view release on metacpan or  search on metacpan

README  view on Meta::CPAN

           has no output test), it seems to insist on adding a timestamp.

   2.44, 2013-10-02 A major change--all threads are gone.
       ENHANCEMENT
           Fsdb is now thread free and only uses  processes  for  parallelism.
           This  change  is a big change--the entire motivation for Fsdb-2 was
           to exploit parallelism via threading.  Parallelism--good, but  perl
           threading--bad  for  performance.   Horribly  bad  for performance.
           About 20x worse than pipes on my box.  (See perl  bug  #119445  for
           the discussion.)

       NEW "Fsdb::Support::Freds"  provides  a  thread-like  abstraction  over
           forking, with some nice support for callbacks in  the  parent  upon
           child termination.

       ENHANCEMENT
           Details  about  removing  threads: "dbpipeline" is thread free, and
           new tests to  verify  each  of  its  parts.   The  easy  cases  are
           "dbcolpercentile",   "dbcolstats",   "dbfilepivot",  "dbjoin",  and
           "dbcolstatscores",  each  of  which   use   it   in   simple   ways
           (2013-09-09).  "dbmerge" is now thread free (2013-09-13), but was a
           significant  rewrite,  which brought "dbsort" along.  "dbmapreduce"
           is partly thread free (2013-09-21), again  as  a  rewrite,  and  it
           brings  "dbmultistats" along.  Full "dbmapreduce" support took much
           longer (2013-10-02).

       BUG FIX
           When running with user-only output ("-n"), dbroweval now resets the
           output vector $ofref after it has been output.

       NEW dbcolcreate will create all columns at the head of  each  row  with
           the "--first" option.

       NEW dbfilecat  will concatenate two files, verifying that they have the
           same schema.

       ENHANCEMENT
           dbmapreduce now passes comments through, rather than eating them as
           before.

           Also,  dbmapreduce  now  supports  a   "--"   option   to   prevent
           misinterpreting sub-program parameters as for dbmapreduce.

       INCOMPATIBLE CHANGE
           dbmapreduce no longer figures out if it needs to add the key to the
           output.   For multi-key-aware reducers, it never does (and cannot).
           For non-multi-key-aware reducers, it defaults to add  the  key  and
           will now fail if the reducer adds the key (with error "dbcolcreate:
           attempt  to  create  pre-existing  column...").  In such cases, one
           must disable adding the key with the new option "--no-prepend-key".

       INCOMPATIBLE CHANGE
           dbmapreduce no longer copies the input field separator by  default.
           For multi-key-aware reducers, it never does (and cannot).  For non-
           multi-key-aware  reducers,  it  defaults  to  not copying the field
           separator,  but  it  will  copy  it  (the  old  default)  with  the
           "--copy-fs" option

   2.45, 2013-10-07 cleanup from de-thread-ification
       BUG FIX
           Corrected a fast busy-wait in dbmerge.

       ENHANCEMENT
           Endgame  mode  enabled  in  dbmerge;  it  (and  also large cases of
           dbsort) should now exploit greater parallelism.

       BUG FIX
           Test case with "Fsdb::BoundedQueue" (gone since 2.44) now removed.

   2.46, 2013-10-08 continuing cleanup of our no-threads version
       BUG FIX
           Fixed some packaging  details.   (Really,  threads  are  no  longer
           required, missing tests in the MANIFEST.)

       IMPROVEMENT
           dbsort  now  better  communicates  with  the merge process to avoid
           bursty parallelism.

           Fsdb::IO::Writer now can take "-autoflush =" 1>  for  line-buffered
           IO.

   2.47, 2013-10-12 test suite cleanup for non-threaded perls
       BUG FIX
           Removed  some  stray  "use  threads" in some test cases.  We didn't
           need them, and these were breaking non-threaded perls.

       BUG FIX
           Better  handling  of  Fred   cleanup;   should   fix   intermittent
           dbmapreduce failures on BSD.

       ENHANCEMENT
           Improved  test  framework  to  show  output when tests fail.  (This
           time, for real.)

   2.48, 2014-01-03 small bugfixes and improved release engineering
       ENHANCEMENT
           Test suites now skip tests for libraries that are missing.   (Patch
           for missing "IO::Compresss:Xz" contributed by Calvin Ardi.)

       ENHANCEMENT
           Removed  references to Jdb in the package specification.  Since the
           name was changed in  2008,  there's  no  longer  a  huge  need  for
           backwards compatibility.  (Suggestion form Petr  abata.)

       ENHANCEMENT
           Test   suites   now   invoke   the   perl   using   the  path  from
           $Config{perlpath}.  Hopefully this helps  testing  in  environments
           where  there  are  multiple installed perls and the default perl is
           not   the   same   as   the   perl-under-test   (as   happens    in
           cpantesters.org).

       BUG FIX
           Added  specific  encoding  to  this manpage to account for Unicode.
           Required to build correctly against perl-5.18.

   2.49, 2014-01-04  bugfix  to  unicode  handling  in  Fsdb  IO  (plus  minor
       packaging fixes)
       BUG FIX
           Restored a line in the .spec to chmod g-s.

       BUG FIX