Legacy ImageMagick Discussions Archive

In some environment or older version, I have experienced identify generate extra newline at the end. So "head -1" is safer. In Windows CMD (without extra install), it becomes tricky, > identify image.tiff | findstr /v /r "^$" | find /c /v "" (findstr is for sure to remove an empty line.) Its unix ...

fmw42 wrote:just use identify on the first page

Code: Select all

identify -format "%n\n" yourimage.tiff[0]

It returns always 1 (in 6.9.1-10 on Windows)

I don't know when it changes, but even if I use older version, only "%n" was special - "%n\n" or even "%n " produced multiple output. Probably developer has changed it for keeping consistency. [EDIT] I checked 6.7.6-3 on Cygwin, 6.5.4-7 on CentOS. And for tiff, check this also. http://www ...

I rechecked after newly installed to C: What does convert -list format | grep PNG identify -list format | grep PNG say? Are PNG00, PNG48, and PNG64 among the listed formats for both commands? I suppose they are, because you said that PNG48 worked for you when directing the output to a file. Yes, all ...

c:\>set MAGICK_TMPDIR=C:/WINDOWS/TEMP c:\>convert -version Version: ImageMagick 6.9.1-10 Q16 x86 2015-07-25 http://www.imagemagick.org Copyright: Copyright (C) 1999-2015 ImageMagick Studio LLC License: http://www.imagemagick.org/script/license.php Features: Cipher DPC HDRI Modules OpenMP Delegates ...

dlemstra wrote:The property will be set on the image and all its layers so that means you can get something like this:

truetruetrue

I am aware of it. Here, I wanted to show convenient one-liner, (although tail is unix specific.)

I understand why. So checking multi page or multi flame is archived easier. Thank you again. identify -quiet -format "%n %[tiff:has-layers]\n" unknown.tif | tail -1 outputs: 1 for single-page & no layer tiff; 2 for 2 page tiff; 3 true for 3 layer tiff. (Indeed 2 layers. As [0] is automatically added ...

It looks OK if provided multi-layer tiff, but with 6.9.1-10 >identify -format "%[tiff:has-layers]\n" Hello.tif identify.exe: Incompatible type for "RichTIFFIPTC"; tag ignored. `TIFFFetchNormalTag' @ warning/tiff.c TIFFWarnings/856. identify.exe: unknown image property "%[tiff:has-layers]" @ warning ...

snibgo wrote:There may be something funny in your text file. With a file that contains ASCII 32 to 256, plus a new-line every 32 characters (file http://snibgo.com/imforums/ALPHABET.TXT )

Your ALPHABET.TXT binary seems to be broken at 0x91 0x92 0x93 0x94.

Took the new lines out, saved it from NotePad with encoding: ANSI, ran your convert and got what appears to be the desired glyphs, that is *without* anything prepended. Now I'll try to make a strike font out of it. If your charset range is limited to 0x0000 - 0x00FF, that would be OK. [EDIT] At a ...

dnorton wrote:There seems to be something prepended but I can deal with that.

It must be BOM. Try to save the text UTF-8 without BOM.

If you save text as UTF-8 everything should work fine.
I guess your text is UTF-16 LE or something like.

6.9.1-9 Q16 on Windows: convert -background transparent -resize 1024x768 -depth 8 -fill "rgba(170,170,170,184)" -font Arial -pointsize 50 label:"Test Stamp" PNG:- | identify -verbose - ... Colors: 257 Histogram: 140491: ( 0, 0, 0, 0) #00000000 graya(0,0) 1034: (170,170,170,255) #AAAAAAFF graya(170,1 ...

Your input pdf is CMYK. Output is RGB.
Current release of ImageMagick (6.9.1-9) does not change colorspace with your command, i.e. CMYK jpg is generated. So if you want CMYK jpg, upgrade is recommended.

Looks like the old sRGB versus linear problem is contributing to this. Ah, so if the text is black the problem doesn't occur (checked with 6.9.1-3 on Cygwin). So workaround may be: convert -background transparent -resize 1024x768 -fill '#000000B8' -font Arial -pointsize 50 label:'Test Stamp' -alpha ...