On 25 October 2015 at 22:34, Lubomir I. Ivanov <[email protected]> wrote: > On 25 October 2015 at 22:08, Lubomir I. Ivanov <[email protected]> wrote: >> On 25 October 2015 at 17:28, Dirk Hohndel <[email protected]> wrote: >>> >>> Lubomir - this does sound vaguely like the problem you fixed before, but >>> even the Windows 4.5.0 binary was built against a libzip that contained >>> your fix... >>> >>> Summary: I'm at a loss what's failing this time. And I'm a bit dismayed >>> that our code does such a poor job at checking for errors and trying to >>> give reasonable error messages. >>> >> >> i forgot the divelogs.de credentials for testing, what were they? >> you can send them to me in a private email. >> > > the upload works fine for me with libzip 56d3e8a2b3a8 (which is an > older one) - trying with the latest libzip and zlib now.
alright, apparently normal human beings won't be able to build the latest libzip (ede515cd6a7a) /zlib (1.2.8) pair without modifying cmake and makefiles on native Windows anymore. i managed to solve these issues and to be able to put some sort of a patched-build... i'm uploading dives to divelogs.de without any problem on Win7 with the latest libzip/zlib, so i cannot reproduce the issue. for Windows, the ZIP_ER_RENAME error can only be throw in the _win32_read_file()...if this only happens on Win10 there could be a regression in the WINAPI for Win10, but i seriously doubt that. i have no solution ATM and i need to ask on the mailing list if they have tried Win10 at all and to stare the libzip source for a while. said source code is very convoluted and there are only 3 people or so, who really understand it. lubomir -- _______________________________________________ subsurface mailing list [email protected] http://lists.subsurface-divelog.org/cgi-bin/mailman/listinfo/subsurface
