1 2007-09-12 Ben Pfaff <blp@gnu.org>
3 * crosstabs.q (postcalc): Free sorted_tab and the structures that
4 it points to, to plug a memory leak. Fixes bug #20910. Thanks to
5 John Darrington for reporting this bug and for review.
7 2007-09-04 Ben Pfaff <blp@gnu.org>
9 * crosstabs.q (cmd_crosstabs): Free xtab and the structures that
10 it points to, to plug a memory leak. Fixes bug #18315.
12 2007-08-15 Jason Stover <jhs@wonko.gcsu.edu>
14 * regression.q (identify_indep_vars): Print an error if dependent
15 and independent variables are the same. Fixes bug 19819.
17 2007-08-12 Ben Pfaff <blp@gnu.org>
19 * flip.c: Drop use of dict_get_compacted_dict_index_to_case_index
20 and just use the ordinary case indexes. There seemed to be no
21 reason for the former method.
23 2007-08-03 Ben Pfaff <blp@gnu.org>
25 * rank.q (rank_cmd): Instead of sorting by SPLIT FILE vars, group
26 by them. Fixes bug #17239.
27 Reviewed by John Darrington.
29 2007-08-01 Ben Pfaff <blp@gnu.org>
31 Clean up handling of median, by treating it almost like any other
32 percentile. Fixes bug #17424. Thanks to John Darrington for
34 * frequencies.q (internal_cmd_frequencies): Fix handling of bit
35 masks for `stats' variable. If median is selected, turn it off
36 and add a 50th percentile.
37 (add_percentile): Simplify code a little.
38 (calc_stats): Drop special casing of median.
39 (dump_statistics): Ditto, except that we label the 50th percentile
40 as "50 (Median)" to make it clear that it's also the median.
42 2007-07-31 Ben Pfaff <blp@gnu.org>
44 Remove integer mode from FREQUENCIES and incidentally fix bug
45 #17421. Reviewed by John Darrington.
46 * frequencies.q (int_pool): Rename data_pool.
47 (gen_pool): Rename syntax_pool.
48 (enum FRQM_*): Removed.
49 (struct freq_tab): Removed `mode', `vector', `min', `max',
50 `out_of_range', `sysmis' members.
51 (calc): Delete support for integer mode.
53 (postprocess_freq_tab): Ditto.
54 (cleanup_freq_tab): Ditto.
55 (frq_custom_variables): Ditto.
57 2007-07-28 John Darrington <john@darrington.wattle.id.au>
59 * t-test.q: Moved the order in which groups are displayed in the
60 independent samples case, where a cut point is given.
62 2007-07-27 Ben Pfaff <blp@gnu.org>
64 * regression.q (run_regression): Move casereader_destroy call so
65 that it always gets called, not just if there was some valid
66 data. Fixes bug #19581.
67 Reviewed by Jason Stover.
69 2007-07-24 Ben Pfaff <blp@gnu.org>
71 * flip.c (struct flip_pgm): Remove `case_size' member (now
73 (cmd_flip): Pass var_cnt as number of cases instead of case_cnt,
74 to fix bug #20494. Don't assign to `case_size' member, which was
75 unused after assignment.
76 (build_dictionary): When NEWNAMES not used, get the number of
77 variables right, to fix bug #20493.
79 2007-07-10 Jason Stover <jhs@math.gcsu.edu>
81 * glm.q: Initial version of the GLM procedure.
83 2007-06-06 Ben Pfaff <blp@gnu.org>
85 Adapt case sources, sinks, and clients of procedure code to the
88 * aggregate.c: Simplify greatly since everything is more uniform
91 * autorecode.c: Adapt to new procedure code.
95 * descriptives.c: Ditto.
97 * npar-summary.c: Ditto.
98 * frequencies.q: Ditto.
101 * regression.q: Ditto.
102 * sort-cases.c: Ditto.
105 * sort-criteria.c: Rewrite to output a struct case_ordering.
107 * flip.c: Rewrite to be a casereader.
109 * rank.q: Simplify greatly since casereaders are much more
110 flexible than what we had before.
112 2007-05-15 Jason Stover <jhs@math.gcsu.edu>
114 * regression.q (run_regression): Tell the user when the data
115 contain no valid cases.
117 2007-05-08 Jason Stover <jhs@math.gcsu.edu>
119 * regression.q: Partial fix of bug which caused a crash if
120 dependent variable and independent variable were the same.
122 2007-04-16 John Darrington <john@darrington.wattle.id.au>
124 * t-test.q: Changed the output width of reported counts and
125 degrees of freedom, to avoid truncating these values. Thanks
126 to Seth Woolley for reporting this problem. A proper fix involves
127 re-thinking the output driver.
129 2007-04-12 Jason Stover <jhs@math.gcsu.edu>
131 * regression.q (run_regression): Added if (n_data >0) to fix bug
134 2007-03-29 Jason Stover <jhs@math.gcsu.edu>
136 * regression.q (prepare_data): New function.
138 * regression.q (compute_moments): New function.
140 2007-03-18 Ben Pfaff <blp@gnu.org>
142 * crosstabs.q (static var write): Rename write_style to avoid
143 conflict with POSIX function of same name.
145 2007-03-16 Jason Stover <jhs@math.gcsu.edu>
147 * regression.q (run_regression): Added support for moments.
149 Sat Feb 17 08:16:00 2007 Ben Pfaff <blp@gnu.org>
151 * flip.c (flip_sink_create): Improve error message when temporary
152 file cannot be created.
154 Tue Feb 6 19:58:03 2007 Ben Pfaff <blp@gnu.org>
156 * flip.c (flip_file): Give better error message on end-of-file.
158 2007-02-04 Jason Stover <jhs@math.gcsu.edu>
160 * regression.q: Fixed p-value computation in the test for
161 individual regression coefficients.
163 Mon Jan 15 11:03:20 2007 Ben Pfaff <blp@gnu.org>
165 Fix bugs found by valgrind when --enable-debug is used with the
166 new case code. These bugs are hidden when the data set is small
167 enough to find in memory; if a bigger data set that would overflow
168 to disk were used, then data corruption would occur.
170 * chisquare.c (create_freq_hash): Pass free_freq_mutable_hash to
171 hsh_create as free function. Make copy of data put into hash.
173 * oneway.q (free_value): New function.
174 (run_oneway): Use free_value as arg to hsh_create. Make copy of
177 * rank.q (rank_cases): Don't access data in case after we've given
180 Tue Jan 9 19:16:11 2007 Ben Pfaff <blp@gnu.org>
184 * aggregate.c (parse_aggregate_functions) Initialize
187 Fri Dec 22 14:04:09 2006 Ben Pfaff <blp@gnu.org>
189 Simplify missing value handling.
191 * aggregate.c (struct agr_var): Remove `bool include_missing', add
192 `enum mv_class exclude'. Remove `int missing', add `bool
193 saw_missing'. Update users.
195 * descriptives.c (struct dsc_trns): Removed `int
196 include_user_missing', add `enum mv_class exclude'. Update users.
197 (struct dsc_proc): Ditto.
199 * examine.q: (static var value_is_missing): Rename
200 `exclude_values', change type to `enum mv_class'. Update users.
204 Fri Dec 22 19:22:18 WST 2006 John Darrington <john@darrington.wattle.id.au>
206 * frequencies.q : Fixed bug #17420, where the table bounds were overun
207 when /FORMAT=nolabels was given.
209 Wed Dec 20 18:45:31 WST 2006 John Darrington <john@darrington.wattle.id.au>
211 * binomial.c binomial.h : New files. Thanks to Jason Stover
212 for assistance with these.
214 * chisquare.c chisquare.h freq.c freq.h npar-summary.c npar-summary.h
215 npar.h npar.q: New files. Implementing NPAR TESTS.
217 * frequencies.q : Moved structure definitions into freq.[ch]
219 Sat Dec 16 22:26:44 2006 Ben Pfaff <blp@gnu.org>
221 Make it possible to pull cases from the active file with a
222 function call, instead of requiring indirection through a callback
225 * aggregate.c (cmd_aggregate): Take advantage of new procedure
227 (agr_to_active_file): Removed.
228 (presorted_agr_to_sysfile): Removed.
230 * autorecode.c (cmd_autorecode): Take advantage of new procedure
232 (autorecode_proc_func): Removed.
234 * flip.c (struct flip_pgm): New members to allow conformance with
235 new case_source_class interface.
236 (cmd_flip): Adapt to new case_source_class interface.
237 (flip_source_read): Ditto.
238 (flip_source_destroy): Ditto.
240 Sat Dec 16 12:54:27 2006 Ben Pfaff <blp@gnu.org>
242 * rank.q (rank_custom_variables): Allow grouping variables to be
243 strings. Fixes bug #18533. Thanks to John Darrington for review.
245 Sat Dec 9 18:47:51 2006 Ben Pfaff <blp@gnu.org>
247 * regression.q (is_depvar): Compare variable pointers instead of
250 Thu Dec 7 15:26:25 WST 2006 John Darrington <john@darrington.wattle.id.au>
252 * examine.q: Allocated the categorical values for the dependent and
253 independent variables, on the heap. Hence they can be of any width.
255 Wed Dec 6 21:14:26 2006 Ben Pfaff <blp@gnu.org>
257 * regression.q (reg_inserted): Compare variable pointers instead
260 Mon Dec 4 22:33:46 2006 Ben Pfaff <blp@gnu.org>
262 * crosstabs.q (insert_summary): Use var_to_string for labeling.
263 (output_pivot_table) Ditto.
266 * frequencies.q (setup_z_trns): Ditto.
268 (dump_condensed) Ditto.
269 (dump_statistics) Ditto.
272 Sun Nov 5 08:31:42 WST 2006 John Darrington <john@darrington.wattle.id.au>
274 * t-test.q, oneway.q: Changed to use the new casefilter structure.
276 Sat Oct 14 16:52:28 2006 Ben Pfaff <blp@gnu.org>
278 * rank.q: (rank_sorted_casefile) Add some missing case_destroy()
279 calls to fix a memory leak.
281 Sun Oct 8 09:45:40 WST 2006 John Darrington <john@darrington.wattle.id.au>
283 * rank.q: Plugged a small memory leak which occurred under error
286 Sat Oct 7 11:06:01 WST 2006 John Darrington <john@darrington.wattle.id.au>
288 * rank.q: Implemented most of the RANK command.
290 2006-07-14 Jason Stover <jhs@math.gcsu.edu>
292 * regression.q (run_regression): New function to move knowledge of
293 pspp_linreg_cache out of math/coefficient.[ch].
295 Sat Jul 1 17:41:46 2006 Ben Pfaff <blp@gnu.org>
297 Fix bug #11612, "q2c documentation does not agree with code".
299 * examine.q: Audit use of q2c "+" prefixes that indicate that a
300 command may appear multiple times.
302 * frequencies.q: Ditto.
306 * regression.q: Ditto.
310 Fri Jun 23 14:18:22 2006 Ben Pfaff <blp@gnu.org>
312 Support long string variables on FREQUENCIES, as
313 an extension when in enhanced algorithms mode. For Greg Hunt
314 <greg@firmansyah.com>.
316 * frequencies.q: (struct freq) Change `v' member from union value
317 to union value *. Update all references.
318 (struct var_freqs) Add width, print members to represent effective
319 variable width and display format.
320 (calc) Copy entire long string value into the hash table.
321 (frq_custom_variables) Set new width, print members.
322 (hash_value_alpha) Get width from var_freqs.
323 (compare_value_alpha_a) Ditto.
324 (compare_freq_alpha_a) Ditto.
325 (compare_freq_alpha_d) Ditto.
326 (dump_full) Get display format from var_freqs.
327 (dump_condensed) Ditto.
329 Mon Jun 19 22:07:13 2006 Ben Pfaff <blp@gnu.org>
331 * frequencies.q: (dump_full) Only put the first MAX_SHORT_STRING
332 bytes of string variables into the output cells, seeing as we only
334 (dump_condensed) Ditto.
336 Mon Jun 19 21:52:05 2006 Ben Pfaff <blp@gnu.org>
338 Fixes a bug reported by Greg Hunt <greg@firmansyah.com>.
340 * frequencies.q: (hsh_hash_bytes) We only copy the first
341 MAX_SHORT_STRING bytes of string variables, so we must only
342 compare that many bytes, even if the string variable is longer.
343 (compare_value_alpha_a) Ditto.
344 (compare_freq_alpha_a) Ditto.
345 (compare_freq_alpha_d) Ditto.
347 2006-05-11 Jason Stover <jhs@math.gcsu.edu>
349 * regression.q: Adjusted code to account for cache->coeff being a
350 pspp_linreg_coeff **.
352 Sun May 7 18:31:25 2006 Ben Pfaff <blp@gnu.org>
356 * aggregate.c (cmd_aggregate): Free default_dict before replacing
359 Sun May 7 17:09:19 2006 Ben Pfaff <blp@gnu.org>
361 * flip.c (flip_file): Check return value of pool_fclose().
363 Sat May 6 16:00:13 2006 Ben Pfaff <blp@gnu.org>
365 Get rid of `char *c' member in union value, for cleanliness.
367 * aggregate.c: (union agr_argument) New union.
368 (struct agr_var) Change element type of arg[] from union value to
370 (parse_aggregate_functions) Change local variable types likewise.
372 * autorecode.c: (union arc_value) New union.
373 (struct arc_item) Change "from" from union value to union
375 (recode) Change local variable from union value to union
377 (autorecode_trns_proc) Ditto.
378 (compare_alpha_value) Ditto.
379 (hash_alpha_value) Ditto.
380 (compare_numeric_value) Ditto.
381 (hash_numeric_value) Ditto.
382 (autorecode_proc_func) Ditto.
384 Sat May 6 10:43:33 2006 Ben Pfaff <blp@gnu.org>
386 Continue reforming procedure execution. In this phase, get rid of
387 the output code for SPLIT FILE groups in procedure.c, which really
388 shouldn't be doing any output. Move it into the individual
389 procedures instead. This also adds some flexibility.
391 * crosstabs.q (precalc): Call output_split_file_values().
393 * descriptives.c (calc_descriptives): Ditto.
395 * examine.q (run_examine): Ditto.
397 * frequencies.q (precalc): Ditto.
399 * oneway.q (run_oneway): Ditto.
401 * regression.q (run_regression): Ditto.
403 * t-test.q (calculate): Ditto.
405 Wed May 3 23:05:31 2006 Ben Pfaff <blp@gnu.org>
407 Continue reforming procedure execution. In this phase, get rid of
408 many global variables, consolidating procedure execution in
409 procedure.c. Encapsulate transformations in new "struct
410 trns_chain". Also, change implementation of N OF CASES, FILTER,
411 and PROCESS IF from special cases to transformations.
413 * aggregate.c (cmd_aggregate) Use discard_variables().
415 2006-04-28 Jason Stover <jhs@math.gcsu.edu>
417 * regression.q (regression_trns_resid_proc): Pass only the
418 variables used in the model to (*model->residual)().
420 * regression.q (regression_trns_pred_proc): Pass only the
421 variables used in the model to (*model->pred)().
423 2006-04-26 Jason Stover <jhs@math.gcsu.edu>
425 * regression.q: Added support for multiple transformations.
427 * regression.q (regression_trns_resid_proc): New function.
429 * regression.q (regression_trns_pred_proc): New function.
431 * regression.q (subcommand_save): Added support for saving
434 * regression.q (regression_trns_free): New function.
436 * regression.q (reg_get_name): New function.
438 * regression.q (reg_save_var): New function.
440 Tue Apr 25 13:18:56 2006 Ben Pfaff <blp@gnu.org>
442 * rank.q (parse_rank_function): Use SE instead of ME for parse
445 Tue Apr 25 13:16:28 2006 Ben Pfaff <blp@gnu.org>
447 * flip.c (flip_sink_write): Use snprintf() to simplify a bit of
450 2006-04-21 Jason Stover <jhs@math.gcsu.edu>
452 * regression.q (try_name): New function. (Partly copied from
453 try_name in descriptives.c.)
455 * regression.q (subcommand_save): Choose residual variable names
458 2006-04-20 Jason Stover <jhs@math.gcsu.edu>
460 * regression.q (cmd_regression): Moved call to subcommand_save()
461 outside multipass_procedure_with_splits().
463 * regression.q (regression_trns_proc): Fixed value counter n_vals
464 before calling *model->residual().
466 2006-04-19 Jason Stover <jhs@math.gcsu.edu>
468 * regression.q (regression_trns_proc): Fixed the look-up of the
471 2006-04-18 Jason Stover <jhs@math.gcsu.edu>
473 * regression.q (regression_trns_proc): Look up the residual
474 variable in the linear regression cache.
476 * regression.q (subcommand_save): Set the residual variable in the
477 linear regression cache.
479 2006-04-17 Jason Stover <jason@wonko.gcsu.edu>
481 * regression.q (regression_trns_proc): Accept case_idx as an int
482 to match the definition of trns_proc_func.
484 2006-04-17 Jason Stover <jhs@math.gcsu.edu>
486 * regression.q (regression_trns_proc): New function.
488 * regression.q (subcommand_save): Create variable residuals and
489 add a transformation to assign values to them. Also free the
490 linreg_cache if the SAVE command was not called. Removed the
491 casereading loop. Placed actual computation of residuals in
492 regression_trns_proc.
494 * regression.q (run_regression): Moved call to free
495 pspp_linreg_cache to subcommand_save.
497 Sat Apr 15 18:01:03 2006 Ben Pfaff <blp@gnu.org>
499 * examine.q (output_examine): Add casts to fix warnings.
501 2006-04-07 Jason Stover <jhs@math.gcsu.edu>
503 * regression.q (subcommand_save): New function.
505 2006-04-04 Jason Stover <jhs@math.gcsu.edu>
507 * regression.q: New function reg_has_categorical () to tell
508 whether a pspp_linreg_struct was made with any variables of type
511 * regression.q: (subcommand_export): Call
512 reg_print_categorical_encoding() only if the model uses any
513 categorical variables.
515 Mon Mar 27 16:00:42 2006 Ben Pfaff <blp@gnu.org>
517 * crosstabs.q: (output_pivot_table) Drop spurious space from
520 2006-03-15 Jason Stover <jhs@math.gcsu.edu>
522 * regression.q: Added custom syntax parser for VARIABLES subcommand
524 * regression.q: Moved most instructions for run_regression ()
525 inside the loop over dependent variables.
527 Thu Mar 2 08:40:33 WST 2006 John Darrington <john@darrington.wattle.id.au>
529 * Moved files from src directory