| <<<Back 1 day (to 2017/11/08) | 20171109 |
aleray | hi, could you help me understand this error: http://dpaste.com/32VA351 | 21:20.17 |
| seems like vim expands tabs | 21:29.46 |
| Trying to follow the answer here: https://stackoverflow.com/questions/35596346/ghostscript-convert-pdf-into-cmyk-preserving-pure-black-for-text#35646895 | 21:49.27 |
| I'd linke to convert rgb pdf to cmyk while preserving pure black for text | 21:49.44 |
| in my pdf the text is vetorized. It is thus graphics I guess | 21:50.04 |
| I'm starting from a pdf 1.4 generated by QTWebkit | 21:50.25 |
| it seems to use srgb | 21:50.32 |
| so I used collink like this: collink -v -f sRGB2014.icc FOGRA27L_coated.icc srgb2fogra27-neutrals.icc | 21:51.04 |
| since it is all mysterious to me I'm wondering if my command is right | 21:51.34 |
| then I used this command: env gs -q -sDEVICE=pdfwrite -o out.pdf -sColorConversionStrategy=CMYK -sSourceObjectICC=control.txt 026-027-edm.pdf | 21:52.06 |
pipitas | Hi @artifex-people | 21:54.51 |
| I was looking at https://www.artifex.com/mupdf-technical-information/#sub-item-5 | 21:55.12 |
| Congratulations, it's a really nice page. Didn't visit you there for a long time. When did you publish this new design? | 21:55.53 |
| Anyway â I wanted to ask you: https://www.artifex.com/mupdf-technical-information/#sub-item-5 â didn't you accidentally swap the columns for 600dpi and 1200dpi? Or why is 1200dpi faster than 600dpi? | 21:57.06 |
aleray | found some info here: https://ghostscript.com/irclogs/2016/01/29.html but seems like it didn't work in the end | 21:58.31 |
| or that fellow didn't at least | 21:58.57 |
| mvrhel_laptop, hello. Some idea about that maybe? ^^ | 22:08.10 |
mvrhel_laptop | aleray: So currently it the source color spaces are DeviceRGB rather than ICC Color spaces, pdfwrite will not do the conversion that you want | 22:09.37 |
aleray | mvrhel_laptop, ok. Is there anything to do then? | 22:10.19 |
mvrhel_laptop | I am in some discussions with the primary engineer who manages pdfwrite about making it possible to always do these conversions | 22:10.27 |
| aleray: Do you know if all your source color spaces are ICC based or are they DeviceRGB etc | 22:10.57 |
pipitas | Another question: Shouldn't you add the licensing info in relation to the Linux platform to the table in https://www.ghostscript.com/download/mupdfdnld.html ? | 22:12.47 |
aleray | mvrhel_laptop, don't know much about colors. I read https://ghostscript.com/doc/current/GS9_Color_Management.pdf but it helped so far that it gave me hope I could manage :) I don't think there is any kind of color management in the source pdf. They are generated via #qtwebkit. | 22:13.15 |
| mvrhel_laptop, here is the output of pdfinfo: http://dpaste.com/2X3HK13 | 22:13.43 |
mvrhel_laptop | aleray, I would need to look at the actual file | 22:14.02 |
aleray | ok. just a moment. | 22:14.37 |
| (and here is the output of identify: http://dpaste.com/1J8C0Z7) | 22:14.59 |
mvrhel_laptop | aleray, If I get a bit of time, I will make s small patch and see if Ken would be happy with it. It seems that there is quite a bit of interest in having this level of control in pdfwrite | 22:15.04 |
pipitas | More hints about your new website: On page https://www.ghostscript.com/download/mupdfdnld.html there is a link at the bottom saying "Old releases available here:Â http://mupdf.com/download/archive". That link does not lead anywhere but to an error message "Not Found". | 22:16.10 |
aleray | mvrhel_laptop, my test file in pm | 22:17.12 |
pipitas | hopes that his hints will be read later by anyone of chrisl, Robin_Watts, tor8, mvrhel_laptop who then would inform the webmaster⦠| 22:17.33 |
aleray | mvrhel_laptop, the identify command mentions "sRGB colorspace" and "icc:model: Artifex Software sRGB ICC Profile" and "Profile-icc: 2576 bytes" | 22:22.26 |
| mvrhel_laptop, Thanks. I Know many people who'd be happy with finding a solution to that :) | 22:23.46 |
pipitas | All: forget my (stupid) remark about swapped columns on https://www.artifex.com/mupdf-technical-information/#sub-item-5Â . I had not seen it is probably PPM. Thought it was milliseconds in execution time, or something | 22:42.11 |
| Forward 1 day (to 2017/11/10)>>> | |