1 Sun Oct 8 09:45:40 WST 2006 John Darrington <john@darrington.wattle.id.au>
3 * rank.q: Plugged a small memory leak which occurred under error
6 Sat Oct 7 11:06:01 WST 2006 John Darrington <john@darrington.wattle.id.au>
8 * rank.q: Implemented most of the RANK command.
10 2006-07-14 Jason Stover <jhs@math.gcsu.edu>
12 * regression.q (run_regression): New function to move knowledge of
13 pspp_linreg_cache out of math/coefficient.[ch].
15 Sat Jul 1 17:41:46 2006 Ben Pfaff <blp@gnu.org>
17 Fix bug #11612, "q2c documentation does not agree with code".
19 * examine.q: Audit use of q2c "+" prefixes that indicate that a
20 command may appear multiple times.
22 * frequencies.q: Ditto.
26 * regression.q: Ditto.
30 Fri Jun 23 14:18:22 2006 Ben Pfaff <blp@gnu.org>
32 Support long string variables on FREQUENCIES, as
33 an extension when in enhanced algorithms mode. For Greg Hunt
34 <greg@firmansyah.com>.
36 * frequencies.q: (struct freq) Change `v' member from union value
37 to union value *. Update all references.
38 (struct var_freqs) Add width, print members to represent effective
39 variable width and display format.
40 (calc) Copy entire long string value into the hash table.
41 (frq_custom_variables) Set new width, print members.
42 (hash_value_alpha) Get width from var_freqs.
43 (compare_value_alpha_a) Ditto.
44 (compare_freq_alpha_a) Ditto.
45 (compare_freq_alpha_d) Ditto.
46 (dump_full) Get display format from var_freqs.
47 (dump_condensed) Ditto.
49 Mon Jun 19 22:07:13 2006 Ben Pfaff <blp@gnu.org>
51 * frequencies.q: (dump_full) Only put the first MAX_SHORT_STRING
52 bytes of string variables into the output cells, seeing as we only
54 (dump_condensed) Ditto.
56 Mon Jun 19 21:52:05 2006 Ben Pfaff <blp@gnu.org>
58 Fixes a bug reported by Greg Hunt <greg@firmansyah.com>.
60 * frequencies.q: (hsh_hash_bytes) We only copy the first
61 MAX_SHORT_STRING bytes of string variables, so we must only
62 compare that many bytes, even if the string variable is longer.
63 (compare_value_alpha_a) Ditto.
64 (compare_freq_alpha_a) Ditto.
65 (compare_freq_alpha_d) Ditto.
67 2006-05-11 Jason Stover <jhs@math.gcsu.edu>
69 * regression.q: Adjusted code to account for cache->coeff being a
72 Sun May 7 18:31:25 2006 Ben Pfaff <blp@gnu.org>
76 * aggregate.c (cmd_aggregate): Free default_dict before replacing
79 Sun May 7 17:09:19 2006 Ben Pfaff <blp@gnu.org>
81 * flip.c (flip_file): Check return value of pool_fclose().
83 Sat May 6 16:00:13 2006 Ben Pfaff <blp@gnu.org>
85 Get rid of `char *c' member in union value, for cleanliness.
87 * aggregate.c: (union agr_argument) New union.
88 (struct agr_var) Change element type of arg[] from union value to
90 (parse_aggregate_functions) Change local variable types likewise.
92 * autorecode.c: (union arc_value) New union.
93 (struct arc_item) Change "from" from union value to union
95 (recode) Change local variable from union value to union
97 (autorecode_trns_proc) Ditto.
98 (compare_alpha_value) Ditto.
99 (hash_alpha_value) Ditto.
100 (compare_numeric_value) Ditto.
101 (hash_numeric_value) Ditto.
102 (autorecode_proc_func) Ditto.
104 Sat May 6 10:43:33 2006 Ben Pfaff <blp@gnu.org>
106 Continue reforming procedure execution. In this phase, get rid of
107 the output code for SPLIT FILE groups in procedure.c, which really
108 shouldn't be doing any output. Move it into the individual
109 procedures instead. This also adds some flexibility.
111 * crosstabs.q (precalc): Call output_split_file_values().
113 * descriptives.c (calc_descriptives): Ditto.
115 * examine.q (run_examine): Ditto.
117 * frequencies.q (precalc): Ditto.
119 * oneway.q (run_oneway): Ditto.
121 * regression.q (run_regression): Ditto.
123 * t-test.q (calculate): Ditto.
125 Wed May 3 23:05:31 2006 Ben Pfaff <blp@gnu.org>
127 Continue reforming procedure execution. In this phase, get rid of
128 many global variables, consolidating procedure execution in
129 procedure.c. Encapsulate transformations in new "struct
130 trns_chain". Also, change implementation of N OF CASES, FILTER,
131 and PROCESS IF from special cases to transformations.
133 * aggregate.c (cmd_aggregate) Use discard_variables().
135 2006-04-28 Jason Stover <jhs@math.gcsu.edu>
137 * regression.q (regression_trns_resid_proc): Pass only the
138 variables used in the model to (*model->residual)().
140 * regression.q (regression_trns_pred_proc): Pass only the
141 variables used in the model to (*model->pred)().
143 2006-04-26 Jason Stover <jhs@math.gcsu.edu>
145 * regression.q: Added support for multiple transformations.
147 * regression.q (regression_trns_resid_proc): New function.
149 * regression.q (regression_trns_pred_proc): New function.
151 * regression.q (subcommand_save): Added support for saving
154 * regression.q (regression_trns_free): New function.
156 * regression.q (reg_get_name): New function.
158 * regression.q (reg_save_var): New function.
160 Tue Apr 25 13:18:56 2006 Ben Pfaff <blp@gnu.org>
162 * rank.q (parse_rank_function): Use SE instead of ME for parse
165 Tue Apr 25 13:16:28 2006 Ben Pfaff <blp@gnu.org>
167 * flip.c (flip_sink_write): Use snprintf() to simplify a bit of
170 2006-04-21 Jason Stover <jhs@math.gcsu.edu>
172 * regression.q (try_name): New function. (Partly copied from
173 try_name in descriptives.c.)
175 * regression.q (subcommand_save): Choose residual variable names
178 2006-04-20 Jason Stover <jhs@math.gcsu.edu>
180 * regression.q (cmd_regression): Moved call to subcommand_save()
181 outside multipass_procedure_with_splits().
183 * regression.q (regression_trns_proc): Fixed value counter n_vals
184 before calling *model->residual().
186 2006-04-19 Jason Stover <jhs@math.gcsu.edu>
188 * regression.q (regression_trns_proc): Fixed the look-up of the
191 2006-04-18 Jason Stover <jhs@math.gcsu.edu>
193 * regression.q (regression_trns_proc): Look up the residual
194 variable in the linear regression cache.
196 * regression.q (subcommand_save): Set the residual variable in the
197 linear regression cache.
199 2006-04-17 Jason Stover <jason@wonko.gcsu.edu>
201 * regression.q (regression_trns_proc): Accept case_idx as an int
202 to match the definition of trns_proc_func.
204 2006-04-17 Jason Stover <jhs@math.gcsu.edu>
206 * regression.q (regression_trns_proc): New function.
208 * regression.q (subcommand_save): Create variable residuals and
209 add a transformation to assign values to them. Also free the
210 linreg_cache if the SAVE command was not called. Removed the
211 casereading loop. Placed actual computation of residuals in
212 regression_trns_proc.
214 * regression.q (run_regression): Moved call to free
215 pspp_linreg_cache to subcommand_save.
217 Sat Apr 15 18:01:03 2006 Ben Pfaff <blp@gnu.org>
219 * examine.q (output_examine): Add casts to fix warnings.
221 2006-04-07 Jason Stover <jhs@math.gcsu.edu>
223 * regression.q (subcommand_save): New function.
225 2006-04-04 Jason Stover <jhs@math.gcsu.edu>
227 * regression.q: New function reg_has_categorical () to tell
228 whether a pspp_linreg_struct was made with any variables of type
231 * regression.q: (subcommand_export): Call
232 reg_print_categorical_encoding() only if the model uses any
233 categorical variables.
235 Mon Mar 27 16:00:42 2006 Ben Pfaff <blp@gnu.org>
237 * crosstabs.q: (output_pivot_table) Drop spurious space from
240 2006-03-15 Jason Stover <jhs@math.gcsu.edu>
242 * regression.q: Added custom syntax parser for VARIABLES subcommand
244 * regression.q: Moved most instructions for run_regression ()
245 inside the loop over dependent variables.
247 Thu Mar 2 08:40:33 WST 2006 John Darrington <john@darrington.wattle.id.au>
249 * Moved files from src directory