Rudiments
|
#include <process.h>
The process class provides static methods for accessing information about and controlling processes, including methods for forking, spawning and executing child processes.
Checks for filename "filename" and reads the process id out of it, if it exists. Returns the process id on success or -1 on failure.
|
static |
Creates a new session, sets the calling process to be the process group leader and detaches from the controlling terminal. Returns the session id on success or -1 if an error occurred.
Create's file "filename" with permissions "permissions" and puts the current process id in it. Note that when you delete this file during shutdown you must use the full pathname since the process::detach() method changes directories to "/". Returns true on success and false on failure.
Detaches the current process from the controlling tty, creates a new session, changes directory to / and sets the file creation mask to 0. Use this method to "fork your process into the background." Returns true on success and false on failure.
Runs "command" with arguments "args", replacing the current running process with this new process. Whether the new process retains the process id of the current process is platform-specific. Note that the first element of array "args" should be the name of the command that you want to run, typically the same as "command". "args" should be NULL terminated. Returns false if an error occurred and otherwise does not return at all.
Terminates the calling process and sets the exit status to "status".
Terminates the calling process "immediately" (without calling any functions registered to run at exit) and sets the exit status to "status".
Sets up a default handler that calls exitImmediately() if the process crashes with a program error signal - SIGABRT, SIGFPE, SIGILL, SIGSEGV, SIGBUS, SIGIOT, SIGEMT or SIGSYS.
Calls exitOnShutDown(), exitOnCrash() below.
Sets up a default handler that calls exitImmediately() when the process receives with a shutdown signal:
|
static |
Creates a child process. The child is a duplicate of the parent inheriting file descriptors and a copy of the parent's address space. The child does not have access to the parent's address space. In the parent process, the process id of the child is returned. In the child process, 0 is returned. -1 is returned if an error occurred and no child is forked.
|
static |
This method causes the process to get information about a change in process state for the specified child process id "pid" (ie. whether it was stopped, continued or killed).
Returns the process id of the child that changed state, -1 if an error occurred, and 0 if "wait" is set false and no child has changed state.
Setting "pid" to -1 causes the method to get information on any child of the current process.
Waiting on child processes to exit prevents so-called "zombie" processes from occurring. However, this method only waits when called. To configure a process to automatically wait on and respond when any child process exits, use setWaitForChildren(true).
If "wait" is set true then the method waits until a child state-change occurs. Otherwise the method returns immediately.
If "ignorestop" is set true then the method ignores when a child process has been stopped.
If "ignorecontinue" is set true then the method ignores when a child process has been continued.
On exit, if "newstate" is non-NULL then it is populated with one of the members of the childstatechange enum, indicating the new state of the child process.
If "newstate" is EXIT_CHILDSTATECHANGE and "exitstatus" is non-null then "exitstatus" is populated with the exit status of the child process.
If "newstate" is TERMINATED_CHILDSTATECHANGE or STOPPED_CHILDSTATECHANGE and "signum" is non-null then "signum" is populated with the signum that terminated or stopped the child process.
If "newstate" is TERMINATED_CHILDSTATECHANGE and "coredump" is non-null then "coredump" is set true if a core dump was produced and false otherwise.
|
static |
Returns the signalhandler used internally to handle a crash. Calls to exitOnCrash(), exitOnCrashOrShutDown(), setShutDownFlagOnCrash(), or setShutDownFlagOnCrashOrShutDown() configure this set to include a standard set of crash signals. If you would like to add signals, or change the signal handler from the default to a custom handler, then you can use this method to get the signalhandler.
|
static |
Returns the effective group id of the current process.
|
static |
Returns the effective user id of the current process.
|
static |
Returns the real group id of the current process.
|
static |
Returns the process id of the parent process of the current process.
|
static |
Returns the process group id of the current process.
|
static |
Returns the process group id of the process pid.
|
static |
Returns the process id of the current process.
Returns true if failed fork() calls will be retried and false otherwise.
|
static |
Returns the session id of the current process.
|
static |
Returns the session id of the process pid.
Returns the value of a global shutdown flag indicating whether or not the process has been asked to shut down.
For example...
The "shutdownhandler" of setShutDownHandler() may simply call setShutDownFlag(true) and then return.
The process may then call getShutDownFlag() at strategic points to see if a shutdown has been requested and shut down gracefully if it has.
All rudiments methods that retry interrupted system calls (see retryInterrupted*() methods of various classes) exit if this flag is set upon return from the system call.
Returns the value of a global variable indicating which signal was received that initiated the process shut down.
For example...
The "shutdownhandler" of setShutDownHandler() may call setShutDownSignal(signum), passing it the value of its "signum" parameter.
Once the process determines that a shutdown has been initiated (perhaps by checking getShutDownFlag(), it may then call getShutDownSignal() to query which signal was received that initiated the shutdown and respond differently to different signals.
|
static |
Returns the signalhandler used internally to handle a shut down. Calls to exitOnShutDown(), exitOnCrashOrShutDown(), setShutDownFlagOnShutDown(), or setShutDownFlagOnCrashOrShutDown() configure this set to include a standard set of shutdown signals. If you would like to add signals, or change the signal handler from the default to a custom handler, then you can use this method to get the signalhandler.
|
static |
Returns the real user id of the current process.
Send signal "signum" to self. Returns true on success and false on failure.
Allows you to register a function to run when the process exits normally.
May be called multiple times to register multiple functions to be run at exit. The functions will be called in the reverse order of their registration.
Note that there is no way to deregister a function from running at exit once it has been registered.
Returns true on success and false on failure.
Send signal "signum" to process "processid". Returns true on success and false on failure.
Allows you to designate a function to run if the process crashes with a program error signal - SIGABRT, SIGFPE, SIGILL, SIGSEGV, SIGBUS, SIGIOT, SIGEMT or SIGSYS.
Each time setCrashHandler() is called, the crash handler replaces the previously designated handler.
Sets the effective group id of the current process to the group id of "groupname". Does not set the real or saved group ids.
Returns true on success and false on failure.
Sets the effective group id of the current process to gid. Does not set the real or saved group ids.
Returns true on success and false on failure.
Sets the effective user id of the current process to the user id of "username". Does not set the real or saved user ids.
Returns true on success and false on failure.
Sets the effective user id of the current process to uid. Does not set the real or saved user ids.
Returns true on success and false on failure.
|
static |
Set file/directory creation mode mask to "mask". Returns the mask that was previously in effect.
Sets the effective group id of the current process to the group id of "groupname". If the effective group id is root, the real and saved group ids are also set.
Returns true on success and false on failure.
Sets the effective group id of the current process to gid. If the effective group id is root, the real and saved group ids are also set.
Returns true on success and false on failure.
Sets the process group id of the current process to the current process id.
Sets the process group id of the current process to pgid.
Sets the process group id of the process pid to pgid.
Sets the real group id of the current process to gid and the effective group id of the current process to egid. If the real group id is set or the effective group id is set to a value not equal to the previous real group id, the saved group id is set to the new effective group id.
Returns true on success and false on failure.
Sets the real user id of the current process to uid and the effective user id of the current process to euid. If the real user id is set or the effective user id is set to a value not equal to the previous real user id, the saved user id is set to the new effective user id.
Returns true on success and false on failure.
Sets a global shutdown flag indicating whether or not the process has been asked to shut down.
For example...
The "shutdownhandler" of setShutDownHandler() may simply call setShutDownFlag(true) and then return.
The process may then call getShutDownFlag() at strategic points to see if a shutdown has been requested and shut down gracefully if it has.
All rudiments methods that retry interrupted system calls (see retryInterrupted*() methods of various classes) exit if this flag is set upon return from the system call.
Sets up a default handler that sets the global shutdown flag (see setShutDownFlag()) to true the process crashes with a program error signal - SIGABRT, SIGFPE, SIGILL, SIGSEGV, SIGBUS, SIGIOT, SIGEMT or SIGSYS.
Calls setShutDownFlagOnShutDown(), setShutDownFlagOnCrash() below.
Sets up a default handler that sets the global shutdown flag (see setShutDownFlag()) to true when the process receives a termination signal:
Allows you to designate a function to run when the process receives a termination signal:
Each time setShutDownHandler() is called, the shut down handler replaces the previously designated handler.
Sets a global variable indicating which signal was received that initiated the process shut down.
For example...
The "shutdownhandler" of setShutDownHandler() may call setShutDownSignal(signum), passing it the value of its "signum" parameter.
Once the process determines that a shutdown has been initiated (perhaps by checking getShutDownFlag(), it may then call getShutDownSignal() to query which signal was received that initiated the shutdown and respond differently to different signals.
Sets the effective user id of the current process to the user id of "username". If the effective user id is root, the real and saved user ids are also set.
Returns true on success and false on failure.
Sets the effective user id of the current process to uid. If the effective user id is root, the real and saved user ids are also set.
Returns true on success and false on failure.
If "wait" is true then this method causes the process will automatically wait on child processes which have exited, preventing so-called "zombie" processes from occurring.
If "wait" is false then this method causes the process not to wait on child processes which have exited. Ordinarily, you'd want to wait on child processes, but that interferes with the behavior of WEXITSTATUS() after a call to system() (and possibly other calls). This method allows you to disable waiting on child processes.
Defaults to false.
|
static |
Runs "command" with arguments "args" as a new process. If "detached" is true then the process is started in the background. Use this instead of a combinination of fork() and exec() as it is more efficient on some platforms. Note that the first element of array "args" should be the name of the command that you want to run, typically the same as "command". "args" should be NULL terminated. Returns the process id of the child process or -1 if an error occurred.
Returns true of this platform supports fork() and false otherwise.
Returns true if the platform supports waiting for a child process to exit or change state and false otherwise.
This methods causes the process to wait until child process "pid" exits. Returns true on success and false on failure.
This methods causes the process to wait until child process "pid" exits. If "exitstatus" is non-null then it is populated with the exit status of the child process. Returns true on success and false on failure.
Appends the backtrace for the current thread to "filename".
If "filename" doesn't already exist then it will be created with rw----— permissions.
A maximum of 128 stack frames will be included in the backtrace.
(Not supported on all platforms.)
|
static |
Appends the backtrace for the current thread to "filename".
If "filename" doesn't already exist then it will be created with "perms" permissions.
A maximum of "maxframes" stack frames will be included in the backtrace.
(Not supported on all platforms.)
Writes the backtrace for the current thread to "buffer".
A maximum of 128 stack frames will be included in the backtrace.
(Not supported on all platforms.)