X-Git-Url: https://pintos-os.org/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=src%2Flanguage%2Fdictionary%2Fsys-file-info.c;h=7f5216134c8e1d869700ce3437db2ea7e7209845;hb=c179ef1e8c8c036b741068240995df634559290a;hp=3327a2c4ca0019066b421d6ec5e0d4aacf8f0f9a;hpb=41f83a8f1c88489e4fc458fb7661430ff91d8a61;p=pspp
diff --git a/src/language/dictionary/sys-file-info.c b/src/language/dictionary/sys-file-info.c
index 3327a2c4ca..7f5216134c 100644
--- a/src/language/dictionary/sys-file-info.c
+++ b/src/language/dictionary/sys-file-info.c
@@ -1,5 +1,5 @@
/* PSPP - a program for statistical analysis.
- Copyright (C) 1997-9, 2000, 2006, 2009, 2010, 2011, 2012, 2013 Free Software Foundation, Inc.
+ Copyright (C) 1997-9, 2000, 2006, 2009, 2010, 2011, 2012, 2013, 2014 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
@@ -13,12 +13,14 @@
You should have received a copy of the GNU General Public License
along with this program. If not, see . */
-
#include
#include
+#include
+#include
#include
+#include "data/any-reader.h"
#include "data/attributes.h"
#include "data/casereader.h"
#include "data/dataset.h"
@@ -26,7 +28,6 @@
#include "data/file-handle-def.h"
#include "data/format.h"
#include "data/missing-values.h"
-#include "data/sys-file-reader.h"
#include "data/value-labels.h"
#include "data/variable.h"
#include "data/vector.h"
@@ -35,162 +36,244 @@
#include "language/lexer/lexer.h"
#include "language/lexer/variable-parser.h"
#include "libpspp/array.h"
+#include "libpspp/hash-functions.h"
+#include "libpspp/i18n.h"
#include "libpspp/message.h"
#include "libpspp/misc.h"
+#include "libpspp/pool.h"
#include "libpspp/string-array.h"
-#include "output/tab.h"
+#include "output/pivot-table.h"
+#include "output/output-item.h"
+#include "gl/count-one-bits.h"
+#include "gl/localcharset.h"
+#include "gl/intprops.h"
#include "gl/minmax.h"
#include "gl/xalloc.h"
#include "gettext.h"
#define _(msgid) gettext (msgid)
+#define N_(msgid) (msgid)
/* Information to include in displaying a dictionary. */
-enum
+enum
{
- DF_DICT_INDEX = 1 << 0,
- DF_FORMATS = 1 << 1,
- DF_VALUE_LABELS = 1 << 2,
- DF_VARIABLE_LABELS = 1 << 3,
- DF_MISSING_VALUES = 1 << 4,
- DF_AT_ATTRIBUTES = 1 << 5, /* Attributes whose names begin with @. */
- DF_ATTRIBUTES = 1 << 6, /* All other attributes. */
- DF_MISC = 1 << 7,
- DF_ALL = (1 << 8) - 1
+ /* Variable table. */
+ DF_NAME = 1 << 0,
+ DF_POSITION = 1 << 1,
+ DF_LABEL = 1 << 2,
+ DF_MEASUREMENT_LEVEL = 1 << 3,
+ DF_ROLE = 1 << 4,
+ DF_WIDTH = 1 << 5,
+ DF_ALIGNMENT = 1 << 6,
+ DF_PRINT_FORMAT = 1 << 7,
+ DF_WRITE_FORMAT = 1 << 8,
+ DF_MISSING_VALUES = 1 << 9,
+#define DF_ALL_VARIABLE ((1 << 10) - 1)
+
+ /* Value labels table. */
+ DF_VALUE_LABELS = 1 << 10,
+
+ /* Attribute table. */
+ DF_AT_ATTRIBUTES = 1 << 11, /* Attributes whose names begin with @. */
+ DF_ATTRIBUTES = 1 << 12, /* All other attributes. */
};
-static int describe_variable (const struct variable *v, struct tab_table *t,
- int r, int pc, int flags);
+static void display_variables (const struct variable **, size_t, int flags);
+static void display_value_labels (const struct variable **, size_t);
+static void display_attributes (const struct attrset *,
+ const struct variable **, size_t, int flags);
+
+static void report_encodings (const struct file_handle *, struct pool *,
+ char **titles, bool *ids,
+ char **strings, size_t n_strings);
+
+static char *get_documents_as_string (const struct dictionary *);
+
+static void
+add_row (struct pivot_table *table, const char *attribute,
+ struct pivot_value *value)
+{
+ int row = pivot_category_create_leaf (table->dimensions[0]->root,
+ pivot_value_new_text (attribute));
+ if (value)
+ pivot_table_put1 (table, row, value);
+}
/* SYSFILE INFO utility. */
int
-cmd_sysfile_info (struct lexer *lexer, struct dataset *ds UNUSED)
+cmd_sysfile_info (struct lexer *lexer, struct dataset *ds)
{
+ struct any_reader *any_reader;
struct file_handle *h;
struct dictionary *d;
- struct tab_table *t;
struct casereader *reader;
- struct sfm_read_info info;
- int r, i;
-
- lex_match_id (lexer, "FILE");
- lex_match (lexer, T_EQUALS);
+ struct any_read_info info;
+ char *encoding;
- h = fh_parse (lexer, FH_REF_FILE, NULL);
- if (!h)
- return CMD_FAILURE;
-
- reader = sfm_open_reader (h, NULL, &d, &info);
- if (!reader)
+ h = NULL;
+ encoding = NULL;
+ for (;;)
{
- fh_unref (h);
- return CMD_FAILURE;
- }
- casereader_destroy (reader);
+ lex_match (lexer, T_SLASH);
- t = tab_create (2, 11 + (info.product_ext != NULL));
- r = 0;
- tab_vline (t, TAL_GAP, 1, 0, 8);
+ if (lex_match_id (lexer, "FILE") || lex_is_string (lexer))
+ {
+ lex_match (lexer, T_EQUALS);
- tab_text (t, 0, r, TAB_LEFT, _("File:"));
- tab_text (t, 1, r++, TAB_LEFT, fh_get_file_name (h));
+ fh_unref (h);
+ h = fh_parse (lexer, FH_REF_FILE, NULL);
+ if (h == NULL)
+ goto error;
+ }
+ else if (lex_match_id (lexer, "ENCODING"))
+ {
+ lex_match (lexer, T_EQUALS);
- tab_text (t, 0, r, TAB_LEFT, _("Label:"));
- {
- const char *label = dict_get_label (d);
- if (label == NULL)
- label = _("No label.");
- tab_text (t, 1, r++, TAB_LEFT, label);
- }
+ if (!lex_force_string (lexer))
+ goto error;
- tab_text (t, 0, r, TAB_LEFT, _("Created:"));
- tab_text_format (t, 1, r++, TAB_LEFT, "%s %s by %s",
- info.creation_date, info.creation_time, info.product);
+ free (encoding);
+ encoding = ss_xstrdup (lex_tokss (lexer));
- if (info.product_ext)
+ lex_get (lexer);
+ }
+ else
+ break;
+ }
+
+ if (h == NULL)
{
- tab_text (t, 0, r, TAB_LEFT, _("Product:"));
- tab_text (t, 1, r++, TAB_LEFT, info.product_ext);
+ lex_sbc_missing ("FILE");
+ goto error;
}
- tab_text (t, 0, r, TAB_LEFT, _("Integer Format:"));
- tab_text (t, 1, r++, TAB_LEFT,
- info.integer_format == INTEGER_MSB_FIRST ? _("Big Endian")
- : info.integer_format == INTEGER_LSB_FIRST ? _("Little Endian")
- : _("Unknown"));
-
- tab_text (t, 0, r, TAB_LEFT, _("Real Format:"));
- tab_text (t, 1, r++, TAB_LEFT,
- info.float_format == FLOAT_IEEE_DOUBLE_LE ? _("IEEE 754 LE.")
- : info.float_format == FLOAT_IEEE_DOUBLE_BE ? _("IEEE 754 BE.")
- : info.float_format == FLOAT_VAX_D ? _("VAX D.")
- : info.float_format == FLOAT_VAX_G ? _("VAX G.")
- : info.float_format == FLOAT_Z_LONG ? _("IBM 390 Hex Long.")
- : _("Unknown"));
-
- tab_text (t, 0, r, TAB_LEFT, _("Variables:"));
- tab_text_format (t, 1, r++, TAB_LEFT, "%zu", dict_get_var_cnt (d));
-
- tab_text (t, 0, r, TAB_LEFT, _("Cases:"));
- if (info.case_cnt == -1)
- tab_text (t, 1, r, TAB_LEFT, _("Unknown"));
- else
- tab_text_format (t, 1, r, TAB_LEFT, "%ld", (long int) info.case_cnt);
- r++;
+ any_reader = any_reader_open (h);
+ if (!any_reader)
+ {
+ free (encoding);
+ return CMD_FAILURE;
+ }
- tab_text (t, 0, r, TAB_LEFT, _("Type:"));
- tab_text (t, 1, r++, TAB_LEFT, _("System File"));
+ if (encoding && !strcasecmp (encoding, "detect"))
+ {
+ char **titles, **strings;
+ struct pool *pool;
+ size_t n_strings;
+ bool *ids;
- tab_text (t, 0, r, TAB_LEFT, _("Weight:"));
- {
- struct variable *weight_var = dict_get_weight (d);
- tab_text (t, 1, r++, TAB_LEFT,
- (weight_var != NULL
- ? var_get_name (weight_var) : _("Not weighted.")));
- }
+ pool = pool_create ();
+ n_strings = any_reader_get_strings (any_reader, pool,
+ &titles, &ids, &strings);
+ any_reader_close (any_reader);
- tab_text (t, 0, r, TAB_LEFT, _("Mode:"));
- tab_text_format (t, 1, r++, TAB_LEFT,
- _("Compression %s."), info.compressed ? _("on") : _("off"));
+ report_encodings (h, pool, titles, ids, strings, n_strings);
+ fh_unref (h);
+ pool_destroy (pool);
+ free (encoding);
+ return CMD_SUCCESS;
+ }
- tab_text (t, 0, r, TAB_LEFT, _("Charset:"));
- tab_text (t, 1, r++, TAB_LEFT, dict_get_encoding (d));
+ reader = any_reader_decode (any_reader, encoding, &d, &info);
+ if (!reader)
+ goto error;
+ casereader_destroy (reader);
- tab_submit (t);
+ struct pivot_table *table = pivot_table_create (N_("File Information"));
+ pivot_dimension_create (table, PIVOT_AXIS_ROW, N_("Attribute"));
- t = tab_create (4, 1 + 2 * dict_get_var_cnt (d));
- tab_headers (t, 0, 0, 1, 0);
- tab_text (t, 0, 0, TAB_LEFT | TAT_TITLE, _("Variable"));
- tab_joint_text (t, 1, 0, 2, 0, TAB_LEFT | TAT_TITLE, _("Description"));
- tab_text (t, 3, 0, TAB_LEFT | TAT_TITLE, _("Position"));
- tab_hline (t, TAL_2, 0, 3, 1);
- for (r = 1, i = 0; i < dict_get_var_cnt (d); i++)
- r = describe_variable (dict_get_var (d, i), t, r, 3,
- DF_ALL & ~DF_AT_ATTRIBUTES);
+ add_row (table, N_("File"),
+ pivot_value_new_user_text (fh_get_file_name (h), -1));
- tab_box (t, TAL_1, TAL_1, -1, -1, 0, 0, 3, r);
- tab_vline (t, TAL_1, 1, 0, r);
- tab_vline (t, TAL_1, 3, 0, r);
+ const char *label = dict_get_label (d);
+ add_row (table, N_("Label"),
+ label ? pivot_value_new_user_text (label, -1) : NULL);
- tab_resize (t, -1, r);
- tab_submit (t);
+ add_row (table, N_("Created"),
+ pivot_value_new_user_text_nocopy (
+ xasprintf ("%s %s by %s", info.creation_date,
+ info.creation_time, info.product)));
- dict_destroy (d);
+ if (info.product_ext)
+ add_row (table, N_("Product"),
+ pivot_value_new_user_text (info.product_ext, -1));
+
+ add_row (table, N_("Integer Format"),
+ pivot_value_new_text (
+ info.integer_format == INTEGER_MSB_FIRST ? N_("Big Endian")
+ : info.integer_format == INTEGER_LSB_FIRST ? N_("Little Endian")
+ : N_("Unknown")));
+
+ add_row (table, N_("Real Format"),
+ pivot_value_new_text (
+ info.float_format == FLOAT_IEEE_DOUBLE_LE ? N_("IEEE 754 LE.")
+ : info.float_format == FLOAT_IEEE_DOUBLE_BE ? N_("IEEE 754 BE.")
+ : info.float_format == FLOAT_VAX_D ? N_("VAX D.")
+ : info.float_format == FLOAT_VAX_G ? N_("VAX G.")
+ : info.float_format == FLOAT_Z_LONG ? N_("IBM 390 Hex Long.")
+ : N_("Unknown")));
+
+ add_row (table, N_("Variables"),
+ pivot_value_new_integer (dict_get_n_vars (d)));
+
+ add_row (table, N_("Cases"),
+ (info.n_cases == -1
+ ? pivot_value_new_text (N_("Unknown"))
+ : pivot_value_new_integer (info.n_cases)));
+
+ add_row (table, N_("Type"),
+ pivot_value_new_text (info.klass->name));
+
+ struct variable *weight_var = dict_get_weight (d);
+ add_row (table, N_("Weight"),
+ (weight_var
+ ? pivot_value_new_variable (weight_var)
+ : pivot_value_new_text (N_("Not weighted"))));
+
+ add_row (table, N_("Compression"),
+ (info.compression == ANY_COMP_NONE
+ ? pivot_value_new_text (N_("None"))
+ : pivot_value_new_user_text (
+ info.compression == ANY_COMP_SIMPLE ? "SAV" : "ZSAV", -1)));
+
+ add_row (table, N_("Encoding"),
+ pivot_value_new_user_text (dict_get_encoding (d), -1));
+
+ if (dict_get_document_n_lines (d) > 0)
+ add_row (table, N_("Documents"),
+ pivot_value_new_user_text_nocopy (get_documents_as_string (d)));
+
+ pivot_table_submit (table);
+
+ size_t n_vars = dict_get_n_vars (d);
+ const struct variable **vars = xnmalloc (n_vars, sizeof *vars);
+ for (size_t i = 0; i < dict_get_n_vars (d); i++)
+ vars[i] = dict_get_var (d, i);
+ display_variables (vars, n_vars, DF_ALL_VARIABLE);
+ display_value_labels (vars, n_vars);
+ display_attributes (dict_get_attributes (dataset_dict (ds)),
+ vars, n_vars, DF_ATTRIBUTES);
+ free (vars);
+
+ dict_unref (d);
fh_unref (h);
- sfm_read_info_destroy (&info);
+ free (encoding);
+ any_read_info_destroy (&info);
return CMD_SUCCESS;
+
+error:
+ fh_unref (h);
+ free (encoding);
+ return CMD_FAILURE;
}
/* DISPLAY utility. */
static void display_macros (void);
static void display_documents (const struct dictionary *dict);
-static void display_variables (const struct variable **, size_t, int);
static void display_vectors (const struct dictionary *dict, int sorted);
-static void display_data_file_attributes (struct attrset *, int flags);
int
cmd_display (struct lexer *lexer, struct dataset *ds)
@@ -210,12 +293,16 @@ cmd_display (struct lexer *lexer, struct dataset *ds)
{
if (!lex_force_match_id (lexer, "LABEL"))
return CMD_FAILURE;
- if (dict_get_label (dataset_dict (ds)) == NULL)
- tab_output_text (TAB_LEFT,
- _("The active dataset does not have a file label."));
- else
- tab_output_text_format (TAB_LEFT, _("File label: %s"),
- dict_get_label (dataset_dict (ds)));
+
+ const char *label = dict_get_label (dataset_dict (ds));
+
+ struct pivot_table *table = pivot_table_create (N_("File Label"));
+ pivot_dimension_create (table, PIVOT_AXIS_ROW, N_("Label"),
+ N_("Label"));
+ pivot_table_put1 (table, 0,
+ (label ? pivot_value_new_user_text (label, -1)
+ : pivot_value_new_text (N_("(none)"))));
+ pivot_table_submit (table);
}
else
{
@@ -228,31 +315,36 @@ cmd_display (struct lexer *lexer, struct dataset *ds)
display_vectors (dataset_dict(ds), sorted);
return CMD_SUCCESS;
}
- else if (lex_match_id (lexer, "SCRATCH"))
+ else if (lex_match_id (lexer, "SCRATCH"))
{
dict_get_vars (dataset_dict (ds), &vl, &n, DC_ORDINARY);
- flags = 0;
+ flags = DF_NAME;
}
- else
+ else
{
- struct subcommand
+ struct subcommand
{
const char *name;
int flags;
};
- static const struct subcommand subcommands[] =
+ static const struct subcommand subcommands[] =
{
{"@ATTRIBUTES", DF_ATTRIBUTES | DF_AT_ATTRIBUTES},
{"ATTRIBUTES", DF_ATTRIBUTES},
- {"DICTIONARY", DF_ALL & ~DF_AT_ATTRIBUTES},
- {"INDEX", DF_DICT_INDEX},
- {"LABELS", DF_DICT_INDEX | DF_VARIABLE_LABELS},
- {"NAMES", 0},
- {"VARIABLES",
- DF_DICT_INDEX | DF_FORMATS | DF_MISSING_VALUES | DF_MISC},
+ {"DICTIONARY", (DF_NAME | DF_POSITION | DF_LABEL
+ | DF_MEASUREMENT_LEVEL | DF_ROLE | DF_WIDTH
+ | DF_ALIGNMENT | DF_PRINT_FORMAT
+ | DF_WRITE_FORMAT | DF_MISSING_VALUES
+ | DF_VALUE_LABELS)},
+ {"INDEX", DF_NAME | DF_POSITION},
+ {"LABELS", DF_NAME | DF_POSITION | DF_LABEL},
+ {"NAMES", DF_NAME},
+ {"VARIABLES", (DF_NAME | DF_POSITION | DF_PRINT_FORMAT
+ | DF_WRITE_FORMAT | DF_MISSING_VALUES)},
{NULL, 0},
};
const struct subcommand *sbc;
+ struct dictionary *dict = dataset_dict (ds);
flags = 0;
for (sbc = subcommands; sbc->name != NULL; sbc++)
@@ -268,32 +360,38 @@ cmd_display (struct lexer *lexer, struct dataset *ds)
if (lex_token (lexer) != T_ENDCMD)
{
- if (!parse_variables_const (lexer, dataset_dict (ds), &vl, &n,
- PV_NONE))
+ if (!parse_variables_const (lexer, dict, &vl, &n, PV_NONE))
{
free (vl);
return CMD_FAILURE;
}
}
else
- dict_get_vars (dataset_dict (ds), &vl, &n, 0);
+ dict_get_vars (dict, &vl, &n, 0);
}
- if (n > 0)
+ if (n > 0)
{
- sort (vl, n, sizeof *vl,
- (sorted
- ? compare_var_ptrs_by_name
- : compare_var_ptrs_by_dict_index), NULL);
- display_variables (vl, n, flags);
+ sort (vl, n, sizeof *vl, (sorted
+ ? compare_var_ptrs_by_name
+ : compare_var_ptrs_by_dict_index), NULL);
+
+ int variable_flags = flags & DF_ALL_VARIABLE;
+ if (variable_flags)
+ display_variables (vl, n, variable_flags);
+
+ if (flags & DF_VALUE_LABELS)
+ display_value_labels (vl, n);
+
+ int attribute_flags = flags & (DF_ATTRIBUTES | DF_AT_ATTRIBUTES);
+ if (attribute_flags)
+ display_attributes (dict_get_attributes (dataset_dict (ds)),
+ vl, n, attribute_flags);
}
else
msg (SW, _("No variables to display."));
- free (vl);
- if (flags & (DF_ATTRIBUTES | DF_AT_ATTRIBUTES))
- display_data_file_attributes (dict_get_attributes (dataset_dict (ds)),
- flags);
+ free (vl);
}
return CMD_SUCCESS;
@@ -302,399 +400,700 @@ cmd_display (struct lexer *lexer, struct dataset *ds)
static void
display_macros (void)
{
- tab_output_text (TAB_LEFT, _("Macros not supported."));
+ msg (SW, _("Macros not supported."));
+}
+
+static char *
+get_documents_as_string (const struct dictionary *dict)
+{
+ const struct string_array *documents = dict_get_documents (dict);
+ struct string s = DS_EMPTY_INITIALIZER;
+ for (size_t i = 0; i < documents->n; i++)
+ {
+ if (i)
+ ds_put_byte (&s, '\n');
+ ds_put_cstr (&s, documents->strings[i]);
+ }
+ return ds_steal_cstr (&s);
}
static void
display_documents (const struct dictionary *dict)
{
- const struct string_array *documents = dict_get_documents (dict);
+ struct pivot_table *table = pivot_table_create (N_("Documents"));
+ struct pivot_dimension *d = pivot_dimension_create (
+ table, PIVOT_AXIS_COLUMN, N_("Documents"), N_("Document"));
+ d->hide_all_labels = true;
- if (string_array_is_empty (documents))
- tab_output_text (TAB_LEFT, _("The active dataset dictionary does not "
- "contain any documents."));
+ if (!dict_get_documents (dict)->n)
+ pivot_table_put1 (table, 0, pivot_value_new_text (N_("(none)")));
else
{
- size_t i;
-
- tab_output_text (TAB_LEFT | TAT_TITLE,
- _("Documents in the active dataset:"));
- for (i = 0; i < dict_get_document_line_cnt (dict); i++)
- tab_output_text (TAB_LEFT | TAB_FIX, dict_get_document_line (dict, i));
+ char *docs = get_documents_as_string (dict);
+ pivot_table_put1 (table, 0, pivot_value_new_user_text_nocopy (docs));
}
+
+ pivot_table_submit (table);
}
static void
display_variables (const struct variable **vl, size_t n, int flags)
{
- struct tab_table *t;
- int nc; /* Number of columns. */
- int pc; /* `Position column' */
- int r; /* Current row. */
- size_t i;
+ struct pivot_table *table = pivot_table_create (N_("Variables"));
- /* One column for the name,
- two columns for general description,
- one column for dictionary index. */
- nc = 1;
- if (flags & ~DF_DICT_INDEX)
- nc += 2;
- pc = nc;
- if (flags & DF_DICT_INDEX)
- nc++;
-
- t = tab_create (nc, n + 5);
- tab_headers (t, 0, 0, 1, 0);
- tab_hline (t, TAL_2, 0, nc - 1, 1);
- tab_text (t, 0, 0, TAB_LEFT | TAT_TITLE, _("Variable"));
- if (flags & ~DF_DICT_INDEX)
- tab_joint_text (t, 1, 0, 2, 0, TAB_LEFT | TAT_TITLE,
- (flags & ~(DF_DICT_INDEX | DF_VARIABLE_LABELS)
- ? _("Description") : _("Label")));
- if (flags & DF_DICT_INDEX)
- tab_text (t, pc, 0, TAB_LEFT | TAT_TITLE, _("Position"));
-
- r = 1;
- for (i = 0; i < n; i++)
- r = describe_variable (vl[i], t, r, pc, flags);
- tab_hline (t, flags & ~DF_DICT_INDEX ? TAL_2 : TAL_1, 0, nc - 1, 1);
- if (flags)
+ struct pivot_dimension *attributes = pivot_dimension_create (
+ table, PIVOT_AXIS_COLUMN, N_("Attributes"));
+
+ struct heading
+ {
+ int flag;
+ const char *title;
+ };
+ static const struct heading headings[] = {
+ { DF_POSITION, N_("Position") },
+ { DF_LABEL, N_("Label") },
+ { DF_MEASUREMENT_LEVEL, N_("Measurement Level") },
+ { DF_ROLE, N_("Role") },
+ { DF_WIDTH, N_("Width") },
+ { DF_ALIGNMENT, N_("Alignment") },
+ { DF_PRINT_FORMAT, N_("Print Format") },
+ { DF_WRITE_FORMAT, N_("Write Format") },
+ { DF_MISSING_VALUES, N_("Missing Values") },
+ };
+ for (size_t i = 0; i < sizeof headings / sizeof *headings; i++)
+ if (flags & headings[i].flag)
+ pivot_category_create_leaf (attributes->root,
+ pivot_value_new_text (headings[i].title));
+
+ struct pivot_dimension *names = pivot_dimension_create (
+ table, PIVOT_AXIS_ROW, N_("Name"));
+ names->root->show_label = true;
+
+ for (size_t i = 0; i < n; i++)
+ {
+ const struct variable *v = vl[i];
+
+ struct pivot_value *name = pivot_value_new_variable (v);
+ name->variable.show = SETTINGS_VALUE_SHOW_VALUE;
+ int row = pivot_category_create_leaf (names->root, name);
+
+ int x = 0;
+ if (flags & DF_POSITION)
+ pivot_table_put2 (table, x++, row, pivot_value_new_integer (
+ var_get_dict_index (v) + 1));
+
+ if (flags & DF_LABEL)
+ {
+ const char *label = var_get_label (v);
+ if (label)
+ pivot_table_put2 (table, x, row,
+ pivot_value_new_user_text (label, -1));
+ x++;
+ }
+
+ if (flags & DF_MEASUREMENT_LEVEL)
+ pivot_table_put2 (
+ table, x++, row,
+ pivot_value_new_text (measure_to_string (var_get_measure (v))));
+
+ if (flags & DF_ROLE)
+ pivot_table_put2 (
+ table, x++, row,
+ pivot_value_new_text (var_role_to_string (var_get_role (v))));
+
+ if (flags & DF_WIDTH)
+ pivot_table_put2 (
+ table, x++, row,
+ pivot_value_new_integer (var_get_display_width (v)));
+
+ if (flags & DF_ALIGNMENT)
+ pivot_table_put2 (
+ table, x++, row,
+ pivot_value_new_text (alignment_to_string (
+ var_get_alignment (v))));
+
+ if (flags & DF_PRINT_FORMAT)
+ {
+ const struct fmt_spec *print = var_get_print_format (v);
+ char s[FMT_STRING_LEN_MAX + 1];
+
+ pivot_table_put2 (
+ table, x++, row,
+ pivot_value_new_user_text (fmt_to_string (print, s), -1));
+ }
+
+ if (flags & DF_WRITE_FORMAT)
+ {
+ const struct fmt_spec *write = var_get_write_format (v);
+ char s[FMT_STRING_LEN_MAX + 1];
+
+ pivot_table_put2 (
+ table, x++, row,
+ pivot_value_new_user_text (fmt_to_string (write, s), -1));
+ }
+
+ if (flags & DF_MISSING_VALUES)
+ {
+ char *s = mv_to_string (var_get_missing_values (v),
+ var_get_encoding (v));
+ if (s)
+ pivot_table_put2 (
+ table, x, row,
+ pivot_value_new_user_text_nocopy (s));
+
+ x++;
+ }
+ }
+
+ pivot_table_submit (table);
+}
+
+static bool
+any_value_labels (const struct variable **vars, size_t n_vars)
+{
+ for (size_t i = 0; i < n_vars; i++)
+ if (val_labs_count (var_get_value_labels (vars[i])))
+ return true;
+ return false;
+}
+
+static void
+display_value_labels (const struct variable **vars, size_t n_vars)
+{
+ if (!any_value_labels (vars, n_vars))
+ return;
+
+ struct pivot_table *table = pivot_table_create (N_("Value Labels"));
+
+ pivot_dimension_create (table, PIVOT_AXIS_COLUMN,
+ N_("Label"), N_("Label"));
+
+ struct pivot_dimension *values = pivot_dimension_create (
+ table, PIVOT_AXIS_ROW, N_("Variable Value"));
+ values->root->show_label = true;
+
+ struct pivot_footnote *missing_footnote = pivot_table_create_footnote (
+ table, pivot_value_new_text (N_("User-missing value")));
+
+ for (size_t i = 0; i < n_vars; i++)
{
- tab_box (t, TAL_1, TAL_1, -1, -1, 0, 0, nc - 1, r - 1);
- tab_vline (t, TAL_1, 1, 0, r - 1);
+ const struct val_labs *val_labs = var_get_value_labels (vars[i]);
+ size_t n_labels = val_labs_count (val_labs);
+ if (!n_labels)
+ continue;
+
+ struct pivot_category *group = pivot_category_create_group__ (
+ values->root, pivot_value_new_variable (vars[i]));
+
+ const struct val_lab **labels = val_labs_sorted (val_labs);
+ for (size_t j = 0; j < n_labels; j++)
+ {
+ const struct val_lab *vl = labels[j];
+
+ struct pivot_value *value = pivot_value_new_var_value (
+ vars[i], &vl->value);
+ if (value->type == PIVOT_VALUE_NUMERIC)
+ value->numeric.show = SETTINGS_VALUE_SHOW_VALUE;
+ else
+ value->string.show = SETTINGS_VALUE_SHOW_VALUE;
+ if (var_is_value_missing (vars[i], &vl->value) == MV_USER)
+ pivot_value_add_footnote (value, missing_footnote);
+ int row = pivot_category_create_leaf (group, value);
+
+ struct pivot_value *label = pivot_value_new_var_value (
+ vars[i], &vl->value);
+ char *escaped_label = xstrdup (val_lab_get_escaped_label (vl));
+ if (label->type == PIVOT_VALUE_NUMERIC)
+ {
+ free (label->numeric.value_label);
+ label->numeric.value_label = escaped_label;
+ label->numeric.show = SETTINGS_VALUE_SHOW_LABEL;
+ }
+ else
+ {
+ free (label->string.value_label);
+ label->string.value_label = escaped_label;
+ label->string.show = SETTINGS_VALUE_SHOW_LABEL;
+ }
+ pivot_table_put2 (table, 0, row, label);
+ }
+ free (labels);
}
- if (flags & ~DF_DICT_INDEX)
- tab_vline (t, TAL_1, nc - 1, 0, r - 1);
- tab_resize (t, -1, r);
- tab_submit (t);
+ pivot_table_submit (table);
}
static bool
-is_at_name (const char *name)
+is_at_name (const char *name)
{
return name[0] == '@' || (name[0] == '$' && name[1] == '@');
}
static size_t
-count_attributes (const struct attrset *set, int flags)
+count_attributes (const struct attrset *set, int flags)
{
struct attrset_iterator i;
struct attribute *attr;
size_t n_attrs;
-
+
n_attrs = 0;
for (attr = attrset_first (set, &i); attr != NULL;
- attr = attrset_next (set, &i))
+ attr = attrset_next (set, &i))
if (flags & DF_AT_ATTRIBUTES || !is_at_name (attribute_get_name (attr)))
n_attrs += attribute_get_n_values (attr);
return n_attrs;
}
static void
-display_attributes (struct tab_table *t, const struct attrset *set, int flags,
- int c, int r)
+display_attrset (struct pivot_table *table, struct pivot_value *set_name,
+ const struct attrset *set, int flags)
{
- struct attribute **attrs;
- size_t n_attrs;
- size_t i;
+ size_t n_total = count_attributes (set, flags);
+ if (!n_total)
+ {
+ pivot_value_destroy (set_name);
+ return;
+ }
+
+ struct pivot_category *group = pivot_category_create_group__ (
+ table->dimensions[1]->root, set_name);
- n_attrs = attrset_count (set);
- attrs = attrset_sorted (set);
- for (i = 0; i < n_attrs; i++)
+ size_t n_attrs = attrset_count (set);
+ struct attribute **attrs = attrset_sorted (set);
+ for (size_t i = 0; i < n_attrs; i++)
{
const struct attribute *attr = attrs[i];
const char *name = attribute_get_name (attr);
- size_t n_values;
- size_t i;
if (!(flags & DF_AT_ATTRIBUTES) && is_at_name (name))
continue;
- n_values = attribute_get_n_values (attr);
- for (i = 0; i < n_values; i++)
+ size_t n_values = attribute_get_n_values (attr);
+ for (size_t j = 0; j < n_values; j++)
{
- if (n_values > 1)
- tab_text_format (t, c, r, TAB_LEFT, "%s[%zu]", name, i + 1);
- else
- tab_text (t, c, r, TAB_LEFT, name);
- tab_text (t, c + 1, r, TAB_LEFT, attribute_get_value (attr, i));
- r++;
+ int row = pivot_category_create_leaf (
+ group,
+ (n_values > 1
+ ? pivot_value_new_user_text_nocopy (xasprintf (
+ "%s[%zu]", name, j + 1))
+ : pivot_value_new_user_text (name, -1)));
+ pivot_table_put2 (table, 0, row,
+ pivot_value_new_user_text (
+ attribute_get_value (attr, j), -1));
}
}
free (attrs);
}
static void
-display_data_file_attributes (struct attrset *set, int flags)
+display_attributes (const struct attrset *dict_attrset,
+ const struct variable **vars, size_t n_vars, int flags)
{
- struct tab_table *t;
- size_t n_attrs;
+ struct pivot_table *table = pivot_table_create (
+ N_("Variable and Dataset Attributes"));
- n_attrs = count_attributes (set, flags);
- if (!n_attrs)
- return;
+ pivot_dimension_create (table, PIVOT_AXIS_COLUMN,
+ N_("Value"), N_("Value"));
+
+ struct pivot_dimension *variables = pivot_dimension_create (
+ table, PIVOT_AXIS_ROW, N_("Variable and Name"));
+ variables->root->show_label = true;
- t = tab_create (2, n_attrs + 1);
- tab_headers (t, 0, 0, 1, 0);
- tab_box (t, TAL_1, TAL_1, -1, TAL_1, 0, 0, tab_nc (t) - 1, tab_nr (t) - 1);
- tab_hline (t, TAL_2, 0, 1, 1);
- tab_text (t, 0, 0, TAB_LEFT | TAT_TITLE, _("Attribute"));
- tab_text (t, 1, 0, TAB_LEFT | TAT_TITLE, _("Value"));
- display_attributes (t, set, flags, 0, 1);
- tab_title (t, "Custom data file attributes.");
- tab_submit (t);
+ display_attrset (table, pivot_value_new_text (N_("(dataset)")),
+ dict_attrset, flags);
+ for (size_t i = 0; i < n_vars; i++)
+ display_attrset (table, pivot_value_new_variable (vars[i]),
+ var_get_attributes (vars[i]), flags);
+
+ if (pivot_table_is_empty (table))
+ pivot_table_unref (table);
+ else
+ pivot_table_submit (table);
}
-/* Puts a description of variable V into table T starting at row
- R. The variable will be described in the format given by
- FLAGS. Returns the next row available for use in the
- table. */
-static int
-describe_variable (const struct variable *v, struct tab_table *t, int r,
- int pc, int flags)
+/* Display a list of vectors. If SORTED is nonzero then they are
+ sorted alphabetically. */
+static void
+display_vectors (const struct dictionary *dict, int sorted)
{
- size_t n_attrs = 0;
- int need_rows;
-
- /* Make sure that enough rows are allocated. */
- need_rows = 1;
- if (flags & ~(DF_DICT_INDEX | DF_VARIABLE_LABELS))
- need_rows += 16;
- if (flags & DF_VALUE_LABELS)
- need_rows += val_labs_count (var_get_value_labels (v));
- if (flags & (DF_ATTRIBUTES | DF_AT_ATTRIBUTES))
+ size_t n_vectors = dict_get_n_vectors (dict);
+ if (n_vectors == 0)
{
- n_attrs = count_attributes (var_get_attributes (v), flags);
- need_rows += n_attrs;
- }
- if (r + need_rows > tab_nr (t))
- {
- int nr = MAX (r + need_rows, tab_nr (t) * 2);
- tab_realloc (t, -1, nr);
+ msg (SW, _("No vectors defined."));
+ return;
}
- /* Put the name, var label, and position into the first row. */
- tab_text (t, 0, r, TAB_LEFT, var_get_name (v));
- if (flags & DF_DICT_INDEX)
- tab_text_format (t, pc, r, 0, "%zu", var_get_dict_index (v) + 1);
+ const struct vector **vectors = xnmalloc (n_vectors, sizeof *vectors);
+ for (size_t i = 0; i < n_vectors; i++)
+ vectors[i] = dict_get_vector (dict, i);
+ if (sorted)
+ qsort (vectors, n_vectors, sizeof *vectors, compare_vector_ptrs_by_name);
- if (flags & DF_VARIABLE_LABELS && var_has_label (v))
- {
- tab_joint_text (t, 1, r, 2, r, TAB_LEFT, var_get_label (v));
- r++;
- }
+ struct pivot_table *table = pivot_table_create (N_("Vectors"));
+ pivot_dimension_create (table, PIVOT_AXIS_COLUMN, N_("Attributes"),
+ N_("Variable"), N_("Print Format"));
+ struct pivot_dimension *vector_dim = pivot_dimension_create (
+ table, PIVOT_AXIS_ROW, N_("Vector and Position"));
+ vector_dim->root->show_label = true;
- /* Print/write format, or print and write formats. */
- if (flags & DF_FORMATS)
+ for (size_t i = 0; i < n_vectors; i++)
{
- const struct fmt_spec *print = var_get_print_format (v);
- const struct fmt_spec *write = var_get_write_format (v);
+ const struct vector *vec = vectors[i];
- if (fmt_equal (print, write))
- {
- char str[FMT_STRING_LEN_MAX + 1];
- tab_joint_text_format (t, 1, r, 2, r, TAB_LEFT,
- _("Format: %s"), fmt_to_string (print, str));
- r++;
- }
- else
+ struct pivot_category *group = pivot_category_create_group__ (
+ vector_dim->root, pivot_value_new_user_text (
+ vector_get_name (vectors[i]), -1));
+
+ for (size_t j = 0; j < vector_get_n_vars (vec); j++)
{
- char str[FMT_STRING_LEN_MAX + 1];
- tab_joint_text_format (t, 1, r, 2, r, TAB_LEFT,
- _("Print Format: %s"),
- fmt_to_string (print, str));
- r++;
- tab_joint_text_format (t, 1, r, 2, r, TAB_LEFT,
- _("Write Format: %s"),
- fmt_to_string (write, str));
- r++;
+ struct variable *var = vector_get_var (vec, j);
+
+ int row = pivot_category_create_leaf (
+ group, pivot_value_new_integer (j + 1));
+
+ pivot_table_put2 (table, 0, row, pivot_value_new_variable (var));
+ char fmt_string[FMT_STRING_LEN_MAX + 1];
+ fmt_to_string (var_get_print_format (var), fmt_string);
+ pivot_table_put2 (table, 1, row,
+ pivot_value_new_user_text (fmt_string, -1));
}
}
-
- /* Measurement level, role, display width, alignment. */
- if (flags & DF_MISC)
- {
- enum var_role role = var_get_role (v);
- tab_joint_text_format (t, 1, r, 2, r, TAB_LEFT,
- _("Measure: %s"),
- measure_to_string (var_get_measure (v)));
- r++;
+ pivot_table_submit (table);
+
+ free (vectors);
+}
+
+/* Encoding analysis. */
+
+static const char *encoding_names[] = {
+ /* These encodings are from http://encoding.spec.whatwg.org/, as retrieved
+ February 2014. Encodings not supported by glibc and encodings relevant
+ only to HTML have been removed. */
+ "utf-8",
+ "windows-1252",
+ "iso-8859-2",
+ "iso-8859-3",
+ "iso-8859-4",
+ "iso-8859-5",
+ "iso-8859-6",
+ "iso-8859-7",
+ "iso-8859-8",
+ "iso-8859-10",
+ "iso-8859-13",
+ "iso-8859-14",
+ "iso-8859-16",
+ "macintosh",
+ "windows-874",
+ "windows-1250",
+ "windows-1251",
+ "windows-1253",
+ "windows-1254",
+ "windows-1255",
+ "windows-1256",
+ "windows-1257",
+ "windows-1258",
+ "koi8-r",
+ "koi8-u",
+ "ibm866",
+ "gb18030",
+ "big5",
+ "euc-jp",
+ "iso-2022-jp",
+ "shift_jis",
+ "euc-kr",
+
+ /* Added by user request. */
+ "ibm850",
+ "din_66003",
+};
+#define N_ENCODING_NAMES (sizeof encoding_names / sizeof *encoding_names)
+
+struct encoding
+ {
+ uint64_t encodings;
+ char **utf8_strings;
+ unsigned int hash;
+ };
+
+static char **
+recode_strings (struct pool *pool,
+ char **strings, bool *ids, size_t n,
+ const char *encoding)
+{
+ char **utf8_strings;
+ size_t i;
+
+ utf8_strings = pool_alloc (pool, n * sizeof *utf8_strings);
+ for (i = 0; i < n; i++)
+ {
+ struct substring utf8;
+ int error;
- if (role != ROLE_INPUT)
+ error = recode_pedantically ("UTF-8", encoding, ss_cstr (strings[i]),
+ pool, &utf8);
+ if (!error)
{
- tab_joint_text_format (t, 1, r, 2, r, TAB_LEFT,
- _("Role: %s"), var_role_to_string (role));
- r++;
+ ss_rtrim (&utf8, ss_cstr (" "));
+ utf8.string[utf8.length] = '\0';
+
+ if (ids[i] && !id_is_plausible (utf8.string, false))
+ error = EINVAL;
}
- tab_joint_text_format (t, 1, r, 2, r, TAB_LEFT,
- _("Display Alignment: %s"),
- alignment_to_string (var_get_alignment (v)));
- r++;
+ if (error)
+ return NULL;
- tab_joint_text_format (t, 1, r, 2, r, TAB_LEFT,
- _("Display Width: %d"),
- var_get_display_width (v));
- r++;
+ utf8_strings[i] = utf8.string;
}
-
- /* Missing values if any. */
- if (flags & DF_MISSING_VALUES && var_has_missing_values (v))
+
+ return utf8_strings;
+}
+
+static struct encoding *
+find_duplicate_encoding (struct encoding *encodings, size_t n_encodings,
+ char **utf8_strings, size_t n_strings,
+ unsigned int hash)
+{
+ struct encoding *e;
+
+ for (e = encodings; e < &encodings[n_encodings]; e++)
{
- const struct missing_values *mv = var_get_missing_values (v);
- char buf[128];
- char *cp;
- int cnt = 0;
int i;
- cp = stpcpy (buf, _("Missing Values: "));
+ if (e->hash != hash)
+ goto next_encoding;
- if (mv_has_range (mv))
- {
- double x, y;
- mv_get_range (mv, &x, &y);
- if (x == LOWEST)
- cp += sprintf (cp, "LOWEST THRU %g", y);
- else if (y == HIGHEST)
- cp += sprintf (cp, "%g THRU HIGHEST", x);
- else
- cp += sprintf (cp, "%g THRU %g", x, y);
- cnt++;
- }
- for (i = 0; i < mv_n_values (mv); i++)
- {
- const union value *value = mv_get_value (mv, i);
- if (cnt++ > 0)
- cp += sprintf (cp, "; ");
- if (var_is_numeric (v))
- cp += sprintf (cp, "%g", value->f);
- else
- {
- int width = var_get_width (v);
- int mv_width = MIN (width, MV_MAX_STRING);
-
- *cp++ = '"';
- memcpy (cp, value_str (value, width), mv_width);
- cp += mv_width;
- *cp++ = '"';
- *cp = '\0';
- }
- }
+ for (i = 0; i < n_strings; i++)
+ if (strcmp (utf8_strings[i], e->utf8_strings[i]))
+ goto next_encoding;
- tab_joint_text (t, 1, r, 2, r, TAB_LEFT, buf);
- r++;
+ return e;
+ next_encoding:;
}
- /* Value labels. */
- if (flags & DF_VALUE_LABELS && var_has_value_labels (v))
- {
- const struct val_labs *val_labs = var_get_value_labels (v);
- size_t n_labels = val_labs_count (val_labs);
- const struct val_lab **labels;
- int orig_r = r;
- size_t i;
+ return NULL;
+}
-#if 0
- tab_text (t, 1, r, TAB_LEFT, _("Value"));
- tab_text (t, 2, r, TAB_LEFT, _("Label"));
- r++;
-#endif
+static bool
+all_equal (const struct encoding *encodings, size_t n_encodings,
+ size_t string_idx)
+{
+ const char *s0;
+ size_t i;
- tab_hline (t, TAL_1, 1, 2, r);
+ s0 = encodings[0].utf8_strings[string_idx];
+ for (i = 1; i < n_encodings; i++)
+ if (strcmp (s0, encodings[i].utf8_strings[string_idx]))
+ return false;
- labels = val_labs_sorted (val_labs);
- for (i = 0; i < n_labels; i++)
- {
- const struct val_lab *vl = labels[i];
+ return true;
+}
- tab_value (t, 1, r, TAB_NONE, &vl->value, v, NULL);
- tab_text (t, 2, r, TAB_LEFT, val_lab_get_escaped_label (vl));
- r++;
- }
- free (labels);
+static int
+equal_prefix (const struct encoding *encodings, size_t n_encodings,
+ size_t string_idx)
+{
+ const char *s0;
+ size_t prefix;
+ size_t i;
- tab_vline (t, TAL_1, 2, orig_r, r - 1);
+ s0 = encodings[0].utf8_strings[string_idx];
+ prefix = strlen (s0);
+ for (i = 1; i < n_encodings; i++)
+ {
+ const char *si = encodings[i].utf8_strings[string_idx];
+ size_t j;
+
+ for (j = 0; j < prefix; j++)
+ if (s0[j] != si[j])
+ {
+ prefix = j;
+ if (!prefix)
+ return 0;
+ break;
+ }
}
- if (flags & (DF_ATTRIBUTES | DF_AT_ATTRIBUTES) && n_attrs)
+ while (prefix > 0 && s0[prefix - 1] != ' ')
+ prefix--;
+ return prefix;
+}
+
+static int
+equal_suffix (const struct encoding *encodings, size_t n_encodings,
+ size_t string_idx)
+{
+ const char *s0;
+ size_t s0_len;
+ size_t suffix;
+ size_t i;
+
+ s0 = encodings[0].utf8_strings[string_idx];
+ s0_len = strlen (s0);
+ suffix = s0_len;
+ for (i = 1; i < n_encodings; i++)
{
- tab_joint_text (t, 1, r, 2, r, TAB_LEFT, "Custom attributes:");
- r++;
+ const char *si = encodings[i].utf8_strings[string_idx];
+ size_t si_len = strlen (si);
+ size_t j;
- display_attributes (t, var_get_attributes (v), flags, 1, r);
- r += n_attrs;
+ if (si_len < suffix)
+ suffix = si_len;
+ for (j = 0; j < suffix; j++)
+ if (s0[s0_len - j - 1] != si[si_len - j - 1])
+ {
+ suffix = j;
+ if (!suffix)
+ return 0;
+ break;
+ }
}
- /* Draw a line below the last row of information on this variable. */
- tab_hline (t, TAL_1, 0, tab_nc (t) - 1, r);
-
- return r;
+ while (suffix > 0 && s0[s0_len - suffix] != ' ')
+ suffix--;
+ return suffix;
}
-/* Display a list of vectors. If SORTED is nonzero then they are
- sorted alphabetically. */
static void
-display_vectors (const struct dictionary *dict, int sorted)
+report_encodings (const struct file_handle *h, struct pool *pool,
+ char **titles, bool *ids, char **strings, size_t n_strings)
{
- const struct vector **vl;
- int i;
- struct tab_table *t;
- size_t nvec;
- size_t nrow;
- size_t row;
-
- nvec = dict_get_vector_cnt (dict);
- if (nvec == 0)
- {
- msg (SW, _("No vectors defined."));
- return;
- }
+ struct encoding encodings[N_ENCODING_NAMES];
+ size_t n_encodings, n_unique_strings;
- vl = xnmalloc (nvec, sizeof *vl);
- nrow = 0;
- for (i = 0; i < nvec; i++)
+ n_encodings = 0;
+ for (size_t i = 0; i < N_ENCODING_NAMES; i++)
{
- vl[i] = dict_get_vector (dict, i);
- nrow += vector_get_var_cnt (vl[i]);
- }
- if (sorted)
- qsort (vl, nvec, sizeof *vl, compare_vector_ptrs_by_name);
-
- t = tab_create (4, nrow + 1);
- tab_headers (t, 0, 0, 1, 0);
- tab_box (t, TAL_1, TAL_1, -1, -1, 0, 0, 3, nrow);
- tab_box (t, -1, -1, -1, TAL_1, 0, 0, 3, nrow);
- tab_hline (t, TAL_2, 0, 3, 1);
- tab_text (t, 0, 0, TAT_TITLE | TAB_LEFT, _("Vector"));
- tab_text (t, 1, 0, TAT_TITLE | TAB_LEFT, _("Position"));
- tab_text (t, 2, 0, TAT_TITLE | TAB_LEFT, _("Variable"));
- tab_text (t, 3, 0, TAT_TITLE | TAB_LEFT, _("Print Format"));
-
- row = 1;
- for (i = 0; i < nvec; i++)
- {
- const struct vector *vec = vl[i];
- size_t j;
+ char **utf8_strings;
+ struct encoding *e;
+ unsigned int hash;
- tab_joint_text (t, 0, row, 0, row + vector_get_var_cnt (vec) - 1,
- TAB_LEFT, vector_get_name (vl[i]));
+ utf8_strings = recode_strings (pool, strings, ids, n_strings,
+ encoding_names[i]);
+ if (!utf8_strings)
+ continue;
- for (j = 0; j < vector_get_var_cnt (vec); j++)
- {
- struct variable *var = vector_get_var (vec, j);
- char fmt_string[FMT_STRING_LEN_MAX + 1];
- fmt_to_string (var_get_print_format (var), fmt_string);
+ /* Hash utf8_strings. */
+ hash = 0;
+ for (size_t j = 0; j < n_strings; j++)
+ hash = hash_string (utf8_strings[j], hash);
- tab_text_format (t, 1, row, TAB_RIGHT, "%zu", j + 1);
- tab_text (t, 2, row, TAB_LEFT, var_get_name (var));
- tab_text (t, 3, row, TAB_LEFT, fmt_string);
- row++;
+ /* If there's a duplicate encoding, just mark it. */
+ e = find_duplicate_encoding (encodings, n_encodings,
+ utf8_strings, n_strings, hash);
+ if (e)
+ {
+ e->encodings |= UINT64_C (1) << i;
+ continue;
}
- tab_hline (t, TAL_1, 0, 3, row);
+
+ e = &encodings[n_encodings++];
+ e->encodings = UINT64_C (1) << i;
+ e->utf8_strings = utf8_strings;
+ e->hash = hash;
+ }
+ if (!n_encodings)
+ {
+ msg (SW, _("No valid encodings found."));
+ return;
}
- tab_submit (t);
+ /* Table of valid encodings. */
+ struct pivot_table *table = pivot_table_create__ (
+ pivot_value_new_text_format (N_("Usable encodings for %s."),
+ fh_get_name (h)), "Usable Encodings");
+ pivot_table_set_caption (
+ table, pivot_value_new_text_format (
+ N_("Encodings that can successfully read %s (by specifying the encoding "
+ "name on the GET command's ENCODING subcommand). Encodings that "
+ "yield identical text are listed together."),
+ fh_get_name (h)));
+
+ pivot_dimension_create (table, PIVOT_AXIS_COLUMN, N_("Encodings"),
+ N_("Encodings"));
+ struct pivot_dimension *number = pivot_dimension_create__ (
+ table, PIVOT_AXIS_ROW, pivot_value_new_user_text ("#", -1));
+ number->root->show_label = true;
+
+ for (size_t i = 0; i < n_encodings; i++)
+ {
+ struct string s = DS_EMPTY_INITIALIZER;
+ for (size_t j = 0; j < 64; j++)
+ if (encodings[i].encodings & (UINT64_C (1) << j))
+ ds_put_format (&s, "%s, ", encoding_names[j]);
+ ds_chomp (&s, ss_cstr (", "));
+
+ int row = pivot_category_create_leaf (number->root,
+ pivot_value_new_integer (i + 1));
+ pivot_table_put2 (
+ table, 0, row, pivot_value_new_user_text_nocopy (ds_steal_cstr (&s)));
+ }
+ pivot_table_submit (table);
+
+ n_unique_strings = 0;
+ for (size_t i = 0; i < n_strings; i++)
+ if (!all_equal (encodings, n_encodings, i))
+ n_unique_strings++;
+ if (!n_unique_strings)
+ return;
+
+ /* Table of alternative interpretations. */
+ table = pivot_table_create__ (
+ pivot_value_new_text_format (N_("%s Encoded Text Strings"),
+ fh_get_name (h)),
+ "Alternate Encoded Text Strings");
+ pivot_table_set_caption (
+ table, pivot_value_new_text (
+ N_("Text strings in the file dictionary that the previously listed "
+ "encodings interpret differently, along with the interpretations.")));
+
+ pivot_dimension_create (table, PIVOT_AXIS_COLUMN, N_("Text"), N_("Text"));
+
+ number = pivot_dimension_create__ (table, PIVOT_AXIS_ROW,
+ pivot_value_new_user_text ("#", -1));
+ number->root->show_label = true;
+ for (size_t i = 0; i < n_encodings; i++)
+ pivot_category_create_leaf (number->root,
+ pivot_value_new_integer (i + 1));
+
+ struct pivot_dimension *purpose = pivot_dimension_create (
+ table, PIVOT_AXIS_ROW, N_("Purpose"));
+ purpose->root->show_label = true;
+
+ for (size_t i = 0; i < n_strings; i++)
+ if (!all_equal (encodings, n_encodings, i))
+ {
+ int prefix = equal_prefix (encodings, n_encodings, i);
+ int suffix = equal_suffix (encodings, n_encodings, i);
+
+ int purpose_idx = pivot_category_create_leaf (
+ purpose->root, pivot_value_new_user_text (titles[i], -1));
+
+ for (size_t j = 0; j < n_encodings; j++)
+ {
+ const char *s = encodings[j].utf8_strings[i] + prefix;
+
+ if (prefix || suffix)
+ {
+ size_t len = strlen (s) - suffix;
+ struct string entry;
+
+ ds_init_empty (&entry);
+ if (prefix)
+ ds_put_cstr (&entry, "...");
+ ds_put_substring (&entry, ss_buffer (s, len));
+ if (suffix)
+ ds_put_cstr (&entry, "...");
+
+ pivot_table_put3 (table, 0, j, purpose_idx,
+ pivot_value_new_user_text_nocopy (
+ ds_steal_cstr (&entry)));
+ }
+ else
+ pivot_table_put3 (table, 0, j, purpose_idx,
+ pivot_value_new_user_text (s, -1));
+ }
+ }
- free (vl);
+ pivot_table_submit (table);
}