Legacy ImageMagick Discussions Archive

Sorry, no I didnt get the time allocated.

It seems that newer IM has slightly improved support, according to the changelog.

Thanks for your reply.

You are absolutely right, using +ping makes newest IM also identify all 3 layers in the original file.

$ /usr/local/im/im6.5.4/bin/identify +ping nem_154049_fc_JNSMALL.psd
nem_154049_fc_JNSMALL.psd[0] PSD 248x239 248x239+0+0 8-bit DirectClass 346kb
nem_154049_fc_JNSMALL ...

Hi rvanderkooy,
Apperently never :-|

However I have posted the Java code. And its working great for us.

Anyone familiar with C and IM could insert this code in IM source and create a patch.

So, the real question migt be: When will someone pay a C prof do it.
Im sure IM would accept a patch ready ...

See also
viewtopic.php?f=3&t=10410

Hi IonSwitz

This post about PSD might be of interest you too:
http://www.imagemagick.org/discourse-server/viewtopic.php?f=2&t=14103

I am hoping I get some time to dive into psd.c by end August.

Perhaps we could do some co-development to fix some PSD file reading/writing bugs?

Yours,
Jacob ...

Using ImageMagick 6.4.5 2009-06-04 Q16 OpenMP on Ubuntu Linux and a PSD file with 3 layers:
(availabable at http://javabog.dk/filer/psd/nem_154049_fc_JNSMALL.psd)

$ convert nem_154049_fc_JNSMALL.psd x.psd
$ convert x.psd y.psd
convert: maximum channels exceeded `x.psd' @ coders/psd.c/ReadPSDImage ...

So, the problem is
- when I use the cmd: convert test_cmyk.jpg test_cmyk.psd
then the cmyk icc-profile in test_cmyk.jpg is gone in test_cmyk.psd

identify -verbose test_cmyk.jpg gives:
Profile-icc: 557168 bytes
U.S. Web Coated (SWOP) v2

this is gone in test_cmyk.psd (also on Linux using ...

We would definitely use it.
As its not available now we use ImageJ ( http://rsbweb.nih.gov/ij/ ) for now.

I repeat my question:
magick, please please clarify what you meant by the

'Ok, easy fix. We need to propagate the extra samples type, in this case associated alpha.'

and how to do this also on new images.
What code did you change? What attributes (og whatever?) was too much/missing in the TIFF ...

magick, please please clarify what you meant by the

'Ok, easy fix. We need to propagate the extra samples type, in this case associated alpha.'

and how to do this also on new images.
What code did you change? What attributes (og whatever?) was too much/missing in the TIFF image to make it Adobe ...

On the same image saved (from Photoshop) as TIFF and JPG image,
the following difference in identify -verbose is shown:


on the JPG image identify -verbose shows:
Profile-8bim: 74358 bytes
unknown[2,0]:
unknown[109,102]:
unknown[0,98]:
unknown[31,0]: b
unknown[0,192]:
Profile-exif: 2447 ...

If convert raw file from f.ex. a canon eos 350D(CR2-file) to a jpg
file by using Camera raw from Adobe Photoshop and then use the command:

identify -verbose _mg_3100.jpg

I get this result all these exif data:

Exif:ApertureValue: 6643856/1000000
Exif:Artist: unknown
Exif:CustomRendered: 0 ...

When converting from RAW format, for example

convert CR2_MG_2959.CR2 x.tif

the image gets very distorted in its colors (using ImageMagick 6.3.8 02/16/08 Q8).
I suppose its becaurse color depth is interpreted wrongly.

I noticed that IM is using this comand behind the scenes to create e PNM image ...

I repeat my request (with hopes of a reaction):

Could you please state the correct names to be used in this method call?

Thank you
Jacob Nordfalk

Sorry, a small bug had sneaked in in case of a pathname with oneven length.

I have edited the previous post to include de fix:
if (count % 2 == 1) n--; //XXX NEW


And regarding to

I really hope you cat get the time to get the fix into the C code at some time. Please let me know if you do ...