(spoiler: i have read only the mail subject.) Is this Battlestar Galactica or something? :) "All this has happened before, and all of it will happen again."
https://www.mail-archive.com/darktable-dev%40lists.darktable.org/msg00344.html Please do actually read what is written there, and do understand it. There is absolutely nothing to add to that mail. And nothing has changed since it was written, all and every point still applies. Roman. On Mon, Oct 3, 2016 at 9:56 PM, Jan Ingwer Baer <jib...@web.de> wrote: > Hi, > > i have done some work to build darktable for windows. Now i can build it > (without maps and lua) on windows and first test show no problems. I have > done this in first for my private use. But i share this, so other can test > and use it. At this time it is only for developres, and not ready for > end-users. > > I hope that i can find enough other interested developers who can help to > test and stabilize it. > > You find my changes i the branch 'darktable-2.0.x-win' in my github fork > 'https://github.com/jibaer/darktable.git'. > > While building the windows version i have found some issues in the > darktable-source: > > - In common/film.c function film_recursive_get_files() there is a mistake in > the call to function g_build_filename(). It is called with the same > parameters as g_build_pathname(), but according to the glib manual it does > not need the first parameter (G_DIR_SEPARATOR_S). Under windows this will > produce an invalid filename with a leading slash (like /c:/user/...). Under > *ix this no problem because it will simple double the leading slash of an > absolute path what the os ignores. > > - In common/darktable.c function strip_semicolons_from_keymap() there is > only a test for the result of fgetc() !=eof. But on mingw this doesnt work, > and it must test for feof() ==0. Without the test for feof() the result is > an infinite loop... > > - With gcc 6.x the search-path for include-files is very important and the > includes <cmath> and <cstdlib> will not work if the default search order for > the system-include-directorys is changed. Because the cmake-files insert > some include-paths with the system-attributes they change the > default-search-order and the build will fail with the error : file > <stdlib.h> not found. On my mingw-build i have found a solution in an edit > to src/CMakeLists.txt that avoids to do use the system-attribute while > adding the include-path for PThreads. > > I hope you will accept my contribution and dont take it hostile. And with > the help from more developers some day there will be an official version of > darktable for windows. To use another raw-photo application is not an > alternative because DT has at least two outstanding features: > - The profile-based noise-reduction is really great > - The mask feature is also really great > > Greetings from Berlin > > Jan Ingwer Baer > ___________________________________________________________________________ > darktable developer mailing list > to unsubscribe send a mail to darktable-dev+unsubscr...@lists.darktable.org > ___________________________________________________________________________ darktable developer mailing list to unsubscribe send a mail to darktable-dev+unsubscr...@lists.darktable.org
