This was the only dive_site_table function that accessed
to global divelog, which is odd. Make it consistent with
the others.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
After all it doesn't access any dive_site structure.
Moreover, rename it, since we use mostly snake_case in core.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
This gives the distance between to location_t objects. It is
unclear why this was in divesite.cpp.
Moreover pass by value, not raw pointer.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
There were a number of free standing functions acting on a
dive-site-table. Make them member functions. This allows
for shorter names. Use the get_idx() function of the base
class, which returns a size_t instead of an int (since that
is what the standard, somewhat unfortunately, uses).
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
This makes memory management more simple, as not explicit deletion
is necessary.
A rather large commit, because changing QVector<> to std::vector<>
is propagated up the call chain.
Adds a new range_contains() helper function for collection
types such as std::vector<>. I didn't want to call it
contains(), since we already have a contains function
for strings and let's keep argument overloading simple.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
This is a long commit, because it introduces a new abstraction:
a general std::vector<> of std::unique_ptrs<>.
Moreover, it replaces a number of pointers by C++ references,
when the callee does not suppoert null objects.
This simplifies memory management and makes ownership more
explicit. It is a proof-of-concept and a test-bed for
the other core data structrures.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
This the C++ version of membuffer. Since everything is C++, it can
just be made the default.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
Also remove the UNUSED() macro, as there were no users left.
The macro was silly anyway - there were many falso positives.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
We can now return mutable/imutable depending on const-ness of
the parameter, owing to parameter overloading.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
Makes the code much nicer to read.
Default initialize cylinder_t to the empty cylinder.
This produces lots of warnings, because most structure are now
not PODs anymore and shouldn't be erased using memset().
These memset()s will be removed one-by-one and replaced by
proper constructors.
The whole ordeal made it necessary to add a constructor to
struct event. To simplify things the whole optimization of
the variable-size event names was removed. In upcoming commits
this will be replaced by std::string anyway.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
To my understanding, declaring empty parameter lists using "(void)"
is an artifact from the bad old K&R times, when functions were
declared without(!) parameters. Which in hindsight was an absolute
recipe for disaster. So for backwards compatibility, functions
without parameters had to be declared using "(void)" as "()"
could also mean "any function".
That was 40 years ago. Meanwhile, C++ introduced references,
which made it a necessity to declare the function parameters.
So "(void)" is redundant and inconsistent in C++ code and
just makes no sense.
Remove it.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
The divesite-helper.cpp only existed because C-string manipulation
was too tedious. Now that divesite.cpp is C++ anyway, the split
is not necessary anymore.
Moreover, return an std::string, since this is a core-function.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
Since this is now in C++, we don't have to use our crazy
TABLE_* macros.
This contains a logic change: the dives associated to a
dive site are now unsorted.
The old code was subtly buggy: dives were added in a sorted
manner, but when the dive was edited the list was not
resorted. Very unlikely that this leads to a serious
problem, still not good.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
Since the taxonomy is now a real C++ struct with constructor
and destructor, dive_site has to be converted to C++ as well.
A bit hairy for now, but will ultimately be distinctly simpler.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
Use std::vector<> instead of fixed size array.
Doesn't do any logic change, even though the back-translation
logic is ominous.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
Use more C++ style memory management for plot_info: Use std::vector
for array data. Return the plot_info instead of filling an output
parameter. Add a constructor/destructor pair so that the caller
isn't bothered with memory management.
The bulk of the commit is replacement of pointers with references,
which is kind of gratuitous. But I started and then went on...
Default initializiation of gas_pressures made it necessary to convert
gas.c to c++, though with minimal changes to the code.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
Replace malloc/free of one structure by C++ idioms and add a
destructor to the struct. Otherwise, don't touch the code.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
It seems that smartrak was the only part of the code that cared
about freeing the dc_descriptor. Make that a general feature
of the new device_data_t destructor, which we could implement
now that things are in C++.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
It was never clear what was a pointer to a static string from
libdivecomputer and what was allocated.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
These can all just be local objects.
Also, don't overwrite them with 0. We later want to convert the
string to std::string, where this would be very sketchy.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
Convert both files simultanously, because the SI_UNITS define works
either under C or under C++.
This was painful, because initialization of struct-members has to
be done in order of definition in C++. And it was completely out
of order. However, as long as not all is C++, we can't use
default initialization directly in the struct definition. :(
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
The old code was wild: For the yearly statistics it would allocate
one entry per dive in the log. Of course, it would also leak
C-style strings.
Convert the whole thing to somewhat idiomatic C++.
Somewhat wasted work, because I'd like to convert the whole thing
to the new statistics code. But let's finish the conversion to C++
first.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
All have been converted to std::min() and std::max().
Moreover, this was Windows only and since we cross-compile, it
is not even clear if this is needed.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
This was the umpteenth inefficient reinvention of a trivial
map. Replace by a hash-map (std::unordered_map). Might just
as well use a balanced binary tree or a sorted array. In the
end, it probably doesn't matter at all.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
uemis_get_divenr() returns maxdiveid and passes mindiveid as a
global variable.
Make this more reasonable by returning a min, max pair.
The way mindiveid is an unsigned int and then reinterpreted as
int is very sketchy. This commit attempts to not change that
behavior.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
This is only initialized and used in one loop. Very mysterious
why this should be a global variable.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
This global variable is used in two function independently of
each other. I don't see how there should be transport of this
value from one function to the other. Ominous.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
uemis_get_answer() would put the raw response into a global variable.
This could be anywhere in the call stack and thus you never knew
when the existing buffer was removed under your feet.
Instead, return the buffer explicitly from uemis_get_answer().
I'm nit perfectly happy about the new interface: an error is
indicated by an empty buffer, which is awkward to test for.
If an empty buffer turns out to be a valid response, this
should be replaced by an std::optional<> or std::expected<>.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
The string code of uemis-downloader.cpp was broken in more ways
than can be listed here. Notably, it brazenly refused to free any
memory allocated for the parameters buffer.
Using std::string and std::string_view should plug all those
memory holes. That made it necessary to do some major refactoring.
This was done blind and therefore will break.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
newmax was an integer variable kept as a string. Very ominous.
Moreover, memory management seems to be broken:
1) The string is never freed.
2) The string is passed as value from do_uemis_import() to
get_matching_dives(), which passes it as reference to
process_raw_buffer(), which may reallocate it, which means
that do_uemis_import() now possesses a pointer to a free()d
string.
Simplify all that by making newmax an integer variable and
passing it as a reference from do_uemis_import() to
get_matching_dives().
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
There were a number of fprintf()s that escaped the conversion
to report_info(), because they used "debugfile" instead of
"stderr" as target. However, debugfile was just #defined to
be stderr, so we might just use report_info() for consistency.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
The uemis code is wild. It simply doesn't deallocate memory
and uses global variables. To get this under control, create
a "struct uemis" and make the functions exported by "uemis.h"
members of "struct uemis". Thus, we don't have to carry around
a parameter for the state of the importing process.
Turn a linked list of "helper" structures (one per imported dive)
into a std::unordered_map, to fix leaking of the helper structures.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
This was very obscure: the function that parses into a struct
dive was passed a void-pointer instead of a struct dive-pointer.
Why? Just pass the correct type.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
The SHA1() helper function was only used when calculating a
SHA1 hash and taking the first four bytes of it as uint32.
Make that explicit by renaming the function into SHA1_uint32()
and directly returning an uint32_t.
Note that the usage in cochran.cpp is sketchy: it generates
a four-byte hash out of two-byte data. Why!?
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
The strtod_flags() function allowed for fine control of how to
parse strings. However, only two different modes were actually
used: ascii mode ("C" locale) and permissive mode (accept ","
and "." as decimal separator).
The former had already its own function name (ascii_strtod).
Make the latter a separatge function as well (permissive_strtod)
and remove all the flags rigmarole.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
Owing to bitrot, this union only contained a float and therefore
is pointless. Let's remove it.
That makes the function name "integer_or_float()" non-sensical.
Call it "parse_float()" instead.
Moreover, change the output-arguments of "parse_float()" from
pointers to references, as null-pointers are not supported.
Finally, remove the "errno" check after "ascii_strtod()". As far as
I can tell, errno is not set in "ascii_strtod()" and using a global
variable for error-reporting it is an incredibly silly interface
anyway.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
Some improvements:
- for constraints where the condition is applied to any one of a list of
items, show the condition as 'any X';
- when filtering for tags, do not include divemode as a 'tag' - this is
hard to impossible for users to understand, and we do have a dedicated
condition for the dive mode.
Signed-off-by: Michael Keller <github@ike.ch>
Fix loading of the negation of filter conditions. Unlike other
conditions that are persisted as `<key>="<value>"`, this is persisted
to git as `negate`.
This fix remediates this for all cases where the condition has already
been saved to the cloud storage.
Saving to XML takes a different approach and indicates negated
conditions with `negate="1"`, making it identical to all other
attributes. The question is if this approach should be implemented in
addition to the above fix, in order to unify the storage format.
Fixes#4246.
Signed-off-by: Michael Keller <github@ike.ch>
Fix an issue introduced in #4148.
Essentially the refactoring missed the fact that in the imperial system
tank size is tracked as the free gas volume, but in the metric system
(which is the one used in most of Subsurface's calculations) tank size
is tracked as water capacity.
So when updating a tank template tracking imperial measurements, the
given (metric) volume in l has to be multiplied by the working pressure,
and vice versa.
This also combines all the logic dealing with `tank_info` data in one
place, hopefully making it less likely that this will be broken by
inconsistencies in the future.
Fixes#4239.
Signed-off-by: Michael Keller <github@ike.ch>
- show the correct gasmix in the profile;
- make gases available for gas switches in the profile after they have
been added;
- persist gas changes;
- add air as a default gas when adding a dive.
This still has problems when undoing a gas switch - instead of
completely removing the gas switch it is just moved to the next point in the
profile.
Signed-off-by: Michael Keller <github@ike.ch>
The last use of these functions was removed in ae299d5e66.
And that's a good thing, because snprintf-style interfaces
make zero sense in times of variable-length character
encodings.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
A long standing issue: the dives_to_add, etc. tables need to be
manually freed. This kind of problem wouldn't arise with proper
C++ data structures.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
- standardise the naming;
- use it consistently;
- apply the 'samples < 50' only when putting manually added dives into
edit mode - everywhere else manually added dives should be treated as
such;
- do not show a warning before editing a manually added dive in planner.
Signed-off-by: Michael Keller <github@ike.ch>
Error introduced in da7ea17b66: the INFO() and ERROR() macros
pass stdout instead of the format string as first parameter
to report_error(). Ooooops. How did this ever pass the
compile tests!?
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
46cf2fc086 fixed a bug where clearing of a divelog, such as the one
used for import, would erase dives in the global(!) divelog.
However, the new code used the function clear_dive_table(), which
only cleared the table without unregistering the dives. In particular,
the dives were not removed from the trips, which means that the trips
were not free()d.
This reinstates the old code, but now passes a divelog paremeter
to delete_single_dive() instead of accessing the global divelog.
Moreover, delete dives from the back to avoid unnecessary
copying.
An alternative and definitely simpler solution might be to just
add a "clear_trip_table()" after "clear_dive_table()".
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
Currently editing of planned dives that have been merged with actual
(logged) dives only works if the 'Planned dive' divecomputer is the
first divecomputer, and this divecomputer is selected when clicking
'Edit planned dive'. In other cases the profile of the first
divecomputer is overlaid with the profile of the planned dive, and the
first divecomputer's profile is overwritten when saving the dive plan.
Fix this problem.
Triggered by @SeppoTakalo's comment (https://github.com/subsurface/subsurface/issues/1913#issuecomment-2075562119): Users don't like that planned dives show up as their own entries in the dive list, so being able to merge them with the actual dive after it has been executed is a good feature - but this wasn't working well until now.
Signed-off-by: Michael Keller <github@ike.ch>
Convert some C-style strings in uemis-downloader.cpp to std::string.
This has the side effect of fixing builds on Debian Trixie, which
currently fail with the (rather silly) error:
/build/subsurface-beta-202405060411/core/uemis-downloader.cpp: In function 'char* build_ans_path(const char*, int)':
/build/subsurface-beta-202405060411/core/uemis-downloader.cpp:290:32: error: '%s' directive output between 0 and 12 bytes may cause result to exceed 'INT_MAX' [-Werror=format-truncation=]
290 | snprintf(buf, len, "%s/%s", path, name);
| ^~
......
529 | ans_path = build_filename(intermediate, fl);
| ~~
cc1plus: some warnings being treated as errors
Signed-off-by: Richard Fuchs <dfx@dfx.at>
Fix a bug causing the 'Download from dive computer' dialogue to hang
when the user attempts to cancel the dialogue after successfully
downloading one or more dives.
Fixes#4176.
Signed-off-by: Michael Keller <github@ike.ch>
ae299d5e66 introduced a format-
string bug by splitting a format-string in two and splitting
the arguments at the wrong place.
The compiler doesn't warn in this case, because the format-
string is passed through translate(...).
This should have crashed, but for some reason didn't, at least
on Linux.
Fix the arguments.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
Add logging of the libdivecomputer return code for errors. Also, switch
logging of errors in the background thread to callback based logging to
make it visible.
Signed-off-by: Michael Keller <github@ike.ch>
Fix the filters for planned (i.e. has at least one dive plan attached)
and logged (i.e. has at least one dive computer log attached) dives.
Also refactor the respective functions for improved readability.
Signed-off-by: Michael Keller <github@ike.ch>
Fix some findings in a Coverity scan in `core/planner.cpp` and
`core/profile.cpp`, that were reported as new after the changes
in #4126 (likely because of the rename from .c to .cpp).
Results: https://scan4.scan.coverity.com/#/project-view/60459/13160
Signed-off-by: Michael Keller <mikeller@042.ch>
Allows us to remove the strndup.h header. This code will be
even more simple, once core is fully converted away from C-strings.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
The old code was leaking memory. Use std::unique_ptr<> for
ownership management.
This is still very primitive and divetags are kept during
application lifetime. There should probably be some form
of reference counting. And the taglist should not be global,
but attached to the divelog.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
The only case left is in android.cpp, though that is only compiled
when compiling the full desktop app on Android. I.e. never. So
don't bother for now.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
Let's use std::string in the core. Notably, I'd like to make
the numerous main() functions mostly independent of Qt. Some
things will have to remain, such as argument parsing, of course.
This changes the API: instead of returning an error code and
taking a pointer to the actual return-value, return an
std::optional<std::string>> that is set if the function succeeds.
Returning an empty string in the error case might be simpler,
but oh well...
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
Use the C++-version of membuffer.
This fixes two memory leaks: report_info() on every(!) invocation
and report_error() before the error callback is set.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
In C++ files, replace MIN and MAX by std::min and std::max,
respectively. There are still a few C files using these
macros. Convert them in due course.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
Had to rewrite the thing, because gcc's warnings don't work
with templatized var-args. Since there is no string-format.cpp
and I didn't want to inline it, moved it to format.cpp.
String formatting is distributed around at least four
headers: membuffer.h, subsurface-string.h, format.h
and format-string.h. This really should be unified!
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
Quite a bit of fallout in users of this structure.
Conveniently, since git-access.cpp is now C++ we can move
some helpers from the monstrous qthelper.cpp to git-access.cpp.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
This may appear a bit ominous, as it doesn't generate a string,
but a vector of strings (one for each line). However, that is
in preparation for the QtQuickification of the profile, where
the text-items take such a list.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
Since this is the only caller, onvert the get_file_name() function
to return an std::string.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
This used to have multiple values, but is currently only checked for
true/false. Reflect that in the type.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
The deco timestep is a parameter to the plan() function. There
seems no need to define this as a global macro. Probably some
code reshuffeling artifact.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
We had locale aware formatting functions that generated QStrings.
Create an alternative that creates std::string, since we want that
in the core.
This commit is unfortunate for two reasons:
- The function is called "casprintf()" for analogy with the QString
version. However, the non locale aware function is called
"format_string_std()" for analogy with "format_string()".
Ultimately these names should be unified. Probably, once there
are no membuffer users left.
- This does UTF-16->UTF-8->UTF-16 roundtrips. The core formatting
functions should render UTF-8 and only convert to UTF-16, in
the UI layer.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
The combo-boxes (cylinder type, weightsystem, etc.) were controlled
by global models. Keeping these models up-to-date was very combersome
and buggy.
Create a new model everytime a combobox is opened. Ultimately it
might even be better to create a copy of the strings and switch
to simple QStringListModel. Set data in the core directly and
don't do this via the models.
The result is much simpler and easier to handle.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
Remove the options to expand entities and so continue when encountering invalid /
malformed XML, as both of these can be exploited by supplying
maliciously crafted XML.
Signed-off-by: Michael Keller <mikeller@042.ch>
Q_FOREACH and foreach are anachronisms.
Range based for may cause a performance regression: it can
lead to a copy of shared containers (one reason why Qt's
COW containers are broken). However, as long as there is no
user noticeable delay, there is no point in analyzing each case.
And also no point in slapping an 'asConst' on every container
that is looped over.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
To avoid memory management woes. These shouldn't be global
variables, but let's fix that later.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
The memory managements for DeviceDetails was very sketchy.
First of all, sharing a pointer to a structure between threads
seems like a recipe for disaster. Secondly, the structure was
a QObject and when first generated included in the (silly)
Qt object tree, but when generated in the threads it was not.
Clearly, this leaks.
Instead, use value semantics and use local copies of the
structure. I didn't go full length and use std::move to
move the data, because this doesn't work through signals
(which are the wrong abstraction here, but OK) and secondly
I didn't have time to analyze whether the caller still
needs the data after passing it down to the worker thread.
To be able to pass an object through signals, the class
has to be registered in the Qt MetaType system. Super
ugly, but fine for now. Ultimately, this whole thing should
probably be replaced by futures, co-routines, or whatever.
Moreover, this removes the prefix from number of "m_*"
function parameters. By convention, "m_" marks member
variables, which function parameters are not.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
make DeviceDetails a metatype
So that we can pass it as value through the signal/slot system.
(squash with original commit)
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
Downloder builds pull in show_computer_list() from
downloadfromdcthread.cpp, but it's declared as extern "C". With 76c2069f
having converted subsurfacestartup.c to .cpp, we can remove the extern
"C"
Signed-off-by: Richard Fuchs <dfx@dfx.at>
Add 'Country' to the fields that are indexed for fulltext search - this
seems to be a quite intuitive choice as 'Country' is also a field that
is available in the dive list view.
Fixes#4134.
Signed-off-by: Michael Keller <mikeller@042.ch>
Opportunistically fix some problems newly raised by a recent Coverity
scan.
Not touching any of the string memory allocation issues as this is being
handled by the move towards C++ strings.
Signed-off-by: Michael Keller <mikeller@042.ch>
printf() is a horrible interface as it does no type checking.
Let's at least use the compiler to check format strings and
arguments. This obviously doesn't work for translated strings
and using report_error on translated strings is dubious. But OK.
Had to convert a number of report_error() calls to supress
warnings.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
qthelper.h is an absolute monstrosity and it is unclear what
report_info and SSRF_INFO have to do with Qt.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
This mimics the code added in commit cf990b0f39 ("preferences: choose language
code with one '-'") and adds some debugging for the mobile case - some people
are being presented with Subsurface-mobile in Korean for some reason.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
When initializing a string with multiple characters, first
comes the length, then the size. Not the other way around.
Fixes#4127.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
On initialization, the old code searched for the first language
code containing a '-'. However, my Qt version gives de-Latn-DE
as the first entry. That messed up the preferences code: it
didn't recognize that entry. Thus, simply opening and closing
the preferences switched the language to Bulgarian.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
There was a pattern of code like
match_action(line, state, dive_action, ARRAY_SIZE(dive_action));
The doubling of the array might cause copy & paste errors, where
only one array is replaced.
Therefore, determine the length of the array with (hopefully
easily understood) template tricksery.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
When iterating over the converted strings of a line, the
first entry of the array would be popped off, leading to
a full copy of the remaining array.
Instead, use an index in the parser state.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
The converted strings were stored in a membuffer and later
converted to std::strings. Generate an std::string directly
to avoid unnecessary copying.
Ultimately, when the core structures are converted to
std::string, there should be no copying of the string data
at all (unless formatting is applied or small string
optimization kicks in, of course).
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
Create a format_string_std function that works like format_string,
but does return a std::string instead of a strdup()ed C string.
Make it a global function to be used in other parts of the code.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
