identify/convert give wrong size of eps file

[mexas@bristol.ac.uk]

I use ImageMagick-6.7.5.7 on FreeBSD 10 -current.

I've a large eps file, converted from a raster image:

% ls -al bile_acid.eps
-rwxr-xr-x 1 mexas wheel 3428323 Mar 22 2011 bile_acid.eps

% identify bile_acid.eps
bile_acid.eps PS 595x842 595x842+0+0 16-bit DirectClass 43.6KB 0.000u 0:00.007
^^^^^^^
The size is wrong:

% head bile_acid.eps
%!PS-Adobe-3.0 EPSF-3.0
%%Creator: GIMP PostScript file plugin V 1.17 by Peter Kirchgessner
%%Title: bile_acid.eps
%%CreationDate: Tue Mar 22 19:44:00 2011
%%DocumentData: Clean7Bit
%%LanguageLevel: 2
%%Pages: 1
%%BoundingBox: 14 14 1599 2379
^^^^^^^^^^^^^^^

The size returned by identify doesn't correspond
to the size given in the eps itself.

When I open the eps file with gv, the whole of
the file is shown and the size is displayed correctly.

When I convert the eps file to something else,
only a 595x842 portion of the original image is converted,
the rest is ignored:

% convert bile_acid.eps bile_acid.jpeg
% identify bile_acid.*
bile_acid.eps PS 595x842 595x842+0+0 16-bit DirectClass 43.6KB 0.000u 0:00.015
bile_acid.jpeg[1] JPEG 595x842 595x842+0+0 8-bit DirectClass 24.5KB 0.000u 0:00.000
%

The source eps: http://seis.bris.ac.uk/~mexas/bile_acid.eps
The resulting jpeg: http://seis.bris.ac.uk/~mexas/bile_acid.jpeg

[anthony]

Postscript images do not have a size! The Bounding box coordinates are not pixels, they are 'points' which is defined as being 72 points per inch.

You need to set a density or resolution which then determines the final size of the image.

[mexas@bristol.ac.uk]

ok, I didn't phrase my report correctly, apologies.

The fact is that I cannot convert this eps to another
format, e.g. jpg, png fully. Only a fragment at the bottom
left corner is converted. About 3/4 of the image is lost
during conversion.

% identify z*
z.eps PS 595x842 595x842+0+0 16-bit DirectClass 43.6KB 0.000u 0:00.000
z.jpg[1] JPEG 413x585 413x585+0+0 8-bit DirectClass 14.5KB 0.000u 0:00.000

% ls -al z*
-rwxr-xr-x 1 mexas wheel 3428323 Mar 10 19:36 z.eps
-rw-r--r-- 1 mexas wheel 14515 Mar 10 19:42 z.jpg

Note that the file size reported by identify for this eps is
clearly wrong: 43.6KB, whereas the real size is >3MB.

The original eps: http://seis.bris.ac.uk/~mexas/z.eps
Converted to jpg: http://seis.bris.ac.uk/~mexas/z.jpg

Can anybody reproduce this?

[fmw42]

identify -verbose z.eps

identify: no decode delegate for this image format `1z.eps' @ error/constitute.c/ReadImage/532.

I would say your eps is corrupt.

[anthony]

More likely it is just missing the EPS commant at the top of the file.

Some programs like to out put a Crtl-D character at the start of Postscript files (for printer reset) So if you image starts with the three characters Ctrl-D % ! remove the Ctrl-D!

[mexas@bristol.ac.uk]

Seems it was a bug.
I updated to Version: ImageMagick 6.7.5-10 2012-03-12 Q16
and the issue has gone away:

% identify z.eps
z.eps PS 1585x2365 1585x2365+0+0 16-bit DirectClass 784KB 0.008u 0:00.007

% identify -verbose z.eps
Image: z.eps
Format: PS (PostScript)
Class: DirectClass
Geometry: 1585x2365+0+0
Resolution: 72x72
Print size: 22.0139x32.8472
Units: Undefined
Type: TrueColorMatte
Endianess: Undefined
Colorspace: sRGB
Depth: 16/8-bit
Channel depth:
red: 8-bit
green: 8-bit
blue: 8-bit
alpha: 1-bit
Channel statistics:
Red:
min: 0 (0)
max: 65535 (1)
mean: 60852.6 (0.928551)
standard deviation: 12060 (0.184024)
kurtosis: 13.6921
skewness: -3.73598
Green:
min: 0 (0)
max: 65535 (1)
mean: 60811.9 (0.92793)
standard deviation: 12206.4 (0.186258)
kurtosis: 13.7476
skewness: -3.74757
Blue:
min: 0 (0)
max: 65535 (1)
mean: 60672.3 (0.9258)
standard deviation: 12303.7 (0.187743)
kurtosis: 13.431
skewness: -3.69447
Alpha:
min: 65535 (1)
max: 65535 (1)
mean: 65535 (1)
standard deviation: 0 (0)
kurtosis: 0
skewness: 0
Image statistics:
Overall:
min: 0 (0)
max: 65535 (1)
mean: 45584.2 (0.69557)
standard deviation: 10557.2 (0.161093)
kurtosis: 96.9406
skewness: -17.8735
Rendering intent: Perceptual
Gamma: 0.45455
Chromaticity:
red primary: (0.64,0.33)
green primary: (0.3,0.6)
blue primary: (0.15,0.06)
white point: (0.3127,0.329)
Interlace: None
Background color: white
Border color: srgba(223,223,223,1)
Matte color: grey74
Transparent color: none
Compose: Over
Page geometry: 1585x2365+0+0
Dispose: Undefined
Iterations: 0
Compression: Undefined
Orientation: Undefined
Properties:
date:create: 2012-03-12T16:33:04+00:00
date:modify: 2012-03-12T16:33:04+00:00
ps:HiResBoundingBox: 1585x2365+14+14
ps:Level: Adobe-3.0 EPSF-3.0

signature: cfde116243dbffe52ffebc59b974ccd53252000b398307f724a6023282d72b03
Profiles:
Profile-icc: 2576 bytes
Description: Artifex Software sRGB ICC Profile
Manufacturer: Artifex Software sRGB ICC Profile
Model: Artifex Software sRGB ICC Profile
Copyright: Copyright Artifex Software 2011
Artifacts:
filename: z.eps
verbose: true
Tainted: False
Filesize: 784KB
Number pixels: 3.749M
Pixels per second: 6.397MB
User time: 0.555u
Elapsed time: 0:01.585
Version: ImageMagick 6.7.5-10 2012-03-12 Q16 http://www.imagemagick.org
%

and I don't see Ctrl-D in the header:

% head -n 30 z.eps

%!PS-Adobe-3.0 EPSF-3.0
%%Creator: GIMP PostScript file plugin V 1.17 by Peter Kirchgessner
%%Title: bile_acid.eps
%%CreationDate: Tue Mar 22 19:44:00 2011
%%DocumentData: Clean7Bit
%%LanguageLevel: 2
%%Pages: 1
%%BoundingBox: 14 14 1599 2379
%%EndComments
%%BeginProlog
% Use own dictionary to avoid conflicts
10 dict begin
%%EndProlog
%%Page: 1 1
% Translate for offset
14.173228346456694 14.173228346456694 translate
% Translate to begin of first scanline
0 2364.730613263971 translate
1584.7370078740157 -2364.730613263971 scale
% Image geometry
2113 3153 8
% Transformation matrix
[ 2113 0 0 3153 0 0 ]
% Strings to hold RGB-samples per scanline
/rstr 2113 string def
/gstr 2113 string def
/bstr 2113 string def
{currentfile /ASCII85Decode filter /RunLengthDecode filter rstr readstring pop}
{currentfile /ASCII85Decode filter /RunLengthDecode filter gstr readstring pop}
{currentfile /ASCII85Decode filter /RunLengthDecode filter bstr readstring pop}

and convert now works as expected:

% convert z.eps z.jpg
% identify z*
z.eps PS 1585x2365 1585x2365+0+0 16-bit DirectClass 784KB 0.000u 0:00.007
z.jpg[1] JPEG 1585x2365 1585x2365+0+0 8-bit DirectClass 371KB 0.008u 0:00.000