later 1.3.2

• Fixed unused variable compiler warning. (@MichaelChirico, #176)

• Fixed #177: The order of includes in later.h could cause compilation errors on some platforms. (@jeroen, #178)

• Closed #181: Fix R CMD check warning re error() format strings (for r-devel). (#133)

later 1.3.1

• For C function declarations that take no parameters, added void parameter. (#172)

later 1.3.0

• Closed #148: When later was attached, parallel::makeForkCluster() would fail. (#149)

• Fixed #150: It was possible for callbacks to execute in the wrong order if the clock time was changed in between the scheduling of two callbacks. (#151)

later 1.2.0

• Closed #138: later is now licensed as MIT. (#139)

• Closed #140: Previously, the event loop stopped running if the R process was forked. (#141)

• Closed #143: Packages which link to later no longer need to take a direct dependency on Rcpp, because later.h no longer includes Rcpp.h. (#144)

• Removed dependency on the BH package. C++11 is now required. (#147)

later 1.1.0.1

• Private event loops are now automatically run by their parent. That is, whenever an event loop is run, its children event loops are automatically run. The create_loop() function has a new parameter parent, which defaults to the current loop. The auto-running behavior can be disabled by using create_loop(parent=NULL). (#119)

• Fixed #73, #109: Previously, later did not build on some platforms, notably ARM, because the -latomic linker was needed on those platforms. A configure script now detects when -latomic is needed. (#114)

• Previously, execLaterNative was initialized when the package was loaded, but not execLaterNative2, resulting in a warning message in some cases. (#116)

later 1.0.0

• Added private event loops: these are event loops that can be run independently from the global event loop. These are useful when you have code that schedules callbacks with later(), and you want to call run_now() block and wait for those callbacks to execute before continuing. Without private event loops, if you call run_now() to wait until a particular callback has finished, you might inadvertantly run other callbacks that were scheduled by other code. With private event loops, you can create a private loop, schedule a callback on it, then call run_now() on that loop until it executes, all without interfering with the global loop. (#84)

later 0.8.0

• Fixed issue #77: On some platforms, the system's C library has support for C11-style threads, but there is no threads.h header file. In this case, later's configure script tried to use the tinycthread, but upon linking, there were function name conflicts between tinycthread and the system's C library. Later no longer tries to use the system's threads.h, and the functions in tinycthread were renamed so that they do not accidentally link to the system C library's C11-style thread functions. PR #79

• Added all argument to run_now(); defaults to TRUE, but if set to FALSE, then run_now will run at most one later operation before returning. PR #75

• Fixed issue #74: Using later with R at the terminal on POSIX could cause 100% CPU. This was caused by later accidentally provoking R to call its input handler continuously. PR #76

• Fixed issue #73: Linking later on ARM failed because boost::atomic requires the -lboost_atomic flag. Now later tries to use std::atomic when available (when the compiler supports C++11), and falls back to boost::atomic if not. PR #80

later 0.7.5

• Fixed issue where the order of callbacks scheduled by native later::later could be nondeterministic if they are scheduled too quickly. This was because callbacks were sorted by the time at which they come due, which could be identical. Later now uses the order of insertion as a tiebreaker. PR #69

later 0.7.4

• Fixed issue #45 and #63: glibc 2.28 and musl (used on Arch and Alpine Linux) added support for C11-style threads.h, which masked functions from the tinycthread library used by later. Later now detects support for threads.h and uses it if available; otherwise it uses tinycthread. PR #64

later 0.7.3

• Fixed issue #57: If a user interrupt occurred when later (internally) called sys.nframe(), the R process would crash. PR #58