This commit tries to do the minimum necessary to move the guile bits from engine
to bindings/guile. As engine is a very central piece in the software, this unfortunately
still touches many other source files:
- A few helper objects have been squashed together:
* engine-helpers-guile.[ch] (of which the c part is extracted from engine-helpers.c)
* gncBusGuile.[ch]
* gnc-hooks-scm.[ch]
- The initialization function of gncmod-engine no longer initializes the scm bits.
Any scm code that wants to interact with the engine code now has to load
the (gnucash engine) scm module, or sometimes (gnucash business-core).
The bulk of changes in this commit actually is updating all the scm consumers to do so.
- scm-scm target has been removed. Instead (gnucash utilities) is part
of scm-engine. A few dependency graphs have been updated for this.
More refinements will be in followup commits.
By properly marking certain parameters as private or public
we can have cmake work out most of the link_libraries and
include_directoris for other targets dependent on core-utils
- first all rules and variables for libgnc-core-utils
- then the rules to generate additional files (like
gncla-dir.h or gnc-vcs-info.h)
- next rules for the python bindings
- finally rules for the dist targets
Note this commit also eliminates variable
core_utils_ALL_SOURCES, replacing its two
uses directly with its value.
* Eliminate redundant variables (core_utils_ALL_INCLUDES, core_utils_ALL_LIBRARIES)
Their values are used directly in target_include_directories and
target_link_libraries respectively.
They have been eliminated from the python target because target_include_directories
and target_link_libraries will correctly propagate these values.
* Use generator expressions for a few conditional compile defines and options
Note core-utils.i is used by both the guile and the python bindings so
it is moved up to the common bindings directory, while guile
specific changes are in bindings/guile.
introduce new API
* gnc_using_unreversed_budgets - queries book's unreversed feature
* gnc_reverse_budget_balance - check if book unreversal status matches
2nd argument. if so, return account's reversal status. else, return
FALSE.
* gnome-budget-view can now show both natural and reversed budgets
* gnome-plugin-page-budget will now read&write both natural and
reversed budgets.
because:
* list? is O(N), because it needs to test for an improper
list. improper lists are lists whose last pair's cdr cell is not
'(). null? and pair? are both O(1).
* avoids reverse which is also O(N): guile has unlimited stack
therefore we can do non-tail-call loop first to pass as parameter to
the tail-call loop. this removes the need for prepend-and-reverse.
Addendum to 8b8c957ed -- if budgeted amounts exclude closing entries,
actual amounts should also exclude closing entries.
Amend documentation, modify budget test to add a closing entry which
will be ignored.
This reverts commit 1a9fcfefad because
on MinGW cmake complains about the paths in pkgconfig files. This can
be addressed by using the MSYS2 cmake instead of the MINGW32 one, but
that requires some other changes... and there's also a path separator
bug in that version of FindPkgConfig.cmake.
Previously de_DE locale will add a *global* 'Tax/Tax Number'
option. This is immediately translated.
This change will upgrade it to be present in all locales, and can be
queried by any report.
with the function/declaration that they substituted.
Note that this doesn't use the recommended new GObject creation macros
because the class names in libgnucash/engine don't follow the gnome
naming convention.
* Ensure all the files included in report.scm are built before
any module calls '(use-modules (gnucash report))'
* Break circular dependency between report.scm and the files it loads via
'(load-from-path ...)'
For example file 'gnucash/report/html-linechart' is loaded via
'(load-from-path ...)' in 'reports.scm' it shouldn't call
'(use-modules (gnucash report))'. That would make it indirectly
depend on its own.
* In the same way 'engine-utilities.scm' was in a circular dependency loop
with 'engine.scm' though even more indirectly via (gnc-module-load ...)'.
The initialization code of libgncmod-engine calls
'(use-modules (gnucash engine))' which in turn tries to load-from-path
'engine-utilities.scm', completing the loop.
This simplifies usage of GoogleTest, since independent handling of
GTEST_LIB and GTEST_INCLUDE_DIR is not necessary anymore.
Additionally CMake creates a dependency now between target gtest and all
test applications using it. This improves build process when building
GoogleTest from source code. When any test application is built,
GoogleTest library is automatically rebuilt if necessary now for
instance.
This simplifies a number of escape sequences.
The generated file to configure the man page was split up a bit further:
- code that extracted the GNC_VCS_REV_Y_M from gnc-vcs-info.h was spun out
into its own function that now sets all parameters in gnc-vcs-info.h
as environment variables.
- this function is now invoked by configure-manpage.cmake to extract
the date to insert into the manpage.
- the manpage in addition now shows the full date rather than only
yyyy-mm. This is how man itself does it as well.
With gncTaxTableGetDefault.
qof_book_get_default_tax_table would have been even better but it
would have created a circular dependency between QofBook and
GncTaxTable.
They are now under gnucash/locale/{us,de_DE}. This introduces a directory to add
any scm code that is locale specific. For now it's limited to tax rules for
the US or Germany.
This simplifies the calls in the rest of gnucash
Note that the locale specific reports themselves don't even load this module any more.
They don't need it, instead they can directly load the locale specific tax scheme modules.
A couple of preference callbacks are setup to track the
'negative_color_preference' and 'auto_raise_lists' which are setup with
a call to 'g_once'. To be able to remove these, the the preference id
is saved so that they can be used in gnc_main_window_remove_prefs to
remove the callback.
Currently when compiling GoogleTest from source code, source file
gtest_main.cc from GoogleTest repository is not compiled into any
library as in GoogleTest repository, where it is compiled into
libgtest_main.a. Instead gtest_main.cc is added to source file list
GTEST_SRC, which is then added to the list of source files of every
single GoogleTest based test application.
To simplify this gtest_main.cc is added to the source file list of
target gtest now. Additionally GTEST_SRC is merged into
lib_gtest_SOURCES, since both variables defined source files for
GoogleTest libraries.
Now target gtest generates library libgtest.a, which already contains
the main function from source file gtest_main.cc. This is different to
GoogleTest build system, where both are separated into two independent
libraries libgtest.a and libgtest_main.a.
gnc_start_of_week
* ICU has a mature C++ api, so prefer that one in our C++ code
* Use PERR instead of fprintf for consistent reporting
* Add the ICU specific linker flags to the test case
this function creates some business data. moved from test-invoice.scm
without the invoice-specific tests. verified all invoices/bills are
created correctly.
it returns a vector-list of the 8 invoices generated.
Changed MRO smallest-fraction to 100 based on information provided
by the reporter. ISO4217 has moved the MRO to the "Historical Currencies"
list that doesn't include CcyMnrUnits.
this can be used instead of delete-duplicates when the list must also
be sorted.
the main reason for this function will be for the upcoming aging.scm
report which will use it heavily to slice APAR splits into owner list.
This function helps to establish UI link from register split to an
owner's split. It should not be used anywhere else. Its use in reports
will be removed.
It is also unnecessarily complex -- it is called from an AP/AR account
register split, it should have all necessary information via split
metadata. e.g. this function *can* return a split owner if originating
from an Asset/Liability account which settles an invoice, but this
functionality is not used because there is no asset/liability register
hook to call it.
The function accepts the name of a deprecated module and optionally a
replacement module.
With that info it will generate a stub guile module that
- will be installed in the gnucash guile load path
- emits a deprecation warning when the old module is used
- if a replacement module is given, will automatically load that module instead
This allows us to gently deprecate complete guile modules without
the burden of manually maintaining their module files.
These files were installed fairly ad-hoc into the share/gnucash/scm directory making
it hard to get an idea of where each file comes from.
The files are now structured as follows:
- any scm file authored by gnucash should go in share/gnucash/scm/gnucash or below
- most scm modules will be directly in that directory
- each module that comes with support files will get a subdirectory named after the
module's base name. For example next to engine.scm there will be directory
named engine for all support files of the engine module
- scm files that are not modules, but are loaded by modules go into
<module-dir>. For example gnc-utils.scm loads gnc-menu-extensions.scm
so that file will be installed in gnc-utils/gnc-menu-extensions.scm
- the report system is our largest module and only part of the restructuring
is done at this point. It will be refined further in future commits.
The same restructuring is also done for the compiled files.
This change was introduced in 2.6. Anyone wishing to migrate from 2.4 to 4 should
first pass via 2.6 and/or 3 anyway so this code will never be used again for 4.x
There are more, but these are most common ones.
There are also a number of urls that don't behave well when https, so those are skipped
At some point I have also started marking non-working URLs as [DEAD LINK], though
that's not a full coverage.
fixing a 17 year old bug.
previous version had defined (save-acc list count) and (save-item list
count) thereby overwriting the inbuilt 'list' keyword, and tried to use
the it later on with (list key)... best rewrite with neater code.
This version is available for all supported platforms and distros
(CentOS can have a version via EPEL, which is required anyway)
Advantages:
- one cmake version for all platforms
- we can drop all conditions based on cmake version
There were several issues:
- we store the compiled test*.go files in a tests subdirectory, but this is not
on the GUILE_LOAD_COMPILED_PATH. Added this directory to the path.
- guile was looking for 'script.scm.go' while we create 'script.go'. This is due
to how we invoke the scripts: guile -l path-to-script.scm -c (something)
we can replace this with -l path-to-script (without extension) as -l won't add
the extension. So I've rewrittin the test command to invoke (load-from-path).
- the test modules foo.scm, bar.scm and baz.scm should go in tests/gnucash
as they are defined as modules (gnucash foo), (gnucash bar) and (gnucash baz)
respectively.
- find html.scm/go on the load path instead of using a relative path.
Instead of random locations only occasionally related to the
corresponding source.
Includes renaming libgnucash/engine/test/test-extras.scm and
gnucash/report/report-system/test/test-extras.scm to avoid a
naming conflict.
The end goal of this cleanup is to have all reports grouped in
subdirectories of the new 'reports' directory.
This first commit moves the standard reports
which mostly involves renaming several modules and targets.
In the Preferences/Accounting Period, there is an option to specify
absolute dates but the end date was being set to start of day instead
of end of day, changed this.
crash when involving foreign currency stocks.
Scheme's inexact->exact function just converts the floating-point
representation of a number into an exact rational (documented in the
API Reference, Simple Generic Data Types, Numerical data types, Exact
and Inexact Numbers), which isn't what we want.
We want the number converted to exact directly from the string and to do
that we have gnc-fq-helper preface it with #e.
And in Windows only with the value from the environment if there is one.
Calling it with "" in Windows ignores the environment and sets it to the
system settings.
Ensure that all includes of swig-runtime.h are *followed* by
including guile-mappings.h so that the defines masking
scm_to_utf8_string and scm_from_utf8_string are undone.
When you have multiple loggers enabled, the log file indentations are
partly based on logger name length and so you can have false
indentations which can cause confusion. This change gets the maximum
name logger length and uses this for all with a minimum default length
of 12 characters.
Caused by the introduction of caching used in gnc_pricedb_nth_price,
when prices were deleted, the cached list was never updated and so the
tree model could reference a deleted price and hence crash. This also
affected adding prices as the new prices would not show due to the
cache not being updated. To fix this a function was added to reset the
cache when prices are updated from the model.
This reverts commit aa53c23239, reversing
changes made to 3c946a8449, because
aa53c23 was based on master and so undid other changes made to maint
since the last merge-to-master.
copied function created by Mark Weaver, core guile dev and augmented
to selectively replace substring indices
This is a much more efficient function than the previous
gnc:substring-replace which will constantly split lists using
substring, and create new strings using string-append.
It also does tail call optimization properly, unlike the previous
functions.
https://lists.gnu.org/archive/html/guile-devel/2013-09/msg00029.html -
original
"Here's an implementation that does this benchmark about 80 times
faster on my machine: (20 milliseconds vs 1.69 seconds)
--8<---------------cut here---------------start------------->8---
(define* (string-replace-substring s substr replacement
#:optional
(start 0)
(end (string-length s)))
(let ((substr-length (string-length substr)))
(if (zero? substr-length)
(error "string-replace-substring: empty substr")
(let loop ((start start)
(pieces (list (substring s 0 start))))
(let ((idx (string-contains s substr start end)))
(if idx
(loop (+ idx substr-length)
(cons* replacement
(substring s start idx)
pieces))
(string-concatenate-reverse (cons (substring s start)
pieces))))))))
--8<---------------cut here---------------end--------------->8---
The reason this is so much faster is because it avoids needless
generation of intermediate strings."
Add tests for libgnucash/scm/utilities.scm functions
- tests for list<->vec
- tests for gnc:substring-replace
- tests for gnc:substring-replace-from-to
The latter confirms that the comment before the function definition
is *incorrect* - it describes that substring-replace-from-to will
start from the 2nd substring for the first substitution, and
performs 2 substitutions. However the comment illustrates only 1
substitution. The test suite performs the test according to code
behaviour, rather than the comment. This issue is moot in practice
because the end-after is always called with negative in the code
base.
original comment:
;; gnc:substring-replace-from-to
;; same as gnc:substring-replace extended by:
;; start: from which occurrence onwards the replacement shall start
;; end-after: max. number times the replacement should executed
;;
;; Example: (gnc:substring-replace-from-to "foobarfoobarfoobar" "bar" "xyz" 2 2)
;; returns "foobarfooxyzfoobar".
When testing for a MS Windows path which has a ":", a gvfs path with the
format '/run/user/1000/gvfs/smb-share:server=192.168.1.11,share=public/
test-xml-file.gnucash' is recognised as a Windows path. To avoid this
expand the test to ":/" and also ":\"
On Windows boost::filesystem::path's string() produces a string
that goes out of scope with the function in which it's called, so
returning its c_str() ptr yields freed memory, usually full of garbage.
This has an interesting side effect in gnc-file.c's check_file_path():
Since the memory is freed, g_path_get_dirname() reuses it after an
iteration or two, writing its result into the same address. The
following strcmp naturally returns 0 because it's comparing two
instances of the same ptr, so check_file_path falsely reports that
the proposed save path is in GNC_USERDATA_DIR.
Windows file uri's can be of the form 'file:///N:/bob.txt' so change
the gnc_uri_get_components to remove a left over '/' at the start so
gnc_resolve_file_path gets the absolute path correctly. Also change
gnc_uri_create_uri to add an extra '/' for Windows file uri's.
Add function to create an absolute file path from a prefix path and a
relative one. If the prefix is null, then the root directory of the
current path is used.
If the original date is an end-of-month date, we take it as an
indicator they always want monthdelta dates to be end-of-months.
This works for monthly/quarterly/halfyearly/annual.
Addendum to commit 65bfeaf5de which was
deemed to be an incomplete fix.
Also I'd forgotten to activate a test in test-date-utilities. Enable it.
this will modify a test which was calibrated to record purchase price
only. fix transaction creation to add prices for both purchase and
sales, and also fix test which was assuming no sale price was bring
recorded.
This is the general case for any transaction creation. Rewrite other
transaction creation routines to use it. All tests still work
unchanged, which confirms this function works well.
This will allow tests to create multisplit transactions, of an
arbitrary number of splits. If the list-of-split's values are not
balanced (i.e. total 0), the engine will create an Imbalance-CUR
split.
The motivation is to allow creation of complex multisplit
multicommodity transactions eg USD50 + GBP20 (USD25) = EUR66 (USD75)
as well as their prices GBP/USD = 25/20 and EUR/USD = 75/66.
* USD -50
* USD -25 = GBP -20
* USD +75 = EUR +66
This will be useful in creating tests for stock-based reports, whereby
stock sales need splits in STOCK/ASSET/INCOME accounts.
the option lookup mechanism will dynamically translate option
names. warn the user if this takes place so that the report writer may
use new option names.
Use the Scheme rationalize method to convert the decimal numbers
from Finance::Quote to ratinal numbers. This avoids rediculous
precision like 8515625000000001/3906250000000000 for 2.18.
If a saved-report with e.g. relative date, multichoice option is
unknown, the report would crash, and the Report-Options would
segfault. This commit fixes both: report-date defaults to 'today',
multichoice-options defaults to default-value.
Following this commit, if a report loads a saved-report or .gcm from a
future version, a gnc:warn will be emitted and the report will not
crash; it will use relative-date today. Multichoice will remain the
default value. Report Options will not segfault.
The user will be notified via a gnc:gui-warn dialog
gnc:gui-[warn|error|msg] are new global functions.
By default they mirror gnc:warn/error/msg. However then gnome is
available, they will display appropriate warn/error/info dialog in
addition to outputting to console.
Instead of recursing the date, we calculate the next month using an
index-based multiplier, and apply modulo/remainder as appropriate to
determine the next month/year.
Then we attempt to create new mktime, and if the resulting mktime's
month is not as expected, reduce the mday by 1 until resulting month
is correct. This fixes monthly intervals for end-of-month days.
Test via monthly/quarterly deltas, and also includes leapyear
calculation.
Revert using boost::locale to generate std::locales as boost::locale-
generated locales don't implement std::locale::facet and there was
a bug in the boost::locale ICU wrapper code that caused the wrong year
to be output for the last 3 days of December.
GCC's libstdc++ supports only the "C" locale on Windows and throws if
one attempts to create any other kind. For dates we work around this
by using wstrftime() to format according to locale and then convert
the UTF16 string to UTF8. wstrftime() interprets the time zone flags
%z, %Z, and %ZP differently so we process those first before calling
strftime. This will have the unfortunate effect of not localizing
timezone names but it's as close as we can get.
