aboutsummaryrefslogtreecommitdiff
path: root/c
AgeCommit message (Collapse)Author
2022-07-19Use static member function instead of friend function.John Ankarström
2022-07-19Add comments.John Ankarström
2022-07-19Remove EBMessageBoxA.John Ankarström
2022-07-19Use Prolog string buffer marks.John Ankarström
Speaking of unclear documentation, it is not obvious whether it is necessary for programs calling into Prolog to manually mark and release strings. I suppose that it should be, if the same logic that applies to terms apply to strings. On the other hand, the stack in which the strings are stored belongs to Prolog, and there is nothing that would prevent Prolog from cleaning up the strings when called at a later time. I am not sure. But better safe than sorry, I guess. The Mark class acts like the Frame class. The constructor and destructor are equivalent to the PL_STRINGS_MARK and PL_STRINGS_RELEASE macros. Unlike for 34c3280, I did not notice any differences in memory usage after this change. Perhaps that is because it has no effect; perhaps it is because Prolog's stack is very big.
2022-07-19Fix Prolog memory leaks.John Ankarström
Apparently foreign frames ARE needed when calling Prolog from C. The official documentation is very terse and could make this clearer. To summarize, whenever a term is created (e.g., PL_new_term_refs), its reference count is increased by one. It is garbage-collected when its reference count hits zero. But the reference count is never decreased unless (a) control returns to Prolog after executing a foreign predicate -- which does not happen in my application -- or (b) the foreign frame in which the term was created is closed. In other words, terms must be created within a foreign frame. This is achieved by initializing a Frame object before creating the term and destroying it once the term has served its purpose. The destructor for Frame does not DISCARD the frame, only CLOSE it. The former would also invalidate all data bound by the terms, which is usually undesirable.
2022-07-19Implement centered message box.John Ankarström
2022-07-18Improve Library::GetProcAddress.John Ankarström
It isn't really more safe, but it removes the need for a confusing function pointer cast, which is easy to get wrong. As far as the compiler is concerned, the result is literally the same, but it does force the caller to (indirectly, via the template parameter) cast the return value, which may be a good thing.
2022-07-18Solve compiler warnings.John Ankarström
2022-07-18Add Unicode support to Win32Error.John Ankarström
2022-07-18Improve exception message.John Ankarström
2022-07-18Fix typo.John Ankarström
2022-07-18Formatting.John Ankarström
2022-07-18Avoid leak in Prolog interface.John Ankarström
PL_new_atom(_wchars) creates an atom with a reference count of one, which is never decreased, and the atom is thus never garbage collected.
2022-07-18Fix 26bc4109.John Ankarström
Apparently, = {0} does not zero a structure in C++.
2022-07-18Ensure that CBT hook is set and unset correctly.John Ankarström
2022-07-17Prefer f() over f(void).John Ankarström
f(void) is a C-ism that is valid but unnecessary in C++.
2022-07-17Make ListView hWnd public.John Ankarström
A getter offers encapsulation, but it is also less transparent in a sense. Thinking of ListView as a struct, it is natural to expose hWnd as a public member variable.
2022-07-17Update type names and variable prefixes.John Ankarström
2022-07-17Name function arguments in headers.John Ankarström
This is obviously a lot less obtuse.
2022-07-17Add const to pointers, update spacing.John Ankarström
It seems that "right-spaced" pointers are more widely used among C++ programmers.
2022-07-17Add const to places.John Ankarström
Note that I did NOT add const to non-pointer/non-reference arguments in function declarations (without a following definition), as they do not mean anything there.
2022-07-17Remove TsmFromSz.John Ankarström
If needed in the future, it is possible to simply copy it from the Git history.
2022-07-17Use SWI-Prolog's wide-character functions.John Ankarström
This avoids the use of TsmFromSz.
2022-07-16Formatting.John Ankarström
2022-07-16Rewrite TszFromSz as TsmFromSz.John Ankarström
I.e. using std::basic_string<TCHAR> instead of TCHAR *. This removes all unmanaged frees.
2022-07-16Add m_hWndParent to ListView.John Ankarström
This avoids g_hWnd.
2022-07-15Split defs.h into separate header files.John Ankarström
This is feasible now that the makedeps script exists to automatically manage build dependencies (see 6034fe2, d00f8b3).
2022-07-15Add Win32Error exception class.John Ankarström
2022-07-15Show error and exit gracefully on uncaught exception.John Ankarström
2022-07-15Add support for Prolog exceptions.John Ankarström
2022-07-15Use std::optional instead of std::unique_ptr for try_make.John Ankarström
This requires C++17.
2022-07-15Replace Library::Load with try_make_unique template.John Ankarström
2022-07-15Don't crash on Prolog exceptions.John Ankarström
In the future, it may be desirable to convert Prolog exceptions to C++ exceptions.
2022-07-15Reimplement Pl with variadic templatesJohn Ankarström
This is a bit safer and about as complex.
2022-07-15Change declaration style.John Ankarström
2022-07-15Add wrapper for LoadLibrary, FreeLibrary.John Ankarström
2022-07-14Code style.John Ankarström
2022-07-14Pass WM_DESTROY to ListView::WndProc.John Ankarström
On WM_DESTROY, the "this" property was removed before it could be retrieved and dereferenced, making it impossible for WM_DESTROY messages to be passed to the WndProc method.
2022-07-14Prevent rating column from being invisible.John Ankarström
Bug introduced in 26f70ab. DataListView was initialized with g_iDPI set to -1, which is obviously incorrect. The default value of -1 is a remnant of earlier code and is no longer necessary.
2022-07-14Minor formal changes.John Ankarström
Height(DLVSIKEY) was incorrect. The argument to Height is supposed to be a boolean value, in this case false. It happened to work because DLVSIKEY is 0 (because Key is the 0th column in the data list view).
2022-07-14Initialize global variables in the right place.John Ankarström
The rules for what messages are sent at window creation -- and in which order -- are neither intuitive or clear. WM_CREATE can NOT be relied upon to initialize global state required by handlers of other messages (such as WM_GETMINMAXINFO, which seems to be sent before WM_CREATE). Thus, the better solution is to initialize everything using a CBT hook before the window procedure is even run. Because CBTProc creates (child): windows of its own, though, one must be careful to only run the initialization once, which is done by checking g_hWnd.
2022-07-11Improve ElvSort.John Ankarström
2022-07-11Refactor, adjust minimum window size.John Ankarström
2022-07-10Convert to C++.John Ankarström
I already hit upon some object-oriented programming patterns in *listview.c, so I felt that it would be natural to use this as an opportunity to learn C++.
2022-07-10Add IDT_TIMER constant.John Ankarström
2022-06-06Add basic threading support.John Ankarström
It is much easier to implement on the Prolog side than on the C side... :-)
2022-06-02Add Plp, Plg functions.John Ankarström
2022-06-02Fix va_list bug.John Ankarström
2022-06-02Fix typo.John Ankarström
2022-06-02Simplify C interface to Prolog.John Ankarström