1 2007-07-31 Ben Pfaff <blp@gnu.org>
3 Remove integer mode from FREQUENCIES and incidentally fix bug
4 #17421. Reviewed by John Darrington.
5 * frequencies.q (int_pool): Rename data_pool.
6 (gen_pool): Rename syntax_pool.
7 (enum FRQM_*): Removed.
8 (struct freq_tab): Removed `mode', `vector', `min', `max',
9 `out_of_range', `sysmis' members.
10 (calc): Delete support for integer mode.
12 (postprocess_freq_tab): Ditto.
13 (cleanup_freq_tab): Ditto.
14 (frq_custom_variables): Ditto.
16 2007-07-28 John Darrington <john@darrington.wattle.id.au>
18 * t-test.q: Moved the order in which groups are displayed in the
19 independent samples case, where a cut point is given.
21 2007-07-27 Ben Pfaff <blp@gnu.org>
23 * regression.q (run_regression): Move casereader_destroy call so
24 that it always gets called, not just if there was some valid
25 data. Fixes bug #19581.
26 Reviewed by Jason Stover.
28 2007-07-24 Ben Pfaff <blp@gnu.org>
30 * flip.c (struct flip_pgm): Remove `case_size' member (now
32 (cmd_flip): Pass var_cnt as number of cases instead of case_cnt,
33 to fix bug #20494. Don't assign to `case_size' member, which was
34 unused after assignment.
35 (build_dictionary): When NEWNAMES not used, get the number of
36 variables right, to fix bug #20493.
38 2007-07-10 Jason Stover <jhs@math.gcsu.edu>
40 * glm.q: Initial version of the GLM procedure.
42 2007-06-06 Ben Pfaff <blp@gnu.org>
44 Adapt case sources, sinks, and clients of procedure code to the
47 * aggregate.c: Simplify greatly since everything is more uniform
50 * autorecode.c: Adapt to new procedure code.
54 * descriptives.c: Ditto.
56 * npar-summary.c: Ditto.
57 * frequencies.q: Ditto.
60 * regression.q: Ditto.
61 * sort-cases.c: Ditto.
64 * sort-criteria.c: Rewrite to output a struct case_ordering.
66 * flip.c: Rewrite to be a casereader.
68 * rank.q: Simplify greatly since casereaders are much more
69 flexible than what we had before.
71 2007-05-15 Jason Stover <jhs@math.gcsu.edu>
73 * regression.q (run_regression): Tell the user when the data
74 contain no valid cases.
76 2007-05-08 Jason Stover <jhs@math.gcsu.edu>
78 * regression.q: Partial fix of bug which caused a crash if
79 dependent variable and independent variable were the same.
81 2007-04-16 John Darrington <john@darrington.wattle.id.au>
83 * t-test.q: Changed the output width of reported counts and
84 degrees of freedom, to avoid truncating these values. Thanks
85 to Seth Woolley for reporting this problem. A proper fix involves
86 re-thinking the output driver.
88 2007-04-12 Jason Stover <jhs@math.gcsu.edu>
90 * regression.q (run_regression): Added if (n_data >0) to fix bug
93 2007-03-29 Jason Stover <jhs@math.gcsu.edu>
95 * regression.q (prepare_data): New function.
97 * regression.q (compute_moments): New function.
99 2007-03-18 Ben Pfaff <blp@gnu.org>
101 * crosstabs.q (static var write): Rename write_style to avoid
102 conflict with POSIX function of same name.
104 2007-03-16 Jason Stover <jhs@math.gcsu.edu>
106 * regression.q (run_regression): Added support for moments.
108 Sat Feb 17 08:16:00 2007 Ben Pfaff <blp@gnu.org>
110 * flip.c (flip_sink_create): Improve error message when temporary
111 file cannot be created.
113 Tue Feb 6 19:58:03 2007 Ben Pfaff <blp@gnu.org>
115 * flip.c (flip_file): Give better error message on end-of-file.
117 2007-02-04 Jason Stover <jhs@math.gcsu.edu>
119 * regression.q: Fixed p-value computation in the test for
120 individual regression coefficients.
122 Mon Jan 15 11:03:20 2007 Ben Pfaff <blp@gnu.org>
124 Fix bugs found by valgrind when --enable-debug is used with the
125 new case code. These bugs are hidden when the data set is small
126 enough to find in memory; if a bigger data set that would overflow
127 to disk were used, then data corruption would occur.
129 * chisquare.c (create_freq_hash): Pass free_freq_mutable_hash to
130 hsh_create as free function. Make copy of data put into hash.
132 * oneway.q (free_value): New function.
133 (run_oneway): Use free_value as arg to hsh_create. Make copy of
136 * rank.q (rank_cases): Don't access data in case after we've given
139 Tue Jan 9 19:16:11 2007 Ben Pfaff <blp@gnu.org>
143 * aggregate.c (parse_aggregate_functions) Initialize
146 Fri Dec 22 14:04:09 2006 Ben Pfaff <blp@gnu.org>
148 Simplify missing value handling.
150 * aggregate.c (struct agr_var): Remove `bool include_missing', add
151 `enum mv_class exclude'. Remove `int missing', add `bool
152 saw_missing'. Update users.
154 * descriptives.c (struct dsc_trns): Removed `int
155 include_user_missing', add `enum mv_class exclude'. Update users.
156 (struct dsc_proc): Ditto.
158 * examine.q: (static var value_is_missing): Rename
159 `exclude_values', change type to `enum mv_class'. Update users.
163 Fri Dec 22 19:22:18 WST 2006 John Darrington <john@darrington.wattle.id.au>
165 * frequencies.q : Fixed bug #17420, where the table bounds were overun
166 when /FORMAT=nolabels was given.
168 Wed Dec 20 18:45:31 WST 2006 John Darrington <john@darrington.wattle.id.au>
170 * binomial.c binomial.h : New files. Thanks to Jason Stover
171 for assistance with these.
173 * chisquare.c chisquare.h freq.c freq.h npar-summary.c npar-summary.h
174 npar.h npar.q: New files. Implementing NPAR TESTS.
176 * frequencies.q : Moved structure definitions into freq.[ch]
178 Sat Dec 16 22:26:44 2006 Ben Pfaff <blp@gnu.org>
180 Make it possible to pull cases from the active file with a
181 function call, instead of requiring indirection through a callback
184 * aggregate.c (cmd_aggregate): Take advantage of new procedure
186 (agr_to_active_file): Removed.
187 (presorted_agr_to_sysfile): Removed.
189 * autorecode.c (cmd_autorecode): Take advantage of new procedure
191 (autorecode_proc_func): Removed.
193 * flip.c (struct flip_pgm): New members to allow conformance with
194 new case_source_class interface.
195 (cmd_flip): Adapt to new case_source_class interface.
196 (flip_source_read): Ditto.
197 (flip_source_destroy): Ditto.
199 Sat Dec 16 12:54:27 2006 Ben Pfaff <blp@gnu.org>
201 * rank.q (rank_custom_variables): Allow grouping variables to be
202 strings. Fixes bug #18533. Thanks to John Darrington for review.
204 Sat Dec 9 18:47:51 2006 Ben Pfaff <blp@gnu.org>
206 * regression.q (is_depvar): Compare variable pointers instead of
209 Thu Dec 7 15:26:25 WST 2006 John Darrington <john@darrington.wattle.id.au>
211 * examine.q: Allocated the categorical values for the dependent and
212 independent variables, on the heap. Hence they can be of any width.
214 Wed Dec 6 21:14:26 2006 Ben Pfaff <blp@gnu.org>
216 * regression.q (reg_inserted): Compare variable pointers instead
219 Mon Dec 4 22:33:46 2006 Ben Pfaff <blp@gnu.org>
221 * crosstabs.q (insert_summary): Use var_to_string for labeling.
222 (output_pivot_table) Ditto.
225 * frequencies.q (setup_z_trns): Ditto.
227 (dump_condensed) Ditto.
228 (dump_statistics) Ditto.
231 Sun Nov 5 08:31:42 WST 2006 John Darrington <john@darrington.wattle.id.au>
233 * t-test.q, oneway.q: Changed to use the new casefilter structure.
235 Sat Oct 14 16:52:28 2006 Ben Pfaff <blp@gnu.org>
237 * rank.q: (rank_sorted_casefile) Add some missing case_destroy()
238 calls to fix a memory leak.
240 Sun Oct 8 09:45:40 WST 2006 John Darrington <john@darrington.wattle.id.au>
242 * rank.q: Plugged a small memory leak which occurred under error
245 Sat Oct 7 11:06:01 WST 2006 John Darrington <john@darrington.wattle.id.au>
247 * rank.q: Implemented most of the RANK command.
249 2006-07-14 Jason Stover <jhs@math.gcsu.edu>
251 * regression.q (run_regression): New function to move knowledge of
252 pspp_linreg_cache out of math/coefficient.[ch].
254 Sat Jul 1 17:41:46 2006 Ben Pfaff <blp@gnu.org>
256 Fix bug #11612, "q2c documentation does not agree with code".
258 * examine.q: Audit use of q2c "+" prefixes that indicate that a
259 command may appear multiple times.
261 * frequencies.q: Ditto.
265 * regression.q: Ditto.
269 Fri Jun 23 14:18:22 2006 Ben Pfaff <blp@gnu.org>
271 Support long string variables on FREQUENCIES, as
272 an extension when in enhanced algorithms mode. For Greg Hunt
273 <greg@firmansyah.com>.
275 * frequencies.q: (struct freq) Change `v' member from union value
276 to union value *. Update all references.
277 (struct var_freqs) Add width, print members to represent effective
278 variable width and display format.
279 (calc) Copy entire long string value into the hash table.
280 (frq_custom_variables) Set new width, print members.
281 (hash_value_alpha) Get width from var_freqs.
282 (compare_value_alpha_a) Ditto.
283 (compare_freq_alpha_a) Ditto.
284 (compare_freq_alpha_d) Ditto.
285 (dump_full) Get display format from var_freqs.
286 (dump_condensed) Ditto.
288 Mon Jun 19 22:07:13 2006 Ben Pfaff <blp@gnu.org>
290 * frequencies.q: (dump_full) Only put the first MAX_SHORT_STRING
291 bytes of string variables into the output cells, seeing as we only
293 (dump_condensed) Ditto.
295 Mon Jun 19 21:52:05 2006 Ben Pfaff <blp@gnu.org>
297 Fixes a bug reported by Greg Hunt <greg@firmansyah.com>.
299 * frequencies.q: (hsh_hash_bytes) We only copy the first
300 MAX_SHORT_STRING bytes of string variables, so we must only
301 compare that many bytes, even if the string variable is longer.
302 (compare_value_alpha_a) Ditto.
303 (compare_freq_alpha_a) Ditto.
304 (compare_freq_alpha_d) Ditto.
306 2006-05-11 Jason Stover <jhs@math.gcsu.edu>
308 * regression.q: Adjusted code to account for cache->coeff being a
309 pspp_linreg_coeff **.
311 Sun May 7 18:31:25 2006 Ben Pfaff <blp@gnu.org>
315 * aggregate.c (cmd_aggregate): Free default_dict before replacing
318 Sun May 7 17:09:19 2006 Ben Pfaff <blp@gnu.org>
320 * flip.c (flip_file): Check return value of pool_fclose().
322 Sat May 6 16:00:13 2006 Ben Pfaff <blp@gnu.org>
324 Get rid of `char *c' member in union value, for cleanliness.
326 * aggregate.c: (union agr_argument) New union.
327 (struct agr_var) Change element type of arg[] from union value to
329 (parse_aggregate_functions) Change local variable types likewise.
331 * autorecode.c: (union arc_value) New union.
332 (struct arc_item) Change "from" from union value to union
334 (recode) Change local variable from union value to union
336 (autorecode_trns_proc) Ditto.
337 (compare_alpha_value) Ditto.
338 (hash_alpha_value) Ditto.
339 (compare_numeric_value) Ditto.
340 (hash_numeric_value) Ditto.
341 (autorecode_proc_func) Ditto.
343 Sat May 6 10:43:33 2006 Ben Pfaff <blp@gnu.org>
345 Continue reforming procedure execution. In this phase, get rid of
346 the output code for SPLIT FILE groups in procedure.c, which really
347 shouldn't be doing any output. Move it into the individual
348 procedures instead. This also adds some flexibility.
350 * crosstabs.q (precalc): Call output_split_file_values().
352 * descriptives.c (calc_descriptives): Ditto.
354 * examine.q (run_examine): Ditto.
356 * frequencies.q (precalc): Ditto.
358 * oneway.q (run_oneway): Ditto.
360 * regression.q (run_regression): Ditto.
362 * t-test.q (calculate): Ditto.
364 Wed May 3 23:05:31 2006 Ben Pfaff <blp@gnu.org>
366 Continue reforming procedure execution. In this phase, get rid of
367 many global variables, consolidating procedure execution in
368 procedure.c. Encapsulate transformations in new "struct
369 trns_chain". Also, change implementation of N OF CASES, FILTER,
370 and PROCESS IF from special cases to transformations.
372 * aggregate.c (cmd_aggregate) Use discard_variables().
374 2006-04-28 Jason Stover <jhs@math.gcsu.edu>
376 * regression.q (regression_trns_resid_proc): Pass only the
377 variables used in the model to (*model->residual)().
379 * regression.q (regression_trns_pred_proc): Pass only the
380 variables used in the model to (*model->pred)().
382 2006-04-26 Jason Stover <jhs@math.gcsu.edu>
384 * regression.q: Added support for multiple transformations.
386 * regression.q (regression_trns_resid_proc): New function.
388 * regression.q (regression_trns_pred_proc): New function.
390 * regression.q (subcommand_save): Added support for saving
393 * regression.q (regression_trns_free): New function.
395 * regression.q (reg_get_name): New function.
397 * regression.q (reg_save_var): New function.
399 Tue Apr 25 13:18:56 2006 Ben Pfaff <blp@gnu.org>
401 * rank.q (parse_rank_function): Use SE instead of ME for parse
404 Tue Apr 25 13:16:28 2006 Ben Pfaff <blp@gnu.org>
406 * flip.c (flip_sink_write): Use snprintf() to simplify a bit of
409 2006-04-21 Jason Stover <jhs@math.gcsu.edu>
411 * regression.q (try_name): New function. (Partly copied from
412 try_name in descriptives.c.)
414 * regression.q (subcommand_save): Choose residual variable names
417 2006-04-20 Jason Stover <jhs@math.gcsu.edu>
419 * regression.q (cmd_regression): Moved call to subcommand_save()
420 outside multipass_procedure_with_splits().
422 * regression.q (regression_trns_proc): Fixed value counter n_vals
423 before calling *model->residual().
425 2006-04-19 Jason Stover <jhs@math.gcsu.edu>
427 * regression.q (regression_trns_proc): Fixed the look-up of the
430 2006-04-18 Jason Stover <jhs@math.gcsu.edu>
432 * regression.q (regression_trns_proc): Look up the residual
433 variable in the linear regression cache.
435 * regression.q (subcommand_save): Set the residual variable in the
436 linear regression cache.
438 2006-04-17 Jason Stover <jason@wonko.gcsu.edu>
440 * regression.q (regression_trns_proc): Accept case_idx as an int
441 to match the definition of trns_proc_func.
443 2006-04-17 Jason Stover <jhs@math.gcsu.edu>
445 * regression.q (regression_trns_proc): New function.
447 * regression.q (subcommand_save): Create variable residuals and
448 add a transformation to assign values to them. Also free the
449 linreg_cache if the SAVE command was not called. Removed the
450 casereading loop. Placed actual computation of residuals in
451 regression_trns_proc.
453 * regression.q (run_regression): Moved call to free
454 pspp_linreg_cache to subcommand_save.
456 Sat Apr 15 18:01:03 2006 Ben Pfaff <blp@gnu.org>
458 * examine.q (output_examine): Add casts to fix warnings.
460 2006-04-07 Jason Stover <jhs@math.gcsu.edu>
462 * regression.q (subcommand_save): New function.
464 2006-04-04 Jason Stover <jhs@math.gcsu.edu>
466 * regression.q: New function reg_has_categorical () to tell
467 whether a pspp_linreg_struct was made with any variables of type
470 * regression.q: (subcommand_export): Call
471 reg_print_categorical_encoding() only if the model uses any
472 categorical variables.
474 Mon Mar 27 16:00:42 2006 Ben Pfaff <blp@gnu.org>
476 * crosstabs.q: (output_pivot_table) Drop spurious space from
479 2006-03-15 Jason Stover <jhs@math.gcsu.edu>
481 * regression.q: Added custom syntax parser for VARIABLES subcommand
483 * regression.q: Moved most instructions for run_regression ()
484 inside the loop over dependent variables.
486 Thu Mar 2 08:40:33 WST 2006 John Darrington <john@darrington.wattle.id.au>
488 * Moved files from src directory