1 2007-06-06 Ben Pfaff <blp@gnu.org>
3 Adapt case sources, sinks, and clients of procedure code to the
6 * aggregate.c: Simplify greatly since everything is more uniform
9 * autorecode.c: Adapt to new procedure code.
13 * descriptives.c: Ditto.
15 * npar-summary.c: Ditto.
16 * frequencies.q: Ditto.
19 * regression.q: Ditto.
20 * sort-cases.c: Ditto.
23 * sort-criteria.c: Rewrite to output a struct case_ordering.
25 * flip.c: Rewrite to be a casereader.
27 * rank.q: Simplify greatly since casereaders are much more
28 flexible than what we had before.
30 2007-05-15 Jason Stover <jhs@math.gcsu.edu>
32 * regression.q (run_regression): Tell the user when the data
33 contain no valid cases.
35 2007-05-08 Jason Stover <jhs@math.gcsu.edu>
37 * regression.q: Partial fix of bug which caused a crash if
38 dependent variable and independent variable were the same.
40 2007-04-16 John Darrington <john@darrington.wattle.id.au>
42 * t-test.q: Changed the output width of reported counts and
43 degrees of freedom, to avoid truncating these values. Thanks
44 to Seth Woolley for reporting this problem. A proper fix involves
45 re-thinking the output driver.
47 2007-04-12 Jason Stover <jhs@math.gcsu.edu>
49 * regression.q (run_regression): Added if (n_data >0) to fix bug
52 2007-03-29 Jason Stover <jhs@math.gcsu.edu>
54 * regression.q (prepare_data): New function.
56 * regression.q (compute_moments): New function.
58 2007-03-18 Ben Pfaff <blp@gnu.org>
60 * crosstabs.q (static var write): Rename write_style to avoid
61 conflict with POSIX function of same name.
63 2007-03-16 Jason Stover <jhs@math.gcsu.edu>
65 * regression.q (run_regression): Added support for moments.
67 Sat Feb 17 08:16:00 2007 Ben Pfaff <blp@gnu.org>
69 * flip.c (flip_sink_create): Improve error message when temporary
70 file cannot be created.
72 Tue Feb 6 19:58:03 2007 Ben Pfaff <blp@gnu.org>
74 * flip.c (flip_file): Give better error message on end-of-file.
76 2007-02-04 Jason Stover <jhs@math.gcsu.edu>
78 * regression.q: Fixed p-value computation in the test for
79 individual regression coefficients.
81 Mon Jan 15 11:03:20 2007 Ben Pfaff <blp@gnu.org>
83 Fix bugs found by valgrind when --enable-debug is used with the
84 new case code. These bugs are hidden when the data set is small
85 enough to find in memory; if a bigger data set that would overflow
86 to disk were used, then data corruption would occur.
88 * chisquare.c (create_freq_hash): Pass free_freq_mutable_hash to
89 hsh_create as free function. Make copy of data put into hash.
91 * oneway.q (free_value): New function.
92 (run_oneway): Use free_value as arg to hsh_create. Make copy of
95 * rank.q (rank_cases): Don't access data in case after we've given
98 Tue Jan 9 19:16:11 2007 Ben Pfaff <blp@gnu.org>
102 * aggregate.c (parse_aggregate_functions) Initialize
105 Fri Dec 22 14:04:09 2006 Ben Pfaff <blp@gnu.org>
107 Simplify missing value handling.
109 * aggregate.c (struct agr_var): Remove `bool include_missing', add
110 `enum mv_class exclude'. Remove `int missing', add `bool
111 saw_missing'. Update users.
113 * descriptives.c (struct dsc_trns): Removed `int
114 include_user_missing', add `enum mv_class exclude'. Update users.
115 (struct dsc_proc): Ditto.
117 * examine.q: (static var value_is_missing): Rename
118 `exclude_values', change type to `enum mv_class'. Update users.
122 Fri Dec 22 19:22:18 WST 2006 John Darrington <john@darrington.wattle.id.au>
124 * frequencies.q : Fixed bug #17420, where the table bounds were overun
125 when /FORMAT=nolabels was given.
127 Wed Dec 20 18:45:31 WST 2006 John Darrington <john@darrington.wattle.id.au>
129 * binomial.c binomial.h : New files. Thanks to Jason Stover
130 for assistance with these.
132 * chisquare.c chisquare.h freq.c freq.h npar-summary.c npar-summary.h
133 npar.h npar.q: New files. Implementing NPAR TESTS.
135 * frequencies.q : Moved structure definitions into freq.[ch]
137 Sat Dec 16 22:26:44 2006 Ben Pfaff <blp@gnu.org>
139 Make it possible to pull cases from the active file with a
140 function call, instead of requiring indirection through a callback
143 * aggregate.c (cmd_aggregate): Take advantage of new procedure
145 (agr_to_active_file): Removed.
146 (presorted_agr_to_sysfile): Removed.
148 * autorecode.c (cmd_autorecode): Take advantage of new procedure
150 (autorecode_proc_func): Removed.
152 * flip.c (struct flip_pgm): New members to allow conformance with
153 new case_source_class interface.
154 (cmd_flip): Adapt to new case_source_class interface.
155 (flip_source_read): Ditto.
156 (flip_source_destroy): Ditto.
158 Sat Dec 16 12:54:27 2006 Ben Pfaff <blp@gnu.org>
160 * rank.q (rank_custom_variables): Allow grouping variables to be
161 strings. Fixes bug #18533. Thanks to John Darrington for review.
163 Sat Dec 9 18:47:51 2006 Ben Pfaff <blp@gnu.org>
165 * regression.q (is_depvar): Compare variable pointers instead of
168 Thu Dec 7 15:26:25 WST 2006 John Darrington <john@darrington.wattle.id.au>
170 * examine.q: Allocated the categorical values for the dependent and
171 independent variables, on the heap. Hence they can be of any width.
173 Wed Dec 6 21:14:26 2006 Ben Pfaff <blp@gnu.org>
175 * regression.q (reg_inserted): Compare variable pointers instead
178 Mon Dec 4 22:33:46 2006 Ben Pfaff <blp@gnu.org>
180 * crosstabs.q (insert_summary): Use var_to_string for labeling.
181 (output_pivot_table) Ditto.
184 * frequencies.q (setup_z_trns): Ditto.
186 (dump_condensed) Ditto.
187 (dump_statistics) Ditto.
190 Sun Nov 5 08:31:42 WST 2006 John Darrington <john@darrington.wattle.id.au>
192 * t-test.q, oneway.q: Changed to use the new casefilter structure.
194 Sat Oct 14 16:52:28 2006 Ben Pfaff <blp@gnu.org>
196 * rank.q: (rank_sorted_casefile) Add some missing case_destroy()
197 calls to fix a memory leak.
199 Sun Oct 8 09:45:40 WST 2006 John Darrington <john@darrington.wattle.id.au>
201 * rank.q: Plugged a small memory leak which occurred under error
204 Sat Oct 7 11:06:01 WST 2006 John Darrington <john@darrington.wattle.id.au>
206 * rank.q: Implemented most of the RANK command.
208 2006-07-14 Jason Stover <jhs@math.gcsu.edu>
210 * regression.q (run_regression): New function to move knowledge of
211 pspp_linreg_cache out of math/coefficient.[ch].
213 Sat Jul 1 17:41:46 2006 Ben Pfaff <blp@gnu.org>
215 Fix bug #11612, "q2c documentation does not agree with code".
217 * examine.q: Audit use of q2c "+" prefixes that indicate that a
218 command may appear multiple times.
220 * frequencies.q: Ditto.
224 * regression.q: Ditto.
228 Fri Jun 23 14:18:22 2006 Ben Pfaff <blp@gnu.org>
230 Support long string variables on FREQUENCIES, as
231 an extension when in enhanced algorithms mode. For Greg Hunt
232 <greg@firmansyah.com>.
234 * frequencies.q: (struct freq) Change `v' member from union value
235 to union value *. Update all references.
236 (struct var_freqs) Add width, print members to represent effective
237 variable width and display format.
238 (calc) Copy entire long string value into the hash table.
239 (frq_custom_variables) Set new width, print members.
240 (hash_value_alpha) Get width from var_freqs.
241 (compare_value_alpha_a) Ditto.
242 (compare_freq_alpha_a) Ditto.
243 (compare_freq_alpha_d) Ditto.
244 (dump_full) Get display format from var_freqs.
245 (dump_condensed) Ditto.
247 Mon Jun 19 22:07:13 2006 Ben Pfaff <blp@gnu.org>
249 * frequencies.q: (dump_full) Only put the first MAX_SHORT_STRING
250 bytes of string variables into the output cells, seeing as we only
252 (dump_condensed) Ditto.
254 Mon Jun 19 21:52:05 2006 Ben Pfaff <blp@gnu.org>
256 Fixes a bug reported by Greg Hunt <greg@firmansyah.com>.
258 * frequencies.q: (hsh_hash_bytes) We only copy the first
259 MAX_SHORT_STRING bytes of string variables, so we must only
260 compare that many bytes, even if the string variable is longer.
261 (compare_value_alpha_a) Ditto.
262 (compare_freq_alpha_a) Ditto.
263 (compare_freq_alpha_d) Ditto.
265 2006-05-11 Jason Stover <jhs@math.gcsu.edu>
267 * regression.q: Adjusted code to account for cache->coeff being a
268 pspp_linreg_coeff **.
270 Sun May 7 18:31:25 2006 Ben Pfaff <blp@gnu.org>
274 * aggregate.c (cmd_aggregate): Free default_dict before replacing
277 Sun May 7 17:09:19 2006 Ben Pfaff <blp@gnu.org>
279 * flip.c (flip_file): Check return value of pool_fclose().
281 Sat May 6 16:00:13 2006 Ben Pfaff <blp@gnu.org>
283 Get rid of `char *c' member in union value, for cleanliness.
285 * aggregate.c: (union agr_argument) New union.
286 (struct agr_var) Change element type of arg[] from union value to
288 (parse_aggregate_functions) Change local variable types likewise.
290 * autorecode.c: (union arc_value) New union.
291 (struct arc_item) Change "from" from union value to union
293 (recode) Change local variable from union value to union
295 (autorecode_trns_proc) Ditto.
296 (compare_alpha_value) Ditto.
297 (hash_alpha_value) Ditto.
298 (compare_numeric_value) Ditto.
299 (hash_numeric_value) Ditto.
300 (autorecode_proc_func) Ditto.
302 Sat May 6 10:43:33 2006 Ben Pfaff <blp@gnu.org>
304 Continue reforming procedure execution. In this phase, get rid of
305 the output code for SPLIT FILE groups in procedure.c, which really
306 shouldn't be doing any output. Move it into the individual
307 procedures instead. This also adds some flexibility.
309 * crosstabs.q (precalc): Call output_split_file_values().
311 * descriptives.c (calc_descriptives): Ditto.
313 * examine.q (run_examine): Ditto.
315 * frequencies.q (precalc): Ditto.
317 * oneway.q (run_oneway): Ditto.
319 * regression.q (run_regression): Ditto.
321 * t-test.q (calculate): Ditto.
323 Wed May 3 23:05:31 2006 Ben Pfaff <blp@gnu.org>
325 Continue reforming procedure execution. In this phase, get rid of
326 many global variables, consolidating procedure execution in
327 procedure.c. Encapsulate transformations in new "struct
328 trns_chain". Also, change implementation of N OF CASES, FILTER,
329 and PROCESS IF from special cases to transformations.
331 * aggregate.c (cmd_aggregate) Use discard_variables().
333 2006-04-28 Jason Stover <jhs@math.gcsu.edu>
335 * regression.q (regression_trns_resid_proc): Pass only the
336 variables used in the model to (*model->residual)().
338 * regression.q (regression_trns_pred_proc): Pass only the
339 variables used in the model to (*model->pred)().
341 2006-04-26 Jason Stover <jhs@math.gcsu.edu>
343 * regression.q: Added support for multiple transformations.
345 * regression.q (regression_trns_resid_proc): New function.
347 * regression.q (regression_trns_pred_proc): New function.
349 * regression.q (subcommand_save): Added support for saving
352 * regression.q (regression_trns_free): New function.
354 * regression.q (reg_get_name): New function.
356 * regression.q (reg_save_var): New function.
358 Tue Apr 25 13:18:56 2006 Ben Pfaff <blp@gnu.org>
360 * rank.q (parse_rank_function): Use SE instead of ME for parse
363 Tue Apr 25 13:16:28 2006 Ben Pfaff <blp@gnu.org>
365 * flip.c (flip_sink_write): Use snprintf() to simplify a bit of
368 2006-04-21 Jason Stover <jhs@math.gcsu.edu>
370 * regression.q (try_name): New function. (Partly copied from
371 try_name in descriptives.c.)
373 * regression.q (subcommand_save): Choose residual variable names
376 2006-04-20 Jason Stover <jhs@math.gcsu.edu>
378 * regression.q (cmd_regression): Moved call to subcommand_save()
379 outside multipass_procedure_with_splits().
381 * regression.q (regression_trns_proc): Fixed value counter n_vals
382 before calling *model->residual().
384 2006-04-19 Jason Stover <jhs@math.gcsu.edu>
386 * regression.q (regression_trns_proc): Fixed the look-up of the
389 2006-04-18 Jason Stover <jhs@math.gcsu.edu>
391 * regression.q (regression_trns_proc): Look up the residual
392 variable in the linear regression cache.
394 * regression.q (subcommand_save): Set the residual variable in the
395 linear regression cache.
397 2006-04-17 Jason Stover <jason@wonko.gcsu.edu>
399 * regression.q (regression_trns_proc): Accept case_idx as an int
400 to match the definition of trns_proc_func.
402 2006-04-17 Jason Stover <jhs@math.gcsu.edu>
404 * regression.q (regression_trns_proc): New function.
406 * regression.q (subcommand_save): Create variable residuals and
407 add a transformation to assign values to them. Also free the
408 linreg_cache if the SAVE command was not called. Removed the
409 casereading loop. Placed actual computation of residuals in
410 regression_trns_proc.
412 * regression.q (run_regression): Moved call to free
413 pspp_linreg_cache to subcommand_save.
415 Sat Apr 15 18:01:03 2006 Ben Pfaff <blp@gnu.org>
417 * examine.q (output_examine): Add casts to fix warnings.
419 2006-04-07 Jason Stover <jhs@math.gcsu.edu>
421 * regression.q (subcommand_save): New function.
423 2006-04-04 Jason Stover <jhs@math.gcsu.edu>
425 * regression.q: New function reg_has_categorical () to tell
426 whether a pspp_linreg_struct was made with any variables of type
429 * regression.q: (subcommand_export): Call
430 reg_print_categorical_encoding() only if the model uses any
431 categorical variables.
433 Mon Mar 27 16:00:42 2006 Ben Pfaff <blp@gnu.org>
435 * crosstabs.q: (output_pivot_table) Drop spurious space from
438 2006-03-15 Jason Stover <jhs@math.gcsu.edu>
440 * regression.q: Added custom syntax parser for VARIABLES subcommand
442 * regression.q: Moved most instructions for run_regression ()
443 inside the loop over dependent variables.
445 Thu Mar 2 08:40:33 WST 2006 John Darrington <john@darrington.wattle.id.au>
447 * Moved files from src directory