bitmap (imPil):
    wxImage = Pil2WxImage (imPil)
    bitmap = WxImage2Bitmap (wxImage)
    return bitmap
#end def

def Pil2WxImage (imPil):
    wxImage = wx.EmptyImage (imPil.size[0], imPil.size[1])
    wxImage.SetData (imPil.convert('RGB').tostring())
    return wxImage
#end def

def WxImage2Bitmap (image):
    bitmap = image.ConvertToBitmap()
    return bitmap
#end def

#--------

def Bitmap2Pil (bitmap):
    WxImage = Bitmap2WxImage (bitmap)
    imPil = WxImage2Pil (WxImage)
    return imPil
#end def

def Bitmap2WxImage (bitmap):
    wximage = wx.ImageFromBitmap (bitmap)
    return wximage
#end def

def WxImage2Pil (wxImage):
    imPil = Image.new ('RGB', (wxImage.GetWidth(), wxImage.GetHeight()))
    imPil.fromstring (wxImage.GetData())
    return imPil
#end def

#----------------------------

-------------- next part --------------
##  Colorspace conversions library

import sys


def Rgb2Cmyk (rgbtuple):

    def Rgb2Cmy (rgbtuple):
    
        # rgb tuple values are in [0 ... 255]
        r, g, b = rgbtuple[0], rgbtuple[1], rgbtuple[2],
    
        c = 1 - (r / 255.0)
        m = 1 - (g / 255.0)
        y = 1 - (b / 255.0)
    
        return c, m, y
    
    #end def Rgb2Cmy

    def Cmy2Cmyk (cmytuple):
    
        # cmy tuple values are in [0.0 ... 1.0]
        c, m, y = cmytuple[0], cmytuple[1], cmytuple[2]
    
        if (c == 1.0) and (m == 1.0) and (y == 1.0):
            K = 1.0         # special dispensation for absolute black to avoid division-by-zero
    
            C = 0.0
            M = 0.0
            Y = 0.0
    
        else:
            K = min (1.0, c, m, y)
    
            C = (c - K) / (1.0 - K)
            M = (m - K) / (1.0 - K)
            Y = (y - K) / (1.0 - K)
        #end if
    
        return C, M, Y, K
    
    #end def Cmy2Cmyk
    
    #------------------------

    # rgb tuple values are in [0...255]
    cmytuple   = Rgb2Cmy  (rgbtuple)
    C, M, Y, K = Cmy2Cmyk (cmytuple)

    return C, M, Y, K

#end def Rgb2Cmyk

def Cmyk2Rgb (cmyktuple):

    def Cmyk2Cmy (cmyktuple):
    
        # CMYK tuple values are in [0.0 ... 1.0]
        C, M, Y, K = cmyktuple[0], cmyktuple[1], cmyktuple[2], cmyktuple[3]
    
        c = ((C * (1.0 - K)) + K)
        m = ((M * (1.0 - K)) + K)
        y = ((Y * (1.0 - K)) + K)
    
        return c, m, y
    
    #end def Cmyk2Cmy
    
    def Cmy2Rgb (cmytuple):
    
        # cmy tuple values are in [0.0 .. 1.0]
        c, m, y = cmytuple[0], cmytuple[1], cmytuple[2]
    
        r = int (((1.0 - c) * 255.0) + 0.5)
        g = int (((1.0 - m) * 255.0) + 0.5)
        b = int (((1.0 - y) * 255.0) + 0.5)
    
        return r, g, b
    
    #end def Cmy2Rgb

    #------------------------

    C, M, Y, K = cmyktuple

    # Make sure that CMYK values are be in [0.0 ... 1.0]
    C = min (max (0.0, C), 1.0)
    M = min (max (0.0, M), 1.0)
    Y = min (max (0.0, Y), 1.0)
    K = min (max (0.0, K), 1.0)

    cmytuple = Cmyk2Cmy (cmyktuple)
    r, g, b  = Cmy2Rgb  (cmytuple)

    return r, g, b

#end def Cmyk2Rgb

def Rgb2Hsl (rgbtuple):

    # H, S, L will be in [0.0 ... 1.0]

    r, g, b = rgbtuple
    rScaled = r/255.0
    gScaled = g/255.0
    bScaled = b/255.0

    rgbMin = min (rScaled, gScaled, bScaled)    # Min RGB value
    rgbMax = max (rScaled, gScaled, bScaled)    # Max RGB value
    deltaRgb = rgbMax - rgbMin                  # Delta RGB value

    L = (rgbMax + rgbMin) / 2.0

    if (deltaRgb == 0.0):   # This is a gray, no chroma.
       H = 0.0
       S = 0.0              # Done !
    
    else:                   # Chromatic data...
        if (L < 0.5):
            S = deltaRgb / (rgbMax + rgbMin)
        else:
            S = deltaRgb / (2.0 - rgbMax - rgbMin)
        #end if
    
        deltaR = (((rgbMax - rScaled)/6.0) + (deltaRgb/2.0)) / deltaRgb
        deltaG = (((rgbMax - gScaled)/6.0) + (deltaRgb/2.0)) / deltaRgb
        deltaB = (((rgbMax - bScaled)/6.0) + (deltaRgb/2.0)) / deltaRgb
    
        if   (rScaled == rgbMax): 
            H = deltaB - deltaG

        elif (gScaled == rgbMax): 
            H = (1.0/3.0) + deltaR - deltaB

        elif (bScaled == rgbMax): 
            H = (2.0/3.0) + deltaG - deltaR

        #end if

        H = H % 1.0

    #end if

    return (H, S, L)

#end def Rgb2Hsl

def Hsl2Rgb (hsltuple):

    def HslHue2Rgb (v1, v2, vH):
       if (vH < 0.0):    vH += 1.0
       if (vH > 1.0):    vH -= 1.0
       if ((6.0 * vH) < 1.0):    return (v1 + (v2 - v1) * 6.0 * vH)
       if ((2.0 * vH) < 1.0):    return (v2)
       if ((3.0 * vH) < 2.0):    return (v1 + (v2 - v1) * ((2.0/3.0) - vH) * 6.0)

       return v1

    #end def HslHue2Rgb

    #------------------------
 
    # H, S, L in [0.0 .. 1.0]
    H, S, L = hsltuple

    if (S == 0.0):          # RGB grayscale

       r = L * 255.0        # R, G, B in [0 .. 255]
       g = L * 255.0
       b = L * 255.0

    else:

        if (L < 0.5): 
            var_2 = L * (1.0 + S)
        else:           
            var_2 = (L + S) - (S * L)
        #end if
    
        var_1 = (2.0 * L) - var_2
    
        r = 255 * HslHue2Rgb (var_1, var_2, H + (1.0/3.0)) 
        g = 255 * HslHue2Rgb (var_1, var_2, H            )
        b = 255 * HslHue2Rgb (var_1, var_2, H - (1.0/3.0))

    #end if

    r = int (r + 0.5)
    g = int (g + 0.5)
    b = int (b + 0.5)

    return (r, g, b)

#end def Hsl2Rgb

def Rgb2Hsv (rgbtuple):

    # H, S, V will be in [0.0 ... 1.0]

    r, g, b = rgbtuple

    r = r / 255.0
    g = g / 255.0
    b = b / 255.0

    minRgb = min (r, g, b)
    maxRgb = max (r, g, b)
    deltaRgb = maxRgb - minRgb

    v = maxRgb

    if (deltaRgb == 0):     # This is a gray, no chroma...
       h = 0.0
       s = 0.0

    else:                   # Chromatic data
        s = deltaRgb / maxRgb

        del_R = (((maxRgb - r) / 6.0) + (deltaRgb / 2.0)) / deltaRgb
        del_G = (((maxRgb - g) / 6.0) + (deltaRgb / 2.0)) / deltaRgb
        del_B = (((maxRgb - b) / 6.0) + (deltaRgb / 2.0)) / deltaRgb

        if   (r == maxRgb):
            h = del_B - del_G

        elif (g == maxRgb):
            h = (1.0 / 3.0) + del_R - del_B

        elif (b == maxRgb):
            h = (2.0 / 3.0) + del_G - del_R
        #end if

        h = h % 1.0
    #end if

    return (h, s, v)

#end def Rgb2Hsv

def Hsv2Rgb (hsvtuple):

    h, s, v = hsvtuple

    if (s == 0.0):
       r = v * 255.0
       g = v * 255.0
       b = v * 255.0

    else:
        h = h * 6.0
        I = int (h)          # floor function
        F = h - I
        P = v * (1.0 - s)
        Q = v * (1.0 - s * F)
        T = v * (1.0 - s * (1.0 - F))

        if   (I == 4):
            r = T
            g = P
            b = v

        elif (I == 5):
            r = v
            g = P
            b = Q

        elif (I == 0):
            r = v
            g = T
            b = P

        elif (I == 1):
            r = Q
            g = v
            b = P

        elif (I == 2):
            r = P
            g = v
            b = T

        elif (I == 3):
            r = P
            g = Q
            b = v
        #end if

        r = int ((r * 255.0) + 0.5)
        g = int ((g * 255.0) + 0.5)
        b = int ((b * 255.0) + 0.5)
    #end if

    return (r, g, b)

#end def Hsv2Rgb

#--------------------------------------

# Stand-alone test for colorspaces.py

if __name__ == '__main__':

    import Image

    if len(sys.argv) < 4:
        print
        print 'Colorspaces:   Values for R, G and B must be given, each in the range [0..255]'
        print
        sys.exit (1)
    #end if

    r = eval (sys.argv [1])
    g = eval (sys.argv [2])
    b = eval (sys.argv [3])
    rgbtuple = (r, g, b)

    #---------  HSL  --------------------------------------

    hsltuple = Rgb2Hsl (rgbtuple)

    h, s, l = hsltuple
    print
    print 'Colorspaces:    RGB (%3d, %3d, %3d)  ==>  HSL (%1.3f, %1.3f, %1.3f)' % (r, g, b, h, s, l)

    print
    rPrime, gPrime, bPrime = Hsl2Rgb (hsltuple)

    print 'Colorspaces:    HSL (%1.3f, %1.3f, %1.3f)  ==>  RGB (%3d, %3d, %3d)' % (h, s, l, rPrime, gPrime, bPrime)
    print

    #---------  CMYK  -------------------------------------

    cmyktuple = Rgb2Cmyk (rgbtuple)

    c, y, m, k = cmyktuple
    print
    print 'Colorspaces:    RGB (%3d, %3d, %3d)  ==>  CMYK (%1.3f, %1.3f, %1.3f, %1.3f)' % (r, g, b, c, y, m, k)

    print
    rPrime, gPrime, bPrime = Cmyk2Rgb (cmyktuple)

    print 'Colorspaces:    CMYK (%1.3f, %1.3f, %1.3f, %1.3f)  ==>  RGB (%3d, %3d, %3d)' % (c, m, y, k, rPrime, gPrime, bPrime)
    print

    #---------  HSV  --------------------------------------

    hsvtuple = Rgb2Hsv (rgbtuple)

    h, s, v = hsvtuple
    print
    print 'Colorspaces:    RGB (%3d, %3d, %3d)  ==>  HSV (%1.3f, %1.3f, %1.3f)' % (r, g, b, h, s, v)

    print
    rPrime, gPrime, bPrime = Hsv2Rgb (hsvtuple)

    print 'Colorspaces:    HSL (%1.3f, %1.3f, %1.3f)  ==>  RGB (%3d, %3d, %3d)' % (h, s, v, rPrime, gPrime, bPrime)
    print

#end if

-------------- next part --------------

import wx
import Image
import os, sys

from colorspaces import *           # pil <--> hsl, hsv, cmyk
from imageconverts import *         # pilIm --> bitmap

#----------------------------

import inspect, os
def pressenter (prompt='\npress return...'):   raw_input (prompt)
def rangelen(listobj):  return xrange(len(listobj))
def showStack():
    for stack in inspect.stack():    print stack[2:4]
#end def
exit = os._exit
def delFileExist (filename):
    if os.path.exists (filename):    os.remove (filename)
#end def

#----------------------------

global filpathname;    filepathname = None      # create the global instance
global thumbnailXsize;    thumbnailXsize = 0    # instantiate
global thumbnailYsize;    thumbnailYsize = 0

global colorspaces;    colorspaces = ['HSL-', 'HSV-', 'CMYK']
global colorspaceStr;    colorspaceStr = colorspaces [0]
global tctrlVar1;    tctrlVar1 = None   # instantiate
global tctrlVar2;    tctrlVar2 = None
global tctrlVar3;    tctrlVar3 = None
global tctrlVar4;    tctrlVar4 = None

# These default values cause no modifications to be made to the transformed image's vector space values.
#   Thus, when the inverse transform is applied, the resulting image will be identical 
#   to the original image.
global x1Var1;    x1Var1 = 0.0      # initial Y=0.0 intercept (the X axis intercept)
global x2Var1;    x2Var1 = 1.0      # initial Y=1.0 intercept
global x1Var2;    x1Var2 = 0.0
global x2Var2;    x2Var2 = 1.0
global x1Var3;    x1Var3 = 0.0
global x2Var3;    x2Var3 = 1.0
global x1Var4;    x1Var4 = 0.0
global x2Var4;    x2Var4 = 1.0

class AppFrame (wx.Frame):

    def __init__ (self, parent, id, title, position, size):

        global filepathname
        self.parentFrame = self
        self.mainframePsn = position
        self.mainframeSize = size

        wx.Frame.__init__ (self, parent, id, title, position, size)

        # Create the File menu
        self.CreateStatusBar()      # A StatusBar in the bottom of the window

        # Setting up the menu.
        filemenu= wx.Menu()

        idOpen = wx.NewId()
        filemenu.Append (idOpen, "&Open"," Open a Text File")
        wx.EVT_MENU (self, idOpen, self.OnOpen)

        filemenu.AppendSeparator()

        idAbout = wx.NewId()
        filemenu.Append (idAbout, "&About"," Information about this program")
        wx.EVT_MENU (self, idAbout, self.OnAbout)

        filemenu.AppendSeparator()

        idExit = wx.NewId()
        filemenu.Append (idExit,"E&xit"," Terminate the program")
        wx.EVT_MENU (self, idExit, self.OnExit)

        # Creating the menubar.
        menuBar = wx.MenuBar()
        menuBar.Append (filemenu,"&File") # Adding the "filemenu" to the MenuBar
        self.SetMenuBar (menuBar)  # Adding the MenuBar to the Frame content.

        #----------

        AppPanel (self, -1, position, size)

        self.Show (True)            # show self frame

    #end def __init__

    def OnOpen (self, event):       # Open a file

        global filepathname

        # 'wx.CHANGE_DIR' is necessary to return a full pathname in Win98
        dlg = wx.FileDialog (self, "Choose a file",
                             style=wx.OPEN|wx.FILE_MUST_EXIST|wx.CHANGE_DIR, wildcard='*.*')

        if dlg.ShowModal() == wx.ID_OK:
            filename = dlg.GetFilename()
            dirname = dlg.GetDirectory()
            filepathname = os.path.join (dirname, filename)
        #end if

        if (type (filepathname) == type ('abc')) or (type (filepathname) == type (u'abc')):
            try:
                self.referenceFrame.Destroy()
            except:
                pass
            #end if

            self.ReferenceFrame()
        #end if

        dlg.Destroy()
    #end def OnOpen

    def ReferenceFrame (self):

        global filepathname

        referenceframeXpsn = self.mainframePsn[0] + self.mainframeSize[0]
        referenceframepsn = (referenceframeXpsn, self.mainframePsn[1])
        referenceframesize = (300, 300)         # is also the transformed image frame's size
        self.referenceFrame = wx.Frame (self.parentFrame, -1, "Original Image ", referenceframepsn, referenceframesize)

        # Create the reference frame's panel
        referencePanel = wx.Panel (self.referenceFrame, -1)
        referencePanel.SetBackgroundColour ('#C9E3E5')
    
        # Create the reference bitmap
        staticBitmap = wx.StaticBitmap (referencePanel, -1, wx.EmptyBitmap(5, 5))
    
        # Get and resize the original image to fit in the reference panel and store it to a bitmap
        bitmap = GetBitmapFromFile (filepathname, referenceframesize)
        staticBitmap.SetBitmap (bitmap)
    
        boxsizer = wx.BoxSizer (wx.VERTICAL)
        expand = False
        boxsizer.Add (staticBitmap, expand)

        referencePanel.SetAutoLayout (True)
        referencePanel.SetSizer (boxsizer)

        # Show the reference frame
        self.referenceFrame.Show (True)

    #end def ReferenceFrame

    def OnAbout (self, event):      # Create a message dialog box
        dlg = wx.MessageDialog (self, 'A colorspace manipulation\nprogram in wxPython', 
                                ' About Colorspace Transform', wx.OK)

        dlg.ShowModal()   # Shows it
        dlg.Destroy()     # finally destroy it when finished.
    #end def

    def OnExit (self, event):
        self.Close (True)           # Close the application.
    #end def

#end class AppFrame

class AppPanel (wx.Panel):

    def __init__ (self, parentframe, id, mainframePsn, mainframeSize):

        global colorspaces
        global colorspaceStr
        global filepathname
        global x1Var1, x2Var1

        self.parentFrame = parentframe
        self.mainframePsn = mainframePsn
        self.mainframeSize = mainframeSize

        wx.Panel.__init__ (self, parentframe, id)
        self.SetBackgroundColour ('#C9E3E5')

        boxsizerV = wx.BoxSizer (wx.VERTICAL)   # all panel widgets go in here
        expand = False
        hgt = 40
        padding = 0

        #--------------------

        rbId = wx.NewId()
        numRbRows = 1
        # Display the colorspace strings without any trailing dash characters;
        #   Create a new string list for this purpose.
        modcolorspaces = []
        for cspaceStr in colorspaces:
            if cspaceStr[-1] == '-':
                modcolorspaces.append (cspaceStr[:-1])      # up to the dash char
            else:
                modcolorspaces.append (cspaceStr[:])        # the wholeoriginal string
            #end if
        #end for
        radbox = wx.RadioBox (self, rbId, "Color Space", wx.DefaultPosition, wx.DefaultSize,
                              modcolorspaces, numRbRows, wx.RA_SPECIFY_ROWS)
        radbox.SetBackgroundColour (self.GetBackgroundColour())
        radbox.SetBackgroundColour (wx.WHITE)
        radbox.SetToolTip (wx.ToolTip ('Select a colorspace model:'))
        radbox.SetLabel ('Colorspace Model Selection')
        wx.EVT_RADIOBOX (self, rbId, self.EvtRadioBox)
        # Apparently the first radiobutton is always selected by default,
        #   so, adjust the program's setting to reflect this.
        colorspaceStr = colorspaces [0]

        boxsizerX1X2var1 = self.VarControlsVar1 (self, colorspaceStr[0].upper())
        boxsizerX1X2var2 = self.VarControlsVar2 (self, colorspaceStr[1].upper())
        boxsizerX1X2var3 = self.VarControlsVar3 (self, colorspaceStr[2].upper())
        boxsizerX1X2var4 = self.VarControlsVar4 (self, colorspaceStr[3].upper())

        #--------------------

        boxsizerV.Add (radbox, expand, wx.CENTER|wx.ALL, 10)
        bordersize = 0
        boxsizerV.Add (boxsizerX1X2var1, expand, wx.CENTER|wx.ALL, bordersize)
        boxsizerV.Add (boxsizerX1X2var2, expand, wx.CENTER|wx.ALL, bordersize)
        boxsizerV.Add (boxsizerX1X2var3, expand, wx.CENTER|wx.ALL, bordersize)
        boxsizerV.Add (boxsizerX1X2var4, expand, wx.CENTER|wx.ALL, bordersize)

        xformbtn = wx.Button (self, -1, "Show Transformed Image")
        wx.EVT_BUTTON (self, xformbtn.GetId(), self.xformBtnHandler)
        boxsizerV.Add (xformbtn, expand, wx.CENTER|wx.ALL, 10)    # a 10-pixel border all around

        self.SetAutoLayout (True)
        self.SetSizer (boxsizerV)
        self.Show (True)

    #end def __init__

    def EvtRadioBox (self, event):

        global tctrlVar1
        global tctrlVar2
        global tctrlVar3
        global tctrlVar4

        colorspaceIndex = event.GetInt()
        colorspaceStr = colorspaces [colorspaceIndex]

        # Write the colorspace string chars into their text controls.

        tctrlVar1.Clear()
        tctrlVar1.SetInsertionPoint (0)
        tctrlVar1.AppendText (colorspaceStr[0].upper())

        tctrlVar2.Clear()
        tctrlVar2.SetInsertionPoint (0)
        tctrlVar2.AppendText (colorspaceStr[1].upper())

        tctrlVar3.Clear()
        tctrlVar3.SetInsertionPoint (0)
        tctrlVar3.AppendText (colorspaceStr[2].upper())

        tctrlVar4.Clear()
        tctrlVar4.SetInsertionPoint (0)
        tctrlVar4.AppendText (colorspaceStr[3].upper())

     #end def

    def VarControlsVar1 (self, parentPanel, varLabelChar):

        global tctrlVar1

        def OnSpinX1 (event):
            global x1Var1

            x1Var1 = (event.GetPosition() - 200) / 200.0     # raw spinbutton value
            self.textX1Var1.SetValue ('%1.3f' % (x1Var1))
        #end def

        def OnSpinX2 (event):
            global x2Var1

            x2Var1 = (event.GetPosition() - 200) / 200.0     # raw spinbutton value
            self.textX2Var1.SetValue ('%1.3f' % (x2Var1))
       #end def

        def ResetX1BtnHandler (event):
            global x1Var1

            self.spinX1.SetValue (200)                        # 0.000
            x1Var1 = 0.00
            self.textX1Var1.SetValue ('%1.3f' % (x1Var1))
        #end def ResetX1BtnHandler

        def ResetX2BtnHandler (event):
            global x2Var1

            self.spinX2.SetValue (400)                        # 0.000
            x2Var1 = 1.00
            self.textX2Var1.SetValue ('%1.3f' % (x2Var1))
        #end def ResetX1BtnHandler

        #--------------------

        expand = False
        hgt = 40
        padding = 0

        boxsizerX1X2 = wx.BoxSizer (wx.HORIZONTAL)

        boxsizerX1V = wx.BoxSizer (wx.VERTICAL)

        labelX1 = wx.StaticText (self, -1, "X1:  Y=0.0 Intercept:", wx.DefaultPosition)
        boxsizerX1V.Add (labelX1, expand, wx.CENTER|wx.ALL, padding)

        boxsizerX1 = wx.BoxSizer (wx.HORIZONTAL)
        self.textX1Var1 = wx.TextCtrl (self, -1, "0.000", wx.DefaultPosition, wx.Size(60, -1),
                                   style=wx.TE_READONLY)
        boxsizerX1.Add (self.textX1Var1, expand, wx.CENTER|wx.ALL, padding)
        self.spinX1 = wx.SpinButton (self, 20, wx.Point(0, 0), wx.Size(hgt, hgt), wx.SP_HORIZONTAL)
        self.spinX1.SetRange (0, 600)                   # -1.000 .. 2.000
        self.spinX1.SetValue (200)                      # x1 = 0.000
        wx.EVT_SPIN (self.spinX1, 20, OnSpinX1)
        boxsizerX1.Add (self.spinX1, expand, 0, padding)

        boxsizerX1V.Add (boxsizerX1, expand, 0, padding)

        resetX1btn = wx.Button (self, -1, "Reset to 0.000")
        boxsizerX1V.Add (resetX1btn, expand, 0, padding)
        wx.EVT_BUTTON (self, resetX1btn.GetId(), ResetX1BtnHandler)

        boxsizerX2V = wx.BoxSizer (wx.VERTICAL)

        labelX2 = wx.StaticText (self, -1, "X2:  Y=1.0 Intercept:", wx.DefaultPosition)
        boxsizerX2V.Add (labelX2, expand, wx.CENTER|wx.ALL, padding)

        boxsizerX2 = wx.BoxSizer (wx.HORIZONTAL)
        self.textX2Var1 = wx.TextCtrl (self, -1, "1.000", wx.DefaultPosition, wx.Size(60, -1),
                                   style=wx.TE_READONLY)
        boxsizerX2.Add (self.textX2Var1, expand, wx.CENTER|wx.ALL, padding)
        self.spinX2 = wx.SpinButton (self, 20, wx.DefaultPosition, wx.Size(hgt, hgt), wx.SP_HORIZONTAL)
        self.spinX2.SetRange (0, 600)                   # -1.000 .. 2.000
        self.spinX2.SetValue (400)                      # x2 = 1.000
        wx.EVT_SPIN (self.spinX2, 20, OnSpinX2)
        boxsizerX2.Add (self.spinX2, expand, wx.CENTER|wx.ALL, padding)

        boxsizerX2V.Add (boxsizerX2, expand, 0, padding)

        resetX2btn = wx.Button (self, -1, "Reset to 1.000")
        boxsizerX2V.Add (resetX2btn, expand, 0, padding)
        wx.EVT_BUTTON (self, resetX2btn.GetId(), ResetX2BtnHandler)

        tctrlVar1 = wx.TextCtrl (self, -1, varLabelChar.upper(), wx.DefaultPosition, wx.Size(20, -1), 
                                 style=wx.TE_READONLY)
        boxsizerX1X2.Add (tctrlVar1, expand, wx.CENTER|wx.ALL, 10)
        boxsizerX1X2.Add (boxsizerX1V, expand, wx.CENTER|wx.ALL, 10)
        boxsizerX1X2.Add (boxsizerX2V, expand, wx.CENTER|wx.ALL, 10)

        return boxsizerX1X2

    #end def VarControlsVar1

    def VarControlsVar2 (self, parentPanel, varLabelChar):

        global tctrlVar2

        def OnSpinX1 (event):
            global x1Var2

            x1Var2 = (event.GetPosition() - 200) / 200.0     # raw spinbutton value
            self.textX1Var2.SetValue ('%1.3f' % (x1Var2))
        #end def

        def OnSpinX2 (event):
            global x2Var2

            x2Var2 = (event.GetPosition() - 200) / 200.0     # raw spinbutton value
            self.textX2Var2.SetValue ('%1.3f' % (x2Var2))
       #end def

        def ResetX1BtnHandler (event):
            global x1Var2
            self.spinX1.SetValue (200)                        # 0.000
            x1Var2 = 0.00
            self.textX1Var2.SetValue ('%1.3f' % (x1Var2))
        #end def ResetX1BtnHandler

        def ResetX2BtnHandler (event):
            global x2Var2

            self.spinX2.SetValue (400)                        # 0.000
            x2Var2 = 1.00
            self.textX2Var2.SetValue ('%1.3f' % (x2Var2))
        #end def ResetX1BtnHandler

        #--------------------

        expand = False
        hgt = 40
        padding = 0

        boxsizerX1X2 = wx.BoxSizer (wx.HORIZONTAL)

        boxsizerX1V = wx.BoxSizer (wx.VERTICAL)

        labelX1 = wx.StaticText (self, -1, "X1:  Y=0.0 Intercept:", wx.DefaultPosition)
        boxsizerX1V.Add (labelX1, expand, wx.CENTER|wx.ALL, padding)

        boxsizerX1 = wx.BoxSizer (wx.HORIZONTAL)
        self.textX1Var2 = wx.TextCtrl (self, -1, "0.000", wx.DefaultPosition, wx.Size(60, -1),
                                   style=wx.TE_READONLY)
        boxsizerX1.Add (self.textX1Var2, expand, wx.CENTER|wx.ALL, padding)
        self.spinX1 = wx.SpinButton (self, 20, wx.DefaultPosition, wx.Size(hgt, hgt), wx.SP_HORIZONTAL)
        self.spinX1.SetRange (0, 600)                   # -1.000 .. 2.000
        self.spinX1.SetValue (200)                        # x1 = 0.000
        wx.EVT_SPIN (self.spinX1, 20, OnSpinX1)
        boxsizerX1.Add (self.spinX1, expand, 0, padding)

        boxsizerX1V.Add (boxsizerX1, expand, 0, padding)

        resetX1btn = wx.Button (self, -1, "Reset to 0.000")
        boxsizerX1V.Add (resetX1btn, expand, 0, padding)
        wx.EVT_BUTTON (self, resetX1btn.GetId(), ResetX1BtnHandler)

        boxsizerX2V = wx.BoxSizer (wx.VERTICAL)

        labelX2 = wx.StaticText (self, -1, "X2:  Y=1.0 Intercept:", wx.DefaultPosition)
        boxsizerX2V.Add (labelX2, expand, wx.CENTER|wx.ALL, padding)

        boxsizerX2 = wx.BoxSizer (wx.HORIZONTAL)
        self.textX2Var2 = wx.TextCtrl (self, -1, "1.000", wx.DefaultPosition, wx.Size(60, -1),
                                   style=wx.TE_READONLY)
        boxsizerX2.Add (self.textX2Var2, expand, wx.CENTER|wx.ALL, padding)
        self.spinX2 = wx.SpinButton (self, 20, wx.DefaultPosition, wx.Size(hgt, hgt), wx.SP_HORIZONTAL)
        self.spinX2.SetRange (0, 600)                   # -1.000 .. 2.000
        self.spinX2.SetValue (400)                        # x2 = 1.000
        wx.EVT_SPIN (self.spinX2, 20, OnSpinX2)
        boxsizerX2.Add (self.spinX2, expand, wx.CENTER|wx.ALL, padding)

        boxsizerX2V.Add (boxsizerX2, expand, 0, padding)

        resetX2btn = wx.Button (self, -1, "Reset to 1.000")
        boxsizerX2V.Add (resetX2btn, expand, 0, padding)
        wx.EVT_BUTTON (self, resetX2btn.GetId(), ResetX2BtnHandler)

        tctrlVar2 = wx.TextCtrl (self, -1, varLabelChar.upper(), 
                                 wx.DefaultPosition, wx.Size(20, -1), style=wx.TE_READONLY)

        boxsizerX1X2.Add (tctrlVar2, expand, wx.CENTER|wx.ALL, 10)    # a 10-pixel border all around
        boxsizerX1X2.Add (boxsizerX1V, expand, wx.CENTER|wx.ALL, 10)    # a 10-pixel border all around
        boxsizerX1X2.Add (boxsizerX2V, expand, wx.CENTER|wx.ALL, 10)    # a 10-pixel border all around

        return boxsizerX1X2

    #end def VarControlsVar2

    def VarControlsVar3 (self, parentPanel, varLabelChar):

        global tctrlVar3

        def OnSpinX1 (event):
            global x1Var3

            x1Var3 = (event.GetPosition() - 200) / 200.0     # raw spinbutton value
            self.textX1Var3.SetValue ('%1.3f' % (x1Var3))
        #end def

        def OnSpinX2 (event):
            global x2Var3

            x2Var3 = (event.GetPosition() - 200) / 200.0     # raw spinbutton value
            self.textX2Var3.SetValue ('%1.3f' % (x2Var3))
       #end def

        def ResetX1BtnHandler (event):
            global x1Var3

            self.spinX1.SetValue (200)                        # 0.000
            x1Var3 = 0.00
            self.textX1Var3.SetValue ('%1.3f' % (x1Var3))
        #end def ResetX1BtnHandler

        def ResetX2BtnHandler (event):
            global x2Var3

            self.spinX2.SetValue (400)                        # 0.000
            x2Var3 = 1.00
            self.textX2Var3.SetValue ('%1.3f' % (x2Var3))
        #end def ResetX1BtnHandler

        #--------------------

        expand = False
        hgt = 40
        padding = 0

        boxsizerX1X2 = wx.BoxSizer (wx.HORIZONTAL)

        boxsizerX1V = wx.BoxSizer (wx.VERTICAL)

        labelX1 = wx.StaticText (self, -1, "X1:  Y=0.0 Intercept:", wx.DefaultPosition)
        boxsizerX1V.Add (labelX1, expand, wx.CENTER|wx.ALL, padding)

        boxsizerX1 = wx.BoxSizer (wx.HORIZONTAL)
        self.textX1Var3 = wx.TextCtrl (self, -1, "0.000", wx.DefaultPosition, wx.Size(60, -1), 
                                   style=wx.TE_READONLY)
        boxsizerX1.Add (self.textX1Var3, expand, wx.CENTER|wx.ALL, padding)
        self.spinX1 = wx.SpinButton (self, 20, wx.DefaultPosition, wx.Size(hgt, hgt), wx.SP_HORIZONTAL)
        self.spinX1.SetRange (0, 600)                   # -1.000 .. 2.000
        self.spinX1.SetValue (200)                        # x1 = 0.000
        wx.EVT_SPIN (self.spinX1, 20, OnSpinX1)
        boxsizerX1.Add (self.spinX1, expand, 0, padding)

        boxsizerX1V.Add (boxsizerX1, expand, 0, padding)

        resetX1btn = wx.Button (self, -1, "Reset to 0.000")
        boxsizerX1V.Add (resetX1btn, expand, 0, padding)
        wx.EVT_BUTTON (self, resetX1btn.GetId(), ResetX1BtnHandler)

        boxsizerX2V = wx.BoxSizer (wx.VERTICAL)

        labelX2 = wx.StaticText (self, -1, "X2:  Y=1.0 Intercept:", wx.DefaultPosition)
        boxsizerX2V.Add (labelX2, expand, wx.CENTER|wx.ALL, padding)

        boxsizerX2 = wx.BoxSizer (wx.HORIZONTAL)
        self.textX2Var3 = wx.TextCtrl (self, -1, "1.000", wx.DefaultPosition, wx.Size(60, -1), 
                                   style=wx.TE_READONLY)
        boxsizerX2.Add (self.textX2Var3, expand, wx.CENTER|wx.ALL, padding)
        self.spinX2 = wx.SpinButton (self, 20, wx.DefaultPosition, wx.Size(hgt, hgt), wx.SP_HORIZONTAL)
        self.spinX2.SetRange (0, 600)                   # -1.000 .. 2.000
        self.spinX2.SetValue (400)                        # x2 = 1.000
        wx.EVT_SPIN (self.spinX2, 20, OnSpinX2)
        boxsizerX2.Add (self.spinX2, expand, wx.CENTER|wx.ALL, padding)

        boxsizerX2V.Add (boxsizerX2, expand, 0, padding)

        resetX2btn = wx.Button (self, -1, "Reset to 1.000")
        boxsizerX2V.Add (resetX2btn, expand, 0, padding)
        wx.EVT_BUTTON (self, resetX2btn.GetId(), ResetX2BtnHandler)

        tctrlVar3 = wx.TextCtrl (self, -1, varLabelChar.upper(), 
                                      wx.DefaultPosition, wx.Size(20, -1), style=wx.TE_READONLY)
        boxsizerX1X2.Add (tctrlVar3, expand, wx.CENTER|wx.ALL, 10)    # a 10-pixel border all around
        boxsizerX1X2.Add (boxsizerX1V, expand, wx.CENTER|wx.ALL, 10)
        boxsizerX1X2.Add (boxsizerX2V, expand, wx.CENTER|wx.ALL, 10)

        return boxsizerX1X2

    #end def VarControlsVar3

    def VarControlsVar4 (self, parentPanel, varLabelChar):

        global tctrlVar4

        def OnSpinX1 (event):
            global x1Var4

            x1Var4 = (event.GetPosition() - 200) / 200.0     # raw spinbutton value
            self.textX1Var4.SetValue ('%1.3f' % (x1Var4))
        #end def

        def OnSpinX2 (event):
            global x2Var4

            x2Var4 = (event.GetPosition() - 200) / 200.0     # raw spinbutton value
            self.textX2Var4.SetValue ('%1.3f' % (x2Var4))
       #end def

        def ResetX1BtnHandler (event):
            global x1Var4

            self.spinX1.SetValue (200)                        # 0.000
            x1Var4 = 0.00
            self.textX1Var4.SetValue ('%1.3f' % (x1Var4))
        #end def ResetX1BtnHandler

        def ResetX2BtnHandler (event):
            global x2Var4

            self.spinX2.SetValue (400)                        # 0.000
            x2Var4 = 1.00
            self.textX2Var4.SetValue ('%1.3f' % (x2Var4))
        #end def ResetX1BtnHandler

        #--------------------

        expand = False
        hgt = 40
        padding = 0

        boxsizerX1X2 = wx.BoxSizer (wx.HORIZONTAL)

        boxsizerX1V = wx.BoxSizer (wx.VERTICAL)

        labelX1 = wx.StaticText (self, -1, "X1:  Y=0.0 Intercept:", wx.DefaultPosition)
        boxsizerX1V.Add (labelX1, expand, wx.CENTER|wx.ALL, padding)

        boxsizerX1 = wx.BoxSizer (wx.HORIZONTAL)
        self.textX1Var4 = wx.TextCtrl (self, -1, "0.000", wx.DefaultPosition, wx.Size(60, -1), 
                                   style=wx.TE_READONLY)
        boxsizerX1.Add (self.textX1Var4, expand, wx.CENTER|wx.ALL, padding)
        self.spinX1 = wx.SpinButton (self, 20, wx.DefaultPosition, wx.Size(hgt, hgt), wx.SP_HORIZONTAL)
        self.spinX1.SetRange (0, 600)                   # -1.000 .. 2.000
        self.spinX1.SetValue (200)                      # x1 = 0.000
        wx.EVT_SPIN (self.spinX1, 20, OnSpinX1)
        boxsizerX1.Add (self.spinX1, expand, 0, padding)

        boxsizerX1V.Add (boxsizerX1, expand, 0, padding)

        resetX1btn = wx.Button (self, -1, "Reset to 0.000")
        boxsizerX1V.Add (resetX1btn, expand, 0, padding)
        wx.EVT_BUTTON (self, resetX1btn.GetId(), ResetX1BtnHandler)

        boxsizerX2V = wx.BoxSizer (wx.VERTICAL)

        labelX2 = wx.StaticText (self, -1, "X2:  Y=1.0 Intercept:", wx.DefaultPosition)
        boxsizerX2V.Add (labelX2, expand, wx.CENTER|wx.ALL, padding)

        boxsizerX2 = wx.BoxSizer (wx.HORIZONTAL)
        self.textX2Var4 = wx.TextCtrl (self, -1, "1.000", wx.DefaultPosition, wx.Size(60, -1), 
                                   style=wx.TE_READONLY)
        boxsizerX2.Add (self.textX2Var4, expand, wx.CENTER|wx.ALL, padding)
        self.spinX2 = wx.SpinButton (self, 20, wx.DefaultPosition, wx.Size(hgt, hgt), wx.SP_HORIZONTAL)
        self.spinX2.SetRange (0, 600)                   # -1.000 .. 2.000
        self.spinX2.SetValue (400)                      # x2 = 1.000
        wx.EVT_SPIN (self.spinX2, 20, OnSpinX2)
        boxsizerX2.Add (self.spinX2, expand, wx.CENTER|wx.ALL, padding)

        boxsizerX2V.Add (boxsizerX2, expand, 0, padding)

        resetX2btn = wx.Button (self, -1, "Reset to 1.000")
        boxsizerX2V.Add (resetX2btn, expand, 0, padding)
        wx.EVT_BUTTON (self, resetX2btn.GetId(), ResetX2BtnHandler)

        tctrlVar4 = wx.TextCtrl (self, -1, varLabelChar.upper(), 
                                 wx.Point(0, 0), wx.Size(20, -1), style=wx.TE_READONLY)
        boxsizerX1X2.Add (tctrlVar4, expand, wx.CENTER|wx.ALL, 10)    # a 10-pixel border all around
        boxsizerX1X2.Add (boxsizerX1V, expand, wx.CENTER|wx.ALL, 10)    # a 10-pixel border all around
        boxsizerX1X2.Add (boxsizerX2V, expand, wx.CENTER|wx.ALL, 10)    # a 10-pixel border all around

        return boxsizerX1X2

    #end def VarControlsVar4

    def xformBtnHandler (self, event):

        global filepathname

        if (type (filepathname)==type('abc')) or (type(filepathname)==type(u'abc')):
            try:
                self.childFrame.Destroy()
            except:
                pass
            #end if

            # Transform the original image into a new one.
            self.ChildFrame()
        #end if

    #end def xformBtnHandler

    def ChildFrame (self):

        global filepathname
        global thumbnailXsize
        global thumbnailYsize
        global x1Var1
        global x2Var1

        childframeXpsn = self.mainframePsn[0] + 2*self.mainframeSize[0]
        childframepsn = (childframeXpsn, self.mainframePsn[1])
        childframesize = (300, 300)
        childFrame = wx.Frame (self.parentFrame, -1, "Image Reverse-Transformed", childframepsn, childframesize)

        # Create the child panel
        self.childPanel = wx.Panel (childFrame, -1)
        self.childPanel.SetBackgroundColour ('#C9E3E5')

        # Create the child's bitmap
        staticBitmap = wx.StaticBitmap (self.childPanel, -1, wx.EmptyBitmap(5, 5))

        # Get and resize the original image to fit in the child panel and store it to a bitmap
        imPil = Image.open (filepathname)
        imPil = imPil.resize ( (thumbnailXsize, thumbnailYsize) )

        # Transform the thumbnail image.  This is where all the hard stuff is performed.
        imRgbPrime = Brighten (imPil)           # !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
        bitmap = Pil2Bitmap (imRgbPrime)
        staticBitmap.SetBitmap (bitmap)

        # Add the static bitmap to the panel.
        boxsizer = wx.BoxSizer (wx.VERTICAL)
        expand = False
        boxsizer.Add (staticBitmap, expand)

        self.childPanel.SetAutoLayout (True)
        self.childPanel.SetSizer (boxsizer)

        # Show the child frame
        childFrame.Show (True)

        self.childFrame = childFrame

    #end def ChildFrame

#end class AppPanel

def GetBitmapFromFile (filepathname, framesize):

    global thumbnailXsize
    global thumbnailYsize

    # If too large, resize the image to fit in the panel space.
    imPil = Image.open (filepathname)
    imXsize, imYsize = imPil.size

    panelXextent = framesize[0] - 10    # border size estimations for MS Windows
    panelYextent = framesize[1] - 24
    
    if (imXsize > panelXextent) or (imYsize > panelYextent):

        # The "primary" axis is the one that must be scaled down to fit its extent.
        #   The "secondary" axis will automatically scale to fit within its extent.
        # Find which axis is the primary.
        x2yExtentRatio = 1.0 * panelXextent / panelYextent
        imageXYratio = 1.0 * imXsize / imYsize

        if imageXYratio > x2yExtentRatio:       # The X axis is the primary axis

            # Find the image scale factor using the X axis.
            scalefactor = (1.0 * panelXextent) / imXsize
        else:
            scalefactor = (1.0 * panelYextent) / imYsize
        #end if

        # Resize the original image
        newXsize = int (scalefactor * imXsize)
        newYsize = int (scalefactor * imYsize)
        imPil = imPil.resize ( (newXsize, newYsize) )
    #end if

    # Convert the Pil image to a bitmap and install it into the displayed static bitmap
    bitmap = Pil2Bitmap (imPil)

    thumbnailXsize = newXsize   # save globally; child frame will use these
    thumbnailYsize = newYsize

    return bitmap

#end def GetBitmapFromFile

def Brighten (imRgb):

    global colorspaceStr
    global x1Var1, x2Var1
    global x1Var2, x2Var2
    global x1Var3, x2Var3
    global x1Var4, x2Var4

    # Extract the data from the input image and convert it to the particular colorspace data.
    rgbData = list (imRgb.getdata())    # must make a list of tuples out of the special sequence

    #print 'Transforming RGB data to a new colorspace ...',
    spaceData = []
    for rgbtuple in rgbData:

        if   colorspaceStr.lower() == 'hsl-':
            spaceData.append (Rgb2Hsl (rgbtuple))

        elif colorspaceStr.lower() == 'hsv-':
            spaceData.append (Rgb2Hsv (rgbtuple))

        elif colorspaceStr.lower() == 'cmyk':
            spaceData.append (Rgb2Cmyk (rgbtuple))

        else:
            print
            print 'Brighten(1):    Colorspace string [%s] not known.' % (colorspaceStr.upper())
            print
            os._exit (1)
        #end if
    #end for
    #print 'Finished'

    # Calculate the slope and offset (derived from y = mx + b).
    slopeVar1 = 1.0e6       # arbitrary maximum value of slope: avoid division-by-zero
    if x1Var1 != x2Var1:    slopeVar1 = 1.0/(x2Var1 - x1Var1)
    offsetVar1 = -1.0 * slopeVar1 * x1Var1

    slopeVar2 = 1.0e6       # arbitrary maximum
    if x1Var2 != x2Var2:    slopeVar2 = 1.0/(x2Var2 - x1Var2)
    offsetVar2 = -1.0 * slopeVar2 * x1Var2

    slopeVar3 = 1.0e6       # arbitrary maximum
    if x1Var3 != x2Var3:    slopeVar3 = 1.0/(x2Var3 - x1Var3)
    offsetVar3 = -1.0 * slopeVar3 * x1Var3

    slopeVar4 = 1.0e6       # arbitrary maximum
    if x1Var4 != x2Var4:    slopeVar4 = 1.0/(x2Var4 - x1Var4)
    offsetVar4 = -1.0 * slopeVar4 * x1Var4

    # Modify the particular colorspace values

    if   colorspaceStr.lower() == 'hsl-':
        for index in xrange (len(spaceData)):
            h, s, l = spaceData [index]

            # Apply an arbitrary algorithm to modify the colorspace vectors.
            h = (h * slopeVar1) + offsetVar1
            s = (s * slopeVar2) + offsetVar2
            l = (l * slopeVar3) + offsetVar3

            spaceData [index] = (h, s, l)
        #end for

    elif colorspaceStr.lower() == 'hsv-':
        for index in xrange (len(spaceData)):
            h, s, v = spaceData [index]

            # Apply an arbitrary algorithm to modify the colorspace vectors.
            h = (h * slopeVar1) + offsetVar1
            s = (s * slopeVar2) + offsetVar2
            v = (v * slopeVar3) + offsetVar3

            spaceData [index] = (h, s, v)
        #end for

    elif colorspaceStr.lower() == 'cmyk':
        for index in xrange (len(spaceData)):
            c, m, y, k = spaceData [index]

            # Apply an arbitrary algorithm to modify the colorspace vectors.
            c = (c * slopeVar1) + offsetVar1
            m = (m * slopeVar2) + offsetVar2
            y = (y * slopeVar3) + offsetVar3
            k = (k * slopeVar4) + offsetVar4

            spaceData [index] = (c, m, y, k)      # replace old values with new in-place
        #end for

    else:
        print
        print 'Brighten(2):    Colorspace string [%s] not known.' % (colorspaceStr.upper())
        print
        os._exit (1)
    #end if

    # Convert the modified colorspace data back into RGB data
    #print 'Returning modified colorspace data to RGB ...',
    rgbPrimedata = []
    for spacetuple in spaceData:

        if   colorspaceStr.lower() == 'cmyk':
            rgbPrimetuple = Cmyk2Rgb (spacetuple)

        elif colorspaceStr.lower() == 'hsl-':
            rgbPrimetuple = Hsl2Rgb (spacetuple)

        elif colorspaceStr.lower() == 'hsv-':
            rgbPrimetuple = Hsv2Rgb (spacetuple)

        else:
            print
            print 'Brighten(3):   Colorspace string not recognized = [%s]' % (colorspaceStr.upper())
            print
            os._exit (1)
        #end if

        rgbPrimedata.append (rgbPrimetuple)
    #end for
    #print 'Finished'

    # Create a new image from the new RGB data
    imRgbPrime = Image.new ('RGB', imRgb.size)
    imRgbPrime.putdata (rgbPrimedata)

    return imRgbPrime

#end def Brighten

#------------------

if __name__ == "__main__":

    app = wx.PySimpleApp()
    wx.InitAllImageHandlers()

    framepsn = (100, 100)
    framesize = (300, 575)          # master frame size
    AppFrame (None, -1, "Colorspace Transformer", framepsn, framesize)

    app.MainLoop()

#end if

From bob at redivi.com  Sat Apr  9 11:03:30 2005
From: bob at redivi.com (Bob Ippolito)
Date: Sat Apr  9 11:03:33 2005
Subject: [Image-SIG] Python 2.4?
In-Reply-To: <3C80DCC2E816D94EA3D5DFA9D3B798D80512CC@OslHkBl0818.ls.no>
References: <3C80DCC2E816D94EA3D5DFA9D3B798D80512CC@OslHkBl0818.ls.no>
Message-ID: <c06584c691dce195bcdbab07fc28e152@redivi.com>


On Mar 21, 2005, at 3:49 AM, Sigmund Lahn wrote:

> The installer complains thet I do not have python 2.3 installed. Does 
> this mean python 2.4 is not supported, or is the installer just 
> outdated?
>
> Ie, should I tweak my register so that python 2.4 is also registered 
> as 2.3?

Extensions built for one minor Python revision are not compatible with 
another.  If you want to use PIL with Python 2.4, you must download an 
installer that is for Python 2.4.

Tweaking your registry is a recipe for disaster.

-bob

From steve at holdenweb.com  Sat Apr  9 16:19:52 2005
From: steve at holdenweb.com (Steve Holden)
Date: Sat Apr  9 16:19:40 2005
Subject: [Image-SIG] Installation Guide
In-Reply-To: <20050312195315.45930.qmail@web20922.mail.yahoo.com>
References: <20050312195315.45930.qmail@web20922.mail.yahoo.com>
Message-ID: <4257E488.6080902@holdenweb.com>

Online Support wrote:
> Hello,
>  
> I'm in critical need to install Python Imaging Library on my 
> linux/cpanel server. I'm not very experienced and I almost always have 
> to use instalation guides. I have downloaded the software, unzipped it, 
> and now I think I just need to install it somehow. Are there files that 
> need to be modified first? I think there is a file called setup.py but 
> I'm not sure how I can run that file.
>  
> I would really appreciate your reply, or if you could point me to an 
> installation giude online for PIL
>  
> Best Regards,
> - Kevin
> 
By "unzipped" I presume you mean "untarred the .tgz file"? Just make 
sure your text files will have Linux line endings not windows (just in 
case: sorry of this seems obvious).

There should be a README file int eh top-level directory. Look for the 
section that starts

--------------------------------------------------------------------
Build instructions (all platforms)
--------------------------------------------------------------------


Or have you seen this and need more?

regards
  Steve
-- 
Steve Holden        +1 703 861 4237  +1 800 494 3119
Holden Web LLC             http://www.holdenweb.com/
Python Web Programming  http://pydish.holdenweb.com/
From jwt at onjapan.net  Mon Apr 11 10:07:06 2005
From: jwt at onjapan.net (Jim Tittsler)
Date: Mon Apr 11 10:06:45 2005
Subject: [Image-SIG] c code to python using gd
In-Reply-To: <423A3DB0.9080706@demir.web.tr>
References: <423A3DB0.9080706@demir.web.tr>
Message-ID: <6376fbbd5a9c289a3e6b6a347737a49a@onjapan.net>

On Mar 18, 2005, at 11:32, Necati DEMiR wrote:
[Your message seems to have gotten stuck in the python.org mailing 
lists for rather a long time... probably because the attachments made 
the message so huge.  You would have been better served by putting the 
images on the web somewhere and given URLs to them.]

> i converted a c code to python. two of them uses gd, but they dont have
> the sama output :(

But they aren't the same code.  Your values of p1 and p2 are quite 
different.
>     int p1 = (255/3);
>     int p2 = 2 * (255/3);
[...]
> p1 = 255+1 / 3
> p2 = 2 * (255+1 / 3)

If you use the same values for p1 and p2 in both C and Python, the 
images will be the same.

-- 
Jim Tittsler, Mitaka, Tokyo, JAPAN             http://www.OnJapan.net/
Python Starship           http://Starship.Python.net/
Ringo MUG Tokyo           http://www.ringo.net/rss.html
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 2091 bytes
Desc: not available
Url : http://mail.python.org/pipermail/image-sig/attachments/20050411/4d94dbce/smime.bin
From M.Singh at exeter.ac.uk  Mon Apr 11 17:39:52 2005
From: M.Singh at exeter.ac.uk (Maneesha)
Date: Mon Apr 11 17:38:56 2005
Subject: [Image-SIG] ICAPR2005 - Deadline 20th April,2005
Message-ID: <425B0101@minerva2.ex.ac.uk>

Dear Colleagues,

We would like to circulate the Call for Participation of ICAPR'05 and its
related workshops to your mailing list. Please circulate it (if appropriate)
to the members of your group.

Many thanks,
Maneesha Singh
Organising Chair (ICAPR'05)

----------------------------------------------------------------------------
3rd International Conference on Advances in Pattern Recognition (ICAPR 2005),
BATH, UK
published by Springer-Verlag

22-25 August, 2005

NEW PAPER SUBMISSION DEADLINE: 20 April, 2005
CONFERENCE WEBSITE: http://www.icapr2005.org.uk
PAPER SUBMISSION WEBSITE: http://www.paaonline.net/cps

Research Topics and Tracks

Track 1: Pattern Recognition Methods
Theoretical developments in machine learning and pattern recognition tools and
techniques;  Statistical pattern recognition; Data Mining; Artificial neural
networks; Bayesian networks; Case-based reasoning; Classifier combination;
Computational models of human learning; Computational learning theory;
Cooperative learning; Decision trees; Evolutionary computation; Feature
selection; Statistical relational learning; Grammatical inference; Incremental
induction and on-line learning; Inductive logic programming; Information
retrieval and learning; Instance based learning; Kernel methods; Level sets

Track 2: Knowledge and Learning
Knowledge acquisition and learning; Knowledge base refinement; Knowledge
intensive learning; Learning from text and web; Evaluation metrics and
methodologies; Machine learning of natural language; Meta learning;
Multi-agent learning; Multi-strategy learning; Planning and learning;
Reinforcement learning; Revision and restructuring; Statistical approaches;
Unsupervised learning; Vision and learning; Visualisation techniques;
Algorithms and techniques; Classification; Clustering; Frequent patterns; Rule
discovery; Statistical techniques and mixture models; Constraint-based mining;
incremental algorithms; Scalable algorithms; Distributed and parallel
algorithms; Privacy preserving data mining; Multi-relational data mining;
Innovative applications; Mining bio-medical data; Web content, structure and
usage mining; Semantic web mining; Mining governmental data, mining for the
public administration; Personalization; Adaptive data mining architectures

Track 3: Data Mining
Database integration; Inductive databases; Data mining query languages; Data
mining query optimisation; Dimensionality reduction; Data reduction;
Discretization; Uncertain and missing information; Complexity issues;
knowledge (pattern) representation; Global vs. local patterns; Logic for data
mining; Statistical inference/ probabilistic modelling; Collaborative data
mining; Vertical data mining environments; Mining different forms of data;
Graph, tree, sequence mining; Semi-structured and XML data mining; Text
mining; Temporal, spatial, and spatio-temporal data mining; Data stream
mining; Multimedia mining; Pattern post-processing; Quality assessment;
Visualization; knowledge interpretation and use

Track 4: Applications
Image analysis; Video analysis; Speech recognition and classification; Text
analysis; Medical imaging; Robotics and Autonomous Systems; Multimedia;
Information fusion; Computational signal processing; Geotechnical
applications; Astronomy applications; Remote Sensing; Biomedical applications;
Environmental sciences; Inspection technologies; Manufacturing applications;
Computational Biology; Bioinformatics; Health sciences

Papers should be submitted through http://www.paaonline.net/cps

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


International Workshop on Document Image Analysis and Recognition
(Chair: Prof. Adnan Amin, University of New South Wales, Australia)
Extended Paper submission deadline: 20 April, 2005
CONFERENCE WEBSITE: http://www.icapr2005.org.uk
Papers should be submitted through http://www.paaonline.net/cps


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


International Workshop on Information Fusion for Biometric Processing
(Chair: Prof. Mike Fairhurst, University of Kent, UK)
Extended Paper submission deadline: 20 April, 2005
CONFERENCE WEBSITE: http://www.icapr2005.org.uk
Papers should be submitted through http://www.paaonline.net/cps


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


International Workshop on Pattern Recognition for Crime Prevention, Security
and Surveillance
(Chair: Prof. Sameer Singh, Loughborough University, UK)
Extended Paper submission deadline: 20 April, 2005
CONFERENCE WEBSITE: http://www.icapr2005.org.uk
Papers should be submitted through http://www.paaonline.net/cps


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

From klimek at grc.nasa.gov  Mon Apr 11 21:27:52 2005
From: klimek at grc.nasa.gov (Bob Klimek)
Date: Mon Apr 11 21:27:25 2005
Subject: [Image-SIG] HSL adjustment, and others
In-Reply-To: <4255445E.1010707@hotpop.com>
References: <4255445E.1010707@hotpop.com>
Message-ID: <425ACFB8.6080703@grc.nasa.gov>

Ray Pasco wrote:

> I've been playing around with just this. This program converts to/from 
> several different colorspaces and changes the alternate colorspace's 
> "ramp" slope and offset. The program only does linear adjustments.
>
> I'd appreciate any comments/suggestions you may have.

Hi Ray,

I've taken a quick look at your rgb-hsl (and hsl-rgb) and rgb-hsv (and 
hsv-rgb) transformations and from the few tests I've run, both 
algorithms seem to be working quite well. And both successfully convert 
to and from. I don't see any difference really, although I don't have 
the time to do a thorough comparison.

My interest is only in hue so I've looked at that component only. I have 
a test image of a type of "rainbow" in which the hue changes from 1.0 
down to 0.0 in a linear fashion as the image is scanned from left to 
right. I use an old algorithm which produces a linear relationship 
between hue and posistion in this image. To my surprise both of these 
algorithms also produce a linear relationship.

BTW, the algorithm that I use (see below) come from Computer Graphics 
Quarterly Report SIGGRAPH-ACM, vol 13, Number 3, Aug. 1979. The hue 
range is from 0 to 2pi. One interesting thing we've noticed is that some 
hue values repeat them selves. That is an 8-bit deep RGB image has 16 
million rgb combinations but our hue algorithm produces only a few 
thousand unique values, not 16 million. All others are repeated hue values.

Anyway, it would be very nice to have a fast image-wide HSI 
transformation done in C with the output as a float array. Maybe some day...

    def calcHue(self, r, g, b):
        rf = float(r)
        gf = float(g)
        bf = float(b)
        rf = rf/255.0
        gf = gf/255.0
        bf = bf/255.0
        inten = (rf+gf+bf)/3.0
        if(not((r==b) and (r==g))):
            min = rf
            if(min>gf): min = gf
            if(min>bf): min = bf
            sat=1.0-(min/inten)
            t1 = rf-gf
            t2 = rf-bf
            t3 = gf-bf
            t4 = math.sqrt((t1*t1)+(t2*t3))
            t5 = (t1+t2)/2.0
            t5 = t5/t4
            hue = math.acos(t5)
            if(gf<bf): hue=2.0*math.pi-hue
        else:
            hue=-1000.0
            sat=-1.0
        return (hue, sat)

Bob

From etherson at yahoo.com  Tue Apr 12 02:11:44 2005
From: etherson at yahoo.com (Eric Etheridge)
Date: Tue Apr 12 02:11:47 2005
Subject: [Image-SIG] PSDraw: Floating Point Typo
Message-ID: <20050412001144.6330.qmail@web31711.mail.mud.yahoo.com>

I have finally tracked down the cause of images
sometimes not showing up when using PSDraw's
im.image() method.  The problem is the following
lines, 101 and 102 in PSDraw.ph:

            sx = x / im.size[0]
            sy = y / im.size[1]

These lines can occasionally perform integer division
rather than floating point division.  This is
partially prevented by the use of float() on lines
87-88, but then lines 90-95 cause problems again:

        xmax = box[2] - box[0]
        ymax = box[3] - box[1]
        if x > xmax:
            y = y * xmax / x; x = xmax
        if y > ymax:
            x = x * ymax / y; y = ymax

x or y may hold an integer value after these lines,
depending on the size and shape of the box versus the
image.  The fix is to add a pair of float() in one of
three places, either around xmax and ymax's creation,
their assignment to x and y, or in lines 101-102:

        xmax = float(box[2] - box[0])
        ymax = float(box[3] - box[1])
or
        if x > xmax:
            y = y * xmax / x; x = float(xmax)
        if y > ymax:
            x = x * ymax / y; y = float(ymax)
or
            sx = float(x) / im.size[0]
            sy = float(y) / im.size[1]

Any of those ways works.  This is an extremely
frustrating bug and its existance can render the
PSDraw module relatively useless, so please include
the fix in the distribution as soon as possible.  I
think it has been in existance for several years.

- Eric Etheridge



		
__________________________________ 
Do you Yahoo!? 
Yahoo! Small Business - Try our new resources site!
http://smallbusiness.yahoo.com/resources/
From klimek at grc.nasa.gov  Tue Apr 12 16:18:45 2005
From: klimek at grc.nasa.gov (Bob Klimek)
Date: Tue Apr 12 16:18:14 2005
Subject: [Image-SIG] HSL adjustment, and others
In-Reply-To: <425AE4F4.9020802@hotpop.com>
References: <4255445E.1010707@hotpop.com> <425ACFB8.6080703@grc.nasa.gov> <4
	25AE4F4.9020802@hotpop.com>
Message-ID: <425BD8C5.4000708@grc.nasa.gov>

No. The paper mentioned below has the pseudocode but I've never 
implemented it.
Bob


Ray Pasco wrote:

> Bob,
>
> So, that's HSI ?  OK, but it seems rather bizarre.  Anyhow, do you 
> have the inverse transform ?
>
> Ray
>
>
> Bob Klimek wrote:
>
>> Ray Pasco wrote:
>>
>>> I've been playing around with just this. This program converts 
>>> to/from several different colorspaces and changes the alternate 
>>> colorspace's "ramp" slope and offset. The program only does linear 
>>> adjustments.
>>>
>>> I'd appreciate any comments/suggestions you may have.
>>
>>
>>
>> Hi Ray,
>>
>> I've taken a quick look at your rgb-hsl (and hsl-rgb) and rgb-hsv 
>> (and hsv-rgb) transformations and from the few tests I've run, both 
>> algorithms seem to be working quite well. And both successfully 
>> convert to and from. I don't see any difference really, although I 
>> don't have the time to do a thorough comparison.
>>
>> My interest is only in hue so I've looked at that component only. I 
>> have a test image of a type of "rainbow" in which the hue changes 
>> from 1.0 down to 0.0 in a linear fashion as the image is scanned from 
>> left to right. I use an old algorithm which produces a linear 
>> relationship between hue and posistion in this image. To my surprise 
>> both of these algorithms also produce a linear relationship.
>>
>> BTW, the algorithm that I use (see below) come from Computer Graphics 
>> Quarterly Report SIGGRAPH-ACM, vol 13, Number 3, Aug. 1979. The hue 
>> range is from 0 to 2pi. One interesting thing we've noticed is that 
>> some hue values repeat them selves. That is an 8-bit deep RGB image 
>> has 16 million rgb combinations but our hue algorithm produces only a 
>> few thousand unique values, not 16 million. All others are repeated 
>> hue values.
>>
>> Anyway, it would be very nice to have a fast image-wide HSI 
>> transformation done in C with the output as a float array. Maybe some 
>> day...
>>
>>    def calcHue(self, r, g, b):
>>        rf = float(r)
>>        gf = float(g)
>>        bf = float(b)
>>        rf = rf/255.0
>>        gf = gf/255.0
>>        bf = bf/255.0
>>        inten = (rf+gf+bf)/3.0
>>        if(not((r==b) and (r==g))):
>>            min = rf
>>            if(min>gf): min = gf
>>            if(min>bf): min = bf
>>            sat=1.0-(min/inten)
>>            t1 = rf-gf
>>            t2 = rf-bf
>>            t3 = gf-bf
>>            t4 = math.sqrt((t1*t1)+(t2*t3))
>>            t5 = (t1+t2)/2.0
>>            t5 = t5/t4
>>            hue = math.acos(t5)
>>            if(gf<bf): hue=2.0*math.pi-hue
>>        else:
>>            hue=-1000.0
>>            sat=-1.0
>>        return (hue, sat)
>>
>> Bob
>>
>>
>>
>
>
>

From aconrad.tlv at magic.fr  Tue Apr 12 16:27:54 2005
From: aconrad.tlv at magic.fr (Alexandre CONRAD)
Date: Tue Apr 12 16:27:55 2005
Subject: [Image-SIG] image 2 MPEG
Message-ID: <425BDAEA.2000408@magic.fr>

Hello,

I'm new to this list. I've been using the python PIL a little bit, and I 
pretty much like it.

This might be a little off-topic: I'm looking for a way to be able to 
convert an image (jpg, gif, any kind that PIL can read) to a MPEG 
I-Frame. I know PIL doesn't support MPEG encoding. But I've seen some 
tools around, like "mpeg2enc" that can encode an image (PPM) to an MPEG 
file.

Has anyone tested this ?

Regards,
-- 
Alexandre CONRAD - TLV
Research & Development
tel : +33 1 30 80 55 05
fax : +33 1 30 80 55 06
6, rue de la plaine
78860 - SAINT NOM LA BRETECHE
FRANCE

From magoldfish at gmail.com  Tue Apr 12 21:41:07 2005
From: magoldfish at gmail.com (Marcus Goldfish)
Date: Tue Apr 12 21:41:13 2005
Subject: [Image-SIG] Loading a DDB into PIL Image
Message-ID: <5e183f3d050412124122d50608@mail.gmail.com>

Hi,

I use ctypes to capture a device-dependent bitmap in Win32 (~600x400,
32bpp), and I would like to process it using PIL.  I need a fast way
to convert the DDB to a PIL Image-- any solutions or suggestions
(ctypes, win32 solutions acceptable])?

"Fast" means that I need to be able to convert the image in << 1 second.

Thanks!
Marcus
From fredrik at pythonware.com  Thu Apr 14 09:56:53 2005
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu Apr 14 09:58:16 2005
Subject: [Image-SIG] Re: Loading a DDB into PIL Image
References: <5e183f3d050412124122d50608@mail.gmail.com>
Message-ID: <d3l7hs$au0$1@sea.gmane.org>

Marcus Goldfish wrote:

> I use ctypes to capture a device-dependent bitmap in Win32 (~600x400,
> 32bpp), and I would like to process it using PIL.  I need a fast way
> to convert the DDB to a PIL Image-- any solutions or suggestions
> (ctypes, win32 solutions acceptable])?
>
> "Fast" means that I need to be able to convert the image in << 1 second.

if you can convert it to a DIB, you can use the BmpImageFile.DibImageFile
constructor.  see the ImageGrab module for sample code.

</F> 



From magoldfish at gmail.com  Thu Apr 14 13:38:55 2005
From: magoldfish at gmail.com (Marcus Goldfish)
Date: Thu Apr 14 13:39:11 2005
Subject: [Image-SIG] Re: Loading a DDB into PIL Image
In-Reply-To: <d3l7hs$au0$1@sea.gmane.org>
References: <5e183f3d050412124122d50608@mail.gmail.com>
	<d3l7hs$au0$1@sea.gmane.org>
Message-ID: <5e183f3d05041404385add0fa6@mail.gmail.com>

> > I use ctypes to capture a device-dependent bitmap in Win32 (~600x400,
> > 32bpp), and I would like to process it using PIL.  I need a fast way
> > to convert the DDB to a PIL Image-- any solutions or suggestions
> > (ctypes, win32 solutions acceptable])?
> > "Fast" means that I need to be able to convert the image in << 1 second.
> 
> if you can convert it to a DIB, you can use the BmpImageFile.DibImageFile
> constructor.  see the ImageGrab module for sample code.
Thanks Fredrik, but I'm not sure how to convert it to DIB.

It seems like the ImageWin module is what I want to use, but it is
missing a constructor which takes a DDB bitmap and produces a DIB or
Image.  Right now, the two best solutions I can see are: (1) trying to
convert to DIB, but I'm not sure how to do this, and (2) pasting my
DDB to the clipboard then using PIL to grab it.  Any ideas on the
relative speeds of the two approaches?

Marcus
From areopagus125 at yahoo.com  Fri Apr 15 00:43:13 2005
From: areopagus125 at yahoo.com (David Smith)
Date: Fri Apr 15 00:43:16 2005
Subject: [Image-SIG] problem with gifmaker.py 
Message-ID: <20050414224313.11689.qmail@web31709.mail.mud.yahoo.com>

I downloaded gifmaker.py from
http://olympus.het.brown.edu/doc/python-imaging/examples/gifmaker.py
 and tried it out.  It writes a file all right, but
when I try to view it with the Windows XP Picture and
Fax viewer, it tells me, "Drawing failed."  I created
a PowerPoint slide and clicked
Insert|Picture|FromFile, and it shows the first frame.
 But it won't cycle through the frames when I put it
into slideshow mode.  What's missing?  

Oh, and I would like to be able to set the frame rate,
as well.

  -- David Smith



		
__________________________________ 
Do you Yahoo!? 
Yahoo! Small Business - Try our new resources site!
http://smallbusiness.yahoo.com/resources/
From jdunck at gmail.com  Sun Apr 17 02:58:59 2005
From: jdunck at gmail.com (Jeremy Dunck)
Date: Sun Apr 17 02:59:02 2005
Subject: [Image-SIG] The mysterious PIL Plus
Message-ID: <2545a92c0504161758143f2174@mail.gmail.com>

I've just recently started using PIL, and just banged my head on the
fact that _imageft is not included in the free PIL.
From jdunck at gmail.com  Sun Apr 17 03:04:06 2005
From: jdunck at gmail.com (Jeremy Dunck)
Date: Sun Apr 17 03:04:09 2005
Subject: [Image-SIG] Re: The mysterious PIL Plus
In-Reply-To: <2545a92c0504161758143f2174@mail.gmail.com>
References: <2545a92c0504161758143f2174@mail.gmail.com>
Message-ID: <2545a92c05041618046bf767b5@mail.gmail.com>

On 4/16/05, Jeremy Dunck <jdunck@gmail.com> wrote:
> I've just recently started using PIL, and just banged my head on the
> fact that _imageft is not included in the free PIL.
> 

Very sorry about that.  Sent accidentally.

To continue: I have googled around for info on PIL Plus (what's it
cost, where can I get it, what license is it under, etc.), and have
come up empty.

Does it exist?  :)
From postmaster at quost.com  Mon Apr 18 14:34:08 2005
From: postmaster at quost.com (dnsrequest.com PostMaster)
Date: Mon Apr 18 14:34:19 2005
Subject: [Image-SIG] Error sending message [1113827660403.3228.pegasus] from
	[dnsrequest.com]
Message-ID: <20050418123417.1FD3D1E4005@bag.python.org>

[<00>] XMail bounce: Rcpt=[claudia@steelkey.com.au];Error=[550 Mailbox unavailable <claudia@steelkey.com.au>]


[<01>] Error sending message [1113827660403.3228.pegasus] from [dnsrequest.com].

ID:        <S68B1A>
Mail From: <image-sig@python.org>
Rcpt To:   <claudia@steelkey.com.au>
Server:    <mail3.dnsrequest.com> [12.96.165.198]


[<02>] The reason of the delivery failure was:

550 Mailbox unavailable <claudia@steelkey.com.au>


[<05>] Here is listed the initial part of the message:

Received: from python.org (84.204.14.21:3177)
	by mail2.dnsrequest.com with [XMail 1.17 (Win32/Ix86) ESMTP Server]
	id <S68B1A> for <claudia@steelkey.com.au> from <image-sig@python.org>;
	Mon, 18 Apr 2005 12:33:54 -0000
From: image-sig@python.org
To: claudia@steelkey.com.au
Subject: HELLO
Date: Mon, 18 Apr 2005 16:33:45 +0400
MIME-Version: 1.0
Content-Type: multipart/mixed;
	boundary="----=_NextPart_000_0002_EB09F4CD.666EB70D"
X-Priority: 3
X-MSMail-Priority: Normal

From fredrik at pythonware.com  Mon Apr 18 18:06:59 2005
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Mon Apr 18 18:10:22 2005
Subject: [Image-SIG] Re: The mysterious PIL Plus
References: <2545a92c0504161758143f2174@mail.gmail.com>
Message-ID: <d40lnf$rqc$1@sea.gmane.org>

Jeremy Dunck wrote:

> I've just recently started using PIL, and just banged my head on the
> fact that _imageft is not included in the free PIL.

if you mean _imagingft, it's been included in PIL since early 2003.  if you're
building PIL yourself, you need FreeType.  if you're using a prebuilt version,
most 1.1.4 and 1.1.5 builds should come with a working _imagingft module.

</F>

From bbaxter at wadsworth.org  Mon Apr 18 19:35:43 2005
From: bbaxter at wadsworth.org (William Baxter)
Date: Mon Apr 18 19:35:55 2005
Subject: [Image-SIG] busy cursor stores up events
Message-ID: <4263EFEF.E09E8C56@wadsworth.org>

Is there a way to get the BusyManager to prevent the widgets from
servicing events accumulated during the busy period?

I've been looking at the 2nd example at

http://effbot.org/zone/tkinter-busy.htm

Although the widgets do not respond to user events while the system is
'busy', they seem to register user events. When the system returns to
the 'not busy' state, all the events are then processed. For instance,
if I try to click the button repeatedly while 'busy', the gui doesn't
respond (that's good). But when it's no longer busy, the button command
prints out
a whole column of hello's from the stored up events.


Thanks,
Bill Baxter
From cohen at slac.stanford.edu  Mon Apr 18 20:30:01 2005
From: cohen at slac.stanford.edu (Johann Cohen Tanugi)
Date: Mon Apr 18 20:30:11 2005
Subject: [Image-SIG] fits format in PIL
Message-ID: <4263FCA9.2060108@slac.stanford.edu>

I googled a bit without much luck so I figured I would post the question 
directly here : does PIL support FITS image? If not yet, is there a team 
working on it?

thanks in advance,
Johann
From rowen at cesmail.net  Mon Apr 18 21:11:37 2005
From: rowen at cesmail.net (Russell E. Owen)
Date: Mon Apr 18 21:13:16 2005
Subject: [Image-SIG] Re: fits format in PIL
References: <4263FCA9.2060108@slac.stanford.edu>
Message-ID: <rowen-095BC5.12113718042005@sea.gmane.org>

In article <4263FCA9.2060108@slac.stanford.edu>,
 Johann Cohen Tanugi <cohen@slac.stanford.edu> wrote:

> I googled a bit without much luck so I figured I would post the question 
> directly here : does PIL support FITS image? If not yet, is there a team 
> working on it?

It does not directly support FITS. One simple technique is to use pyfits 
to read in the data, then display it using PIL. At least it's simple in 
regards parsing the FITS data; if you a have high dynamic range image, 
as is typical for FITS data, then scaling the data to display anything 
meaningful to the user is another story.

Some resources:
- Astronomical Python (link to pyfits and other utilties): 
http://www.astro.washington.edu/rowen/AstroPy.html>
- Sample grayscale image viewer (still being worked on but quite 
functional) RO.Wdg.GrayImageDispWdg in RO package: 
<http://www.astro.washington.edu/rowen/ROPython.html>
- matplotlib includes an image viewer. It is rather slow but might prove 
useful.

-- Russell

From cohen at slac.stanford.edu  Mon Apr 18 21:24:22 2005
From: cohen at slac.stanford.edu (Johann Cohen Tanugi)
Date: Mon Apr 18 21:24:30 2005
Subject: [Image-SIG] Re: fits format in PIL
In-Reply-To: <rowen-095BC5.12113718042005@sea.gmane.org>
References: <4263FCA9.2060108@slac.stanford.edu>
	<rowen-095BC5.12113718042005@sea.gmane.org>
Message-ID: <42640966.4010305@slac.stanford.edu>

thanks for your answer. From your email I infer that PIL displays 
numarray objects, because this is the output format of a pyfits IO call 
on a fits file.
Is there any thought about creating a plugin in PIL for FITS access?

cheers,
Johann

Russell E. Owen wrote:

>In article <4263FCA9.2060108@slac.stanford.edu>,
> Johann Cohen Tanugi <cohen@slac.stanford.edu> wrote:
>
>  
>
>>I googled a bit without much luck so I figured I would post the question 
>>directly here : does PIL support FITS image? If not yet, is there a team 
>>working on it?
>>    
>>
>
>It does not directly support FITS. One simple technique is to use pyfits 
>to read in the data, then display it using PIL. At least it's simple in 
>regards parsing the FITS data; if you a have high dynamic range image, 
>as is typical for FITS data, then scaling the data to display anything 
>meaningful to the user is another story.
>
>Some resources:
>- Astronomical Python (link to pyfits and other utilties): 
>http://www.astro.washington.edu/rowen/AstroPy.html>
>- Sample grayscale image viewer (still being worked on but quite 
>functional) RO.Wdg.GrayImageDispWdg in RO package: 
><http://www.astro.washington.edu/rowen/ROPython.html>
>- matplotlib includes an image viewer. It is rather slow but might prove 
>useful.
>
>-- Russell
>
>_______________________________________________
>Image-SIG maillist  -  Image-SIG@python.org
>http://mail.python.org/mailman/listinfo/image-sig
>  
>
From jdunck at gmail.com  Tue Apr 19 06:00:30 2005
From: jdunck at gmail.com (Jeremy Dunck)
Date: Tue Apr 19 06:00:33 2005
Subject: [Image-SIG] PIL Plus?
Message-ID: <2545a92c050418210073a53848@mail.gmail.com>

I've seen some references to PIL Plus both on this list and
Pythonware's PIL documents.

However, I've been unable to find a place to buy or download this software.

Can someone give me a pointer?  Why so secretive?
From NadavH at VisionSense.com  Tue Apr 19 08:08:21 2005
From: NadavH at VisionSense.com (Nadav Horesh)
Date: Tue Apr 19 08:00:52 2005
Subject: [Image-SIG] PIL Plus?
In-Reply-To: <2545a92c050418210073a53848@mail.gmail.com>
References: <2545a92c050418210073a53848@mail.gmail.com>
Message-ID: <4264A055.5030106@VisionSense.com>

It is a commercial package  (www.pythonware.com)

  Nadav

Jeremy Dunck wrote:

>I've seen some references to PIL Plus both on this list and
>Pythonware's PIL documents.
>
>However, I've been unable to find a place to buy or download this software.
>
>Can someone give me a pointer?  Why so secretive?
>_______________________________________________
>Image-SIG maillist  -  Image-SIG@python.org
>http://mail.python.org/mailman/listinfo/image-sig
>
>  
>
From bbaxter at wadsworth.org  Tue Apr 19 16:05:18 2005
From: bbaxter at wadsworth.org (William Baxter)
Date: Tue Apr 19 16:11:36 2005
Subject: [Image-SIG] busy cursor stores up events
References: <4263EFEF.E09E8C56@wadsworth.org>
Message-ID: <4265101E.5EE1AF77@wadsworth.org>

Sorry! Wrong discussion group! Sorry about that.

- BB

William Baxter wrote:
> 
> Is there a way to get the BusyManager to prevent the widgets from
> servicing events accumulated during the busy period?
> 
> I've been looking at the 2nd example at
> 
> http://effbot.org/zone/tkinter-busy.htm
> 
> Although the widgets do not respond to user events while the system is
> 'busy', they seem to register user events. When the system returns to
> the 'not busy' state, all the events are then processed. For instance,
> if I try to click the button repeatedly while 'busy', the gui doesn't
> respond (that's good). But when it's no longer busy, the button command
> prints out
> a whole column of hello's from the stored up events.
> 
> Thanks,
> Bill Baxter
> _______________________________________________
> Image-SIG maillist  -  Image-SIG@python.org
> http://mail.python.org/mailman/listinfo/image-sig
From lars.yencken at gmail.com  Wed Apr 20 08:03:43 2005
From: lars.yencken at gmail.com (Lars Yencken)
Date: Wed Apr 20 08:03:49 2005
Subject: [Image-SIG] Rendering Japanese fonts
Message-ID: <e36a4dac0504192303a92f11@mail.gmail.com>
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 p.f.moore at gmail.com  Wed Apr 20 10:35:24 2005
From: p.f.moore at gmail.com (Paul Moore)
Date: Wed Apr 20 10:35:27 2005
Subject: [Image-SIG] Rendering Japanese fonts
In-Reply-To: <e36a4dac0504192303a92f11@mail.gmail.com>
References: <e36a4dac0504192303a92f11@mail.gmail.com>
Message-ID: <79990c6b05042001355f30f3b4@mail.gmail.com>
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From fredrik at pythonware.com  Wed Apr 20 20:58:48 2005
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Wed Apr 20 21:01:06 2005
Subject: [Image-SIG] Re: Rendering Japanese fonts
References: <e36a4dac0504192303a92f11@mail.gmail.com>
Message-ID: <d468gv$teo$1@sea.gmane.org>

Lars Yencken wrote:

> I'm trying to render a Japanese font to an image, but I'm having some problems.
>
> Converting the font from ttf to pil via ttf2bdf and pilfont.py works
> fine. When I try out the rendering code however, none of the Japanese
> characters appear, only the normal ascii ones do.
>
> Here's the offending code:
>
>    font = ImageFont.load('kochi-mincho.pil')
>    image = Image.new('RGB', (512,512), color=color__white)
>    draw = ImageDraw.Draw(image)
>    desiredText = unicode('nothing ??? appears', 'utf8').encode('euc_jp')
>    draw.text((10,10), desiredText, font=font, fill=color__black)
>    image.save('example.jpg')
>
> Notice that the actual japanese characters are originally in unicode
> and then encoded to euc_jp. I tried leaving them in unicode (which
> broke it), and selecting other possible encodings (like shift_jis or
> utf8) but these still only rendered the ascii, i.e. the above renders
> to "nothingappears".

can you make the BDF file available somewhere? (either post it somewhere
and mail me the URL, or, if it's not too large, mail the file to me).

</F> 


From lars.yencken at gmail.com  Thu Apr 21 03:22:02 2005
From: lars.yencken at gmail.com (Lars Yencken)
Date: Thu Apr 21 03:22:04 2005
Subject: [Image-SIG] Re: Rendering Japanese fonts
In-Reply-To: <d468gv$teo$1@sea.gmane.org>
References: <e36a4dac0504192303a92f11@mail.gmail.com>
	<d468gv$teo$1@sea.gmane.org>
Message-ID: <e36a4dac050420182256bfad54@mail.gmail.com>

Hi Fredrik,

On 4/21/05, Fredrik Lundh <fredrik@pythonware.com> wrote:
> Lars Yencken wrote:
> > I'm trying to render a Japanese font to an image, but I'm having some problems.
> >
> > Converting the font from ttf to pil via ttf2bdf and pilfont.py works
> > fine. When I try out the rendering code however, none of the Japanese
> > characters appear, only the normal ascii ones do.
> 
> can you make the BDF file available somewhere? (either post it somewhere
> and mail me the URL, or, if it's not too large, mail the file to me).

I've put it up at:
    http://www.csse.unimelb.edu.au/~lljy/kochi-mincho.bdf.bz2

Lars
From lars.yencken at gmail.com  Thu Apr 21 03:25:27 2005
From: lars.yencken at gmail.com (Lars Yencken)
Date: Thu Apr 21 03:26:28 2005
Subject: [Image-SIG] Rendering Japanese fonts
In-Reply-To: <79990c6b05042001355f30f3b4@mail.gmail.com>
References: <e36a4dac0504192303a92f11@mail.gmail.com>
	<79990c6b05042001355f30f3b4@mail.gmail.com>
Message-ID: <e36a4dac05042018254df608f1@mail.gmail.com>

Hi Paul,

On 4/20/05, Paul Moore <p.f.moore@gmail.com> wrote:
> I don't know if you are literally including Japanese characters in
> your source code, as you appear to be doing here, but unless you have
> an encoding declaration in your module, this won't work - by default,
> Python source code only handles ASCII characters.

For this test code, I am including the japanese characters right in,
but python was nice enough to warn me first time around when I forgot
to include the encoding. I'm using utf8 encoding, and I seem to be
able to manipulate these characters fine in other scripts. The
.encode() call should put in the appropriate escapes that are needed.

Thanks for checking though =)

Lars
From charlie at begeistert.org  Thu Apr 21 19:35:55 2005
From: charlie at begeistert.org (charlie clark)
Date: Thu Apr 21 19:36:13 2005
Subject: [Image-SIG] New to PIL and struggling...
Message-ID: <opsple55f0yt02yl@mail.isis.de>

Dear all,

I thought this would be quite easy but it isn't...

I would like to be able to mark areas on a map with round points. I 
already have the round points as a separate graphic with a transparent 
background. When I use Image.paste(im, (10, 10)) the paste works but the 
background of the pasted image is white rather than transparent. I think 
that I should be doing something with a mask but I don't quite understand 
how this works. Thanks for any tips.

Charlie

-- 
Charlie Clark
Helmholtzstr. 20
D?sseldorf
D-40215
Tel: +49-211-938-5360
GSM: +49-178-782-6226
From fredrik at pythonware.com  Thu Apr 21 19:49:10 2005
From: fredrik at pythonware.com (Fredrik Lundh)
Date: Thu Apr 21 19:49:54 2005
Subject: [Image-SIG] Re: New to PIL and struggling...
References: <opsple55f0yt02yl@mail.isis.de>
Message-ID: <d48oq8$g9r$1@sea.gmane.org>

"charlie clark" wrote:

> I would like to be able to mark areas on a map with round points. I 
> already have the round points as a separate graphic with a transparent 
> background. When I use Image.paste(im, (10, 10)) the paste works but the 
> background of the pasted image is white rather than transparent. I think 
> that I should be doing something with a mask but I don't quite understand 
> how this works. Thanks for any tips.

to do a transparent paste, you need to pass in the mask (or matte) as the
third argument.

if you're pasting RGBA data into an RGB image, you can simply pass in the
overlay itself as the third argument:

    image.paste(im, (10, 10), im)

if you're using P images, you have to create a mask first.  something like this
might work:

    lut = [1] * 256
    lut[im.info["transparency"]] = 0
    mask = im.point(lut, "1")
    image.paste(im, (10, 10), mask)

</F>

From charlie at begeistert.org  Fri Apr 22 16:31:41 2005
From: charlie at begeistert.org (charlie clark)
Date: Fri Apr 22 16:31:21 2005
Subject: [Image-SIG] (no subject)
Message-ID: <opspm1a3vyyt02yl@mail.isis.de>

"charlie clark" wrote:

> I would like to be able to mark areas on a map with round points. I 
> already have the round points as a separate graphic with a transparent 
> background. When I use Image.paste(im, (10, 10)) the paste works but the 
> background of the pasted image is white rather than transparent. I think 
> that I should be doing something with a mask but I don't quite 
> understand how this works. Thanks for any tips.

> to do a transparent paste, you need to pass in the mask (or matte) as the
> third argument.

> if you're pasting RGBA data into an RGB image, you can simply pass in the
> overlay itself as the third argument:

     image.paste(im, (10, 10), im)

Yes, I've tried this but it didn't make any difference.

> if you're using P images, you have to create a mask first.  something 
> like this
> might work:

     lut = [1] * 256
     lut[im.info["transparency"]] = 0

     mask = im.point(lut, "1")
     image.paste(im, (10, 10), mask)

Yes, it would when I understood it better! :-/

I'm working with .GIFs which as I understand it are P (palette) images. I 
had hoped that using convert("RGBA") would map the transparency to the 
alpha channel. But it didn't. I need to use RGB as otherwise the palettes 
of both images get mixed up.

I tried your example but got "images don't match" because I'd converted to 
RGBA abd im.point is expecting a Palette. Reloading and not converting the 
image and everything works fine. But I can't admit to understanding 
everything.

Trying to understand things: lut is effectively a palette table all set to 
the same value?
What does lut[im.info["transparency"]] = 0 do?
In my case its would sets lut[14] = 0
So the real magic is im.point() which creates a single band presumably 
reacting to the transparency in the original image.

Sorry for the really simple questions but even though I've now got this 
working, I like to understand how!

Thanks

Charlie
From aureli.soriafrisch at gmail.com  Fri Apr 22 21:21:01 2005
From: aureli.soriafrisch at gmail.com (Aureli Soria Frisch)
Date: Fri Apr 22 21:30:41 2005
Subject: [Image-SIG] JPEG problems by installing ...
Message-ID: <a06210203be8efd3c5981@[212.202.69.143]>

Hi all,

I am using PIL under Mac OS X and trying to display an image with:

im.show()

I got an error message telling me that the jpeg encoder is not 
available so I installed again PIL by making the changes in the 
setup.py (JPEG_ROOT="/sw/lib","/sw/include" (since I use the libjpeg 
from fink). Now the setup.py build gives me the message:

>running build_ext
>--------------------------------------------------------------------
>PIL 1.1.5 BUILD SUMMARY
>--------------------------------------------------------------------
>version       1.1.5
>platform      darwin 2.3 (#1, Sep 13 2003, 00:49:11)
>               [GCC 3.3 20030304 (Apple Computer, Inc. build 1495)]
>--------------------------------------------------------------------
>*** TKINTER support not available
>--- JPEG support ok
>--- ZLIB (PNG/ZIP) support ok
>--- FREETYPE2 support ok
>--------------------------------------------------------------------

but the selftest.py still launches an error on the availability of 
JPEG decoder...

>*****************************************************************
>Failure in example: _info(Image.open("Images/lena.jpg"))
>from line #24 of selftest.testimage
>Exception raised:
>Traceback (most recent call last):
>   File "./doctest.py", line 499, in _run_examples_inner
>     exec compile(source, "<string>", "single") in globs
>   File "<string>", line 1, in ?
>   File "./selftest.py", line 21, in _info
>     im.load()
>   File "PIL/ImageFile.py", line 180, in load
>     d = Image._getdecoder(self.mode, d, a, self.decoderconfig)
>   File "PIL/Image.py", line 328, in _getdecoder
>     raise IOError("decoder %s not available" % decoder_name)
>IOError: decoder jpeg not available
>1 items had failures:
>    1 of  55 in selftest.testimage
>***Test Failed*** 1 failures.
>*** 1 tests of 55 failed.

By doing im.show() still have the original error message that the jpg 
encoder is not available... I am quite lost and would appreciate any 
help...

Thanks in advance
Aureli

-- 
############################################
Aureli Soria Frisch

post: Stargarder Str. 73 SF, 10437 Berlin, Germany
e-mail: aureli.soria-frisch@ieee.org
fon: +49 (0) 174 574 79 74
web: http://klendathu.ipk.fhg.de/~aureli/
############################################
From bob at redivi.com  Fri Apr 22 21:55:27 2005
From: bob at redivi.com (Bob Ippolito)
Date: Fri Apr 22 21:55:41 2005
Subject: [Image-SIG] JPEG problems by installing ...
In-Reply-To: <a06210203be8efd3c5981@[212.202.69.143]>
References: <a06210203be8efd3c5981@[212.202.69.143]>
Message-ID: <d865ebdab3d1fac06b7bf2fa00ca3f3b@redivi.com>


On Apr 22, 2005, at 3:21 PM, Aureli Soria Frisch wrote:

> I am using PIL under Mac OS X and trying to display an image with:
>
> im.show()
>
> I got an error message telling me that the jpeg encoder is not 
> available so I installed again PIL by making the changes in the 
> setup.py (JPEG_ROOT="/sw/lib","/sw/include" (since I use the libjpeg 
> from fink). Now the setup.py build gives me the message:

Why aren't you using the Mac OS X packages from 
http://pythonmac.org/packages/ ?

-bob

From aureli.soriafrisch at gmail.com  Sat Apr 23 15:32:14 2005
From: aureli.soriafrisch at gmail.com (Aureli Soria Frisch)
Date: Sat Apr 23 22:16:33 2005
Subject: [Image-SIG] JPEG problems by installing ...
In-Reply-To: <d865ebdab3d1fac06b7bf2fa00ca3f3b@redivi.com>
References: <a06210203be8efd3c5981@[212.202.69.143]>
	<d865ebdab3d1fac06b7bf2fa00ca3f3b@redivi.com>
Message-ID: <a06210202be8ffebd4c1a@[212.202.69.143]>


>On Apr 22, 2005, at 3:21 PM, Aureli Soria Frisch wrote:
>
>>I am using PIL under Mac OS X and trying to display an image with:
>>
>>im.show()
>>
>>I got an error message telling me that the jpeg encoder is not 
>>available so I installed again PIL by making the changes in the 
>>setup.py (JPEG_ROOT="/sw/lib","/sw/include" (since I use the 
>>libjpeg from fink). Now the setup.py build gives me the message:
>
>Why aren't you using the Mac OS X packages from 
>http://pythonmac.org/packages/ ?


I didn't know about... everything is now OK.

Thank you very much

Best Regards,
Aureli
-- 
############################################
Aureli Soria Frisch

post: Stargarder Str. 73 SF, 10437 Berlin, Germany
e-mail: aureli.soria-frisch@ieee.org
fon: +49 (0) 174 574 79 74
web: http://klendathu.ipk.fhg.de/~aureli/
############################################