John Darrington [Tue, 18 Jan 2005 15:15:12 +0000 (15:15 +0000)]
More memory leaks fixed.
John Darrington [Tue, 18 Jan 2005 11:31:45 +0000 (11:31 +0000)]
Fixed numerous memory leaks. Changed many of the test cases to use a canonical
format for the temporary test file names.
John Darrington [Fri, 14 Jan 2005 04:05:47 +0000 (04:05 +0000)]
Integrated the charts into the HTML stream.
John Darrington [Thu, 13 Jan 2005 13:02:26 +0000 (13:02 +0000)]
Moved the SIGINT handler from casefile.c to main.c
John Darrington [Mon, 10 Jan 2005 06:50:45 +0000 (06:50 +0000)]
Added a signal handler to delete temp casefiles on SIGINT and SIGQUIT.
Inhibited the PERMISSIONS command when SAFER is on.
command.def: Added a lot more unimplemented commands.
Moved --version information to copyleft.c
John Darrington [Sun, 9 Jan 2005 08:54:04 +0000 (08:54 +0000)]
Changed the generation of the test data, so that it's a lot faster.
Ben Pfaff [Sun, 9 Jan 2005 07:59:48 +0000 (07:59 +0000)]
Sat Jan 8 23:58:34 2005 Ben Pfaff <blp@gnu.org>
* sort.c: (compare_initial_runs) Needed additional level of
dereferencing.
(merge_once) Fix plenty of stupid mistakes.
Sat Jan 8 23:55:27 2005 Ben Pfaff <blp@gnu.org>
* casefile.c: (casefile_sleep) Need to flush_buffer() after
calling casefile_to_disk() or we will lose the last block in the
file if the casefile started out as disk-based.
(casefile_get_reader) Initialize reader->destructive to 0.
(cmd_debug_casefile) Add new test pattern.
(test_casefile) Define new test pattern to make sure
casefile_sleep() works properly.
Ben Pfaff [Sun, 9 Jan 2005 00:47:44 +0000 (00:47 +0000)]
* credits.texi: Removed.
* bugs.texi: Rewrote. Moved known bugs to savannah website.
* not-implemented.texi: AGGREGATE and AUTORECODE are implemented.
John Darrington [Fri, 7 Jan 2005 00:17:19 +0000 (00:17 +0000)]
Added an optional --without-libplot flag to configure
John Darrington [Thu, 6 Jan 2005 12:55:44 +0000 (12:55 +0000)]
Added a signal handler for SIGFPE
(Re)Fixed an assertion in sort.c
Added a test for sorting with ENORMOUS input files.
Added a test to ensure that fatal errors are reported properly.
Tidied up a few other tests.
Ben Pfaff [Thu, 6 Jan 2005 05:29:46 +0000 (05:29 +0000)]
Fix assertion for proper Huffman merge pattern: 0 == 1 modulo 1.
John Darrington [Wed, 5 Jan 2005 14:45:54 +0000 (14:45 +0000)]
Fixed bug #11307
John Darrington [Wed, 5 Jan 2005 00:32:16 +0000 (00:32 +0000)]
Added a test for reading/writing of portable files.
Fixed a bug which crashed PSPP if VALUE LABELS had a trailing slash.
Ben Pfaff [Tue, 4 Jan 2005 01:46:48 +0000 (01:46 +0000)]
* pfm-read.c: (read_variables) Remove direct manipulation of v->aux,
which is no longer needed. Fixes Bug #11483.
John Darrington [Sat, 1 Jan 2005 11:03:45 +0000 (11:03 +0000)]
Various bug fixes.
John Darrington [Fri, 31 Dec 2004 08:51:06 +0000 (08:51 +0000)]
Added Boxplots to the EXAMINE subcommand repertoire
Fixed a few bugs in percentiles calculations
John Darrington [Wed, 29 Dec 2004 01:19:57 +0000 (01:19 +0000)]
Implemented calculation of percentiles and Tukey hinges
John Darrington [Fri, 24 Dec 2004 07:22:14 +0000 (07:22 +0000)]
Fixed bug #11227 (T-Test not working with alpha independent variable )
John Darrington [Sat, 11 Dec 2004 03:45:40 +0000 (03:45 +0000)]
Fixed the trimmed mean calculation under various special cases.
John Darrington [Sat, 4 Dec 2004 09:20:59 +0000 (09:20 +0000)]
Added code to calculate the histogram limits
John Darrington [Thu, 2 Dec 2004 05:52:51 +0000 (05:52 +0000)]
Added histograms into examine.
Revised the histogram API.
Revised the piechart API.
Ben Pfaff [Wed, 1 Dec 2004 07:03:42 +0000 (07:03 +0000)]
Update TODO.
Ben Pfaff [Wed, 1 Dec 2004 05:40:23 +0000 (05:40 +0000)]
Use AC_TYPE_OFF_T/HAVE_OFF_T to check for off_t.
Ben Pfaff [Tue, 30 Nov 2004 16:54:03 +0000 (16:54 +0000)]
If seeko() is not available use long int for off_t.
John Darrington [Tue, 30 Nov 2004 08:59:50 +0000 (08:59 +0000)]
Fixed a bug which crept into the npplot function
John Darrington [Mon, 29 Nov 2004 04:26:59 +0000 (04:26 +0000)]
Fixed conflict with stdio.h
Ben Pfaff [Mon, 29 Nov 2004 03:30:09 +0000 (03:30 +0000)]
Change -a to -pR in cp invocations for SUSv3 compliance.
John Darrington [Thu, 25 Nov 2004 11:02:53 +0000 (11:02 +0000)]
Added the skewness and kurtosis calculations to the EXAMINE command.
Added an option to configure to build without the termcap library.
John Darrington [Tue, 23 Nov 2004 05:54:57 +0000 (05:54 +0000)]
Fixed some memory leaks.
John Darrington [Sun, 21 Nov 2004 02:43:46 +0000 (02:43 +0000)]
Documented the EXAMINE subcommands. Fixed up the examine test. Fixed a problem with the totals.
John Darrington [Sat, 20 Nov 2004 12:26:20 +0000 (12:26 +0000)]
Rewrote most of the examine command.
John Darrington [Fri, 19 Nov 2004 06:06:21 +0000 (06:06 +0000)]
Plugged some memory leaks.
Ben Pfaff [Wed, 17 Nov 2004 07:49:12 +0000 (07:49 +0000)]
Update PRNG documentation.
Ben Pfaff [Tue, 16 Nov 2004 08:08:00 +0000 (08:08 +0000)]
Adopt GSL random number generators, paving the way for providing the
complete suite of random number generators on expressions.
Ben Pfaff [Tue, 16 Nov 2004 06:18:00 +0000 (06:18 +0000)]
Fix XDATE.JDAY formula.
John Darrington [Tue, 16 Nov 2004 05:25:48 +0000 (05:25 +0000)]
Added the permissions command.
Ben Pfaff [Mon, 15 Nov 2004 09:51:22 +0000 (09:51 +0000)]
Instead of making system or portable file readers responsible for
dropping and reordering variables, make them read full cases and
let the caller take care of any changes.
Instead of providing a system or portable file writer with a raw
case in the format needed for output, provide it with a regular
case. The writer takes care of any needed translation.
Instead of treating `struct file_handle' as a class to subclass
into data files, system files, and portable files, instead use it
as a helper that coordinates access. Now it is opaque, too.
Break dictionary functions into separate header file.
Get rid of procedure-specific union in struct variable, using
instead a void * pointer and a destructor function.
John Darrington [Fri, 12 Nov 2004 03:35:20 +0000 (03:35 +0000)]
Fixed some problems with value labels not reporting correctly.
John Darrington [Thu, 11 Nov 2004 13:23:44 +0000 (13:23 +0000)]
Added npplot and detrended np plots
John Darrington [Thu, 11 Nov 2004 00:33:46 +0000 (00:33 +0000)]
Fixed bug in the levene calculation. It hadn't been initialised properly,
and the test had been gratiutously pasted from buggy output.
Ben Pfaff [Tue, 9 Nov 2004 06:22:19 +0000 (06:22 +0000)]
Fix comment.
Ben Pfaff [Tue, 9 Nov 2004 05:47:54 +0000 (05:47 +0000)]
Fix a few formatting mistakes.
John Darrington [Tue, 9 Nov 2004 01:50:09 +0000 (01:50 +0000)]
Made makefile aware of texinfo dependencies
John Darrington [Sun, 7 Nov 2004 09:31:05 +0000 (09:31 +0000)]
Added (some of the) calculations for the examine command.
John Darrington [Sat, 6 Nov 2004 13:30:14 +0000 (13:30 +0000)]
Changed the definition of factors to be self referential.
John Darrington [Sat, 6 Nov 2004 12:45:30 +0000 (12:45 +0000)]
Made EXAMINE cope properly with splits.
John Darrington [Sat, 6 Nov 2004 06:57:35 +0000 (06:57 +0000)]
Fixed problem where oneway didn't deal properly with splits.
John Darrington [Fri, 5 Nov 2004 09:50:34 +0000 (09:50 +0000)]
Added a note about SPLIT requiring adjacent cases.
John Darrington [Thu, 4 Nov 2004 03:17:32 +0000 (03:17 +0000)]
Fixed the parsing of subcommands, such that the (n) setting is optional
Added prototypes for the va*_to_string functions
John Darrington [Wed, 3 Nov 2004 03:26:34 +0000 (03:26 +0000)]
Added framework for the EXAMINE command
John Darrington [Mon, 1 Nov 2004 04:51:21 +0000 (04:51 +0000)]
Added test for ntiles. Fixed q2c parsing of integer/double subcommands
John Darrington [Sun, 31 Oct 2004 08:17:08 +0000 (08:17 +0000)]
Updated the documentation and fixed a couple of the test scripts
John Darrington [Sat, 30 Oct 2004 10:14:05 +0000 (10:14 +0000)]
Split pspp.texi into one texi file per chapter
John Darrington [Sat, 30 Oct 2004 04:46:38 +0000 (04:46 +0000)]
Fixed behaviour of oneway when presented with missing values
John Darrington [Sat, 30 Oct 2004 01:21:21 +0000 (01:21 +0000)]
Fixed a bug in the levene test, and added the levene test to the oneway cmd
John Darrington [Fri, 29 Oct 2004 12:07:54 +0000 (12:07 +0000)]
Fixed the t-test model to be consistent with the anova model.
John Darrington [Fri, 29 Oct 2004 06:02:44 +0000 (06:02 +0000)]
Added framework for charts
John Darrington [Tue, 26 Oct 2004 10:45:42 +0000 (10:45 +0000)]
Amalgamated t-test.h and oneway.h into a single group_proc.h file
John Darrington [Mon, 25 Oct 2004 12:54:42 +0000 (12:54 +0000)]
Added remaining calculations for the separate variance contrast tests
John Darrington [Mon, 25 Oct 2004 04:09:51 +0000 (04:09 +0000)]
Fixed namespace clash (stderr)
John Darrington [Mon, 25 Oct 2004 03:47:35 +0000 (03:47 +0000)]
Added this file which I forgot.
John Darrington [Sun, 24 Oct 2004 06:25:14 +0000 (06:25 +0000)]
Added most of the calculations for the ONEWAY command.
John Darrington [Fri, 22 Oct 2004 08:13:13 +0000 (08:13 +0000)]
Updated AUTHORS to reflect JMD's work
John Darrington [Fri, 22 Oct 2004 08:00:39 +0000 (08:00 +0000)]
Added infrastructure for the ONEWAY command.
John Darrington [Fri, 22 Oct 2004 07:59:34 +0000 (07:59 +0000)]
Added framework for the ONEWAY command.
Ben Pfaff [Mon, 7 Jun 2004 06:07:14 +0000 (06:07 +0000)]
(cmd_descriptives) Remove harmless but bogus test in STATISTICS
parsing.
Ben Pfaff [Tue, 1 Jun 2004 04:57:11 +0000 (04:57 +0000)]
Fix memory leaks.
Generalize casefiles to the extent that we can use them for sorting
and other kinds of data transformations. Change cases to be
copy-on-write to improve memory efficiency in common cases. Every
access to a member of a `struct ccase' was changed to be a call to a
case_*() function, especially case_data(), case_num(), case_str(), or
case_data_rw(). Many instances of a local variable named "case_num"
were changed to "case_idx" as a consequence. Many `struct ccase *'
were changed to actual `struct ccase' because of copying semantics of
cases. In several places there was a choice between updating debug
code to work with the new ADTs or just deleting it because it was
useless; I chose to delete it.
The workspace idea didn't work out.
Ben Pfaff [Mon, 31 May 2004 05:50:10 +0000 (05:50 +0000)]
Fully implement arbitrary delimiters on DATA LIST, extending the half
implementation that was already there.
Update our string ADTs, struct string and struct len_string. Get rid
of pool support, which was largely unused. Rename lots of functions
to have more obvious or consistent names.
Fix a few miscellaneous bugs.
pjk [Thu, 13 May 2004 18:57:57 +0000 (18:57 +0000)]
Do not kill trailing spaces - breaks data list list where the last variable(s) on a line are empty
pjk [Thu, 13 May 2004 18:31:13 +0000 (18:31 +0000)]
Adjusted code for DATA LIST (free/list) to be more consistent with SPSS in
the instance where delimiters are explicitly set. In particular, if a delim.
is explicitly set, we treat quotes / apostrophes as normal chars:
From SPSS Syntax Base manual:
For data with explicitly specified value delimiters (for example, DATA LIST FREE (","):
· Multiple delimiters without any intervening space can be used to specify missing data.
· The specified delimiters cannot occur within a data value, even if you enclose the value
in quotation marks or apostrophes.
pjk [Thu, 13 May 2004 17:51:22 +0000 (17:51 +0000)]
Add support for (TAB) to DATA LIST FREE/LIST
pjk [Tue, 4 May 2004 15:27:24 +0000 (15:27 +0000)]
Fix segfault in recode (recode with an int src, string dest, where the src
is SYSMIS
Ben Pfaff [Mon, 3 May 2004 06:10:44 +0000 (06:10 +0000)]
Fix typo.
pjk [Wed, 28 Apr 2004 20:59:26 +0000 (20:59 +0000)]
configure.ac: check for presence / add libgnugetopt to support
getopt_long on FreeBSD
algorithm.c, alloc.h: Wrap includes of alloca.h with #ifdef
HAVE_ALLOCA_H. Some platforms (FreeBSD for example) have alloca in
stdlib.h... Autoconf detects this in AC_FUNC_ALLOCA
pjk [Wed, 28 Apr 2004 16:03:45 +0000 (16:03 +0000)]
- Add test for compute heap munching bug (compute a=upcase(a)) where a is a
large string
pjk [Tue, 27 Apr 2004 22:41:19 +0000 (22:41 +0000)]
- Fixed bug in pool (free_all_gizmos) where gizmos pointer wasn't being
NULLed after we freed all of the gizmos
- Fixed bug in vfm (cancel_transformations) where we didn't NULL t_trns after
freeing it
- Updated routines around variable labels to allow 255 char. variable labels
(this is supported by SPSS 11.0 at least...) This doesn't appear to break
backward compat. Let me know if it does.
Ben Pfaff [Tue, 27 Apr 2004 05:54:31 +0000 (05:54 +0000)]
We're abusing the current ASCII driver by telling it to allocate a
9999-line, 9999-character page in the tests. This causes some systems
to curl up and die because it allocates 20 MB of contiguous RAM. This
change alleviates at least part of the problem. It is mostly a
stop-gap until the new output system is ready.
Also, get rid of unused matrix.[ch].
pjk [Fri, 23 Apr 2004 19:25:53 +0000 (19:25 +0000)]
Fix broken check-in
pjk [Fri, 23 Apr 2004 18:44:42 +0000 (18:44 +0000)]
Add test for bug fixed recently in recode
pjk [Fri, 23 Apr 2004 18:41:02 +0000 (18:41 +0000)]
Fix bug where a copy dest spec would cause a failed assertion in strdup
Fix bug where a copy dest spec would cause the wrong member of the union
to be copied
Ben Pfaff [Thu, 22 Apr 2004 00:30:53 +0000 (00:30 +0000)]
Fix @node line.
Ben Pfaff [Tue, 20 Apr 2004 05:56:02 +0000 (05:56 +0000)]
Done with updating portable file description in docs.
Ben Pfaff [Tue, 20 Apr 2004 05:52:11 +0000 (05:52 +0000)]
Minor updates to data file and portable file descriptions based on
emails from Aapi Hämäläinen <aapi@soho.joensuu.fi>.
Ben Pfaff [Tue, 20 Apr 2004 03:52:33 +0000 (03:52 +0000)]
Update.
Ben Pfaff [Sat, 17 Apr 2004 07:02:10 +0000 (07:02 +0000)]
Contrary to what I'd always understood, there is an efficient
algorithm for deletion from a hash table populated via linear probing.
This change implements it.
Ben Pfaff [Wed, 14 Apr 2004 02:26:37 +0000 (02:26 +0000)]
Adjust calculation of kurtosis to avoid subtracting huge numbers from
huge numbers, on Michael Kiefte's advice.
Ben Pfaff [Sun, 11 Apr 2004 21:32:00 +0000 (21:32 +0000)]
Rework moments routines for improved numerical stability based on
Michael Kiefte's advice. Any bugs or remaining numerical problems are
still mine though.
There is now a struct moments1 for use with one-pass moments. It uses
a provisional means algorithm as an attempt to improve accuracy of
higher moments. The older struct moments now only handles two-pass
moments.
Ben Pfaff [Sat, 10 Apr 2004 03:06:45 +0000 (03:06 +0000)]
Fix assert failure when only MOMENT_MEAN is needed on DESCRIPTIVES.
Michael Kiefte [Fri, 9 Apr 2004 10:01:56 +0000 (10:01 +0000)]
*** empty log message ***
Michael Kiefte [Fri, 9 Apr 2004 09:58:47 +0000 (09:58 +0000)]
* descript.c:
fixed problem with parsing in match_statistic() causing "DESCRIPTIVE
STAT=MEAN." to barf.
"MEAN" is now default if "SORT" given without specification.
fixed infinite loop with "DESCRIPT GIBBERISH=ALL."
parsing is generally less forgiving of syntax errors: better to have
it do nothing and type it in again then to not know what it actually did instead
(problem also related to infinite loop problem above).
z-score transformation now check score for user-missing values and ch
ecks std_dev for SYSMIS.
Ben Pfaff [Wed, 7 Apr 2004 04:57:18 +0000 (04:57 +0000)]
Update what I'm working on.
Michael Kiefte [Tue, 6 Apr 2004 18:50:26 +0000 (18:50 +0000)]
Updated ChangeLog.
Michael Kiefte [Tue, 6 Apr 2004 18:14:36 +0000 (18:14 +0000)]
* Changed dict_get_case_weight() to accept an additional int * flag to complain about system-missing, user-missing, zero, or negative weights.
* Updated existing functions to pass int * to dict_get_case_weight().
John Darrington [Mon, 5 Apr 2004 09:58:14 +0000 (09:58 +0000)]
Fixed configuration problems with gsl
John Darrington [Mon, 5 Apr 2004 00:09:15 +0000 (00:09 +0000)]
Fixed some problems encountered when compiling under Cygwin
Ben Pfaff [Sat, 3 Apr 2004 19:45:09 +0000 (19:45 +0000)]
Fix infinite loop on comment at end of file, add test.
John Darrington [Sat, 3 Apr 2004 07:10:55 +0000 (07:10 +0000)]
Fixed the calculation of percentiles and added --syntax and --algorithm options
John Darrington [Fri, 2 Apr 2004 03:21:08 +0000 (03:21 +0000)]
Converted the t-test to use multipass_with_split.... mechanism
Ben Pfaff [Thu, 1 Apr 2004 06:38:35 +0000 (06:38 +0000)]
Use moments data structure on FREQUENCIES.
Ben Pfaff [Wed, 31 Mar 2004 06:11:10 +0000 (06:11 +0000)]
Add test for multipass procedure bug.