1 2007-03-29 Jason Stover <jhs@math.gcsu.edu>
3 * regression.q (prepare_data): New function.
5 * regression.q (compute_moments): New function.
7 2007-03-18 Ben Pfaff <blp@gnu.org>
9 * crosstabs.q (static var write): Rename write_style to avoid
10 conflict with POSIX function of same name.
12 2007-03-16 Jason Stover <jhs@math.gcsu.edu>
14 * regression.q (run_regression): Added support for moments.
16 Sat Feb 17 08:16:00 2007 Ben Pfaff <blp@gnu.org>
18 * flip.c (flip_sink_create): Improve error message when temporary
19 file cannot be created.
21 Tue Feb 6 19:58:03 2007 Ben Pfaff <blp@gnu.org>
23 * flip.c (flip_file): Give better error message on end-of-file.
25 2007-02-04 Jason Stover <jhs@math.gcsu.edu>
27 * regression.q: Fixed p-value computation in the test for
28 individual regression coefficients.
30 Mon Jan 15 11:03:20 2007 Ben Pfaff <blp@gnu.org>
32 Fix bugs found by valgrind when --enable-debug is used with the
33 new case code. These bugs are hidden when the data set is small
34 enough to find in memory; if a bigger data set that would overflow
35 to disk were used, then data corruption would occur.
37 * chisquare.c (create_freq_hash): Pass free_freq_mutable_hash to
38 hsh_create as free function. Make copy of data put into hash.
40 * oneway.q (free_value): New function.
41 (run_oneway): Use free_value as arg to hsh_create. Make copy of
44 * rank.q (rank_cases): Don't access data in case after we've given
47 Tue Jan 9 19:16:11 2007 Ben Pfaff <blp@gnu.org>
51 * aggregate.c (parse_aggregate_functions) Initialize
54 Fri Dec 22 14:04:09 2006 Ben Pfaff <blp@gnu.org>
56 Simplify missing value handling.
58 * aggregate.c (struct agr_var): Remove `bool include_missing', add
59 `enum mv_class exclude'. Remove `int missing', add `bool
60 saw_missing'. Update users.
62 * descriptives.c (struct dsc_trns): Removed `int
63 include_user_missing', add `enum mv_class exclude'. Update users.
64 (struct dsc_proc): Ditto.
66 * examine.q: (static var value_is_missing): Rename
67 `exclude_values', change type to `enum mv_class'. Update users.
71 Fri Dec 22 19:22:18 WST 2006 John Darrington <john@darrington.wattle.id.au>
73 * frequencies.q : Fixed bug #17420, where the table bounds were overun
74 when /FORMAT=nolabels was given.
76 Wed Dec 20 18:45:31 WST 2006 John Darrington <john@darrington.wattle.id.au>
78 * binomial.c binomial.h : New files. Thanks to Jason Stover
79 for assistance with these.
81 * chisquare.c chisquare.h freq.c freq.h npar-summary.c npar-summary.h
82 npar.h npar.q: New files. Implementing NPAR TESTS.
84 * frequencies.q : Moved structure definitions into freq.[ch]
86 Sat Dec 16 22:26:44 2006 Ben Pfaff <blp@gnu.org>
88 Make it possible to pull cases from the active file with a
89 function call, instead of requiring indirection through a callback
92 * aggregate.c (cmd_aggregate): Take advantage of new procedure
94 (agr_to_active_file): Removed.
95 (presorted_agr_to_sysfile): Removed.
97 * autorecode.c (cmd_autorecode): Take advantage of new procedure
99 (autorecode_proc_func): Removed.
101 * flip.c (struct flip_pgm): New members to allow conformance with
102 new case_source_class interface.
103 (cmd_flip): Adapt to new case_source_class interface.
104 (flip_source_read): Ditto.
105 (flip_source_destroy): Ditto.
107 Sat Dec 16 12:54:27 2006 Ben Pfaff <blp@gnu.org>
109 * rank.q (rank_custom_variables): Allow grouping variables to be
110 strings. Fixes bug #18533. Thanks to John Darrington for review.
112 Sat Dec 9 18:47:51 2006 Ben Pfaff <blp@gnu.org>
114 * regression.q (is_depvar): Compare variable pointers instead of
117 Thu Dec 7 15:26:25 WST 2006 John Darrington <john@darrington.wattle.id.au>
119 * examine.q: Allocated the categorical values for the dependent and
120 independent variables, on the heap. Hence they can be of any width.
122 Wed Dec 6 21:14:26 2006 Ben Pfaff <blp@gnu.org>
124 * regression.q (reg_inserted): Compare variable pointers instead
127 Mon Dec 4 22:33:46 2006 Ben Pfaff <blp@gnu.org>
129 * crosstabs.q (insert_summary): Use var_to_string for labeling.
130 (output_pivot_table) Ditto.
133 * frequencies.q (setup_z_trns): Ditto.
135 (dump_condensed) Ditto.
136 (dump_statistics) Ditto.
139 Sun Nov 5 08:31:42 WST 2006 John Darrington <john@darrington.wattle.id.au>
141 * t-test.q, oneway.q: Changed to use the new casefilter structure.
143 Sat Oct 14 16:52:28 2006 Ben Pfaff <blp@gnu.org>
145 * rank.q: (rank_sorted_casefile) Add some missing case_destroy()
146 calls to fix a memory leak.
148 Sun Oct 8 09:45:40 WST 2006 John Darrington <john@darrington.wattle.id.au>
150 * rank.q: Plugged a small memory leak which occurred under error
153 Sat Oct 7 11:06:01 WST 2006 John Darrington <john@darrington.wattle.id.au>
155 * rank.q: Implemented most of the RANK command.
157 2006-07-14 Jason Stover <jhs@math.gcsu.edu>
159 * regression.q (run_regression): New function to move knowledge of
160 pspp_linreg_cache out of math/coefficient.[ch].
162 Sat Jul 1 17:41:46 2006 Ben Pfaff <blp@gnu.org>
164 Fix bug #11612, "q2c documentation does not agree with code".
166 * examine.q: Audit use of q2c "+" prefixes that indicate that a
167 command may appear multiple times.
169 * frequencies.q: Ditto.
173 * regression.q: Ditto.
177 Fri Jun 23 14:18:22 2006 Ben Pfaff <blp@gnu.org>
179 Support long string variables on FREQUENCIES, as
180 an extension when in enhanced algorithms mode. For Greg Hunt
181 <greg@firmansyah.com>.
183 * frequencies.q: (struct freq) Change `v' member from union value
184 to union value *. Update all references.
185 (struct var_freqs) Add width, print members to represent effective
186 variable width and display format.
187 (calc) Copy entire long string value into the hash table.
188 (frq_custom_variables) Set new width, print members.
189 (hash_value_alpha) Get width from var_freqs.
190 (compare_value_alpha_a) Ditto.
191 (compare_freq_alpha_a) Ditto.
192 (compare_freq_alpha_d) Ditto.
193 (dump_full) Get display format from var_freqs.
194 (dump_condensed) Ditto.
196 Mon Jun 19 22:07:13 2006 Ben Pfaff <blp@gnu.org>
198 * frequencies.q: (dump_full) Only put the first MAX_SHORT_STRING
199 bytes of string variables into the output cells, seeing as we only
201 (dump_condensed) Ditto.
203 Mon Jun 19 21:52:05 2006 Ben Pfaff <blp@gnu.org>
205 Fixes a bug reported by Greg Hunt <greg@firmansyah.com>.
207 * frequencies.q: (hsh_hash_bytes) We only copy the first
208 MAX_SHORT_STRING bytes of string variables, so we must only
209 compare that many bytes, even if the string variable is longer.
210 (compare_value_alpha_a) Ditto.
211 (compare_freq_alpha_a) Ditto.
212 (compare_freq_alpha_d) Ditto.
214 2006-05-11 Jason Stover <jhs@math.gcsu.edu>
216 * regression.q: Adjusted code to account for cache->coeff being a
217 pspp_linreg_coeff **.
219 Sun May 7 18:31:25 2006 Ben Pfaff <blp@gnu.org>
223 * aggregate.c (cmd_aggregate): Free default_dict before replacing
226 Sun May 7 17:09:19 2006 Ben Pfaff <blp@gnu.org>
228 * flip.c (flip_file): Check return value of pool_fclose().
230 Sat May 6 16:00:13 2006 Ben Pfaff <blp@gnu.org>
232 Get rid of `char *c' member in union value, for cleanliness.
234 * aggregate.c: (union agr_argument) New union.
235 (struct agr_var) Change element type of arg[] from union value to
237 (parse_aggregate_functions) Change local variable types likewise.
239 * autorecode.c: (union arc_value) New union.
240 (struct arc_item) Change "from" from union value to union
242 (recode) Change local variable from union value to union
244 (autorecode_trns_proc) Ditto.
245 (compare_alpha_value) Ditto.
246 (hash_alpha_value) Ditto.
247 (compare_numeric_value) Ditto.
248 (hash_numeric_value) Ditto.
249 (autorecode_proc_func) Ditto.
251 Sat May 6 10:43:33 2006 Ben Pfaff <blp@gnu.org>
253 Continue reforming procedure execution. In this phase, get rid of
254 the output code for SPLIT FILE groups in procedure.c, which really
255 shouldn't be doing any output. Move it into the individual
256 procedures instead. This also adds some flexibility.
258 * crosstabs.q (precalc): Call output_split_file_values().
260 * descriptives.c (calc_descriptives): Ditto.
262 * examine.q (run_examine): Ditto.
264 * frequencies.q (precalc): Ditto.
266 * oneway.q (run_oneway): Ditto.
268 * regression.q (run_regression): Ditto.
270 * t-test.q (calculate): Ditto.
272 Wed May 3 23:05:31 2006 Ben Pfaff <blp@gnu.org>
274 Continue reforming procedure execution. In this phase, get rid of
275 many global variables, consolidating procedure execution in
276 procedure.c. Encapsulate transformations in new "struct
277 trns_chain". Also, change implementation of N OF CASES, FILTER,
278 and PROCESS IF from special cases to transformations.
280 * aggregate.c (cmd_aggregate) Use discard_variables().
282 2006-04-28 Jason Stover <jhs@math.gcsu.edu>
284 * regression.q (regression_trns_resid_proc): Pass only the
285 variables used in the model to (*model->residual)().
287 * regression.q (regression_trns_pred_proc): Pass only the
288 variables used in the model to (*model->pred)().
290 2006-04-26 Jason Stover <jhs@math.gcsu.edu>
292 * regression.q: Added support for multiple transformations.
294 * regression.q (regression_trns_resid_proc): New function.
296 * regression.q (regression_trns_pred_proc): New function.
298 * regression.q (subcommand_save): Added support for saving
301 * regression.q (regression_trns_free): New function.
303 * regression.q (reg_get_name): New function.
305 * regression.q (reg_save_var): New function.
307 Tue Apr 25 13:18:56 2006 Ben Pfaff <blp@gnu.org>
309 * rank.q (parse_rank_function): Use SE instead of ME for parse
312 Tue Apr 25 13:16:28 2006 Ben Pfaff <blp@gnu.org>
314 * flip.c (flip_sink_write): Use snprintf() to simplify a bit of
317 2006-04-21 Jason Stover <jhs@math.gcsu.edu>
319 * regression.q (try_name): New function. (Partly copied from
320 try_name in descriptives.c.)
322 * regression.q (subcommand_save): Choose residual variable names
325 2006-04-20 Jason Stover <jhs@math.gcsu.edu>
327 * regression.q (cmd_regression): Moved call to subcommand_save()
328 outside multipass_procedure_with_splits().
330 * regression.q (regression_trns_proc): Fixed value counter n_vals
331 before calling *model->residual().
333 2006-04-19 Jason Stover <jhs@math.gcsu.edu>
335 * regression.q (regression_trns_proc): Fixed the look-up of the
338 2006-04-18 Jason Stover <jhs@math.gcsu.edu>
340 * regression.q (regression_trns_proc): Look up the residual
341 variable in the linear regression cache.
343 * regression.q (subcommand_save): Set the residual variable in the
344 linear regression cache.
346 2006-04-17 Jason Stover <jason@wonko.gcsu.edu>
348 * regression.q (regression_trns_proc): Accept case_idx as an int
349 to match the definition of trns_proc_func.
351 2006-04-17 Jason Stover <jhs@math.gcsu.edu>
353 * regression.q (regression_trns_proc): New function.
355 * regression.q (subcommand_save): Create variable residuals and
356 add a transformation to assign values to them. Also free the
357 linreg_cache if the SAVE command was not called. Removed the
358 casereading loop. Placed actual computation of residuals in
359 regression_trns_proc.
361 * regression.q (run_regression): Moved call to free
362 pspp_linreg_cache to subcommand_save.
364 Sat Apr 15 18:01:03 2006 Ben Pfaff <blp@gnu.org>
366 * examine.q (output_examine): Add casts to fix warnings.
368 2006-04-07 Jason Stover <jhs@math.gcsu.edu>
370 * regression.q (subcommand_save): New function.
372 2006-04-04 Jason Stover <jhs@math.gcsu.edu>
374 * regression.q: New function reg_has_categorical () to tell
375 whether a pspp_linreg_struct was made with any variables of type
378 * regression.q: (subcommand_export): Call
379 reg_print_categorical_encoding() only if the model uses any
380 categorical variables.
382 Mon Mar 27 16:00:42 2006 Ben Pfaff <blp@gnu.org>
384 * crosstabs.q: (output_pivot_table) Drop spurious space from
387 2006-03-15 Jason Stover <jhs@math.gcsu.edu>
389 * regression.q: Added custom syntax parser for VARIABLES subcommand
391 * regression.q: Moved most instructions for run_regression ()
392 inside the loop over dependent variables.
394 Thu Mar 2 08:40:33 WST 2006 John Darrington <john@darrington.wattle.id.au>
396 * Moved files from src directory