summaryrefslogtreecommitdiff
path: root/Porting
Commit message (Collapse)AuthorAgeFilesLines
* Bump Perl version from 5.27.2 to 5.27.3Aaron Crane2017-07-202-30/+30
| | | | Including the various pieces of Module::CoreList.
* Release announcement template: next stable is 5.28Aaron Crane2017-07-201-2/+2
|
* perldelta template: add a note about module versionsAaron Crane2017-07-201-0/+6
|
* Tick off 5.27.2 releaseAaron Crane2017-07-201-1/+1
|
* Add epigraph for 5.27.2Aaron Crane2017-07-201-0/+26
|
* Import Encode-2.92 from CPANAaron Crane2017-07-181-2/+1
| | | | This also permits removing the local customisation for the previous version.
* Porting/perldelta_template.pod: tiny grammar tweakAaron Crane2017-07-181-8/+8
|
* Restore Porting/make-rmg-checklist --html optionAaron Crane2017-07-162-3/+22
|
* Suppress irrelevant "MUST SKIP this step" RMG paragraphsAaron Crane2017-07-161-2/+7
| | | | | | Sections that aren't relevant to the current release type are suppressed in their entirety, so the remaining "MUST SKIP" messages are just confusing. Remove them from the content.
* Replace Release Managers Guide (RMG) with new version:Sawyer X2017-07-162-101/+146
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Many of the mistakes made by me during a release has to do with the confusing instructions in the guide. * Some steps are mentioned in different order * Some steps are mentioned (and noted to *NOT* do) * The confusion between "MAINT" and "BLEAD-FINAL", and "BLEAD-FINAL" and "BLEAD-POINT". This generator generates a checklist with only the instruction you *will* have to perform. Any steps that mentions they must be skipped for the release will not be included in the end-result. Unlike the previous guide, you need not know the type of the release you do. Instead, you give the version you want to release and it generates the appropriate one for you. All the following incantations work: perl Porting/make-rmg-checklist --version 5.26.0-RC2 # RC perl Porting/make-rmg-checklist --version 5.26.0 # BLEAD-FINAL perl Porting/make-rmg-checklist --version 5.27.0 # BLEAD-POINT perl Porting/make-rmg-checklist --version 5.27.1 # BLEAD-POINT perl Porting/make-rmg-checklist --version 5.26.1 # MAINT Extra benefit: Apparently it includes additional checklist steps at the top that somehow are not included when you currently generate. Downside: HTML is not yet supported.
* Carthago delenda estChris 'BinGOs' Williams2017-07-151-1/+1
|
* Tick off 5.22.4 and 5.24.2Steve Hay2017-07-151-2/+3
| | | | That was probably the last 5.22. There may or may not be another 5.24.
* Epigraphs for 5.22.4 and 5.24.2Steve Hay2017-07-151-0/+46
|
* Add epigraphs for 5.22.4-RC1 and 5.24.2-RC1Steve Hay2017-07-011-0/+32
|
* Tick off 5.27.1 releaseSteve Hay2017-06-291-1/+1
|
* Update release scheduleSteve Hay2017-06-271-2/+3
| | | | | I'm hoping to get 5.22.4/5.24.2 released in the next few weeks, with 5.26.1 following around a month after that.
* Update Math-BigRat to CPAN version 0.2613Eric Herman2017-06-261-1/+1
| | | | | | | | | | Includes test addressing Bug #131410: "Math::BigRat in perl-5.26.0-RC2 severely broken" https://rt.perl.org/Public/Bug/Display.html?id=131410 See Also: http://code.activestate.com/lists/perl5-porters/236513/ https://rt.cpan.org/Ticket/Display.html?id=121139
* Update Time-HiRes to CPAN version 1.9742Chris 'BinGOs' Williams2017-06-241-1/+1
| | | | | | | | | [DELTA] 1.9742 [2017-04-16] - prefer 3-argument open: blead 1ae6ead9 - fix dist/Time-HiRes/t/*.t that assumed '.' in @INC: blead 465db51d - fix cases where 'do file' should be 'do ./file'.: blead 8b69401c
* upgrade Time-Piece from 1.31 to 1.3201David Mitchell2017-06-241-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 1.3201 2017-06-22 - Fix compile errors on MS - Use macro for buf sizes - Remove some compile warnings - SP of by 1 in _crt_localtime 1.32 2017-06-20 - Since 1.31: - New truncate() method - Add new flags: %F, %T, %P + more - Add rudimentary locale support via use_locale() - More tests - Less bugs (hopefully) 1.31_04 2017-06-16 - Tests for %F and %T - Fix wrong test count on Windows - Use sensible defaults for strptime 1.31_03 2017-06-14 - Don't use localization on default - Bugs with %s in strptime fixed 1.31_02 2017-01-16 - Add better localization between strftime and strptime 1.31_01 2017-01-13 - Respect singular for 1 sec, min, etc. in pretty (choroba) - Add truncate() (openstrike) - Add many more tests (openstrike) - Add more compatibility to strf/strptime
* upgrade Scalar-List-Utils from 1.47 to 1.48David Mitchell2017-06-241-1/+1
| | | | | | | | | | [CHANGES] * Note in documentation that outer function's @_ can be accessed in some blocks, but ought not be (thanks wchristian) [BUGFIXES] * Ensure pairmap extends its stack correctly (thanks davem) * Fix name of List::Util::unpairs in its error messages
* Bump the perl version 5.27.1 -> 5.27.2Eric Herman2017-06-222-30/+30
| | | | | | | | Note that at this stage, porting tests fail: Failed 3 tests out of 2390, 99.87% okay. ../dist/Module-CoreList/t/is_core.t ../dist/Module-CoreList/t/maintainer.t porting/corelist.t
* Upgrade Scalar-List-Utils from 1.46 to 1.47David Mitchell2017-06-221-1/+1
| | | | | | | | | | | | Note that the last few commits have reverted any local customisations to 1.46. These were of two kinds. First, fixes for a tainting issue, the last of which was just a cherry-pick from upstream, which should have made it into 1.47 anyway. The second were fixes to tests for netbsd-vax and Inf/Nan. I don't think blead urgently needs these - we should wait for them to re-appear via the normal upstream merging process. See https://rt.cpan.org/Public/Bug/Display.html?id=118470
* remove Scalar-List-Utils customisation listDavid Mitchell2017-06-221-8/+0
| | | | | | | | .. in preparation for installing new release Revert "Scalar-List-Utils: customize Maintainers.pl" This reverts commit 1476bbfa98fcd0efc3a4d1ccdd0b021ad7403986.
* undo a Scalar-List-Utils customisationDavid Mitchell2017-06-221-1/+0
| | | | | | | | .. in preparation for installing new release Revert "Cherry-pick Scalar-List-Utils tainted.t fix from upstream" This reverts commit a76614debec3c698069bcf0fa012318ed15461cc.
* Porting/bench.pl: test --fields with 1 fieldDavid Mitchell2017-06-211-1/+1
| | | | | When only one field is specified, e.g. --fields=Ir, it prints in a more compact form. Test this format.
* Porting/bench.pl comments: s/perl_name/perl_label/David Mitchell2017-06-211-9/+9
| | | | | | The second-level hash n %results etc is now keyed by the label of the perl executable rather than the name of the executable. So update code comments accordingly.
* Porting/bench.pl: fix up podDavid Mitchell2017-06-211-82/+23
| | | | | | | General tweaking and proof-reading. I also removed the long obscure example for --grindargs, on the grounds that it was long and obscure.
* Porting/bench.pl --bisect: output field valueDavid Mitchell2017-06-211-0/+2
| | | | | As well as giving an exit value indicating whether the field value is in range, display the value on stdout.
* Porting/bench.pl: tweak local varsDavid Mitchell2017-06-211-4/+3
| | | | | rename $perl_args to $cmd_line_args (less ambiguous), and the @putargs var is no longer used.
* Porting/bench.pl: add --autolabel optionDavid Mitchell2017-06-211-4/+95
| | | | | | | | | | | | | | | | | This automatically creates a unique label per perl interpreter. For example Porting/bench.pl --autolabel \ ~/bin/perl-5.20.0 ~/bin/perl-5.22.0 ~/bin/perl-5.24.0 is equivalent to Porting/bench.pl \ ~/bin/perl-5.20.0=5.20.0 ~/bin/perl-5.22.0=5.22.0 \ ~/bin/perl-5.24.0=5.24.0 As part of adding a test, this commit also slightly fixes up the test code which generates a regex from an NNN.NN part of a template.
* Porting/bench.pl: improve 'no such test' err msgDavid Mitchell2017-06-211-3/+11
| | | | | | | | | | | With --tests=A,B,.. if a test isn't found, it gives an error message, and under --vebrose lists all available tests. Make it: not under --verbose: suggest running under verbose to get a list; under --verbose: list available tests one per line rather than all on one line.
* Porting/bench.pl: trivial var changesDavid Mitchell2017-06-211-4/+4
| | | | | | | | $hash->{perls} has been assigned to $read_perls, so use that var rather than $hash. Ditto $read_results. Also, the results/test sub-hash is indexed by label rather than perl executable name, so rename the $perl var to $label.
* Porting/bench.pl: new syntax for perl args and envDavid Mitchell2017-06-211-34/+97
| | | | | | | | | | | | | | | | | | | | | | | | The existing syntax for supplying per-perl-executable command-line arguments and environment was clunky and buggy. Replace it with the following syntax and add tests: One of more perls to execute is listed on the command line after all switches have been processed. Each perl is of the form: some-perl [--args='-Mstrict -Mwarnings' ] \ [ --env='A=a' --env='B=b' .... ] i.e. the name of a perl executable, optionally succeeded by --args and --env switches. E.g. the above invokes 'some-perl -Mstrict -Mwarnings' with the A and B environment variables set. See http://nntp.perl.org/group/perl.perl5.porters/240004
* Porting/bench.pl ban duplicate labels in multireadDavid Mitchell2017-06-211-3/+18
| | | | | | | | | | | | | With multiple --read options, die if the same label appears twice. A duplicate between a --read file and on the command line was already checked for, but not from two reads. Also, improve the error messages. Finally, regenerate the callsub2.json test file using a './perl2' label this time, so that both callsub.json and callsub2.json can still be read without error.
* Porting/bench.pl: ban --benchfile under --readDavid Mitchell2017-06-211-4/+7
| | | | | | In the presence of one or more --reads, the tests to benchmark are decided by what tests were previously run and the benchfile is ignored. make it an error to specify a benchfile in this case.
* Porting/bench.pl: better handle multiple --readDavid Mitchell2017-06-211-11/+48
| | | | | | | | | | | | | | Fix up the code which checks and merges multiple results files: * better error messages; * remove redundant filter_tests() calls; * check each --read file for internal consistency in test names; * add tests * fixup pod
* Porting/bench.pl: only use used sort elemsDavid Mitchell2017-06-211-0/+3
| | | | | The @$order list of test names in sorted order: pare it down to only the tests actually being run. This will make --write files smaller.
* Porting/bench.pl: remove unneeded sort order calcsDavid Mitchell2017-06-211-6/+0
| | | | | | In a couple of places when --reading from previous runs, it calculates a default default order if one wasn't read from the file. Since the file should always contain an 'order', this is redundant.
* Porting/bench.pl: add $done_read to do_grind()David Mitchell2017-06-211-3/+4
| | | | | | | | | It was using $loop_counts to test whether a file had previously been read. Since $loop_counts is set merely as a side-effect of the read, it was slightly confusing to see if ($loop_counts) { ... } when the code in question had nothing particularly to do with loop counts. So add an explicit boolean var.
* Porting/bench.pl: add croak testsDavid Mitchell2017-06-211-3/+6
| | | | | Add tests to t/porting/bench.t for (nearly) every possible die() in Porting/bench.pl, plus a few minor fixups to those dies.
* bench.pl re-organise option processingDavid Mitchell2017-06-211-156/+247
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Conceptually re-organise command-line options into categories such as input, benchmarking, and output; and update the pod accordingly. In particular, The flow is now seen as: read in zero or more old results with --read; combine with zero or more new results generated via benchmarking any perls specified on the command line; output the results in one or more forms. The output part is what has conceptually changed. Now there are fundamentally three ways to output the results, and you can do any or all three if you want: --write - save results for later analysis; --bisect - set the process exit value based on the results; --show - display the results in human readable form. Most of the remaining options, such as --norm, --raw, --compact are then modifiers to --show. --show is on by default apart from in the presence of --write or --bisect. The main functional difference is with bisect; since it is now just another output type, it can be combined with --read, --write and even --show. The changes to the pod are mainly mainly moving individual =items around and adding new sections and sub-lists. The text of the individual items hasn't been changed (that I recall).
* Porting/bench.pl: fix broken --bisect optionDavid Mitchell2017-06-211-2/+2
| | | | | | | This option had bit-rotted: checks for the right number of perl executables weren't taking account of --bisect. Fix, and add tests.
* Porting/bench.pl: allow 1 perl under --rawDavid Mitchell2017-06-211-1/+1
| | | | | If not displaying relative percentages, then there's a valid use case for running with a single perl under test.
* Porting/bench.pl: don't print usage on errDavid Mitchell2017-06-211-2/+2
| | | | | | | | | | If there's a error with the command-line arguments, don't print the full usage information (which drowns out the actual error message); instead, just print this line: Use the -h option for usage information. Also, allow -h as a synonym for --help.
* linkify Perl 5.27.1 epigraphEric Herman2017-06-211-1/+1
|
* cmpVersion skip Module-CoreList modulesChris 'BinGOs' Williams2017-06-201-0/+3
| | | | | Versioning in Module-CoreList should not be dictated by an automated script.
* Ra ra Rasputin lover of the Russian QueenChris 'BinGOs' Williams2017-06-201-1/+1
|
* new perldeltaEric Herman2017-06-201-1/+1
|
* epigraph for 5.27.1Eric Herman2017-06-201-1/+15
| | | | | Also linkify and fixup 5.27.0 epigraph ... waiting on link for 5.27.1
* Update Filter-Util-Call to CPAN version 1.57Chris 'BinGOs' Williams2017-06-161-1/+1
| | | | | | | | | | | | | | | | | | | | | | [DELTA] 1.57 2017-01-22 rurban ---- * Todo the t/exec.t test 2 on cygwin. * Fixed/Todo the t/decrypt.t test 7 utf8 failures. Skip with non UTF-8 locale. 1.56 2017-01-20 rurban ---- * add binmode to the decrypt/encr,decrypt sample scripts * add utf8-encoded testcase to t/decrypt.t [cpan #110921]. use -C * stabilized some tests, add diag to sometimes failing sh tests * moved filter-util.pl to t/ * fixed INSTALLDIRS back to site since 5.12 [gh #2] * fixed exec/sh test races using the same temp. filenames * reversed this Changes file to latest first * added Travis CI
View Lorry Controller Status