--- /dev/null
+/* PSPP - a program for statistical analysis.
+ Copyright (C) 1997-9, 2000, 2006, 2009, 2010, 2011, 2012 Free Software Foundation, Inc.
+
+ This program is free software: you can redistribute it and/or modify
+ it under the terms of the GNU General Public License as published by
+ the Free Software Foundation, either version 3 of the License, or
+ (at your option) any later version.
+
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ GNU General Public License for more details.
+
+ You should have received a copy of the GNU General Public License
+ along with this program. If not, see <http://www.gnu.org/licenses/>. */
+
+#include <config.h>
+
+#include <stdlib.h>
+#include <uniwidth.h>
+
+#include "data/case.h"
+#include "data/dataset.h"
+#include "data/data-out.h"
+#include "data/format.h"
+#include "data/transformations.h"
+#include "data/variable.h"
+#include "language/command.h"
+#include "language/commands/data-writer.h"
+#include "language/commands/file-handle.h"
+#include "language/commands/placement-parser.h"
+#include "language/lexer/format-parser.h"
+#include "language/lexer/lexer.h"
+#include "language/lexer/variable-parser.h"
+#include "libpspp/assertion.h"
+#include "libpspp/compiler.h"
+#include "libpspp/i18n.h"
+#include "libpspp/message.h"
+#include "libpspp/misc.h"
+#include "libpspp/pool.h"
+#include "libpspp/u8-line.h"
+#include "output/driver.h"
+#include "output/pivot-table.h"
+#include "output/output-item.h"
+
+#include "gl/xalloc.h"
+
+#include "gettext.h"
+#define N_(msgid) msgid
+#define _(msgid) gettext (msgid)
+
+/* Describes what to do when an output field is encountered. */
+enum field_type
+ {
+ PRT_LITERAL, /* Literal string. */
+ PRT_VAR /* Variable. */
+ };
+
+/* Describes how to output one field. */
+struct prt_out_spec
+ {
+ /* All fields. */
+ enum field_type type; /* What type of field this is. */
+ int record; /* 1-based record number. */
+ int first_column; /* 0-based first column. */
+ int start_ofs, end_ofs;
+
+ /* PRT_VAR only. */
+ const struct variable *var; /* Associated variable. */
+ struct fmt_spec format; /* Output spec. */
+ bool add_space; /* Add trailing space? */
+ bool sysmis_as_spaces; /* Output SYSMIS as spaces? */
+
+ /* PRT_LITERAL only. */
+ struct substring string; /* String to output. */
+ int width; /* Width of 'string', in display columns. */
+ };
+
+/* PRINT, PRINT EJECT, WRITE private data structure. */
+struct print_trns
+ {
+ struct pool *pool; /* Stores related data. */
+ bool eject; /* Eject page before printing? */
+ bool include_prefix; /* Prefix lines with space? */
+ const char *encoding; /* Encoding to use for output. */
+ struct dfm_writer *writer; /* Output file, NULL=listing file. */
+ struct prt_out_spec *specs;
+ size_t n_specs;
+ size_t n_records; /* Number of records to write. */
+ };
+
+enum which_formats
+ {
+ PRINT,
+ WRITE
+ };
+
+static const struct trns_class print_binary_trns_class;
+static const struct trns_class print_text_trns_class;
+
+static int cmd_print__ (struct lexer *, struct dataset *,
+ enum which_formats, bool eject);
+static bool parse_specs (struct lexer *, struct pool *tmp_pool,
+ struct print_trns *, int records_ofs,
+ struct dictionary *, enum which_formats);
+static void dump_table (struct print_trns *);
+
+static bool print_trns_free (void *trns_);
+
+static const struct prt_out_spec *find_binary_spec (const struct print_trns *);
+\f
+/* Basic parsing. */
+
+/* Parses PRINT command. */
+int
+cmd_print (struct lexer *lexer, struct dataset *ds)
+{
+ return cmd_print__ (lexer, ds, PRINT, false);
+}
+
+/* Parses PRINT EJECT command. */
+int
+cmd_print_eject (struct lexer *lexer, struct dataset *ds)
+{
+ return cmd_print__ (lexer, ds, PRINT, true);
+}
+
+/* Parses WRITE command. */
+int
+cmd_write (struct lexer *lexer, struct dataset *ds)
+{
+ return cmd_print__ (lexer, ds, WRITE, false);
+}
+
+/* Parses the output commands. */
+static int
+cmd_print__ (struct lexer *lexer, struct dataset *ds,
+ enum which_formats which_formats, bool eject)
+{
+ bool print_table = false;
+ struct file_handle *fh = NULL;
+ char *encoding = NULL;
+
+ /* Fill in prt to facilitate error-handling. */
+ struct pool *pool = pool_create ();
+ struct print_trns *trns = pool_alloc (pool, sizeof *trns);
+ *trns = (struct print_trns) { .pool = pool, .eject = eject };
+ struct pool *tmp_pool = pool_create_subpool (trns->pool);
+
+ /* Parse the command options. */
+ int records_ofs = 0;
+ while (lex_token (lexer) != T_SLASH && lex_token (lexer) != T_ENDCMD)
+ {
+ if (lex_match_id (lexer, "OUTFILE"))
+ {
+ lex_match (lexer, T_EQUALS);
+
+ fh = fh_parse (lexer, FH_REF_FILE, NULL);
+ if (fh == NULL)
+ goto error;
+ }
+ else if (lex_match_id (lexer, "ENCODING"))
+ {
+ lex_match (lexer, T_EQUALS);
+ if (!lex_force_string (lexer))
+ goto error;
+
+ free (encoding);
+ encoding = ss_xstrdup (lex_tokss (lexer));
+
+ lex_get (lexer);
+ }
+ else if (lex_match_id (lexer, "RECORDS"))
+ {
+ lex_match (lexer, T_EQUALS);
+ lex_match (lexer, T_LPAREN);
+ if (!lex_force_int_range (lexer, "RECORDS", 0, INT_MAX))
+ goto error;
+ trns->n_records = lex_integer (lexer);
+ records_ofs = lex_ofs (lexer);
+ lex_get (lexer);
+ lex_match (lexer, T_RPAREN);
+ }
+ else if (lex_match_id (lexer, "TABLE"))
+ print_table = true;
+ else if (lex_match_id (lexer, "NOTABLE"))
+ print_table = false;
+ else
+ {
+ lex_error_expecting (lexer, "OUTFILE", "ENCODING", "RECORDS",
+ "TABLE", "NOTABLE");
+ goto error;
+ }
+ }
+
+ /* When PRINT or PRINT EJECT writes to an external file, we
+ prefix each line with a space for compatibility. */
+ trns->include_prefix = which_formats == PRINT && fh != NULL;
+
+ /* Parse variables and strings. */
+ if (!parse_specs (lexer, tmp_pool, trns, records_ofs,
+ dataset_dict (ds), which_formats))
+ goto error;
+
+ /* Are there any binary formats?
+
+ There are real difficulties figuring out what to do when both binary
+ formats and nontrivial encodings enter the picture. So when binary
+ formats are present we fall back to much simpler handling. */
+ const struct prt_out_spec *binary_spec = find_binary_spec (trns);
+ if (binary_spec && !fh)
+ {
+ lex_ofs_error (lexer, binary_spec->start_ofs, binary_spec->end_ofs,
+ _("%s is required when binary formats are specified."),
+ "OUTFILE");
+ goto error;
+ }
+
+ if (lex_end_of_command (lexer) != CMD_SUCCESS)
+ goto error;
+
+ if (fh != NULL)
+ {
+ trns->writer = dfm_open_writer (fh, encoding);
+ if (trns->writer == NULL)
+ goto error;
+ trns->encoding = dfm_writer_get_encoding (trns->writer);
+ }
+ else
+ trns->encoding = UTF8;
+
+ /* Output the variable table if requested. */
+ if (print_table)
+ dump_table (trns);
+
+ /* Put the transformation in the queue. */
+ add_transformation (ds, (binary_spec
+ ? &print_binary_trns_class
+ : &print_text_trns_class), trns);
+
+ pool_destroy (tmp_pool);
+ fh_unref (fh);
+
+ return CMD_SUCCESS;
+
+ error:
+ print_trns_free (trns);
+ fh_unref (fh);
+ return CMD_FAILURE;
+}
+\f
+static bool parse_string_argument (struct lexer *, struct print_trns *,
+ size_t *allocated_specs,
+ int record, int *column);
+static bool parse_variable_argument (struct lexer *, const struct dictionary *,
+ struct print_trns *,
+ size_t *allocated_specs,
+ struct pool *tmp_pool,
+ int *record, int *column,
+ enum which_formats);
+
+/* Parses all the variable and string specifications on a single
+ PRINT, PRINT EJECT, or WRITE command into the prt structure.
+ Returns success. */
+static bool
+parse_specs (struct lexer *lexer, struct pool *tmp_pool,
+ struct print_trns *trns, int records_ofs, struct dictionary *dict,
+ enum which_formats which_formats)
+{
+ int record = 0;
+ int column = 1;
+
+ if (lex_token (lexer) == T_ENDCMD)
+ {
+ trns->n_records = 1;
+ return true;
+ }
+
+ size_t allocated_specs = 0;
+ while (lex_token (lexer) != T_ENDCMD)
+ {
+ if (!parse_record_placement (lexer, &record, &column))
+ return false;
+
+ bool ok = (lex_is_string (lexer)
+ ? parse_string_argument (lexer, trns, &allocated_specs,
+ record, &column)
+ : parse_variable_argument (lexer, dict, trns, &allocated_specs,
+ tmp_pool, &record, &column,
+ which_formats));
+ if (!ok)
+ return 0;
+
+ lex_match (lexer, T_COMMA);
+ }
+
+ if (trns->n_records != 0 && trns->n_records != record)
+ lex_ofs_error (lexer, records_ofs, records_ofs,
+ _("Output calls for %d records but %zu specified on RECORDS "
+ "subcommand."),
+ record, trns->n_records);
+ trns->n_records = record;
+
+ return true;
+}
+
+static struct prt_out_spec *
+add_spec (struct print_trns *trns, size_t *allocated_specs)
+{
+ if (trns->n_specs >= *allocated_specs)
+ trns->specs = pool_2nrealloc (trns->pool, trns->specs, allocated_specs,
+ sizeof *trns->specs);
+ return &trns->specs[trns->n_specs++];
+}
+
+/* Parses a string argument to the PRINT commands. Returns success. */
+static bool
+parse_string_argument (struct lexer *lexer, struct print_trns *trns,
+ size_t *allocated_specs, int record, int *column)
+{
+ struct prt_out_spec *spec = add_spec (trns, allocated_specs);
+ *spec = (struct prt_out_spec) {
+ .type = PRT_LITERAL,
+ .record = record,
+ .first_column = *column,
+ .string = ss_clone_pool (lex_tokss (lexer), trns->pool),
+ .start_ofs = lex_ofs (lexer),
+ };
+ lex_get (lexer);
+
+ /* Parse the included column range. */
+ if (lex_is_number (lexer))
+ {
+ int first_column, last_column;
+ bool range_specified;
+
+ if (!parse_column_range (lexer, 1,
+ &first_column, &last_column, &range_specified))
+ return false;
+
+ spec->first_column = first_column;
+ if (range_specified)
+ {
+ struct string s;
+ ds_init_substring (&s, spec->string);
+ ds_set_length (&s, last_column - first_column + 1, ' ');
+ spec->string = ss_clone_pool (s.ss, trns->pool);
+ ds_destroy (&s);
+ }
+ }
+ spec->end_ofs = lex_ofs (lexer) - 1;
+
+ spec->width = u8_width (CHAR_CAST (const uint8_t *, spec->string.string),
+ spec->string.length, UTF8);
+ *column = spec->first_column + spec->width;
+
+ return true;
+}
+
+/* Parses a variable argument to the PRINT commands by passing it off
+ to fixed_parse_compatible() or fixed_parse_fortran() as appropriate.
+ Returns success. */
+static bool
+parse_variable_argument (struct lexer *lexer, const struct dictionary *dict,
+ struct print_trns *trns, size_t *allocated_specs,
+ struct pool *tmp_pool, int *record, int *column,
+ enum which_formats which_formats)
+{
+ const struct variable **vars;
+ size_t n_vars;
+ if (!parse_variables_const_pool (lexer, tmp_pool, dict,
+ &vars, &n_vars, PV_DUPLICATE))
+ return false;
+
+ struct fmt_spec *formats, *f;
+ size_t n_formats;
+ bool add_space;
+ int formats_start = lex_ofs (lexer);
+ if (lex_is_number (lexer) || lex_token (lexer) == T_LPAREN)
+ {
+ if (!parse_var_placements (lexer, tmp_pool, n_vars, FMT_FOR_OUTPUT,
+ &formats, &n_formats))
+ return false;
+ add_space = false;
+ }
+ else
+ {
+ lex_match (lexer, T_ASTERISK);
+
+ formats = pool_nmalloc (tmp_pool, n_vars, sizeof *formats);
+ n_formats = n_vars;
+ for (size_t i = 0; i < n_vars; i++)
+ {
+ const struct variable *v = vars[i];
+ formats[i] = (which_formats == PRINT
+ ? *var_get_print_format (v)
+ : *var_get_write_format (v));
+ }
+ add_space = which_formats == PRINT;
+ }
+ int formats_end = lex_ofs (lexer) - 1;
+
+ size_t var_idx = 0;
+ for (f = formats; f < &formats[n_formats]; f++)
+ if (!execute_placement_format (f, record, column))
+ {
+ const struct variable *var = vars[var_idx++];
+ char *error = fmt_check_width_compat__ (f, var_get_name (var),
+ var_get_width (var));
+ if (error)
+ {
+ lex_ofs_error (lexer, formats_start, formats_end, "%s", error);
+ free (error);
+ return false;
+ }
+
+ struct prt_out_spec *spec = add_spec (trns, allocated_specs);
+ *spec = (struct prt_out_spec) {
+ .type = PRT_VAR,
+ .record = *record,
+ .first_column = *column,
+ .var = var,
+ .format = *f,
+ .add_space = add_space,
+
+ /* This is a completely bizarre twist for compatibility: WRITE
+ outputs the system-missing value as a field filled with spaces,
+ instead of using the normal format that usually contains a
+ period. */
+ .sysmis_as_spaces = (which_formats == WRITE
+ && var_is_numeric (var)
+ && (fmt_get_category (f->type)
+ != FMT_CAT_BINARY)),
+ };
+
+ *column += f->w + add_space;
+ }
+ assert (var_idx == n_vars);
+
+ return true;
+}
+
+/* Prints the table produced by the TABLE subcommand to the listing
+ file. */
+static void
+dump_table (struct print_trns *trns)
+{
+ struct pivot_table *table = pivot_table_create (N_("Print Summary"));
+
+ pivot_dimension_create (table, PIVOT_AXIS_COLUMN, N_("Attributes"),
+ N_("Record"), N_("Columns"), N_("Format"));
+
+ struct pivot_dimension *variables = pivot_dimension_create (
+ table, PIVOT_AXIS_ROW, N_("Variable"));
+
+ for (size_t i = 0; i < trns->n_specs; i++)
+ {
+ const struct prt_out_spec *spec = &trns->specs[i];
+ if (spec->type != PRT_VAR)
+ continue;
+
+ int row = pivot_category_create_leaf (
+ variables->root, pivot_value_new_variable (spec->var));
+
+ pivot_table_put2 (table, 0, row,
+ pivot_value_new_integer (spec->record));
+ int last_column = spec->first_column + spec->format.w - 1;
+ pivot_table_put2 (table, 1, row, pivot_value_new_user_text_nocopy (
+ xasprintf ("%d-%d",
+ spec->first_column, last_column)));
+
+ char fmt_string[FMT_STRING_LEN_MAX + 1];
+ pivot_table_put2 (table, 2, row, pivot_value_new_user_text (
+ fmt_to_string (&spec->format, fmt_string), -1));
+ }
+
+ int row = pivot_category_create_leaf (
+ variables->root, pivot_value_new_text (N_("N of Records")));
+ pivot_table_put2 (table, 0, row,
+ pivot_value_new_integer (trns->n_records));
+
+ pivot_table_submit (table);
+}
+
+static const struct prt_out_spec *
+find_binary_spec (const struct print_trns *trns)
+{
+ for (size_t i = 0; i < trns->n_specs; i++)
+ {
+ const struct prt_out_spec *spec = &trns->specs[i];
+ if (spec->type == PRT_VAR
+ && fmt_get_category (spec->format.type) == FMT_CAT_BINARY)
+ return spec;
+ }
+ return NULL;
+}
+\f
+/* Transformation, for all-text output. */
+
+static void print_text_flush_records (struct print_trns *, struct u8_line *,
+ int target_record,
+ bool *eject, int *record);
+
+/* Performs the transformation inside print_trns T on case C. */
+static enum trns_result
+print_text_trns_proc (void *trns_, struct ccase **c,
+ casenumber case_num UNUSED)
+{
+ struct print_trns *trns = trns_;
+ struct u8_line line;
+
+ bool eject = trns->eject;
+ int record = 1;
+
+ u8_line_init (&line);
+ for (size_t i = 0; i < trns->n_specs; i++)
+ {
+ const struct prt_out_spec *spec = &trns->specs[i];
+ int x0 = spec->first_column;
+
+ print_text_flush_records (trns, &line, spec->record, &eject, &record);
+
+ u8_line_set_length (&line, spec->first_column);
+ if (spec->type == PRT_VAR)
+ {
+ const union value *input = case_data (*c, spec->var);
+ int x1;
+
+ if (!spec->sysmis_as_spaces || input->f != SYSMIS)
+ {
+ size_t len;
+ int width;
+ char *s;
+
+ s = data_out (input, var_get_encoding (spec->var),
+ &spec->format, settings_get_fmt_settings ());
+ len = strlen (s);
+ width = u8_width (CHAR_CAST (const uint8_t *, s), len, UTF8);
+ x1 = x0 + width;
+ u8_line_put (&line, x0, x1, s, len);
+ free (s);
+ }
+ else
+ {
+ int n = spec->format.w;
+
+ x1 = x0 + n;
+ memset (u8_line_reserve (&line, x0, x1, n), ' ', n);
+ }
+
+ if (spec->add_space)
+ *u8_line_reserve (&line, x1, x1 + 1, 1) = ' ';
+ }
+ else
+ {
+ const struct substring *s = &spec->string;
+
+ u8_line_put (&line, x0, x0 + spec->width, s->string, s->length);
+ }
+ }
+ print_text_flush_records (trns, &line, trns->n_records + 1,
+ &eject, &record);
+ u8_line_destroy (&line);
+
+ if (trns->writer != NULL && dfm_write_error (trns->writer))
+ return TRNS_ERROR;
+ return TRNS_CONTINUE;
+}
+
+/* Advance from *RECORD to TARGET_RECORD, outputting records
+ along the way. If *EJECT is true, then the first record
+ output is preceded by ejecting the page (and *EJECT is set
+ false). */
+static void
+print_text_flush_records (struct print_trns *trns, struct u8_line *line,
+ int target_record, bool *eject, int *record)
+{
+ for (; target_record > *record; (*record)++)
+ {
+ char leader = ' ';
+
+ if (*eject)
+ {
+ *eject = false;
+ if (trns->writer == NULL)
+ output_item_submit (page_break_item_create ());
+ else
+ leader = '1';
+ }
+ *u8_line_reserve (line, 0, 1, 1) = leader;
+
+ if (trns->writer == NULL)
+ output_log ("%s", ds_cstr (&line->s) + 1);
+ else
+ {
+ size_t len = ds_length (&line->s);
+ char *s = ds_cstr (&line->s);
+
+ if (!trns->include_prefix)
+ {
+ s++;
+ len--;
+ }
+
+ dfm_put_record_utf8 (trns->writer, s, len);
+ }
+ }
+}
+\f
+/* Transformation, for output involving binary. */
+
+static void print_binary_flush_records (struct print_trns *,
+ struct string *line, int target_record,
+ bool *eject, int *record);
+
+/* Performs the transformation inside print_trns T on case C. */
+static enum trns_result
+print_binary_trns_proc (void *trns_, struct ccase **c,
+ casenumber case_num UNUSED)
+{
+ struct print_trns *trns = trns_;
+ bool eject = trns->eject;
+ char encoded_space = recode_byte (trns->encoding, C_ENCODING, ' ');
+ int record = 1;
+ struct string line = DS_EMPTY_INITIALIZER;
+
+ ds_put_byte (&line, ' ');
+ for (size_t i = 0; i < trns->n_specs; i++)
+ {
+ const struct prt_out_spec *spec = &trns->specs[i];
+ print_binary_flush_records (trns, &line, spec->record, &eject, &record);
+
+ ds_set_length (&line, spec->first_column, encoded_space);
+ if (spec->type == PRT_VAR)
+ {
+ const union value *input = case_data (*c, spec->var);
+ if (!spec->sysmis_as_spaces || input->f != SYSMIS)
+ data_out_recode (input, var_get_encoding (spec->var),
+ &spec->format, settings_get_fmt_settings (),
+ &line, trns->encoding);
+ else
+ ds_put_byte_multiple (&line, encoded_space, spec->format.w);
+ if (spec->add_space)
+ ds_put_byte (&line, encoded_space);
+ }
+ else
+ {
+ ds_put_substring (&line, spec->string);
+ if (0 != strcmp (trns->encoding, UTF8))
+ {
+ size_t length = spec->string.length;
+ char *data = ss_data (ds_tail (&line, length));
+ char *s = recode_string (trns->encoding, UTF8, data, length);
+ memcpy (data, s, length);
+ free (s);
+ }
+ }
+ }
+ print_binary_flush_records (trns, &line, trns->n_records + 1,
+ &eject, &record);
+ ds_destroy (&line);
+
+ if (trns->writer != NULL && dfm_write_error (trns->writer))
+ return TRNS_ERROR;
+ return TRNS_CONTINUE;
+}
+
+/* Advance from *RECORD to TARGET_RECORD, outputting records
+ along the way. If *EJECT is true, then the first record
+ output is preceded by ejecting the page (and *EJECT is set
+ false). */
+static void
+print_binary_flush_records (struct print_trns *trns, struct string *line,
+ int target_record, bool *eject, int *record)
+{
+ for (; target_record > *record; (*record)++)
+ {
+ char *s = ds_cstr (line);
+ size_t length = ds_length (line);
+ char leader = ' ';
+
+ if (*eject)
+ {
+ *eject = false;
+ leader = '1';
+ }
+ s[0] = recode_byte (trns->encoding, C_ENCODING, leader);
+
+ if (!trns->include_prefix)
+ {
+ s++;
+ length--;
+ }
+ dfm_put_record (trns->writer, s, length);
+
+ ds_truncate (line, 1);
+ }
+}
+\f
+/* Frees TRNS. */
+static bool
+print_trns_free (void *trns_)
+{
+ struct print_trns *trns = trns_;
+ bool ok = true;
+
+ if (trns->writer != NULL)
+ ok = dfm_close_writer (trns->writer);
+ pool_destroy (trns->pool);
+
+ return ok;
+}
+
+static const struct trns_class print_binary_trns_class = {
+ .name = "PRINT",
+ .execute = print_binary_trns_proc,
+ .destroy = print_trns_free,
+};
+
+static const struct trns_class print_text_trns_class = {
+ .name = "PRINT",
+ .execute = print_text_trns_proc,
+ .destroy = print_trns_free,
+};
+
projects / pspp / blobdiff