| 1 |
Revision history for Perl extension IPC::Shareable. |
| 2 |
|
| 3 |
1.13 2022-10-11 |
| 4 |
- In singleton(), do a check whether class was sent in. There was a shifting |
| 5 |
issue if called with IPC::Shareable::singleton() as opposed to |
| 6 |
IPC::Shareable->singleton() |
| 7 |
- When an exclusive collision occurs and both 'graceful' and 'warn' are set, |
| 8 |
the warning thrown now includes the segment/semaphore key in use. This |
| 9 |
makes it easier to clean things up with 'ipcrm' |
| 10 |
|
| 11 |
1.12 2022-03-13 |
| 12 |
- Add tests in t/07-new.t to test how using tied() against a dereferenced |
| 13 |
variable returned from new() allows access to the underlying |
| 14 |
IPC::Shareable object |
| 15 |
- Reworked how spawn() and unspawn() manage things internally |
| 16 |
- POD updates, fixes and clarifications |
| 17 |
- Removed spawn() and unspawn(). There's really no need for these as one |
| 18 |
can simply create but not destroy |
| 19 |
|
| 20 |
1.11 2022-03-07 |
| 21 |
- Fix certain tests so they don't throw if we bail early |
| 22 |
|
| 23 |
1.10 2022-03-07 |
| 24 |
- Fix issue where if segments were created underneath of a parent by an |
| 25 |
external process other than the process that created the parent, the |
| 26 |
global register wasn't being updated, therefore those segments weren't |
| 27 |
being removed when calling clean_up_all() |
| 28 |
- All test files now check number of segments before and after the entire |
| 29 |
script has run, verifying that all segments were cleaned up ok |
| 30 |
- The test suite itself in its entirety makes sure that all segments |
| 31 |
created during the suite run are cleaned up properly |
| 32 |
- Added CI_TESTING=1 to coverage CI tests |
| 33 |
|
| 34 |
1.09 2022-03-06 |
| 35 |
- Made 81-fork_dup_rand_keys.t developer only (but still need to figure out |
| 36 |
why it only fails on FreeBSD) |
| 37 |
|
| 38 |
1.08 2022-03-04 |
| 39 |
- Added 'protected' option and clean_up_protected(). If set, the cleanup |
| 40 |
methods, nor the automatic 'destroy' functionality will remove the segment |
| 41 |
the option was set in, nor any children created underneath of it. A call |
| 42 |
to clean_up_protected() is required to remove them |
| 43 |
- Updated attribute tests to include all available options |
| 44 |
- _shm_key_rand() now checks EXCLUSIVE_CHECK_LIMIT times to find a key for |
| 45 |
a free segment. It croaks if not |
| 46 |
- Added _shm_key_rand_int() so we can mock it to test the dup random key |
| 47 |
checks |
| 48 |
- Added build prereq of Mock::Sub to test the random key function |
| 49 |
|
| 50 |
1.07 2022-03-04 |
| 51 |
- When running in forked environments, _shm_key_rand() was consistently |
| 52 |
returning the same 'random' key to each forked process. We've fixed this |
| 53 |
issue, and added tests for it (fixed by adding a call to srand() before |
| 54 |
rand) |
| 55 |
- Move $SIG{CHLD} handlers to spawn() and unspawn() instead of having them |
| 56 |
global (fixes #15) |
| 57 |
- t/65-seg_size.t fails on 32-bit perls; Add a check and bail if we're on |
| 58 |
a perl < 64-bit (fixes #14) |
| 59 |
|
| 60 |
1.06 2021-09-08 |
| 61 |
- Fix significant bug where at least on MacOS, if we generated a CRC of a |
| 62 |
key that was greater than 0x80000000, it would round down the key to that |
| 63 |
number exactly, causing duplicates, and incorrect segment access |
| 64 |
- Added new() method, returns a reference (default hash) without having to |
| 65 |
do the tie() directly |
| 66 |
- Fix issue in JSON decoder where an empty JSON string may have caused |
| 67 |
decode failure |
| 68 |
- Modify t/65-seg_size.t beyond RAM max test to cover varying error messages |
| 69 |
on differing OSs |
| 70 |
- Renamed t/67-out_of_memory.t as the tests are related to exhausting SHM |
| 71 |
slots, not RAM |
| 72 |
|
| 73 |
1.05 2021-07-16 |
| 74 |
- Bump prereq version of Test::SharedFork to 0.35 |
| 75 |
- Added ability to use JSON as the serializer |
| 76 |
- Added ipcs() |
| 77 |
- Added test to ensure we croak if data size is greater than segment size |
| 78 |
- Added 'tidy' attribute, removes unneeded nested structure segments |
| 79 |
- In SharedMem, added _key attribute, and added set/get methods |
| 80 |
- Changed key generation from using padding, pack and unpack to using |
| 81 |
String::CRC32's crc32(), which allows arbitrary, unpadded strings |
| 82 |
- Added attributes() allows fetching one, or all of the object's atttributes |
| 83 |
- Added 'limit' attribute, by setting false, allows a user to set a segment |
| 84 |
size larger than our internal 1GB default |
| 85 |
- Cleaned up exception throwing (particularly if seg size eats up all |
| 86 |
memory, we throw an appropriate message) |
| 87 |
- Display proper exception if we try to exhaust all available shared memory |
| 88 |
segments (fixes #3) |
| 89 |
|
| 90 |
1.04 2021-06-28 |
| 91 |
- Skip unspawn tests for perls with -Duselongdouble, as Storable is not |
| 92 |
compatible storing/retrieving such numbers (closes #5) |
| 93 |
|
| 94 |
1.03 2021-06-25 |
| 95 |
- Add newline to singleton() warn so that it doesn't print out trace info |
| 96 |
|
| 97 |
1.02 2021-06-25 |
| 98 |
- Remove erroneous listing from MANIFEST |
| 99 |
- Work around issue in t/65-seg_size.t where MacOS and FreeBSD don't have |
| 100 |
the -i flag to ipcs (thanks for the PR, Roland Walker!) |
| 101 |
- Added singleton(), ensures that only a single instance of a script can |
| 102 |
run at any given time |
| 103 |
- Added tests to ensure we croak() if create is not set and the segment |
| 104 |
doesn't yet exist |
| 105 |
- Fix POD issue where we stated exclusive returns undef instead of croaking |
| 106 |
(fixes #10) |
| 107 |
- Change 'perl' shell call to $^X in t/71-unspawn.t (fixes #6) |
| 108 |
|
| 109 |
1.01 2021-06-24 |
| 110 |
- Added 'graceful' flag option. With exclusive, if a second process |
| 111 |
attempts to tie an exclusive segment, we exit gracefully instead of |
| 112 |
croaking |
| 113 |
|
| 114 |
1.00 2021-06-09 |
| 115 |
- Project adopted by Steve Bertrand |
| 116 |
- Added global_register() and process_register() to return the |
| 117 |
registries of active segments and semaphores |
| 118 |
- Added seg() and sem() methods, returns the structure of a shared |
| 119 |
memory segment and a sempahore respectively |
| 120 |
- Added initial "persist" hash variables. Separate processes (even |
| 121 |
multiple files in multiple windows) can share a variable, and it'll |
| 122 |
remain available even after all processes exit. The variable will be |
| 123 |
re-attached if the same shared segment key is used in subsequent runs |
| 124 |
- Reversed order of Changes file |
| 125 |
- Removed trace() and debug() code for ease of reading |
| 126 |
- Improved exception/error messages, added exception testing, reduced |
| 127 |
and minimized the number of exceptions actually uncaught |
| 128 |
- Significant POD cleanup |
| 129 |
- 100% rewrite of all test files, and added a slew of new ones, all |
| 130 |
using Test::More |
| 131 |
- Add build requirement of Test::SharedFork to handle the out of |
| 132 |
sequence fork() tests in t/35-clean.t and t/30-lock_operations |
| 133 |
- Major POD updates |
| 134 |
- Added tests to prove that RT 123057 isn't really an issue (ie. segment |
| 135 |
size parameter works correctly) |
| 136 |
|
| 137 |
0.61 Mon Oct 8 00:27:39 2012 |
| 138 |
- Added patch from Frank Lichtenheld fixing IPC::Shareable's dependence |
| 139 |
on the presence of a perl bug which is no longer present in |
| 140 |
perl >= 5.10 |
| 141 |
- Fixed bug reported by Dan Harbin where the FETCH operation on a tie()d |
| 142 |
string containing HASH, ARRAY, or SCALAR fails because it was using |
| 143 |
the stringification of the data to determine what kind of reference |
| 144 |
it was. Now using Scalar::Util::reftype |
| 145 |
- Added missing dependency on IPC::Semaphore to Makefile.PL (reported |
| 146 |
by Adrian Issott) |
| 147 |
- Added a 'sleep 1' in a test that was hanging on certain systems |
| 148 |
due (possibly) to two alarm signals coming too quickly to the |
| 149 |
child process. From Ton Voon. |
| 150 |
|
| 151 |
0.60 Mon Mar 5 15:20:18 EST 2001 |
| 152 |
- Lee Lindley ([email protected]) added the _was_changed |
| 153 |
optimization, improved the locking functionality, fixed |
| 154 |
numerous bugs, and generally cleaned things up; thanks. |
| 155 |
- Removed support for "no" as a false value in arguments; |
| 156 |
thanks to Dave Rolsky <[email protected]> |
| 157 |
|
| 158 |
0.54 Mon Jan 8 11:52:28 EST 2001 |
| 159 |
- Fix to allow IPC::Shareable to work with 1.0.* versions of |
| 160 |
Storable |
| 161 |
|
| 162 |
0.53 Tue Nov 14 00:33:35 EST 2000 |
| 163 |
- Fixed race condition in test suite causing intermitent |
| 164 |
failures. |
| 165 |
- Better checking for success of calls to Storable::thaw(); |
| 166 |
thanks to Raphael Manfredi <[email protected]>. |
| 167 |
|
| 168 |
0.52 Thu Sep 14 12:30:17 EDT 2000 |
| 169 |
- Now STORE, PUSH, POP, etc all call _thaw() before doing |
| 170 |
their business. |
| 171 |
- Refined SIGALRM handlers in test scripts |
| 172 |
- Fixed concurrency issues affecting tied arrays and hashes; |
| 173 |
thanks to thanks to Robert Emmery <[email protected]>, |
| 174 |
Terry Ewing <[email protected]>, Tim Fries |
| 175 |
<[email protected]>, and Joe Thomas <[email protected]>. |
| 176 |
- Doc fixes thanks to Paul Makepeace |
| 177 |
<[email protected]> |
| 178 |
|
| 179 |
0.51 Fri May 5 23:47:06 EDT 2000 |
| 180 |
- Fixed bug that would cause IPC::Shareable::BUF_SIZ to be |
| 181 |
ignored; thanks to Robert Emmery <[email protected]> and |
| 182 |
Mohammed J. Kabir <[email protected]> for reporting. |
| 183 |
- Stopped tests from leaking shm segments |
| 184 |
- Added test of argument parsing |
| 185 |
- doc fixes |
| 186 |
|
| 187 |
0.50 Tue Mar 21 11:56:32 EST 2000 |
| 188 |
- Complete rewrite incorporating the following changes. |
| 189 |
- Requires 5.00503. This allowed the module to get rid of the |
| 190 |
global cache for shared memory segments; each Shareable |
| 191 |
object now carries around its own data. |
| 192 |
- 5.00503 also allowed tied arrays to be implemented |
| 193 |
- Shared memory segments can no longer be of infinite length |
| 194 |
thereby reducing the amount of code in the module by a |
| 195 |
factor of 2. |
| 196 |
- Uses IPC::Shareable::SharedMem class for accessing |
| 197 |
shared memory. |
| 198 |
- Uses IPC::Semaphore module for accessing semaphores. |
| 199 |
- Completely revisited the way references are dealed with: all |
| 200 |
referenced thingies are now automagically tied to shared |
| 201 |
memory. |
| 202 |
- Constants now imported from IPC::SysV; Shareable.xs is gone |
| 203 |
- Rewrote test suite and moved into t subdirectory |
| 204 |
- Updated man page |
| 205 |
|
| 206 |
0.30 Mon Jan 19 11:13:41 EST 1998 |
| 207 |
- Added SEM_UNDO to semop() calls; Maurice Aubrey. |
| 208 |
- Fixed some bugs in the locking code; Maurice Aubrey. |
| 209 |
- Made calls to debug() conditional for efficency; Maurice Aubrey. |
| 210 |
- Fixed a signal handler in test.pl; Maurice Aubrey. |
| 211 |
|
| 212 |
0.29 Mon Jan 12 13:49:42 EST 1998 |
| 213 |
- $MAXVER patch for when the version semaphore reaches its limit |
| 214 |
and rolls back over to 0; Maurice Aubrey <[email protected]>. |
| 215 |
- patch to quieten things down under -w; Doug MacEachern |
| 216 |
<[email protected]> |
| 217 |
|
| 218 |
0.28 Wed Oct 22 14:59:08 EDT 1997 |
| 219 |
- cleaned up the way thingys are magically tied; Ben Sugars. |
| 220 |
- moved many subroutines so that they are auto-loaded; Ben |
| 221 |
Sugars. |
| 222 |
- updated man page; Ben Sugars. |
| 223 |
|
| 224 |
0.26 Mon Oct 20 10:06:43 EDT 1997 |
| 225 |
- fixed bug regarding assigning a reference to an empty |
| 226 |
hash to a tied variable. Thanks to Jason Stevens. |
| 227 |
|
| 228 |
0.25 Tue Oct 7 14:41:49 EDT 1997 |
| 229 |
- added more checking of sem*() and shm*() return values; |
| 230 |
Michael Stevens and Ben Sugars. |
| 231 |
- added shlock and shunlock; Ben Sugars. |
| 232 |
- fixed bug that would cause modifications of magically |
| 233 |
referenced thingys to fail. Thanks to Maurice Aubrey |
| 234 |
|
| 235 |
0.20 Thu Aug 28 15:13:46 EDT 1997 |
| 236 |
- added ability to magically create ties to implicitly |
| 237 |
referenced thingys; Ben Sugars. |
| 238 |
|
| 239 |
0.18 Thu Aug 28 09:12:30 EDT 1997 |
| 240 |
- fixed garbage collection bug; Ben Sugars. Thanks to |
| 241 |
Michael Stevens <[email protected]> for the patch. |
| 242 |
|
| 243 |
0.17 Wed Aug 27 15:57:11 EDT 1997 |
| 244 |
- fixed some bugs in &parse_argument_hash; Ben Sugars. |
| 245 |
|
| 246 |
0.16 Mon Aug 11 16:10:54 EDT 1997 |
| 247 |
- new shared memory segments now initialized with null |
| 248 |
values; Ben Sugars. |
| 249 |
|
| 250 |
0.15 Fri Aug 8 15:45:29 EDT 1997 |
| 251 |
- implemented locking using semaphores; Ben Sugars. |
| 252 |
|
| 253 |
0.12 Thu Aug 7 14:47:42 EDT 1997 |
| 254 |
- various bug fixes; Ben Sugars. |
| 255 |
|
| 256 |
0.11 Wed Aug 6 10:14:49 EDT 1997 |
| 257 |
- initial implementation of semaphores for versioning and |
| 258 |
caching; Ben Sugars. |
| 259 |
|
| 260 |
0.10 Fri Aug 1 13:32:52 EDT 1997 |
| 261 |
- can now tie both scalars and hashes of arbitrary length; |
| 262 |
Ben Sugars. |
| 263 |
|
| 264 |
0.05 Wed Jul 30 15:02:31 EDT 1997 |
| 265 |
- scalars can now be tied; Ben Sugars. |
| 266 |
|
| 267 |
0.01 Wed Jul 30 09:00:53 1997 |
| 268 |
- original version; created by h2xs 1.18. |