Adding XSetWMProtocols method to X11#665
Adding XSetWMProtocols method to X11#665matthiasblaesing merged 13 commits intojava-native-access:masterfrom
Conversation
| public void testSetWMProtocols() { | ||
| X11.Atom[] protocols = new X11.Atom[]{ X11.INSTANCE.XInternAtom(display, "WM_DELETE_WINDOW", false) }; | ||
| int status = X11.INSTANCE.XSetWMProtocols(display, root, protocols, 1); | ||
| assertEquals("Bad status for XSetWMProtocols", 1, status); |
There was a problem hiding this comment.
If it cannot intern the WM_PROTOCOLS atom, XSetWMProtocols returns a zero status. Otherwise, it returns a nonzero status.
So - would it be better to check for a non-zero status?
|
Please make sure you also add an entry into the CHANGES.md about the new function(s). When you work on this please merge the changesets into logical segments - I'm not a fan of squashing commits just because it can be done, but it helps readability if logical blocks are commited together. In this case I think one changeset can represent the whole change without negative impact on readability. |
| Pointer class_hints); | ||
|
|
||
| int XSetWMProtocols(Display display, Window window, Atom[] atom, int count); | ||
| int XGetWMProtocols(Display display, Window w, Atom[] protocols_return, IntByReference count_return); |
There was a problem hiding this comment.
This won't work - the function is declared as:
Status XGetWMProtocols(Display *display, Window w, Atom
**protocols_return, int *count_return); protocols_return is a pointer to an array of atoms (following the documentation). The array of atoms is allocated on the callee side and not on the java side. As an Atom is basicly a primitive long, I only see the option to bind this as a PointerByReference. From that you have to extract the data itself and in the end free the list:
X11.Atom[] atomList = new X11.Atom[count.getValue()];
for(int i = count.getValue() - 1; i >= 0; i--) {
atomList[i] = new X11.Atom(protocols.getValue().getLong(X11.Atom.SIZE * i));
}
X11.INSTANCE.XFree(protocols.getValue());There was a problem hiding this comment.
Good catch - I didn't realise the array was allocated on the callee side. If I want to add a helper method, would X.java be a good place to put it?
There was a problem hiding this comment.
You can see a pattern in the com.sun.platform.win32 package. In general there is the library binding (Kernel32.java) and there is the abstract helper class (Kernel32Util.java) that has "Util" as suffix and contains helpers as static methods.
|
@zainab-ali thank you for your contribution! I merged your changes into master. It is appreciated! |
Motivation: Due how our code was structured it was possible that we not always handle the lifetime of the underlying quiche connection correctly. Beside this it was also possible to convert a CLOSED channel state to an ACTIVE again. Modifications: - Rewrite the state machine in QuicheQuicChannel to correctly handle the different states and also make it impossible to change a final state like CLOSED - Correctly handle removal of the QuicheQuicChannel from the datastructures in QuicheQuicChannel Result: Correct handling of lifetime and state.
2 participants
Adding
XSetWMProtocolsmethod tocom.sun.jna.platform.unix.X11Adding a single test to
com.sun.jna.platform.unix.X11Teststo check that the command returns the correct status code.