> On May 23, 2015, at 1:42 AM, Geert Janssens <geert.gnuc...@kobaltwit.be> > wrote: > > On Saturday 23 May 2015 10:34:44 Geert Janssens wrote: >> On Friday 22 May 2015 23:59:44 John Ralls wrote: >>> Updated via https://github.com/Gnucash/gnucash/commit/82329272 >>> (commit) from https://github.com/Gnucash/gnucash/commit/847e2c27 >>> (commit) >>> >>> >>> >>> commit 82329272048863c21ff286b5e79516896fab6db3 >>> Author: John Ralls <jra...@ceridwen.us> >>> Date: Fri May 22 19:50:25 2015 -0700 >>> >>> Fix compile failures on latest version of MinGW. >>> >>> Summary of changes: >>> src/app-utils/guile-util.c | 5 +++++ >>> src/backend/dbi/Makefile.am | 1 + >>> src/backend/dbi/gnc-backend-dbi.c | 5 +++++ >>> src/backend/xml/gnc-backend-xml.c | 5 +++++ >>> src/backend/xml/io-example-account.c | 5 +++++ >>> src/backend/xml/io-gncxml-v2.c | 5 +++++ >>> src/core-utils/binreloc.c | 5 +++++ >>> src/core-utils/gnc-filepath-utils.c | 5 +++++ >>> src/core-utils/gnc-gkeyfile-utils.c | 5 +++++ >>> src/engine/Query.c | 5 +++++ >>> src/engine/Split.c | 5 +++++ >>> src/engine/Transaction.c | 5 +++++ >>> src/gnome/assistant-hierarchy.c | 5 +++++ >>> src/html/gnc-html-webkit.c | 6 +++++- >>> src/html/gnc-html.c | 5 +++++ >>> src/import-export/aqb/assistant-ab-initial.c | 5 +++++ >>> src/import-export/aqb/gnc-file-aqb-import.c | 5 +++++ >>> src/import-export/csv-imp/gnc-csv-model.c | 4 ++++ >>> src/import-export/qif-imp/assistant-qif-import.c | 5 +++++ >>> src/libqof/qof/guid.cpp | 5 +++++ >>> src/libqof/qof/qoflog.cpp | 8 +++++++- >>> src/libqof/qof/qofsession.cpp | 5 +++++ >>> 22 files changed, 107 insertions(+), 2 deletions(-) >>> >>> _______________________________________________ >>> gnucash-patches mailing list >>> gnucash-patc...@gnucash.org >>> https://lists.gnucash.org/mailman/listinfo/gnucash-patches >> >> John, >> >> Thanks for working on this. >> >> This patch is puzzling me. I see you are adding this code snippet to >> several source files: >> >> #include <platform.h> >> #if PLATFORM(WINDOWS) >> #include <windows.h> >> #endif >> >> What's particular about these files that they require this added >> snippet ? > > Follow up on my own question: I found yesterday's conversation between > jralls and warlord on irc on this topic: > http://lists.gnucash.org/logs/2015/05/2015-05-22.html#T21:56:25 > >> >> And secondly, it surprises me that these changes were not needed on >> the maint branch. The build server managed to build gnucash maint >> without issues on May 20, also using the new mingw-rt package. >> > This remains a mystery so far. >
This is http://ehc.ac/p/mingw/bugs/2256/ As Marshall points out, the work-around is bogus, but it’s still necessary unless we roll back to a mingwrt without the bad unistd.h. It’s interesting, though, that it indeed doesn’t seem to affect maint. I just ran a build of that locally to check that, and it’s going just fine. The only reason I can think of is that it’s tripped by the C++ interface headers somehow. Regards, John Ralls _______________________________________________ gnucash-devel mailing list gnucash-devel@gnucash.org https://lists.gnucash.org/mailman/listinfo/gnucash-devel