There was no way the account was informed about the lot being deleted.
This was resulting into invalid lot lists in an account while scrubbing.
Which would bomb at later stages like when re-running check & repair or
opening the lot viewer. Also saving would report a number of critical messages
about invalid objects, but these didn't result in a segfault.
day's date. Previous efforts didn't work on Windows and shouldn't have
elsewhere because gnc_localtime and gnc_mktime are inverse functions;
in particular gnc_mktime doesn't adjust the time based on the value
of tm_isdst, though some implementations of STDC mktime do.
Rather than being too clever about trying to find the DST offset this
implementation just uses 1 hour.
The Python includes files for Windows define HAVE_PUTENV to be
empty. GnuCash under MINGW (only) defines it to be 1. This causes
redefinition errors when building the Python bindings on Windows.
HAVE_PUTENV is not used in the GnuCash code base, so the best
solution seems to be to just get rid of it.
Use the date-time functions in gnc-date instead of the guile ones for more consistent behaviour.
Do the same for test test-cash-flow, which was also time-of-day sensitive
This commit makes the test work in all timezones from -11 to +10. Timezone +11 still fails, but that's the compromise we have made in the rest of gnucash as well.
This was an ugly work-around for a bogus warning from gcc-3, a
long-superceded compiler. The way the macro was written caused
portablility issues (see PR 102) and other projects (gstreamer
in particular) removed it several years ago.
Thanks to Julian Ospald for bringing it to our attention.
The fundamental problem is https://sourceforge.net/p/libofx/bugs/39.
Work around it by detecting it at configure time and if present check
to see if the time in question is not daylight time. If that's the case,
determine the daylight time offset and add it back to the timestamp.
The default treeview sort order when created is the name column which
happens to be column 0. When saving the sort order, column 0 is checked
to see if it is the sort column and if it is the sort order is not saved
Changed to check for the column not being 'name' and then save the order
* Add ability to either create SWIG C files or use generated ones
depending on BUILDING_FROM_VCS.
* Add complete support for compiling Python bindings using
WITH_PYTHON=YES.
* Add CMakeLists.txt files to a couple more directories.
Comparing only on the strlen of the imported commodity produces false
matches when the account commodity string begins with the same
characters but is longer. E.g. if the imported commodity is F (Ford
Motor) and Facebook (FB) is encountered first in the account list then
the matcher will select Facebook, but that will fail when the commodity
GUIDs are compared.
On OpenSuSe. They've taken to using libdbi's git repo instead of the last
tarball (0.9.0 in March 2013) and there's a significant change in the
data format.
Oddly, this was fixed in master as part of building the file as C++. Even
more oddly no C compiler caught the type error for 6 years and it's not clear
to me what happened to make clang notice it now.
