1 2008-03-10 Jason Stover <jhs@debs.hoobahooba.net>
3 * regression.q (run_regression): Removed code for EXPORT
4 subcommand. Remove use of coefficient 0 as the intercept.
6 2008-02-14 John Darrington <john@darrington.wattle.id.au>
8 * examine.q: Fixed counts of missing variables. Thanks to
9 Jason Stover for reporting this problem.
11 2008-01-02 John Darrington <john@darrington.wattle.id.au>
13 * binomial.c chisquare.c examine.q frequencies.q oneway.q regression.q : updated
14 all users of var_get_value_name to use replacement function var_append_value_name.
16 2007-12-07 Ben Pfaff <blp@gnu.org>
20 * crosstabs.q (precalc): Initialize data structures even if the
21 first case cannot be read.
23 * frequencies.q (precalc): Ditto.
25 2007-11-03 Ben Pfaff <blp@gnu.org>
27 Allow output files to overwrite input files (bug #21280).
29 * aggregate.c (cmd_aggregate): Manage file handle reference
32 * correlations.q (internal_cmd_frequencies): Ditto.
33 (cor_custom_matrix): Ditto.
35 * regression.q (regression_custom_export): Ditto.
36 (cmd_regression): Ditto.
38 2007-10-12 Ben Pfaff <blp@gnu.org>
40 * flip.c (flip_file): No need to conditionally substitute for
41 "fseeko" and "off_t" manually anymore, as gnulib takes care of it
44 2007-09-21 Jason Stover <jhs@wonko.gcsu.edu>
46 * regression.q (run_regression): Partial fix of memory leak, bug
49 2007-09-19 Ben Pfaff <blp@gnu.org>
53 * aggregate.c (cmd_aggregate): Destroy casereader consistently,
54 even if casereader fails.
56 * examine.q (run_examine): Ditto.
58 * glm.q (run_glm): Ditto.
60 * oneway.q (run_oneway): Ditto.
62 * regression.q (run_regression): Ditto.
64 * t-test.q (calculate): Ditto.
66 * descriptives.c (calc_descriptives): Ditto. Also avoid
67 gratuitous casereader_clone.
69 2007-09-13 Jason Stover <jhs@debs.hoobahooba.net>
71 * regression.q (cmd_regression): Move declaration of models in to
72 definition of cmd_regression.
74 * regression.q (run_regression): Free mom to fix memory leak.
76 2007-09-12 Ben Pfaff <blp@gnu.org>
78 * crosstabs.q (postcalc): Free sorted_tab and the structures that
79 it points to, to plug a memory leak. Fixes bug #20910. Thanks to
80 John Darrington for reporting this bug and for review.
82 2007-09-04 Ben Pfaff <blp@gnu.org>
84 * crosstabs.q (cmd_crosstabs): Free xtab and the structures that
85 it points to, to plug a memory leak. Fixes bug #18315.
87 2007-08-15 Jason Stover <jhs@wonko.gcsu.edu>
89 * regression.q (identify_indep_vars): Print an error if dependent
90 and independent variables are the same. Fixes bug 19819.
92 2007-08-12 Ben Pfaff <blp@gnu.org>
94 * flip.c: Drop use of dict_get_compacted_dict_index_to_case_index
95 and just use the ordinary case indexes. There seemed to be no
96 reason for the former method.
98 2007-08-03 Ben Pfaff <blp@gnu.org>
100 * rank.q (rank_cmd): Instead of sorting by SPLIT FILE vars, group
101 by them. Fixes bug #17239.
102 Reviewed by John Darrington.
104 2007-08-01 Ben Pfaff <blp@gnu.org>
106 Clean up handling of median, by treating it almost like any other
107 percentile. Fixes bug #17424. Thanks to John Darrington for
109 * frequencies.q (internal_cmd_frequencies): Fix handling of bit
110 masks for `stats' variable. If median is selected, turn it off
111 and add a 50th percentile.
112 (add_percentile): Simplify code a little.
113 (calc_stats): Drop special casing of median.
114 (dump_statistics): Ditto, except that we label the 50th percentile
115 as "50 (Median)" to make it clear that it's also the median.
117 2007-07-31 Ben Pfaff <blp@gnu.org>
119 Remove integer mode from FREQUENCIES and incidentally fix bug
120 #17421. Reviewed by John Darrington.
121 * frequencies.q (int_pool): Rename data_pool.
122 (gen_pool): Rename syntax_pool.
123 (enum FRQM_*): Removed.
124 (struct freq_tab): Removed `mode', `vector', `min', `max',
125 `out_of_range', `sysmis' members.
126 (calc): Delete support for integer mode.
128 (postprocess_freq_tab): Ditto.
129 (cleanup_freq_tab): Ditto.
130 (frq_custom_variables): Ditto.
132 2007-07-28 John Darrington <john@darrington.wattle.id.au>
134 * t-test.q: Moved the order in which groups are displayed in the
135 independent samples case, where a cut point is given.
137 2007-07-27 Ben Pfaff <blp@gnu.org>
139 * regression.q (run_regression): Move casereader_destroy call so
140 that it always gets called, not just if there was some valid
141 data. Fixes bug #19581.
142 Reviewed by Jason Stover.
144 2007-07-24 Ben Pfaff <blp@gnu.org>
146 * flip.c (struct flip_pgm): Remove `case_size' member (now
148 (cmd_flip): Pass var_cnt as number of cases instead of case_cnt,
149 to fix bug #20494. Don't assign to `case_size' member, which was
150 unused after assignment.
151 (build_dictionary): When NEWNAMES not used, get the number of
152 variables right, to fix bug #20493.
154 2007-07-10 Jason Stover <jhs@math.gcsu.edu>
156 * glm.q: Initial version of the GLM procedure.
158 2007-06-06 Ben Pfaff <blp@gnu.org>
160 Adapt case sources, sinks, and clients of procedure code to the
163 * aggregate.c: Simplify greatly since everything is more uniform
166 * autorecode.c: Adapt to new procedure code.
168 * chisquare.c: Ditto.
169 * crosstabs.q: Ditto.
170 * descriptives.c: Ditto.
172 * npar-summary.c: Ditto.
173 * frequencies.q: Ditto.
176 * regression.q: Ditto.
177 * sort-cases.c: Ditto.
180 * sort-criteria.c: Rewrite to output a struct case_ordering.
182 * flip.c: Rewrite to be a casereader.
184 * rank.q: Simplify greatly since casereaders are much more
185 flexible than what we had before.
187 2007-05-15 Jason Stover <jhs@math.gcsu.edu>
189 * regression.q (run_regression): Tell the user when the data
190 contain no valid cases.
192 2007-05-08 Jason Stover <jhs@math.gcsu.edu>
194 * regression.q: Partial fix of bug which caused a crash if
195 dependent variable and independent variable were the same.
197 2007-04-16 John Darrington <john@darrington.wattle.id.au>
199 * t-test.q: Changed the output width of reported counts and
200 degrees of freedom, to avoid truncating these values. Thanks
201 to Seth Woolley for reporting this problem. A proper fix involves
202 re-thinking the output driver.
204 2007-04-12 Jason Stover <jhs@math.gcsu.edu>
206 * regression.q (run_regression): Added if (n_data >0) to fix bug
209 2007-03-29 Jason Stover <jhs@math.gcsu.edu>
211 * regression.q (prepare_data): New function.
213 * regression.q (compute_moments): New function.
215 2007-03-18 Ben Pfaff <blp@gnu.org>
217 * crosstabs.q (static var write): Rename write_style to avoid
218 conflict with POSIX function of same name.
220 2007-03-16 Jason Stover <jhs@math.gcsu.edu>
222 * regression.q (run_regression): Added support for moments.
224 Sat Feb 17 08:16:00 2007 Ben Pfaff <blp@gnu.org>
226 * flip.c (flip_sink_create): Improve error message when temporary
227 file cannot be created.
229 Tue Feb 6 19:58:03 2007 Ben Pfaff <blp@gnu.org>
231 * flip.c (flip_file): Give better error message on end-of-file.
233 2007-02-04 Jason Stover <jhs@math.gcsu.edu>
235 * regression.q: Fixed p-value computation in the test for
236 individual regression coefficients.
238 Mon Jan 15 11:03:20 2007 Ben Pfaff <blp@gnu.org>
240 Fix bugs found by valgrind when --enable-debug is used with the
241 new case code. These bugs are hidden when the data set is small
242 enough to find in memory; if a bigger data set that would overflow
243 to disk were used, then data corruption would occur.
245 * chisquare.c (create_freq_hash): Pass free_freq_mutable_hash to
246 hsh_create as free function. Make copy of data put into hash.
248 * oneway.q (free_value): New function.
249 (run_oneway): Use free_value as arg to hsh_create. Make copy of
252 * rank.q (rank_cases): Don't access data in case after we've given
255 Tue Jan 9 19:16:11 2007 Ben Pfaff <blp@gnu.org>
259 * aggregate.c (parse_aggregate_functions) Initialize
262 Fri Dec 22 14:04:09 2006 Ben Pfaff <blp@gnu.org>
264 Simplify missing value handling.
266 * aggregate.c (struct agr_var): Remove `bool include_missing', add
267 `enum mv_class exclude'. Remove `int missing', add `bool
268 saw_missing'. Update users.
270 * descriptives.c (struct dsc_trns): Removed `int
271 include_user_missing', add `enum mv_class exclude'. Update users.
272 (struct dsc_proc): Ditto.
274 * examine.q: (static var value_is_missing): Rename
275 `exclude_values', change type to `enum mv_class'. Update users.
279 Fri Dec 22 19:22:18 WST 2006 John Darrington <john@darrington.wattle.id.au>
281 * frequencies.q : Fixed bug #17420, where the table bounds were overun
282 when /FORMAT=nolabels was given.
284 Wed Dec 20 18:45:31 WST 2006 John Darrington <john@darrington.wattle.id.au>
286 * binomial.c binomial.h : New files. Thanks to Jason Stover
287 for assistance with these.
289 * chisquare.c chisquare.h freq.c freq.h npar-summary.c npar-summary.h
290 npar.h npar.q: New files. Implementing NPAR TESTS.
292 * frequencies.q : Moved structure definitions into freq.[ch]
294 Sat Dec 16 22:26:44 2006 Ben Pfaff <blp@gnu.org>
296 Make it possible to pull cases from the active file with a
297 function call, instead of requiring indirection through a callback
300 * aggregate.c (cmd_aggregate): Take advantage of new procedure
302 (agr_to_active_file): Removed.
303 (presorted_agr_to_sysfile): Removed.
305 * autorecode.c (cmd_autorecode): Take advantage of new procedure
307 (autorecode_proc_func): Removed.
309 * flip.c (struct flip_pgm): New members to allow conformance with
310 new case_source_class interface.
311 (cmd_flip): Adapt to new case_source_class interface.
312 (flip_source_read): Ditto.
313 (flip_source_destroy): Ditto.
315 Sat Dec 16 12:54:27 2006 Ben Pfaff <blp@gnu.org>
317 * rank.q (rank_custom_variables): Allow grouping variables to be
318 strings. Fixes bug #18533. Thanks to John Darrington for review.
320 Sat Dec 9 18:47:51 2006 Ben Pfaff <blp@gnu.org>
322 * regression.q (is_depvar): Compare variable pointers instead of
325 Thu Dec 7 15:26:25 WST 2006 John Darrington <john@darrington.wattle.id.au>
327 * examine.q: Allocated the categorical values for the dependent and
328 independent variables, on the heap. Hence they can be of any width.
330 Wed Dec 6 21:14:26 2006 Ben Pfaff <blp@gnu.org>
332 * regression.q (reg_inserted): Compare variable pointers instead
335 Mon Dec 4 22:33:46 2006 Ben Pfaff <blp@gnu.org>
337 * crosstabs.q (insert_summary): Use var_to_string for labeling.
338 (output_pivot_table) Ditto.
341 * frequencies.q (setup_z_trns): Ditto.
343 (dump_condensed) Ditto.
344 (dump_statistics) Ditto.
347 Sun Nov 5 08:31:42 WST 2006 John Darrington <john@darrington.wattle.id.au>
349 * t-test.q, oneway.q: Changed to use the new casefilter structure.
351 Sat Oct 14 16:52:28 2006 Ben Pfaff <blp@gnu.org>
353 * rank.q: (rank_sorted_casefile) Add some missing case_destroy()
354 calls to fix a memory leak.
356 Sun Oct 8 09:45:40 WST 2006 John Darrington <john@darrington.wattle.id.au>
358 * rank.q: Plugged a small memory leak which occurred under error
361 Sat Oct 7 11:06:01 WST 2006 John Darrington <john@darrington.wattle.id.au>
363 * rank.q: Implemented most of the RANK command.
365 2006-07-14 Jason Stover <jhs@math.gcsu.edu>
367 * regression.q (run_regression): New function to move knowledge of
368 pspp_linreg_cache out of math/coefficient.[ch].
370 Sat Jul 1 17:41:46 2006 Ben Pfaff <blp@gnu.org>
372 Fix bug #11612, "q2c documentation does not agree with code".
374 * examine.q: Audit use of q2c "+" prefixes that indicate that a
375 command may appear multiple times.
377 * frequencies.q: Ditto.
381 * regression.q: Ditto.
385 Fri Jun 23 14:18:22 2006 Ben Pfaff <blp@gnu.org>
387 Support long string variables on FREQUENCIES, as
388 an extension when in enhanced algorithms mode. For Greg Hunt
389 <greg@firmansyah.com>.
391 * frequencies.q: (struct freq) Change `v' member from union value
392 to union value *. Update all references.
393 (struct var_freqs) Add width, print members to represent effective
394 variable width and display format.
395 (calc) Copy entire long string value into the hash table.
396 (frq_custom_variables) Set new width, print members.
397 (hash_value_alpha) Get width from var_freqs.
398 (compare_value_alpha_a) Ditto.
399 (compare_freq_alpha_a) Ditto.
400 (compare_freq_alpha_d) Ditto.
401 (dump_full) Get display format from var_freqs.
402 (dump_condensed) Ditto.
404 Mon Jun 19 22:07:13 2006 Ben Pfaff <blp@gnu.org>
406 * frequencies.q: (dump_full) Only put the first MAX_SHORT_STRING
407 bytes of string variables into the output cells, seeing as we only
409 (dump_condensed) Ditto.
411 Mon Jun 19 21:52:05 2006 Ben Pfaff <blp@gnu.org>
413 Fixes a bug reported by Greg Hunt <greg@firmansyah.com>.
415 * frequencies.q: (hsh_hash_bytes) We only copy the first
416 MAX_SHORT_STRING bytes of string variables, so we must only
417 compare that many bytes, even if the string variable is longer.
418 (compare_value_alpha_a) Ditto.
419 (compare_freq_alpha_a) Ditto.
420 (compare_freq_alpha_d) Ditto.
422 2006-05-11 Jason Stover <jhs@math.gcsu.edu>
424 * regression.q: Adjusted code to account for cache->coeff being a
425 pspp_linreg_coeff **.
427 Sun May 7 18:31:25 2006 Ben Pfaff <blp@gnu.org>
431 * aggregate.c (cmd_aggregate): Free default_dict before replacing
434 Sun May 7 17:09:19 2006 Ben Pfaff <blp@gnu.org>
436 * flip.c (flip_file): Check return value of pool_fclose().
438 Sat May 6 16:00:13 2006 Ben Pfaff <blp@gnu.org>
440 Get rid of `char *c' member in union value, for cleanliness.
442 * aggregate.c: (union agr_argument) New union.
443 (struct agr_var) Change element type of arg[] from union value to
445 (parse_aggregate_functions) Change local variable types likewise.
447 * autorecode.c: (union arc_value) New union.
448 (struct arc_item) Change "from" from union value to union
450 (recode) Change local variable from union value to union
452 (autorecode_trns_proc) Ditto.
453 (compare_alpha_value) Ditto.
454 (hash_alpha_value) Ditto.
455 (compare_numeric_value) Ditto.
456 (hash_numeric_value) Ditto.
457 (autorecode_proc_func) Ditto.
459 Sat May 6 10:43:33 2006 Ben Pfaff <blp@gnu.org>
461 Continue reforming procedure execution. In this phase, get rid of
462 the output code for SPLIT FILE groups in procedure.c, which really
463 shouldn't be doing any output. Move it into the individual
464 procedures instead. This also adds some flexibility.
466 * crosstabs.q (precalc): Call output_split_file_values().
468 * descriptives.c (calc_descriptives): Ditto.
470 * examine.q (run_examine): Ditto.
472 * frequencies.q (precalc): Ditto.
474 * oneway.q (run_oneway): Ditto.
476 * regression.q (run_regression): Ditto.
478 * t-test.q (calculate): Ditto.
480 Wed May 3 23:05:31 2006 Ben Pfaff <blp@gnu.org>
482 Continue reforming procedure execution. In this phase, get rid of
483 many global variables, consolidating procedure execution in
484 procedure.c. Encapsulate transformations in new "struct
485 trns_chain". Also, change implementation of N OF CASES, FILTER,
486 and PROCESS IF from special cases to transformations.
488 * aggregate.c (cmd_aggregate) Use discard_variables().
490 2006-04-28 Jason Stover <jhs@math.gcsu.edu>
492 * regression.q (regression_trns_resid_proc): Pass only the
493 variables used in the model to (*model->residual)().
495 * regression.q (regression_trns_pred_proc): Pass only the
496 variables used in the model to (*model->pred)().
498 2006-04-26 Jason Stover <jhs@math.gcsu.edu>
500 * regression.q: Added support for multiple transformations.
502 * regression.q (regression_trns_resid_proc): New function.
504 * regression.q (regression_trns_pred_proc): New function.
506 * regression.q (subcommand_save): Added support for saving
509 * regression.q (regression_trns_free): New function.
511 * regression.q (reg_get_name): New function.
513 * regression.q (reg_save_var): New function.
515 Tue Apr 25 13:18:56 2006 Ben Pfaff <blp@gnu.org>
517 * rank.q (parse_rank_function): Use SE instead of ME for parse
520 Tue Apr 25 13:16:28 2006 Ben Pfaff <blp@gnu.org>
522 * flip.c (flip_sink_write): Use snprintf() to simplify a bit of
525 2006-04-21 Jason Stover <jhs@math.gcsu.edu>
527 * regression.q (try_name): New function. (Partly copied from
528 try_name in descriptives.c.)
530 * regression.q (subcommand_save): Choose residual variable names
533 2006-04-20 Jason Stover <jhs@math.gcsu.edu>
535 * regression.q (cmd_regression): Moved call to subcommand_save()
536 outside multipass_procedure_with_splits().
538 * regression.q (regression_trns_proc): Fixed value counter n_vals
539 before calling *model->residual().
541 2006-04-19 Jason Stover <jhs@math.gcsu.edu>
543 * regression.q (regression_trns_proc): Fixed the look-up of the
546 2006-04-18 Jason Stover <jhs@math.gcsu.edu>
548 * regression.q (regression_trns_proc): Look up the residual
549 variable in the linear regression cache.
551 * regression.q (subcommand_save): Set the residual variable in the
552 linear regression cache.
554 2006-04-17 Jason Stover <jason@wonko.gcsu.edu>
556 * regression.q (regression_trns_proc): Accept case_idx as an int
557 to match the definition of trns_proc_func.
559 2006-04-17 Jason Stover <jhs@math.gcsu.edu>
561 * regression.q (regression_trns_proc): New function.
563 * regression.q (subcommand_save): Create variable residuals and
564 add a transformation to assign values to them. Also free the
565 linreg_cache if the SAVE command was not called. Removed the
566 casereading loop. Placed actual computation of residuals in
567 regression_trns_proc.
569 * regression.q (run_regression): Moved call to free
570 pspp_linreg_cache to subcommand_save.
572 Sat Apr 15 18:01:03 2006 Ben Pfaff <blp@gnu.org>
574 * examine.q (output_examine): Add casts to fix warnings.
576 2006-04-07 Jason Stover <jhs@math.gcsu.edu>
578 * regression.q (subcommand_save): New function.
580 2006-04-04 Jason Stover <jhs@math.gcsu.edu>
582 * regression.q: New function reg_has_categorical () to tell
583 whether a pspp_linreg_struct was made with any variables of type
586 * regression.q: (subcommand_export): Call
587 reg_print_categorical_encoding() only if the model uses any
588 categorical variables.
590 Mon Mar 27 16:00:42 2006 Ben Pfaff <blp@gnu.org>
592 * crosstabs.q: (output_pivot_table) Drop spurious space from
595 2006-03-15 Jason Stover <jhs@math.gcsu.edu>
597 * regression.q: Added custom syntax parser for VARIABLES subcommand
599 * regression.q: Moved most instructions for run_regression ()
600 inside the loop over dependent variables.
602 Thu Mar 2 08:40:33 WST 2006 John Darrington <john@darrington.wattle.id.au>
604 * Moved files from src directory