12009-12-07 Andreas Kupries <andreas_kupries@users.sourceforge.net> 2 3 * 4 * Released and tagged Tcllib 1.12 ======================== 5 * 6 72008-12-12 Andreas Kupries <andreas_kupries@users.sourceforge.net> 8 9 * 10 * Released and tagged Tcllib 1.11.1 ======================== 11 * 12 132008-10-16 Andreas Kupries <andreas_kupries@users.sourceforge.net> 14 15 * 16 * Released and tagged Tcllib 1.11 ======================== 17 * 18 192008-10-07 Andreas Kupries <andreas_kupries@users.sourceforge.net> 20 21 * bench.tcl: Bumped to version 0.4 for 2008-06-30 commit by 22 * bench.man: myself. Was a major rewrite of the internals, 23 * pkgIndex.tcl: should have been bumped then. 24 252008-06-30 Andreas Kupries <andreask@activestate.com> 26 27 * bench.tcl (::bench::Invoke): Reworked the protocol between 28 * libbench.tcl: manager and execution system to allow for 29 incremental returning of results and proper progress 30 feedback. This enables users to see how a benchmark progresses, 31 and to provide their own notes about conditions and decisions as 32 well. 33 342007-09-12 Andreas Kupries <andreas_kupries@users.sourceforge.net> 35 36 * 37 * Released and tagged Tcllib 1.10 ======================== 38 * 39 402007-08-23 Andreas Kupries <andreask@activestate.com> 41 42 * bench.tcl: Fixed problem with the glob patterns used to query 43 * bench.man: the data array, was not matching the list quoting 44 * pkgIndex.tcl: used to generate the keys. Was fine while we had 45 no keys with spaces in the interp reference, but with -pkgdir 46 this is possible, and broke. Version bumped to 0.3.1. Reported 47 by Rolf Ade. 48 492007-08-21 Andreas Kupries <andreask@activestate.com> 50 51 * bench.tcl (::bench::run): Extended with a new option -pkgdir 52 * bench.man: helping in the handling of multiple versions of a 53 * pkgIndex.tcl: package to benchmark, as suggested and first 54 * libbench.tcl: implemented by Rolf Ade. Moved invokation of 55 libbench to a separate helper procedure. Extended the 56 documentation. Version bumped to 0.3. 57 582007-04-10 Andreas Kupries <andreas_kupries@users.sourceforge.net> 59 60 * bench_lang_intro.man: New files, documentation of the 61 * bench_lang_spec.man: benchmark declaration language, and 62 * bench_read.man: of the supporting packages. 63 * bench_wcsv.man: 64 * bench_wtext.man: 65 662007-01-22 Andreas Kupries <andreas_kupries@users.sourceforge.net> 67 68 * libbench.tcl: Added new benchmark options -ipre, -ipost. Per 69 * pkgIndex.tcl: iteration pre/post scripts, untimed. Version of 70 * bench.cl: package 'bench' is now 0.2. 71 722007-01-21 Andreas Kupries <andreas_kupries@users.sourceforge.net> 73 74 * bench_wcsv.tcl: Fixed sorting of descriptions in text and 75 * bench_wtext.tcl: csv output. Version is now 0.1.2. 76 772007-01-18 Andreas Kupries <andreas_kupries@users.sourceforge.net> 78 79 * bench.tcl (bench::norm): Removed 'split ,' from code, was left 80 * pkgIndex.tcl: in wrongly after the rewrite of the raw 81 representation. The relevant key is a list which we can and have 82 to use directly, no split required. The fixed bug caused the 83 normalization to fail and return the empty string for all 84 cells. Version number bumped to 0.1.1 for this. 85 862006-10-03 Andreas Kupries <andreas_kupries@users.sourceforge.net> 87 88 * 89 * Released and tagged Tcllib 1.9 ======================== 90 * 91 922006-06-13 Andreas Kupries <andreask@activestate.com> 93 94 * bench_read.tcl: Rewrite the internal raw representation, use 95 * bench.tcl: lists as array keys, easier to handle, no 96 * bench_wcsv.tcl: splitting, and quoting is done automatically 97 * bench_wtext.tcl: by Tcl itself. See [Tcllib SF Bug 1414159]. 98 992006-05-27 Andreas Kupries <andreas_kupries@users.sourceforge.net> 100 101 * bench_read.tcl: Fixed typo "-error" --> "-code error". 102 1032006-01-25 Michael Schlenker <mic42@users.sourceforge.net> 104 105 * bench_wcsv.tcl : Fixed bug when trying to format benchs on windows. 106 * bench_wtext.tcl: The interpreter path was truncated due to a misuse of 107 split and lindex, where string first was appropriate. 108 1092005-10-27 Andreas Kupries <andreask@activestate.com> 110 111 * bench.tcl (::bench::norm): Fixed bug leaving time data in 112 non-reference column when the reference is empty. To the unwary 113 the result looks like factors, which have ridiculous values. Now 114 the row is shown, but empty. 115 1162005-10-21 Andreas Kupries <andreask@activestate.com> 117 118 * bench.tcl (::bench::del): New command. Removal of a column from 119 benchmark data. 120 1212005-10-18 Andreas Kupries <andreask@activestate.com> 122 123 * bench_read.tcl: New file. Command and package to read benchmark 124 data in the text, csv, or raw formats. 125 126 * bench.tcl (::bench::edit): New command. Changes specified 127 interpreter path to user specified value. Needed if we wish to 128 merge data coming from the same interpreter, for different 129 revisions of the package under test. 130 1312005-10-17 Andreas Kupries <andreas_kupries@users.sourceforge.net> 132 133 * Package derived from the original code added to Tcllib. 134 1352004-12-29 Jeff Hobbs <jeffh@ActiveState.com> 136 137 * tcl/parse.bench: ensure file size is consistent between interp 138 runs with formatted BOUND string. 139 1402004-12-27 Jeff Hobbs <jeffh@ActiveState.com> 141 142 * doc/runbench.1: fix doc for -throwerrors [Bug 1091766] 143 144 * runbench.tcl (getInterps): use exec << instead of echo [Bug 1091764] 145 1462004-12-24 Miguel Sofer <msofer@users.sf.net> 147 148 * tcl/namespace.bench: new benchmark, measures the cost of calling 149 the same global command alternating different namespaces. 150 1512004-12-20 Jeff Hobbs <jeffh@ActiveState.com> 152 153 * tcl/array.bench (new): array hash benchmarks 154 155 * tcl/file.bench: fix checkall to operate for tclsh <=8.0 156 157 * tcl/string.bench: fix string match -nocase for tclsh <=8.2 158 159 * runbench.tcl (convertVersion): add -globtclsh -globwish file 160 path glob opts (tclsh* and wish* by default). 161 Normalize soft-links. 162 163 * normbench.tcl (normalize-text): harden time norm check 164 1652003-08-06 Jeff Hobbs <jeffh@ActiveState.com> 166 167 * normbench.tcl (normalize): correct normalization of new-style 168 stats where TclX data is present in output. 169 1702003-02-11 Jeff Hobbs <jeffh@ActiveState.com> 171 172 * tcl/list.bench: lsearch -regexp benchmarks 173 174 * tcl/file.bench: updated with more benchmarks 175 1762003-02-08 Jeff Hobbs <jeffh@ActiveState.com> 177 178 * tcl/startup.bench: replaced by file benchmarks 179 * tcl/file.bench: file benchmarks 180 1812002-11-13 Jeff Hobbs <jeffh@ActiveState.com> 182 183 * tcl/regexp.bench: added anchored re tests 184 185 * tcl/klist.bench: allow method filter from command lineinvocation. 186 187 * tcl/list.bench: add lset benchmarks 188 189 * tcl/md5.bench: correct to work with pre-8.2 interps 190 191 * tcl/string.bench: add string growth, remove split benchmarks 192 * tcl/split.bench: more split benchmarks 193 194 * runbench.tcl: allow tclsh*/wish* (no version required) 195 1962002-07-24 Miguel Sofer <msofer@users.sourceforge.net> 197 198 * tcl/base64.bench: added the current code from tcllib. 199 2002002-06-20 Miguel Sofer <msofer@users.sourceforge.net> 201 202 * tcl/read.bench: modified to actually "use" the data being read 203 by setting a local variable. 204 2052002-06-20 Miguel Sofer <msofer@users.sourceforge.net> 206 207 * tcl/md5.bench: added the faster implementation from tcllib 208 2092002-06-12 Jeff Hobbs <jeffh@ActiveState.com> 210 211 * tcl/catch.bench: corrected use of string map in toplevel code 212 213 * tcl/expr.bench: corrected use of string repeat in toplevel code 214 215 * tcl/sha1.bench: correct wideint problem for 8.4 in sha1DF 216 217 * tcl/string.bench: corrected string equality checks to use 218 different variables (objects) 219 220 * tcl/gccont.bench: new benchmark that does some bioinformatics 221 manipulation on dna sequences 222 2232002-06-12 Miguel Sofer <msofer@users.sourceforge.net> 224 225 * tcl/klist.bench: 226 * tcl/heapsort.bench: added algorithms using [lset] 227 2282002-06-11 Miguel Sofer <msofer@users.sourceforge.net> 229 230 * tcl/regexp.bench: made the bench access the match variables, to 231 benchmark also the read access to them. 232 * tcl/vars.bench: added a "VAR ref local" benchmark, to be able to 233 compare the access times of linked variables to those of local 234 variables. 235 2362002-05-29 Jeff Hobbs <jeffh@ActiveState.com> 237 238 * tcl/parse.bench: more complex string parsing benchmark (8.0+) 239 240 * tcl/encoding.bench: start of some encoding benchmarks (8.1+) 241 242 * tcl/expr.bench: added ==/!= expr benchmarks 243 244 * tcl/string.bench: corrected the equality benchmarks to not use 245 the same object unless specified. 246 2472002-04-25 Jeff Hobbs <jeffh@ActiveState.com> 248 249 * runbench.tcl: 250 * libbench.tcl: added ability to set # threads to use if Thread 251 package can be loaded. 252 improved -result error checking 253 254 * tcl/base64.bench: verify result of encode/decode 255 256 * tcl/proc.bench: added empty proc benchmarks 257 258 * tcl/list.bench: added LIST concat benchmarks (hartweg) 259 2602002-03-27 Miguel Sofer <msofer@users.sourceforge.net> 261 262 * tcl/catch.bench: modified the catch benchmarks to allow 263 comparison with catching non-error exceptions; added new 264 "CATCH except" benchmark. 265 2662002-03-15 Miguel Sofer <msofer@users.sourceforge.net> 267 268 * tcl/catch.bench: added benchmark for catch in a body with many 269 nested exception ranges. 270 2712002-02-22 Jeff Hobbs <jeffh@ActiveState.com> 272 273 * tcl/loops.bench: added while 1 benchmark 274 275 * tcl/conditional.bench: added if 1/0 benchmark 276 2772002-02-07 Jeff Hobbs <jeffh@ActiveState.com> 278 279 * runbench.tcl: noted thread option. 280 281 * libbench.tcl: added ability to check result of test 282 283 * tcl/base64.bench: stripped arg stuff out of code to make it work 284 in 8.0 as well. 285 286 * tcl/list.bench: corrected list-2.11 to append to simple var. 287 288 * tcl/map.bench: added http mapReply & simple regsubs benchmarks 289 290 * tcl/read.bench: commented out new changing buffersize benchmarks 291 as they do weird things to various interp versions. 292 293 * tcl/regexp.bench: added static regexp benchmarks 294 295 * tcl/string.bench: added string first utf benchmarks 296 297 * tcl/vars.bench: corrected namespace usage for pre-8 interps. 298 2992001-09-25 Jeff Hobbs <jeffh@ActiveState.com> 300 301 * tcl/string.bench: added exact string match benchmark and fixed 302 other string match benchmarks 303 304 * tcl/list.bench: added simple list benchmark 305 306 * tcl/vars.bench: added mset benchmarks 307 308 * libbench.tcl: 309 * runbench.tcl: added support for -threads option to try and load 310 a thread package and run separate benchmark files simultaneously. 311 3122001-08-29 Jeff Hobbs <jeffh@ActiveState.com> 313 314 * tcl/methods.bench: 315 * tcl/vars.bench: added some more benchmarks 316 3172001-07-18 Andreas Kupries <andreas_kupries@users.sourceforge.net> 318 319 * tcl/read.bench: new "read" benchmarks detailing the effect of 320 the buffersize on IO performance. Created to check out the 321 performance patch associated with SF item #427196. 322 3232001-06-19 Jeff Hobbs <jeffh@ActiveState.com> 324 325 * tcl/binary.bench: new "binary" benchmarks 326 327 * tcl/string.bench: more random split benchmarks 328 3292001-06-03 Jeff Hobbs <jeffh@ActiveState.com> 330 331 * libbench.tcl: 332 * runbench.tcl: reduced default iterations to 1000 (still quite 333 sufficient to remove random noise). 334 3352001-05-31 Jeff Hobbs <jeffh@ActiveState.com> 336 337 * tcl/conditional.bench: added switch/if comparison bench. 338 339 * tcl/base64.bench: new benchmark with base64 code (from tcllib). 340 341 * tcl/md5.bench: new benchmark with Libes' md5 (from tcllib). 342 343 * tcl/sha1.bench: new benchmark with a couple of pure tcl sha1 344 routines (Libes and Fellows). 345 3462001-05-29 Andreas Kupries <andreas_kupries@users.sourceforge.net> 347 348 * doc/libbench.n: 349 * doc/runbench.1: 350 * doc/normbench.1: Added documentation of benchmark library and 351 applications. 352 353 * doc: Added documentation directory. 354 3552001-05-22 Jeff Hobbs <jeffh@ActiveState.com> 356 357 * runbench.tcl: corrected error for reporting errors in sourced files 358 359 * tcl/fcopy.bench: made use of bench_tmpfile for more accurate 360 data (not skewed by network). 361 362 * libbench.tcl (bench_tmpfile): correctly allow multiple calls to 363 bench_tmpfile within one file. 364 365 * normbench.tcl: new file that allows for post-process 366 normalization of the benchmark data. 367 Corrected last minute code checkin bug. 368 Added support for moving left (to higher versions) to normalize 369 when the requested version returned non-double data. 370 371 * tcl/libbench.tcl: 372 * tcl/runbench.tcl: changed -iterations to be a maximum number for 373 timings, to override any larger number the benchmark may set for 374 itself. 375 Rearranged result format of benchmarks to return data by benchmark 376 description. Benchmarks are now always returned in alphabetical 377 order of the benchmark description. 378 Changed benchmarks to rerun the interpreter per benchmark file 379 instead of sourcing all files into the same interpreter. This 380 reduces any skew related to excessive mem usage or other factors 381 that may arise for one benchmark file. 382 Changed midpoint numbers to time elapsed calculation. 383 Added -normalize option that post-processes the time information 384 to normalize against one version as a baseline. 385 Changed -errors <bool> to -throwerrors with no arg, and changed 386 the default to not throw errors in benchmark files. 387 Added version string to verbose run info. 388 389 * tcl/klist.bench: added support for <8.0 to all benchmarks except 390 shuffle0, with notably reduced default run iters due to extreme 391 slowness of <8.0 interps for these tasks. 392 393 * tcl/string.bench: 394 * tcl/regexp.bench: fixed incorrect str-repeat replacement function 395 3962001-05-18 Jeff Hobbs <jeffh@ActiveState.com> 397 398 * tcl/string.bench: added <8.0 compatible rev-recursive benchmark, 399 fixed non-octal escape in ustring instantiation. 400 401 * tcl/wordcount.bench: added <8.1 compatible benchmarks 402 403 * tcl/methods.bench: return for interps <8.0 404 4052001-05-19 Andreas Kupries <andreas_kupries@users.sourceforge.net> 406 407 * tcl/conditional.bench: Changed some descriptions to make them 408 unique and matching to the code. 409 410 * tcl/fcopy.bench: New benchmarks for the [fcopy] command 411 (unsupported0 in older versions of the core). 412 4132001-05-16 Jeff Hobbs <jeffh@ActiveState.com> 414 415 * tcl/string.bench: added static string length benchmarks 416 417 * tcl/wordcount.in: 418 * tcl/wordcount.bench: wordcount benchmarks 419 420 * tcl/heapsort.bench: new file with heapsort benchmark 421 * tcl/string.bench: 422 * tcl/matrix.bench: 423 * tcl/regexp.bench: extended benchmarks 424 4252001-05-11 Jeff Hobbs <jeffh@ActiveState.com> 426 427 * tcl/string.bench: clarified string reverse benchmarks, added 428 more to the string compare benchmarks. 429 430 * tcl/matrix.bench: some new matrix benchmarks. Basically a seed 431 file looking for more. procs courtesy Sofer. 432 433 * tcl/list.bench: added a list-iter benchmark 434 435 * tcl/klist.bench: reduced default iters in klist.bench. Accuracy 436 seems about the same without the wait... 437 438 * libbench.tcl: 439 * runbench.tcl: added support for -rmatch option (regexp match of 440 benchmark description). 441 Added MIDPOINT verbose puts for interim time notes. 442 4432001-04-11 Jeff Hobbs <jeffh@ActiveState.com> 444 445 * tcl/klist.bench: added shuffle5* from wiki. 446 4472001-03-28 Jeff Hobbs <jeffh@activestate.com> 448 449 * tcl/string.bench: fixed str-first proc that had bogus code in it. 450 added more split benchmarks for dkf's split improvement in 8.4. 451 452 * tk/canvas.bench: expanded item draw benchmarks 453 4542001-03-23 <jeffh@activestate.com> 455 456 * tk/canvas.bench: added simple item draw benchmarks 457 4582001-03-15 <jeffh@activestate.com> 459 460 * tcl/klist.bench: improved non-tclbench data output. 461 462 * runbench.tcl: added more error capturing. 463 464 * tcl/string.bench: fixed calls to string repeat to work with 465 <8.1.1 interps. 466 467 * tcl/klist.bench: new file to benchmark various list shuffling 468 techniques (from wiki). 469 * tcl/methods.bench: new file to benchmark various method 470 invocation speeds (petasis). 471 4722000-10-19 Jeff Hobbs <hobbs@ajubasolutions.com> 473 474 * tcl/string.bench (str-append-2): added more append tests 475 4762000-08-30 Jeff Hobbs <hobbs@scriptics.com> 477 478 * tcl/string.bench: made string repeat calls compatible with 479 pre-8.1.1 interpreters. 480 481 * libbench.tcl (bench_tmpfile): add env to global list 482 4832000-08-29 Eric Melski <ericm@ajubasolutions.com> 484 485 * tcl/string.bench: Extended string append benchmarks to exploit 486 new growth algorithm for string objects in Tcl 8.4a2. 487 4882000-05-31 Jeff Hobbs <hobbs@scriptics.com> 489 490 * runbench.tcl: new options -errors (passed to libbench), -verbose 491 (by default we are now quieter on output), -output <text|list|csv> 492 (different output types - csv is char-sep-value for Excel). 493 Added start/finish times (in -verbose mode). 494 * libbench.tcl: libbench now takes -option switches for 495 flexibility, options for -errors BOOL (error suppression), -interp 496 NAME (to specify interp), -match PATTERN (glob pattern to filter 497 tests by desc), -iters NUM (default number of iters to run). 498 Reorganized how data is returned to runbench master. 499 500 * tk/entry.bench (new): 501 * tk/canvas.bench (new): new tests for widget creation, config 502 503 * tcl/array.bench (removed): 504 * tcl/vars.bench: merged array.bench tests into VAR 505 506 * tcl/map.bench: fixed for compatability with Tcl7.4- 507 5082000-05-25 Jeff Hobbs <hobbs@scriptics.com> 509 510 * runbench.tcl: added -match, -notcl, -notk options, restructured 511 startup sequence. 512 513 * libbench.tcl: added ability to return string values from bench 514 tests and support for filtering tests to run. 515 516 * tcl/string.bench: moved string mapping benchmarks and added more 517 string equality benchmarks 518 * tcl/map.bench: added extended string mapping benchmark 519 520 * tcl/read.bench: 521 * tcl/startup.bench: 522 * tk/startup.bench: updated code to reflect proc-oriented tmpfile 523 operations. 524