1 Sun Nov 5 08:31:42 WST 2006 John Darrington <john@darrington.wattle.id.au>
3 * t-test.q, oneway.q: Changed to use the new casefilter structure.
5 Sat Oct 14 16:52:28 2006 Ben Pfaff <blp@gnu.org>
7 * rank.q: (rank_sorted_casefile) Add some missing case_destroy()
8 calls to fix a memory leak.
10 Sun Oct 8 09:45:40 WST 2006 John Darrington <john@darrington.wattle.id.au>
12 * rank.q: Plugged a small memory leak which occurred under error
15 Sat Oct 7 11:06:01 WST 2006 John Darrington <john@darrington.wattle.id.au>
17 * rank.q: Implemented most of the RANK command.
19 2006-07-14 Jason Stover <jhs@math.gcsu.edu>
21 * regression.q (run_regression): New function to move knowledge of
22 pspp_linreg_cache out of math/coefficient.[ch].
24 Sat Jul 1 17:41:46 2006 Ben Pfaff <blp@gnu.org>
26 Fix bug #11612, "q2c documentation does not agree with code".
28 * examine.q: Audit use of q2c "+" prefixes that indicate that a
29 command may appear multiple times.
31 * frequencies.q: Ditto.
35 * regression.q: Ditto.
39 Fri Jun 23 14:18:22 2006 Ben Pfaff <blp@gnu.org>
41 Support long string variables on FREQUENCIES, as
42 an extension when in enhanced algorithms mode. For Greg Hunt
43 <greg@firmansyah.com>.
45 * frequencies.q: (struct freq) Change `v' member from union value
46 to union value *. Update all references.
47 (struct var_freqs) Add width, print members to represent effective
48 variable width and display format.
49 (calc) Copy entire long string value into the hash table.
50 (frq_custom_variables) Set new width, print members.
51 (hash_value_alpha) Get width from var_freqs.
52 (compare_value_alpha_a) Ditto.
53 (compare_freq_alpha_a) Ditto.
54 (compare_freq_alpha_d) Ditto.
55 (dump_full) Get display format from var_freqs.
56 (dump_condensed) Ditto.
58 Mon Jun 19 22:07:13 2006 Ben Pfaff <blp@gnu.org>
60 * frequencies.q: (dump_full) Only put the first MAX_SHORT_STRING
61 bytes of string variables into the output cells, seeing as we only
63 (dump_condensed) Ditto.
65 Mon Jun 19 21:52:05 2006 Ben Pfaff <blp@gnu.org>
67 Fixes a bug reported by Greg Hunt <greg@firmansyah.com>.
69 * frequencies.q: (hsh_hash_bytes) We only copy the first
70 MAX_SHORT_STRING bytes of string variables, so we must only
71 compare that many bytes, even if the string variable is longer.
72 (compare_value_alpha_a) Ditto.
73 (compare_freq_alpha_a) Ditto.
74 (compare_freq_alpha_d) Ditto.
76 2006-05-11 Jason Stover <jhs@math.gcsu.edu>
78 * regression.q: Adjusted code to account for cache->coeff being a
81 Sun May 7 18:31:25 2006 Ben Pfaff <blp@gnu.org>
85 * aggregate.c (cmd_aggregate): Free default_dict before replacing
88 Sun May 7 17:09:19 2006 Ben Pfaff <blp@gnu.org>
90 * flip.c (flip_file): Check return value of pool_fclose().
92 Sat May 6 16:00:13 2006 Ben Pfaff <blp@gnu.org>
94 Get rid of `char *c' member in union value, for cleanliness.
96 * aggregate.c: (union agr_argument) New union.
97 (struct agr_var) Change element type of arg[] from union value to
99 (parse_aggregate_functions) Change local variable types likewise.
101 * autorecode.c: (union arc_value) New union.
102 (struct arc_item) Change "from" from union value to union
104 (recode) Change local variable from union value to union
106 (autorecode_trns_proc) Ditto.
107 (compare_alpha_value) Ditto.
108 (hash_alpha_value) Ditto.
109 (compare_numeric_value) Ditto.
110 (hash_numeric_value) Ditto.
111 (autorecode_proc_func) Ditto.
113 Sat May 6 10:43:33 2006 Ben Pfaff <blp@gnu.org>
115 Continue reforming procedure execution. In this phase, get rid of
116 the output code for SPLIT FILE groups in procedure.c, which really
117 shouldn't be doing any output. Move it into the individual
118 procedures instead. This also adds some flexibility.
120 * crosstabs.q (precalc): Call output_split_file_values().
122 * descriptives.c (calc_descriptives): Ditto.
124 * examine.q (run_examine): Ditto.
126 * frequencies.q (precalc): Ditto.
128 * oneway.q (run_oneway): Ditto.
130 * regression.q (run_regression): Ditto.
132 * t-test.q (calculate): Ditto.
134 Wed May 3 23:05:31 2006 Ben Pfaff <blp@gnu.org>
136 Continue reforming procedure execution. In this phase, get rid of
137 many global variables, consolidating procedure execution in
138 procedure.c. Encapsulate transformations in new "struct
139 trns_chain". Also, change implementation of N OF CASES, FILTER,
140 and PROCESS IF from special cases to transformations.
142 * aggregate.c (cmd_aggregate) Use discard_variables().
144 2006-04-28 Jason Stover <jhs@math.gcsu.edu>
146 * regression.q (regression_trns_resid_proc): Pass only the
147 variables used in the model to (*model->residual)().
149 * regression.q (regression_trns_pred_proc): Pass only the
150 variables used in the model to (*model->pred)().
152 2006-04-26 Jason Stover <jhs@math.gcsu.edu>
154 * regression.q: Added support for multiple transformations.
156 * regression.q (regression_trns_resid_proc): New function.
158 * regression.q (regression_trns_pred_proc): New function.
160 * regression.q (subcommand_save): Added support for saving
163 * regression.q (regression_trns_free): New function.
165 * regression.q (reg_get_name): New function.
167 * regression.q (reg_save_var): New function.
169 Tue Apr 25 13:18:56 2006 Ben Pfaff <blp@gnu.org>
171 * rank.q (parse_rank_function): Use SE instead of ME for parse
174 Tue Apr 25 13:16:28 2006 Ben Pfaff <blp@gnu.org>
176 * flip.c (flip_sink_write): Use snprintf() to simplify a bit of
179 2006-04-21 Jason Stover <jhs@math.gcsu.edu>
181 * regression.q (try_name): New function. (Partly copied from
182 try_name in descriptives.c.)
184 * regression.q (subcommand_save): Choose residual variable names
187 2006-04-20 Jason Stover <jhs@math.gcsu.edu>
189 * regression.q (cmd_regression): Moved call to subcommand_save()
190 outside multipass_procedure_with_splits().
192 * regression.q (regression_trns_proc): Fixed value counter n_vals
193 before calling *model->residual().
195 2006-04-19 Jason Stover <jhs@math.gcsu.edu>
197 * regression.q (regression_trns_proc): Fixed the look-up of the
200 2006-04-18 Jason Stover <jhs@math.gcsu.edu>
202 * regression.q (regression_trns_proc): Look up the residual
203 variable in the linear regression cache.
205 * regression.q (subcommand_save): Set the residual variable in the
206 linear regression cache.
208 2006-04-17 Jason Stover <jason@wonko.gcsu.edu>
210 * regression.q (regression_trns_proc): Accept case_idx as an int
211 to match the definition of trns_proc_func.
213 2006-04-17 Jason Stover <jhs@math.gcsu.edu>
215 * regression.q (regression_trns_proc): New function.
217 * regression.q (subcommand_save): Create variable residuals and
218 add a transformation to assign values to them. Also free the
219 linreg_cache if the SAVE command was not called. Removed the
220 casereading loop. Placed actual computation of residuals in
221 regression_trns_proc.
223 * regression.q (run_regression): Moved call to free
224 pspp_linreg_cache to subcommand_save.
226 Sat Apr 15 18:01:03 2006 Ben Pfaff <blp@gnu.org>
228 * examine.q (output_examine): Add casts to fix warnings.
230 2006-04-07 Jason Stover <jhs@math.gcsu.edu>
232 * regression.q (subcommand_save): New function.
234 2006-04-04 Jason Stover <jhs@math.gcsu.edu>
236 * regression.q: New function reg_has_categorical () to tell
237 whether a pspp_linreg_struct was made with any variables of type
240 * regression.q: (subcommand_export): Call
241 reg_print_categorical_encoding() only if the model uses any
242 categorical variables.
244 Mon Mar 27 16:00:42 2006 Ben Pfaff <blp@gnu.org>
246 * crosstabs.q: (output_pivot_table) Drop spurious space from
249 2006-03-15 Jason Stover <jhs@math.gcsu.edu>
251 * regression.q: Added custom syntax parser for VARIABLES subcommand
253 * regression.q: Moved most instructions for run_regression ()
254 inside the loop over dependent variables.
256 Thu Mar 2 08:40:33 WST 2006 John Darrington <john@darrington.wattle.id.au>
258 * Moved files from src directory