1 2007-07-10 Jason Stover <jhs@math.gcsu.edu>
3 * glm.q: Initial version of the GLM procedure.
5 2007-06-06 Ben Pfaff <blp@gnu.org>
7 Adapt case sources, sinks, and clients of procedure code to the
10 * aggregate.c: Simplify greatly since everything is more uniform
13 * autorecode.c: Adapt to new procedure code.
17 * descriptives.c: Ditto.
19 * npar-summary.c: Ditto.
20 * frequencies.q: Ditto.
23 * regression.q: Ditto.
24 * sort-cases.c: Ditto.
27 * sort-criteria.c: Rewrite to output a struct case_ordering.
29 * flip.c: Rewrite to be a casereader.
31 * rank.q: Simplify greatly since casereaders are much more
32 flexible than what we had before.
34 2007-05-15 Jason Stover <jhs@math.gcsu.edu>
36 * regression.q (run_regression): Tell the user when the data
37 contain no valid cases.
39 2007-05-08 Jason Stover <jhs@math.gcsu.edu>
41 * regression.q: Partial fix of bug which caused a crash if
42 dependent variable and independent variable were the same.
44 2007-04-16 John Darrington <john@darrington.wattle.id.au>
46 * t-test.q: Changed the output width of reported counts and
47 degrees of freedom, to avoid truncating these values. Thanks
48 to Seth Woolley for reporting this problem. A proper fix involves
49 re-thinking the output driver.
51 2007-04-12 Jason Stover <jhs@math.gcsu.edu>
53 * regression.q (run_regression): Added if (n_data >0) to fix bug
56 2007-03-29 Jason Stover <jhs@math.gcsu.edu>
58 * regression.q (prepare_data): New function.
60 * regression.q (compute_moments): New function.
62 2007-03-18 Ben Pfaff <blp@gnu.org>
64 * crosstabs.q (static var write): Rename write_style to avoid
65 conflict with POSIX function of same name.
67 2007-03-16 Jason Stover <jhs@math.gcsu.edu>
69 * regression.q (run_regression): Added support for moments.
71 Sat Feb 17 08:16:00 2007 Ben Pfaff <blp@gnu.org>
73 * flip.c (flip_sink_create): Improve error message when temporary
74 file cannot be created.
76 Tue Feb 6 19:58:03 2007 Ben Pfaff <blp@gnu.org>
78 * flip.c (flip_file): Give better error message on end-of-file.
80 2007-02-04 Jason Stover <jhs@math.gcsu.edu>
82 * regression.q: Fixed p-value computation in the test for
83 individual regression coefficients.
85 Mon Jan 15 11:03:20 2007 Ben Pfaff <blp@gnu.org>
87 Fix bugs found by valgrind when --enable-debug is used with the
88 new case code. These bugs are hidden when the data set is small
89 enough to find in memory; if a bigger data set that would overflow
90 to disk were used, then data corruption would occur.
92 * chisquare.c (create_freq_hash): Pass free_freq_mutable_hash to
93 hsh_create as free function. Make copy of data put into hash.
95 * oneway.q (free_value): New function.
96 (run_oneway): Use free_value as arg to hsh_create. Make copy of
99 * rank.q (rank_cases): Don't access data in case after we've given
102 Tue Jan 9 19:16:11 2007 Ben Pfaff <blp@gnu.org>
106 * aggregate.c (parse_aggregate_functions) Initialize
109 Fri Dec 22 14:04:09 2006 Ben Pfaff <blp@gnu.org>
111 Simplify missing value handling.
113 * aggregate.c (struct agr_var): Remove `bool include_missing', add
114 `enum mv_class exclude'. Remove `int missing', add `bool
115 saw_missing'. Update users.
117 * descriptives.c (struct dsc_trns): Removed `int
118 include_user_missing', add `enum mv_class exclude'. Update users.
119 (struct dsc_proc): Ditto.
121 * examine.q: (static var value_is_missing): Rename
122 `exclude_values', change type to `enum mv_class'. Update users.
126 Fri Dec 22 19:22:18 WST 2006 John Darrington <john@darrington.wattle.id.au>
128 * frequencies.q : Fixed bug #17420, where the table bounds were overun
129 when /FORMAT=nolabels was given.
131 Wed Dec 20 18:45:31 WST 2006 John Darrington <john@darrington.wattle.id.au>
133 * binomial.c binomial.h : New files. Thanks to Jason Stover
134 for assistance with these.
136 * chisquare.c chisquare.h freq.c freq.h npar-summary.c npar-summary.h
137 npar.h npar.q: New files. Implementing NPAR TESTS.
139 * frequencies.q : Moved structure definitions into freq.[ch]
141 Sat Dec 16 22:26:44 2006 Ben Pfaff <blp@gnu.org>
143 Make it possible to pull cases from the active file with a
144 function call, instead of requiring indirection through a callback
147 * aggregate.c (cmd_aggregate): Take advantage of new procedure
149 (agr_to_active_file): Removed.
150 (presorted_agr_to_sysfile): Removed.
152 * autorecode.c (cmd_autorecode): Take advantage of new procedure
154 (autorecode_proc_func): Removed.
156 * flip.c (struct flip_pgm): New members to allow conformance with
157 new case_source_class interface.
158 (cmd_flip): Adapt to new case_source_class interface.
159 (flip_source_read): Ditto.
160 (flip_source_destroy): Ditto.
162 Sat Dec 16 12:54:27 2006 Ben Pfaff <blp@gnu.org>
164 * rank.q (rank_custom_variables): Allow grouping variables to be
165 strings. Fixes bug #18533. Thanks to John Darrington for review.
167 Sat Dec 9 18:47:51 2006 Ben Pfaff <blp@gnu.org>
169 * regression.q (is_depvar): Compare variable pointers instead of
172 Thu Dec 7 15:26:25 WST 2006 John Darrington <john@darrington.wattle.id.au>
174 * examine.q: Allocated the categorical values for the dependent and
175 independent variables, on the heap. Hence they can be of any width.
177 Wed Dec 6 21:14:26 2006 Ben Pfaff <blp@gnu.org>
179 * regression.q (reg_inserted): Compare variable pointers instead
182 Mon Dec 4 22:33:46 2006 Ben Pfaff <blp@gnu.org>
184 * crosstabs.q (insert_summary): Use var_to_string for labeling.
185 (output_pivot_table) Ditto.
188 * frequencies.q (setup_z_trns): Ditto.
190 (dump_condensed) Ditto.
191 (dump_statistics) Ditto.
194 Sun Nov 5 08:31:42 WST 2006 John Darrington <john@darrington.wattle.id.au>
196 * t-test.q, oneway.q: Changed to use the new casefilter structure.
198 Sat Oct 14 16:52:28 2006 Ben Pfaff <blp@gnu.org>
200 * rank.q: (rank_sorted_casefile) Add some missing case_destroy()
201 calls to fix a memory leak.
203 Sun Oct 8 09:45:40 WST 2006 John Darrington <john@darrington.wattle.id.au>
205 * rank.q: Plugged a small memory leak which occurred under error
208 Sat Oct 7 11:06:01 WST 2006 John Darrington <john@darrington.wattle.id.au>
210 * rank.q: Implemented most of the RANK command.
212 2006-07-14 Jason Stover <jhs@math.gcsu.edu>
214 * regression.q (run_regression): New function to move knowledge of
215 pspp_linreg_cache out of math/coefficient.[ch].
217 Sat Jul 1 17:41:46 2006 Ben Pfaff <blp@gnu.org>
219 Fix bug #11612, "q2c documentation does not agree with code".
221 * examine.q: Audit use of q2c "+" prefixes that indicate that a
222 command may appear multiple times.
224 * frequencies.q: Ditto.
228 * regression.q: Ditto.
232 Fri Jun 23 14:18:22 2006 Ben Pfaff <blp@gnu.org>
234 Support long string variables on FREQUENCIES, as
235 an extension when in enhanced algorithms mode. For Greg Hunt
236 <greg@firmansyah.com>.
238 * frequencies.q: (struct freq) Change `v' member from union value
239 to union value *. Update all references.
240 (struct var_freqs) Add width, print members to represent effective
241 variable width and display format.
242 (calc) Copy entire long string value into the hash table.
243 (frq_custom_variables) Set new width, print members.
244 (hash_value_alpha) Get width from var_freqs.
245 (compare_value_alpha_a) Ditto.
246 (compare_freq_alpha_a) Ditto.
247 (compare_freq_alpha_d) Ditto.
248 (dump_full) Get display format from var_freqs.
249 (dump_condensed) Ditto.
251 Mon Jun 19 22:07:13 2006 Ben Pfaff <blp@gnu.org>
253 * frequencies.q: (dump_full) Only put the first MAX_SHORT_STRING
254 bytes of string variables into the output cells, seeing as we only
256 (dump_condensed) Ditto.
258 Mon Jun 19 21:52:05 2006 Ben Pfaff <blp@gnu.org>
260 Fixes a bug reported by Greg Hunt <greg@firmansyah.com>.
262 * frequencies.q: (hsh_hash_bytes) We only copy the first
263 MAX_SHORT_STRING bytes of string variables, so we must only
264 compare that many bytes, even if the string variable is longer.
265 (compare_value_alpha_a) Ditto.
266 (compare_freq_alpha_a) Ditto.
267 (compare_freq_alpha_d) Ditto.
269 2006-05-11 Jason Stover <jhs@math.gcsu.edu>
271 * regression.q: Adjusted code to account for cache->coeff being a
272 pspp_linreg_coeff **.
274 Sun May 7 18:31:25 2006 Ben Pfaff <blp@gnu.org>
278 * aggregate.c (cmd_aggregate): Free default_dict before replacing
281 Sun May 7 17:09:19 2006 Ben Pfaff <blp@gnu.org>
283 * flip.c (flip_file): Check return value of pool_fclose().
285 Sat May 6 16:00:13 2006 Ben Pfaff <blp@gnu.org>
287 Get rid of `char *c' member in union value, for cleanliness.
289 * aggregate.c: (union agr_argument) New union.
290 (struct agr_var) Change element type of arg[] from union value to
292 (parse_aggregate_functions) Change local variable types likewise.
294 * autorecode.c: (union arc_value) New union.
295 (struct arc_item) Change "from" from union value to union
297 (recode) Change local variable from union value to union
299 (autorecode_trns_proc) Ditto.
300 (compare_alpha_value) Ditto.
301 (hash_alpha_value) Ditto.
302 (compare_numeric_value) Ditto.
303 (hash_numeric_value) Ditto.
304 (autorecode_proc_func) Ditto.
306 Sat May 6 10:43:33 2006 Ben Pfaff <blp@gnu.org>
308 Continue reforming procedure execution. In this phase, get rid of
309 the output code for SPLIT FILE groups in procedure.c, which really
310 shouldn't be doing any output. Move it into the individual
311 procedures instead. This also adds some flexibility.
313 * crosstabs.q (precalc): Call output_split_file_values().
315 * descriptives.c (calc_descriptives): Ditto.
317 * examine.q (run_examine): Ditto.
319 * frequencies.q (precalc): Ditto.
321 * oneway.q (run_oneway): Ditto.
323 * regression.q (run_regression): Ditto.
325 * t-test.q (calculate): Ditto.
327 Wed May 3 23:05:31 2006 Ben Pfaff <blp@gnu.org>
329 Continue reforming procedure execution. In this phase, get rid of
330 many global variables, consolidating procedure execution in
331 procedure.c. Encapsulate transformations in new "struct
332 trns_chain". Also, change implementation of N OF CASES, FILTER,
333 and PROCESS IF from special cases to transformations.
335 * aggregate.c (cmd_aggregate) Use discard_variables().
337 2006-04-28 Jason Stover <jhs@math.gcsu.edu>
339 * regression.q (regression_trns_resid_proc): Pass only the
340 variables used in the model to (*model->residual)().
342 * regression.q (regression_trns_pred_proc): Pass only the
343 variables used in the model to (*model->pred)().
345 2006-04-26 Jason Stover <jhs@math.gcsu.edu>
347 * regression.q: Added support for multiple transformations.
349 * regression.q (regression_trns_resid_proc): New function.
351 * regression.q (regression_trns_pred_proc): New function.
353 * regression.q (subcommand_save): Added support for saving
356 * regression.q (regression_trns_free): New function.
358 * regression.q (reg_get_name): New function.
360 * regression.q (reg_save_var): New function.
362 Tue Apr 25 13:18:56 2006 Ben Pfaff <blp@gnu.org>
364 * rank.q (parse_rank_function): Use SE instead of ME for parse
367 Tue Apr 25 13:16:28 2006 Ben Pfaff <blp@gnu.org>
369 * flip.c (flip_sink_write): Use snprintf() to simplify a bit of
372 2006-04-21 Jason Stover <jhs@math.gcsu.edu>
374 * regression.q (try_name): New function. (Partly copied from
375 try_name in descriptives.c.)
377 * regression.q (subcommand_save): Choose residual variable names
380 2006-04-20 Jason Stover <jhs@math.gcsu.edu>
382 * regression.q (cmd_regression): Moved call to subcommand_save()
383 outside multipass_procedure_with_splits().
385 * regression.q (regression_trns_proc): Fixed value counter n_vals
386 before calling *model->residual().
388 2006-04-19 Jason Stover <jhs@math.gcsu.edu>
390 * regression.q (regression_trns_proc): Fixed the look-up of the
393 2006-04-18 Jason Stover <jhs@math.gcsu.edu>
395 * regression.q (regression_trns_proc): Look up the residual
396 variable in the linear regression cache.
398 * regression.q (subcommand_save): Set the residual variable in the
399 linear regression cache.
401 2006-04-17 Jason Stover <jason@wonko.gcsu.edu>
403 * regression.q (regression_trns_proc): Accept case_idx as an int
404 to match the definition of trns_proc_func.
406 2006-04-17 Jason Stover <jhs@math.gcsu.edu>
408 * regression.q (regression_trns_proc): New function.
410 * regression.q (subcommand_save): Create variable residuals and
411 add a transformation to assign values to them. Also free the
412 linreg_cache if the SAVE command was not called. Removed the
413 casereading loop. Placed actual computation of residuals in
414 regression_trns_proc.
416 * regression.q (run_regression): Moved call to free
417 pspp_linreg_cache to subcommand_save.
419 Sat Apr 15 18:01:03 2006 Ben Pfaff <blp@gnu.org>
421 * examine.q (output_examine): Add casts to fix warnings.
423 2006-04-07 Jason Stover <jhs@math.gcsu.edu>
425 * regression.q (subcommand_save): New function.
427 2006-04-04 Jason Stover <jhs@math.gcsu.edu>
429 * regression.q: New function reg_has_categorical () to tell
430 whether a pspp_linreg_struct was made with any variables of type
433 * regression.q: (subcommand_export): Call
434 reg_print_categorical_encoding() only if the model uses any
435 categorical variables.
437 Mon Mar 27 16:00:42 2006 Ben Pfaff <blp@gnu.org>
439 * crosstabs.q: (output_pivot_table) Drop spurious space from
442 2006-03-15 Jason Stover <jhs@math.gcsu.edu>
444 * regression.q: Added custom syntax parser for VARIABLES subcommand
446 * regression.q: Moved most instructions for run_regression ()
447 inside the loop over dependent variables.
449 Thu Mar 2 08:40:33 WST 2006 John Darrington <john@darrington.wattle.id.au>
451 * Moved files from src directory