Glade 3 does not save properties with default values anymore. This
reduces patch sizes when Glade-3/Gtk+ is used with varying versions and
also avoids warnings about properties unknown to older Gtk+ versions.
The version Glade 3 used is 3.4.5.
git-svn-id: svn+ssh://svn.gnucash.org/repo/gnucash/trunk@17966 57a11ea4-9604-0410-9ed3-97b8803252fd
Without the fallback, NULL is printf'ed which crashes GnuCash on
Windows.
BP
git-svn-id: svn+ssh://svn.gnucash.org/repo/gnucash/trunk@17965 57a11ea4-9604-0410-9ed3-97b8803252fd
Take note: this change will *hide* any manually hacked saved reports
that make use of the new report-guid. The new saved-reports-2.4 file
will shadow the old saved-reports-2.0 file and will leave those hacked
reports behind. Any pre-existing saved reports that use report-guid will
have to be manually migrated to the new file.
git-svn-id: svn+ssh://svn.gnucash.org/repo/gnucash/trunk@17942 57a11ea4-9604-0410-9ed3-97b8803252fd
However, libdbi uses the column type name to determine how big it thinks the integer
value is. Therefore, change all 64 bit integer values (including num/denom in numeric
values) to 'bigint'. Tables will be automatically upgraded when an sqlite3 file is
opened.
git-svn-id: svn+ssh://svn.gnucash.org/repo/gnucash/trunk@17927 57a11ea4-9604-0410-9ed3-97b8803252fd
doesn't matter for the file backend, but does for the SQL backend, because the SQL backend
only loads transactions/splits as needed.
2) Partially solve problem where int64 values are loaded incorrectly. This started because
num and denom for prices were seen to be limited to 32 bits. libdbi provides a
libdbi_result_get_longlong() routine, but unfortunately, on my system (with libdbi 0.8.2-3),
it seems to have some bugs a) it returns negative values unsigned and b) it seems limited to
a 32 bit value even when the correct value is in the db.
git-svn-id: svn+ssh://svn.gnucash.org/repo/gnucash/trunk@17925 57a11ea4-9604-0410-9ed3-97b8803252fd
1. create a transaction that has two or more splits in the same account
2. open a register on that account that is in either "Basic Ledger" or
"Auto-Split Ledger" mode
3. change some transaction other than the one with multiple splits in this
register
4. click on the second or succeeding occurrence of the transaction with
multiple splits in this register
This will put the cursor on the first occurrence of the transaction instead of
the one you clicked on.
Patch by Mike Alexander.
BP
git-svn-id: svn+ssh://svn.gnucash.org/repo/gnucash/trunk@17920 57a11ea4-9604-0410-9ed3-97b8803252fd
Otherwise, the scheduled transaction won't be created.
git-svn-id: svn+ssh://svn.gnucash.org/repo/gnucash/trunk@17916 57a11ea4-9604-0410-9ed3-97b8803252fd
gnc_import_TransInfo_delete() was defined but never called!
This led to a memory leak, and erroneous behaviour.
Bug 552275 - Cancelling OFX Import Still Adds Entries
Bug 434944 - Refused transactions from import appear in register
Patch by Alan Jenkins <alan-jenkins@tuffmail.co.uk>
BP
git-svn-id: svn+ssh://svn.gnucash.org/repo/gnucash/trunk@17891 57a11ea4-9604-0410-9ed3-97b8803252fd
Affects all users of the generic import: OFX, HBCI and AqBanking.
Suggested by <http://bugzilla.gnome.org/show_bug.cgi?id=139310>.
"
Assume you have a transaction that transfers money from account 1 to account 2.
Import an OFX statement from account 1 that matches the transaction. Then
import an OFX statement from account 2 that matches the same transaction.
Because each transaction can only store one online_id, when you import from
account 1 again it doesn't remember that it's already been imported.
If online_id was stored in a split slot instead of a transaction slot, it would
allow tracking the online_id of both accounts without being overwritten.
"
For backwards compatibility, we still check for online_id in transactions.
This is more of a convenience than a critical feature, however.
The compatibility code could be removed in future versions if desired.
Patch by Alan Jenkins <alan-jenkins@tuffmail.co.uk>
git-svn-id: svn+ssh://svn.gnucash.org/repo/gnucash/trunk@17889 57a11ea4-9604-0410-9ed3-97b8803252fd
Fix typos and use of apostrophes.
Also make the status of the document clearer.
Patch by Alan Jenkins.
git-svn-id: svn+ssh://svn.gnucash.org/repo/gnucash/trunk@17888 57a11ea4-9604-0410-9ed3-97b8803252fd
Scheme is not common lisp, so rather use 'else', t might be unbound.
Found by Forest Bond.
BP
git-svn-id: svn+ssh://svn.gnucash.org/repo/gnucash/trunk@17885 57a11ea4-9604-0410-9ed3-97b8803252fd
Patch by Alan Jenkins.
Signed-off-by: Christian Stimming <stimming@tuhh.de>
BP
git-svn-id: svn+ssh://svn.gnucash.org/repo/gnucash/trunk@17880 57a11ea4-9604-0410-9ed3-97b8803252fd
This fixes the default currency and description in the new account dialog.
Patch by Alan Jenkins.
Signed-off-by: Christian Stimming <stimming@tuhh.de>
BP
git-svn-id: svn+ssh://svn.gnucash.org/repo/gnucash/trunk@17879 57a11ea4-9604-0410-9ed3-97b8803252fd
Most of the source tree is UTF8, but there are some latin-1 hangovers.
Apply "iconv --from-code ISO_8859-1" to these files.
This affects bullet point characters in src/optional/xsl/README, and
Benoit Grégoire's name in source files (and hence the Doxygen output).
E.g. it should fix <http://svn.gnucash.org/docs/HEAD/reference.html>
which claims to be UTF8, but encodes Benoit's name in latin-1.
Patch by Alan Jenkins.
git-svn-id: svn+ssh://svn.gnucash.org/repo/gnucash/trunk@17877 57a11ea4-9604-0410-9ed3-97b8803252fd
