This fix prevents empty strings as tokens and removes duplicated tokens.
Function tokenize_string() is used for bayesian import matching, where
empty token strings or duplicated tokens lead to wrong results within
probability calculation for matching of a transaction to an account.
Empty token strings can occur if (see function g_strsplit())
* two or more spaces occur directly after another
* the string begins or ends with spaces
In get_first_pass_probabilities() function
qof_instance_foreach_slot_prefix() is called with a prefix path
including closing slash after token now. This avoids, that also entries
with token as a substring are included in token info, where key only
starts with token.
Finally function build_token_info() checks, if the key suffix after the
token consists only of the GUID. This avoids, that also entries with the
same prefix and slashes are included in token info.
Provided function func is now called with key suffix only instead of
full key (prefix is omitted). This is neccessary for fixing function
build_token_info() in the next commit.
This starts by setting the gnucash version number in the 'project' call.
This will result in a number of variables set by cmake. The remainder
of this commit is to reuse the auto-generated
PROJECT_NAME, PROJECT_VERSION, PROJECT_VERSION_MAJOR
PROJECT_VERSION_MINOR, PROJECT_VERSION_PATCH and PROJECT_VERSION_TWEAK
instead of all the various instances of PACKAGE, GNUCASH_MAJOR_VERSION
and so on.
One parameter is worth mentioning - GETTEXT_PACKAGE
GnuCash is not using this directly (any more) but it has to be set before
including gi18n-lib.h.
Turned out to be a bug introduced by our port to cmake.
The Windows special casing for Swedish account hierarchies
was never called at all.
This fix includes a fix for a const correctness compiler warning as well.
Algorithms used xaccFooOrder to attempt strict ordering of Foo
objects. Unfortunately the internal definitions of xaccAccountOrder
does not guarantee a strict ordering of accounts.
Also remove unused split<? comparator.
If environment variable LANG is set to 'Swedish_Sweden' (to work around a gettext bug
affecting Swedish locales) the wrong unix locale was substituted previnting
gnucash from finding the associated account hierarchy templates.
Remove closing-entries substring/regex filter.
We will rely on the closing flag created with the Tools>Close Book
tool. This means very *old* closing entries will need to be re-done
via Tools>Close Book using the same date as the old closing entries.
- offer choices for sorting owners in each APAR list
- may be useful for ordering customer aging debts
- this means we remove the default owner->name sorting
instead of list-accessors, use (ice-9 match) to assign names to list
items. remove ugly 'tip-and-rev' hack.
also remove hidden option 'reverse-balance?' and pass it as an
argument to the renderer.
this report handles reversal of income amounts and shouldn't
re-reverse according to originating split's account type.
also friendly-fn should be #f instead of (lambda (a) "") because the
former is handled specifically in trep-engine's add-subheading to
display a blank cell.
If an account name is changed, the autocomplete list store was updated
but the hash table was not so this change checks for the name change
and updates the hash table. Also fixed was changing the account from /
to a placeholder account which will show or hide the account.
INV/PMT txns both derive DESCRIPTION from invoice-posting / payment
txn's ASSET/LIABILITY splits->memos
This seems to produce consistent results in all circumstances.
If a payment txn is modified so that it is funded from >1
asset/liability split, it will concat unique transfer splits memos.
eg. payment transaction
Asset:USDBank -$100 memo "100 USD"
Asset:GBPBank -$100 memo "70 GBP"
A/Payable +$200 memo "orig-memo"
The Payment Description will be "100 USD, 70 GBP"
This fairly complex-looking change is actually simple.
default-calculated-cells is redefined so that the running-balance
column has a custom subheading renderer. The subheading renderer was
formerly used solely for the account->friendly-debit/credit-string
renderer, and is now upgraded to handle hard-coded symbols.
If the friendly-heading-fn is a symbol 'bal-bf, it'll print the
split->account->balance at begindate.
Other columns may also be upgraded for novel subheading renderers. No
forward/backward compatibility issues are expected at all. This change
looks complex because the begindate is not available to
make-split-table and has to be passed on as an argument to be
available for use, and the subheading-renderer will test whether to
display the friendly-fn-renderer at a later stage than previously.
Other meanings of previous " -> " are now:
" -> " conversion
" .. " a range
Other fixes:
1. "File>Save" typo
2. Remove $ from sort order tooltips as it is
a. not relevant,
b. untranslatable for languages used in different regions.
On recommendation of the Translation Project's Benno Schulenberg.
Required re-annotating some comments. Also removed several comments
about leaving a line break to prevent inserting it into gnucash.pot.
That didn't actually work: they were inserted anyway.
Use g_object_weak_ref() to null GncItemEdit::sheet to prevent
a use-after-free crash.
Note that this is a band-aid fix. To fix correctly GncItemEdit and probably
most of the rest of the register must be rewritten with modern GObject
macros and resource management.
* gnucash/report/report-gnome/gnc-plugin-page-report.c:
- replace 'report' by 'report configuration' where appropriate (they are
two different animals).
- correct the menu path to 'Reports > Saved Report Configuration'
The online_id will generated based on bankcode and accountnumber. The
accountnumber may have leading zeros which shall be ignored to make the
creation reliable. In my case the accountnumber of the accountinfo has
no
leading zeros, but the accountnumber of the imported transactions has
leading
zeros. In the result the generated online_id was different and the
widget
to assign the transactions to a account appears always.
NEW API:
gnc:html-table-multirow-col-headers and
gnc:html-table-set-multirow-col-headers!
gnc:html-table col-headers have been augmented to support list of th
rows.
* BACKWARD COMPATIBILITY is offered. old use of single-row headers API
gnc:html-table-col-headers and gnc:html-table-set-col-headers! should
be unchanged; will get/set a single row of th elements.
* NEW functions gnc:html-table-multirow-col-headers and
gnc:html-table-set-multirow-col-headers! will get/set an arbitrary
number of rows of th elements.
* using old API gnc:html-table-col-headers on a table, whose multiple
row headers have been set, will lead to a warning and return the first
row only.
