396685 - paste item of context menu is not updated correctly

Status: RESOLVED INVALID

(Firefox :: Menus, defect)

Description

[son.two]

User-Agent:       Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.8.1.6) Gecko/20061201 Firefox/2.0.0.6 (Ubuntu-feisty)
Build Identifier: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.8.1.6) Gecko/20061201 Firefox/2.0.0.6 (Ubuntu-feisty)

It is possible to have situation when "Paste" item of context menu is disabled but I can paste using Ctrl-V.  Here is an original bug filed against Java -
http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6584971

I have also created small native test which only copy XA_STRING into clipboard and I've got the same result :(

Reproducible: Always

Steps to Reproduce:
1. compile the test 
(gcc -s -O -o xclipboard xclipboard.c -I /usr/X11R6/include -L /usr/X11R6/lib -lX11)
2. run it
3. open context menu in firefox (e.g. for address bar) verify that "paste" item is disabled
4. paste text into address bar using Ctrl-V you should see "It Works"
Actual Results:  
"paste" item is disabled, but it is possible to paste using clipboard

Expected Results:  
"paste" item should be enabled

here is the test:

#include <X11/Xlib.h>
#include <X11/Xatom.h>
#include <assert.h>
#include <unistd.h>
#include <stdio.h>
#include <stdlib.h>

Atom XA_CLIPBOARD;

main()
{
  Display *dpy = XOpenDisplay(NULL);
  assert(dpy);
  Window w = XCreateSimpleWindow(dpy, DefaultRootWindow(dpy), 0, 0,
                                 200, 100, 0, 0, 0);
  XSelectInput(dpy, w, StructureNotifyMask);
  XMapWindow(dpy, w);
  XSelectionRequestEvent *req;
  XEvent e, respond;
  for(;;) {
    XNextEvent(dpy, &e);
    if (e.type == MapNotify) break;
  }
  XFlush(dpy);
  //
  XA_CLIPBOARD = XInternAtom(dpy, "CLIPBOARD", False);
  Atom a1, a2, a3, type;
  XSelectInput(dpy, w, StructureNotifyMask+ExposureMask);
  int format, result;
  unsigned long len, bytes_left, dummy;
  unsigned char *data;
  Window Sown;

  XSetSelectionOwner (dpy, XA_CLIPBOARD, w, CurrentTime);
  XFlush (dpy);

  for (;;) {
    XNextEvent (dpy, &e);
    if (e.type == SelectionRequest)
      //
      // Somebody wants our data
      //
    {
      req=&(e.xselectionrequest);
      printf ("Selection Request from Mr %i I am %i\n",
              (int)e.xselection.requestor, (int)w);
      printf ("prop:%i tar:%i sel:%i\n", req->property,
              req->target, req->selection);
      if (req->target == XA_STRING)
      {
        XChangeProperty (dpy,
                         req->requestor,
                         req->property,
                         XA_STRING,
                         8,
                         PropModeReplace,
                         (unsigned char*) "It Works",
                         8);
        respond.xselection.property=req->property;
      }
      else // Strings only please
      {
        printf ("No String %i\n",
                (int)req->target);
        respond.xselection.property= None;
      }
      respond.xselection.type= SelectionNotify;
      respond.xselection.display= req->display;
      respond.xselection.requestor= req->requestor;
      respond.xselection.selection=req->selection;
      respond.xselection.target= req->target;
      respond.xselection.time = req->time;
      XSendEvent (dpy, req->requestor,0,0,&respond);
      XFlush (dpy);
    }
  }
}

Comment 1

[Henrik Skupin [:whimboo][⌚️UTC+1]]

Reporter, is it still a problem with Firefox 3.5.2?

Comment 2

[Henrik Skupin [:whimboo][⌚️UTC+1]]

No response from reporter. Closing as incomplete.

Comment 3

[son.two]

Sorry for late response, but I have not received update from 2009-09-02 (looks like gmail considered it as a spam :(

I've just retested the problem with 3.5.2 and it is still reproducible, so
I'm reopening this bug.

Comment 4

[Henrik Skupin [:whimboo][⌚️UTC+1]]

So does it happen for all other applications too or only to specific ones? I just need simple steps to reproduce.

Comment 5

[son.two]

The problem is reproducible with this particular test which, as far as I can see,
is a valid X11 program (and mimic behavior of some java programs).

I'm not sure what means "simple steps to reproduce" :(  
What kind of problems do you have with provided test and steps?

Comment 6

[son.two]

BTW if firefox uses gtk (or some other library) and you can provide simple testcase which demonstrates that the problem is in this library, please share this code, so I could file bug against the library.

Comment 7

[Henrik Skupin [:whimboo][⌚️UTC+1]]

I cannot compile this program so that's the reason why I have asked you. It would be fantastic when you can tell us an applications on Ubuntu which makes this behavior visible.

Comment 8

[son.two]

to build the test on Ubuntu you just need gcc (hope you already have it ;) and
X11 headers (package libx11-dev).

Comment 9

[Henrik Skupin [:whimboo][⌚️UTC+1]]

Ok, that did it. I can see that the context menu entry for paste is not enabled but Ctrl+V works. I also checked a couple of other gnome applications and none of them show an enabled context menu entry for paste. So this should really be a gnome issue.

Please file a bug against gnome.

Comment 10

[son.two]

before filing the bug, it would be great to write a simple test, if you have one please share it with me (I'm not expert in gtk, so it will take a time for me to write the test)

Comment 11

[Henrik Skupin [:whimboo][⌚️UTC+1]]

You have your application. That should be enough. Following steps have to be done:

1. Run your application once.
2. Start any other gnome application e.g. the default text editor
3. Use ctrl+v to paste the clipboard entry
4. Open the context menu over the text field => it's not enabled

Comment 12

[son.two]

luckily, I have found an example of pure gtk code which demonstrates the problem

Comment 13

[Henrik Skupin [:whimboo][⌚️UTC+1]]

Thanks. Feel free to comment the url of the gtk bug report.

Comment 14

[son.two]

https://bugzilla.gnome.org/show_bug.cgi?id=594720