1 2008-02-06 John Darrington <john@darrington.wattle.id.au>
3 * get-data.c: Add a /BSIZE subcommand to PSQL reader.
5 2008-02-02 John Darrington <john@darrington.wattle.id.au>
7 * get-data.c (cmd_get_data): Support PSQL type.
9 2007-12-07 Ben Pfaff <blp@gnu.org>
13 * data-parser.c (data_parser_make_active_file): Fix case count
14 argument to casereader_create_sequential, which fixes data reading
16 Provided by John Darrington.
18 2007-12-04 Ben Pfaff <blp@gnu.org>
20 Move DATA LIST parsing into generic infrastructure, and generalize
21 it slightly. Then, use the same infrastructure to implement GET
24 * data-parser.c: New file.
26 * data-parser.h: New file.
28 * data-list.c (struct dls_var_spec): Removed.
29 (ll_to_dls_var_spec): Removed.
30 (enum dls_type): Removed.
31 (struct data_list_pgm): Rename struct data_list_trns. Remove
32 pool, specs, type, record_cnt, delims, skip_records, value_cnt
33 members. Add new `parser' member.
34 (cmd_data_list): Use data-parser infrastructure.
37 (dump_fixed_table): Removed.
38 (dump_free_table): Removed.
40 (read_from_data_list): Removed.
41 (read_from_data_list_fixed): Removed.
42 (read_from_data_list_free): Removed.
43 (read_from_data_list_list): Removed.
44 (data_list_trns_free): Rename arguments for clarity.
45 (data_list_trns_proc): Ditto.
46 (data_list_casereader_read): Removed.
47 (data_list_casereader_destroy): Removed.
48 (data_list_casereader_class): Removed.
50 * get-data.c (cmd_get_data): Support TXT type.
51 (set_type): New function.
52 (parse_get_txt): New function.
54 2007-12-04 Ben Pfaff <blp@gnu.org>
56 * placement-parser.c (parse_column): New function.
57 (parse_column_range): Add `base' argument. Update all callers.
59 2007-12-04 Ben Pfaff <blp@gnu.org>
61 Make GET DATA a separate command, instead of something invoked
64 * automake.mk: Remove get-data.h from sources.
66 * get-data.h: Removed.
68 * get-data.c (parse_get_data_command): Rename cmd_get_data.
70 * get.c (parse_read_command): No longer any need to check for DATA
73 2007-12-04 Ben Pfaff <blp@gnu.org>
75 * src/language/data-io/data-reader.c (struct dfm_reader): New
76 `file_size' member to support dfm_get_percent_read.
77 (dfm_open_reader): Initialize file_size.
78 (dfm_get_percent_read): New function.
80 2007-11-08 Ben Pfaff <blp@gnu.org>
82 Patch #6256: add support for binary, 360 file formats. Reviewed
85 * data-reader.c (struct dfm_reader): New member `block_left'.
86 (dfm_open_reader): Initialize block_left. For FH_MODE_TEXT, open
87 the file in text mode.
88 (read_error): New function.
89 (partial_record): New function.
90 (try_to_read_fully): New function.
91 (enum descriptor_type): New enum.
92 (read_descriptor_word): New function.
93 (corrupt_size): New function.
94 (read_size): New function.
95 (read_file_record): Implement new modes.
96 (read_record): Now take care of tracking line numbers here.
97 (dfm_reader_get_legacy_encoding): New function.
99 * data-writer.c (dfm_put_record): Implement new modes.
100 (dfm_writer_get_legacy_encoding): New function.
102 * file-handle.q: Parse new formats.
103 (cmd_file_handle): Set up new formats.
105 * print.c (struct print_trns): New member `encoding'.
106 (internal_cmd_print): Set encoding.
107 (print_trns_proc): Recode output data if necessary.
108 (flush_records): Recode leader byte.
110 2007-11-03 Ben Pfaff <blp@gnu.org>
112 Allow output files to overwrite input files (bug #21280).
114 * data-list.c (cmd_data_list): Manage file handle reference
117 * data-reader.c (struct dfm_reader): Add `lock' member.
118 (dfm_close_reader): Simplify, as reference counting is now
119 separate from locking.
120 (dfm_open_reader): Lock file.
122 * data-writer.c (struct dfm_writer): Add fh_lock, replace_file
124 (dfm_open_writer): Lock file and prepare for replacement.
125 (dfm_close_writer): Unlock file and replace it.
127 * file-handle.q (cmd_close_file_handle): Use fh_unname.
128 (fh_parse): Don't distinguish existing handles for a given file
129 name from new ones. Manage file handle reference counts.
131 * get.c (parse_read_command): Manage file handle reference counts.
132 (parse_write_command): Ditto.
133 (mtf_close_all_files): Ditto.
135 * inpt-pgm.c (cmd_reread): Manage file handle reference counts.
137 * print-space.c (cmd_print_space): Manage file handle reference
140 * print.c (internal_cmd_print): Manage file handle reference
143 2007-11-03 John Darrington <john@darrington.wattle.id.au>
145 * get.c: Add GET DATA command variant.
147 * get-data.c get-data.h (new files): Added support for
148 GET DATA /TYPE='gnm' command.
150 2007-09-23 Ben Pfaff <blp@gnu.org>
152 Bug #21111. Reviewed by John Darrington.
154 * data-list.c (data_list_trns_proc): Properly set retval when END
155 subcommand is in use.
156 (cmd_data_list): Don't allow END subcommand to be used with DATA
159 2007-09-12 Ben Pfaff <blp@gnu.org>
161 * get.c (get_translate_case): Change input case parameter from
162 const struct ccase * to struct ccase *, to match change in
163 casereader and casewriter translators. Destroy input case, to fix
166 2007-08-12 Ben Pfaff <blp@gnu.org>
168 * get.c (parse_read_command): Compact the values in the target
169 dictionary, to save space.
171 2007-08-12 Ben Pfaff <blp@gnu.org>
173 * get.c (struct case_map): Move into new file src/data/case-map.c.
174 (start_case_map): Ditto, and rename case_map_prepare_dict.
175 (finish_case_map): Ditto, and rename case_map_from_dict.
176 (map_case): Ditto, and rename case_map_execute.
177 (destroy_case_map): Ditto, and rename case_map_destroy.
178 (case_map_get_value_cnt): Ditto.
180 2007-08-12 Ben Pfaff <blp@gnu.org>
182 * get.c (case_map_get_value_cnt): New function.
184 2007-07-25 Ben Pfaff <blp@gnu.org>
187 * data-list.c (read_from_data_list_fixed): Handle multi-record
190 2007-07-11 Ben Pfaff <blp@gnu.org>
192 * get.c (map_case): Create destination case instead of leaving it
193 undefined. Fixes bug #20285.
194 Reviewed by John Darrington.
196 2007-06-06 Ben Pfaff <blp@gnu.org>
198 * get.c: Essentially rewrite MATCH FILES to support FIRST and
201 2007-06-06 Ben Pfaff <blp@gnu.org>
203 Adapt case sources, sinks, and clients of procedure code to the
206 * data-list.c: Make DATA LIST into a casereader.
208 * get.c: Change GET, IMPORT, SAVE, EXPORT to use casereaders,
211 * inpt-pgm.c: Use caseinit code. Turn INPUT PROGRAM into a
214 * list.q: Adapt to new procedure code.
216 2007-05-06 Ben Pfaff <blp@gnu.org>
218 Abstract the documents within a dictionary a little better.
219 Thanks to John Darrington for suggestion, initial version, and
222 * get.c (mtf_merge_dictionary): Simplify creating merged document.
224 * sys-file-info.c (display_documents): Use new
225 dict_get_document_line_cnt and dict_get_document_line functions.
227 Thu Feb 1 16:56:02 2007 Ben Pfaff <blp@gnu.org>
229 * file-handle.q (fh_parse): Update to new fh_create_file
232 Sat Dec 16 22:16:18 2006 Ben Pfaff <blp@gnu.org>
234 Make it possible to pull cases from the active file with a
235 function call, instead of requiring indirection through a callback
238 * automake.mk: Removed matrix-data.c.
240 * matrix-data.c: Removed.
242 * data-list.c (data_list_source_read): Conform with new
243 case_source_class interface.
244 (data_list_source_destroy): Ditto.
246 * get.c (case_reader_source_class): Ditto.
247 (case_reader_source_destroy): Ditto.
248 (parse_output_proc): Take advantage of new procedure interface.
249 (output_proc): Removed.
250 (struct mtf_file): Add "struct ccase *" member to allow use of new
252 (cmd_match_files): Take advantage of new procedure interface.
253 (mtf_processing_finish): Removed.
254 (mtf_read_nonactive_records): Renamed mtf_read_records. Now reads
255 from every file, without any exception for the active file.
256 (mtf_compare_BY_values): Simplify for new interface.
257 (mtf_processing): Simplify for new interface.
259 * inpt-pgm.c (is_valid_state): New function.
260 (input_program_source_read): Conform with new case_source_class
262 (input_program_source_destroy): Ditto.
263 (end_case_trns_proc): Now just needs to return TRNS_END_CASE.
265 Sat Dec 9 18:43:34 2006 Ben Pfaff <blp@gnu.org>
267 * list.q (cmd_list): Use new var_create, var_destroy functions.
269 Thu Nov 30 21:51:58 2006 Ben Pfaff <blp@gnu.org>
271 * inpt-pgm.c (cmd_reread): Always return error code upon detecting
272 syntax error. Fixes bug #18419. Thanks to John Darrington for
275 Sun Nov 19 09:17:45 2006 Ben Pfaff <blp@gnu.org>
277 * data-list.c (parse_free): Follow documented (but odd) rule that
278 N format is treated as F format for free-field input.
280 * data-reader.c (read_file_record): Drop new-line character from
281 input text lines. This is symmetrical with the recently changed
282 dfm_put_record semantics.
284 Thu Nov 2 20:56:03 2006 Ben Pfaff <blp@gnu.org>
286 Implement SKIP keyword on DATA LIST. Fixes bug #17099.
288 * data-list.c: (struct data_list_pgm) Add `skip_records' members.
289 (cmd_data_list) Set skip_records based on user input.
290 (data_list_source_read) Skip records requested by user.
292 Tue Oct 31 20:04:06 2006 Ben Pfaff <blp@gnu.org>
294 * placement-parser.c: (PRS_TYPE_T) Now that struct fmt_spec uses
295 an enum fmt_type for its type member, we can't depend on the
296 ability to put negative values into that member as out-of-band
297 values, because enum fmt_type might be an unsigned type. So use
298 values around SCHAR_MAX instead, because we know that SCHAR_MAX
299 will fit into any type, signed or unsigned, and there aren't
300 nearly that many format types.
301 (parse_var_placements) Add for_input parameter to specify whether
302 we're parsing input or output formats. Update all callers.
303 (fixed_parse_columns) Ditto.
304 (fixed_parse_fortran) Ditto.
306 Tue Oct 31 18:21:48 2006 Ben Pfaff <blp@gnu.org>
308 * print-space.c (print_space_trns_proc): Let dfm_put_record add
309 the new-line character, to match dfm_put_record change below.
311 Sat Oct 28 11:57:19 2006 Ben Pfaff <blp@gnu.org>
313 * data-writer.c (struct dfm_writer): Removed `bounce' member, and
314 all references to it.
315 (dfm_put_record) Change semantics so that it adds formatting
316 itself, such as new-line characters, instead of putting that
317 responsibility on the caller. Also, pad binary records with
318 spaces instead of zeros, for compatibility.
320 * print.c (struct prt_out_spec) New member `sysmis_as_spaces'.
321 (struct print_trns) Remove `omit_new_lines' and all references,
322 since dfm_put_record() is taking care of that. Add
324 (internal_cmd_print) Allow an empty set of data to print. Set
326 (parse_specs) Allow an empty set of data to print.
327 (parse_variable_argument) Only add space with PRINT command. Set
329 (print_trns_proc) Indent records if include_prefix is set, for
330 compatibility. Output SYSMIS as spaces if sysmis_as_spaces is
331 set. Put "1" in first column if PRINT EJECT is used with an
332 external output file.
333 (flush_records) Ditto.
335 Sat Oct 28 16:19:57 WST 2006 John Darrington <john@darrington.wattle.id.au>
337 * data-reader.c: Eliminated references to extern variable getl_buf
339 Sat Aug 5 08:25:07 2006 Ben Pfaff <blp@gnu.org>
341 Fix bug #17329 in REREAD parsing, reported by John Darrington.
343 * inpt-pgm.c (cmd_reread): Fix file handle parsing.
345 Mon Jul 31 10:32:31 2006 Ben Pfaff <blp@gnu.org>
347 * print.c (parse_specs): Allow a comma between specifications.
349 Sun Jul 16 19:57:10 2006 Ben Pfaff <blp@gnu.org>
351 * automake.mk: (src_language_data_io_libdata_io_a_SOURCE) Add
352 print-space.c, placement-parser.c, placement-parser.h.
354 * data-list.c: Basically rewrote the whole thing. Broke out a lot
355 of code into placement-parser.c. Code is much cleaner now.
357 * placement-parser.c: New file.
359 * placement-parser.h: New file.
361 * print.c: Basically rewrote the whole thing. Broke out PRINT
362 SPACE into print-space.c. Code is much cleaner now.
364 * print-space.c: New file.
366 Sat Jul 1 17:39:40 2006 Ben Pfaff <blp@gnu.org>
368 Fix bug #11612, "q2c documentation does not agree with code".
370 * list.q: Audit use of q2c "+" prefixes that indicate that a
371 command may appear multiple times.
373 Sat Jul 1 20:44:22 2006 Ben Pfaff <blp@gnu.org>
375 Fix bug #15786: System File Creation crashes if directoy is
378 * get.c (parse_write_command): Check that the any_writer open
381 Tue Jun 27 22:44:28 2006 Ben Pfaff <blp@gnu.org>
383 Fix regression in command name completion reported by John
384 Darrington. Now completion is again state-dependent and occurs
385 only on the first line of a command.
387 * inpt-pgm.c: (cmd_input_program) Reading of first token in
388 command moved into cmd_parse.
390 Fri Jun 9 13:56:00 2006 Ben Pfaff <blp@gnu.org>
392 Reform string library.
394 * matrix-data.c (context): Use dynamic string.
395 (another_token) Deal with changed dfm_get_record() interface.
399 * data-list.c (struct data_list_pgm) Delete delims, delim_cnt
400 members, replacing them by struct string delims. Update all
401 references to use struct string functions.
402 (cut_field) Change interface to avoid needing "end_blank", by
403 getting the data-reader to remember that state for us. Change
404 internals to use substring. Update both callers.
406 * data-reader.c (read_file_record): Use ds_read_stream().
407 (dfm_get_record) Change interface to return substring. Updated
409 (dfm_expand_tabs) Use ds_find_char(). Now maintain position
410 relative to end-of-line. Use ds_swap().
411 (dfm_reread_record) Don't limit position by line length.
412 (dfm_column_start) Make parameter const.
413 (dfm_columns_past_end) New function.
414 (dfm_get_column) New function.
416 Thu May 25 18:26:26 WST 2006 John Darrington <john@darrington.wattle.id.au>
418 * print.c (print_trns_free): Made the code agree with the comment, by
419 not freeing PRT. Has the side effect that the command no longer
420 crashes on invalid syntax.
422 Tue May 9 20:55:46 2006 Ben Pfaff <blp@gnu.org>
424 * get.c (cmd_match_files): Fix memory leak replacing default_dict.
426 Sat May 6 22:25:09 2006 Ben Pfaff <blp@gnu.org>
430 * list.q (write_fallback_headers): (write_fallback_headers)
431 Properly record width of leader and pass it to write_varname().
433 Sat May 6 19:03:13 2006 Ben Pfaff <blp@gnu.org>
435 * get.c: (mtf_merge_dictionary) Fix value label memory leak.
437 Sat May 6 13:51:16 2006 Ben Pfaff <blp@gnu.org>
439 Use a casefile, instead of a case sink, for MATCH FILES output.
440 It's more straightforward, although it has the same effect.
442 * get.c: (struct mtf_proc) Replace `sink' case sink member by
443 `output' casefile member.
444 (cmd_match_files) Work with casefile instead of sink.
445 (mtf_processing) Add case to casefile instead of sink.
447 Sat May 6 10:43:07 2006 Ben Pfaff <blp@gnu.org>
449 Continue reforming procedure execution. In this phase, get rid of
450 the output code for SPLIT FILE groups in procedure.c, which really
451 shouldn't be doing any output. Move it into the individual
452 procedures instead. This also adds some flexibility.
454 * list.q (write_all_headers): Call output_split_file_values().
456 Wed May 3 23:00:17 2006 Ben Pfaff <blp@gnu.org>
458 Continue reforming procedure execution. In this phase, get rid of
459 many global variables, consolidating procedure execution in
460 procedure.c. Encapsulate transformations in new "struct
461 trns_chain". Also, change implementation of N OF CASES, FILTER,
462 and PROCESS IF from special cases to transformations.
464 * data-list.c: (data_list_trns_proc) Return TRNS_END_FILE at end
465 of file. (Why didn't we do this before?)
466 (cmd_match_files) Direct procedure output to null sink.
467 Use discard_variables() instead of indirect version.
469 * inpt-pgm.c: Use transformation chain.
470 (struct input_program_pgm) Add trns_chain member.
471 (cmd_input_program) Initialize trns_chain member and capture
472 transformations with proc_capture_transformations().
473 (input_program_source_read) Use trns_chain_execute().
474 (destroy_input_program) Destroy input chain.
476 Tue May 2 10:39:56 WST 2006 John Darrington <john@darrington.wattle.id.au>
478 * list.q Changed from using fixed length char buffers to struct
479 string so that any length variables can be used.
481 Mon May 1 18:21:19 2006 Ben Pfaff <blp@gnu.org>
483 Further clean up the CMD_* command result codes.
485 * (enum cmd_result_extensions) New. Add CMD_END_INPUT_PROGRAM and
486 CMD_END_CASE result codes.
487 (struct input_program_pgm) Added case_nr, write_case, wc_data
488 members for use by END CASE transformation.
489 (emit_END_CASE) New function.
490 (cmd_input_program) Interpret CMD_END_CASE by outputting an END
491 CASE transformation. If none is output by the input program
492 itself, add one automatically at the end. Change lack of
493 variables from warning to error.
494 (cmd_end_input_program) Return CMD_END_INPUT_PROGRAM instead of
496 (input_program_source_read) No longer any need to special-case END
497 CASE. Handle TRNS_DROP_CASE properly. Initialize new members in
498 inp for use by END CASE transformation.
499 (destroy_input_program) New function.
500 (input_program_source_destroy) Just call destroy_input_program().
501 (cmd_end_case) Just return CMD_END_CASE.
502 (end_case_trns_proc) No longer a stub handled by
503 input_program_source_read(). Actually output the case and
504 increment the case number.
506 Mon May 1 16:06:30 2006 Ben Pfaff <blp@gnu.org>
508 Remove vestiges of REPEATING DATA support.
510 * data-list.c: (struct rpd_num_or_var) Removed.
511 (struct repeating_data_trns) Removed.
512 (cmd_repeating_data) Removed.
513 (find_variable_input_spec) Removed.
514 (parse_num_or_var) Removed.
515 (parse_repeating_data) Removed.
516 (realize_value) Removed.
517 (struct rpd_parse_info) Removed.
518 (rpd_parse_record) Removed.
519 (repeating_data_trns_proc) Removed.
520 (repeating_data_trns_free) Removed.
521 (repeating_data_set_write_case) Removed.
524 * inpt-pgm.c: (input_program_source_read) Don't deal with
527 * data-list.h: Removed.
529 * automake.mk (src_language_data_io_libdata_io_a_SOURCES): Removed
532 Mon May 1 15:58:28 2006 Ben Pfaff <blp@gnu.org>
534 Remove vestiges of FILE TYPE support.
536 * data-list.c: (cmd_data_list) Don't check for FILE TYPE.
537 (cmd_repeating_data) Ditto.
539 * automake.mk (src_language_data_io_libdata_io_a_SOURCES): Remove
540 file-type.c, file-type.h.
542 * file-type.c: Removed.
544 * file-type.h: Removed.
546 Wed Apr 26 13:16:28 2006 Ben Pfaff <blp@gnu.org>
548 Improve the way we handle the various parsing "states". Until now
549 we've hard-coded the state transitions in the command definition
550 file, but that's error-prone and, worse, it's redundant--we can
551 figure out what state we're in anyhow. We can cleanly handle
552 INPUT PROGRAM and FILE TYPE with a nested command-processing loop.
554 * data-list.c: (cmd_data_list) Use in_file_type() or
555 in_input_program() in place of case_source_is_class() or
556 case_source_is_complex().
558 * file-type.c: NB: Not really fixed except minimally to compile,
559 because it doesn't work anyway.
560 (in_file_type) New function.
561 (cmd_record_type) No need to check that we're in FILE TYPE.
562 (cmd_end_file_type) Ditto.
563 (var file_type_source_class) Make static.
565 * get.c: (cmd_match_files) Check vfm_source instead of pgm_state.
567 * inpt-pgm.c: (in_input_program) New function.
568 (cmd_input_program) Rewrite to include nested command processing
570 (cmd_end_input_program) Just return CMD_END_SUBLOOP.
571 (var input_program_source_class) Make static.
572 (cmd_end_case) No need to check that we're in INPUT PROGRAM.
573 (cmd_end_file) Ditto.
575 * automake.mk (src_language_data_io_libdata_io_a_SOURCES): Add
576 file-type.h, inpt-pgm.h.
578 * file-type.h: New file.
580 * inpt-pgm.h: New file.
582 Tue Apr 25 13:11:55 2006 Ben Pfaff <blp@gnu.org>
584 * print.c: Don't special-case MS-DOS line terminators.
585 (macro LINE_END_WIDTH) Removed.
586 (alloc_line) Line ends are 1 byte.
587 (print_trns_proc) Just output \n for line end.
589 Sun Apr 23 22:05:58 2006 Ben Pfaff <blp@gnu.org>
591 Continue reforming error message support. In this phase, get rid
592 of message "titles" and put the message text in `struct error'.
593 Now `struct error' encapsulates a message more properly.
595 * data-list.c: (macro RPD_ERR) Removed.
596 (rpd_msg) New function. Updated all references to tmsg() to call
597 this function instead.
599 Sat Apr 15 19:38:13 2006 Ben Pfaff <blp@gnu.org>
601 Remove last users of struct file_ext so we can get rid of it
604 * data-reader.c: (struct dfm_reader) Change file member from
605 struct file_ext to FILE *. Updated all references.
606 (dfm_close_reader) Close file with fn_close() instead of
607 fn_close_ext(). Also, make a copy of the file name from the file
608 handle before closing it, because we can't extract it after we
610 (dfm_open_reader) Open file with fn_open() instead of
613 * data-writer.c: (struct dfm_writer) Change file member
614 struct file_ext to FILE *. Updated all references.
615 (dfm_close_writer) Close file with fn_close() instead of
616 fn_close_ext(). Also, make a copy of the file name from the file
617 handle before closing it, because we can't extract it after we
619 (dfm_open_writer) Open file with fn_open() instead of
622 Sat Apr 15 18:00:32 2006 Ben Pfaff <blp@gnu.org>
624 * data-list.c: Add prototype to suppress warning for
625 cmd_repeating_data().
627 Thu Mar 2 08:40:33 WST 2006 John Darrington <john@darrington.wattle.id.au>
629 * Moved files from src directory