X-Git-Url: https://pintos-os.org/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=src%2Fdata%2Fsys-file-writer.c;h=0aeb93a0a3137bacb0fff97a0d5bf1fcfac5cac7;hb=c3d6b68809037b48ff51dd693a0c8e75bcc30d7a;hp=5daea89dc5a908ac385a5f19dc8c2c62db7c0e20;hpb=cb586666724d5fcbdb658ce471b85484f0a7babe;p=pspp-builds.git diff --git a/src/data/sys-file-writer.c b/src/data/sys-file-writer.c index 5daea89d..0aeb93a0 100644 --- a/src/data/sys-file-writer.c +++ b/src/data/sys-file-writer.c @@ -1,5 +1,5 @@ /* PSPP - a program for statistical analysis. - Copyright (C) 1997-9, 2000, 2006, 2007, 2009 Free Software Foundation, Inc. + Copyright (C) 1997-9, 2000, 2006, 2007, 2009, 2010, 2011 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 @@ -16,8 +16,8 @@ #include -#include "sys-file-writer.h" -#include "sys-file-private.h" +#include "data/sys-file-writer.h" +#include "data/sys-file-private.h" #include #include @@ -26,32 +26,34 @@ #include #include -#include -#include -#include -#include -#include -#include -#include - -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -#include "minmax.h" -#include "unlocked-io.h" -#include "xalloc.h" +#include "data/attributes.h" +#include "data/case.h" +#include "data/casewriter-provider.h" +#include "data/casewriter.h" +#include "data/dictionary.h" +#include "data/file-handle-def.h" +#include "data/file-name.h" +#include "data/format.h" +#include "data/make-file.h" +#include "data/missing-values.h" +#include "data/mrset.h" +#include "data/settings.h" +#include "data/short-names.h" +#include "data/value-labels.h" +#include "data/variable.h" +#include "libpspp/float-format.h" +#include "libpspp/i18n.h" +#include "libpspp/integer-format.h" +#include "libpspp/message.h" +#include "libpspp/misc.h" +#include "libpspp/str.h" +#include "libpspp/string-array.h" +#include "libpspp/version.h" + +#include "gl/xmemdup0.h" +#include "gl/minmax.h" +#include "gl/unlocked-io.h" +#include "gl/xalloc.h" #include "gettext.h" #define _(msgid) gettext (msgid) @@ -95,10 +97,11 @@ struct sfm_writer static const struct casewriter_class sys_file_casewriter_class; static void write_header (struct sfm_writer *, const struct dictionary *); -static void write_variable (struct sfm_writer *, const struct variable *, const struct dictionary *); -static void write_value_labels (struct sfm_writer *, - struct variable *, int idx, const struct dictionary *); -static void write_integer_info_record (struct sfm_writer *); +static void write_variable (struct sfm_writer *, const struct variable *); +static void write_value_labels (struct sfm_writer *, struct variable *, + int idx); +static void write_integer_info_record (struct sfm_writer *, + const struct dictionary *); static void write_float_info_record (struct sfm_writer *); static void write_longvar_table (struct sfm_writer *w, @@ -113,6 +116,9 @@ static void write_vls_length_table (struct sfm_writer *w, static void write_long_string_value_labels (struct sfm_writer *, const struct dictionary *); +static void write_mrsets (struct sfm_writer *, const struct dictionary *, + bool pre_v14); + static void write_variable_display_parameters (struct sfm_writer *w, const struct dictionary *dict); @@ -127,6 +133,12 @@ static void write_int (struct sfm_writer *, int32_t); static inline void convert_double_to_output_format (double, uint8_t[8]); static void write_float (struct sfm_writer *, double); static void write_string (struct sfm_writer *, const char *, size_t); +static void write_utf8_string (struct sfm_writer *, const char *encoding, + const char *string, size_t width); +static void write_utf8_record (struct sfm_writer *, const char *encoding, + const struct string *content, int subtype); +static void write_string_record (struct sfm_writer *, + const struct substring content, int subtype); static void write_bytes (struct sfm_writer *, const void *, size_t); static void write_zeros (struct sfm_writer *, size_t); static void write_spaces (struct sfm_writer *, size_t); @@ -140,8 +152,8 @@ static void put_cmp_opcode (struct sfm_writer *, uint8_t); static void put_cmp_number (struct sfm_writer *, double); static void put_cmp_string (struct sfm_writer *, const void *, size_t); -bool write_error (const struct sfm_writer *); -bool close_writer (struct sfm_writer *); +static bool write_error (const struct sfm_writer *); +static bool close_writer (struct sfm_writer *); /* Returns default options for writing a system file. */ struct sfm_write_options @@ -155,8 +167,7 @@ sfm_writer_default_options (void) } /* Opens the system file designated by file handle FH for writing - cases from dictionary D according to the given OPTS. If - COMPRESS is nonzero, the system file will be compressed. + cases from dictionary D according to the given OPTS. No reference to D is retained, so it may be modified or destroyed at will after this function returns. D is not @@ -205,14 +216,14 @@ sfm_open_writer (struct file_handle *fh, struct dictionary *d, goto error; /* Create the file on disk. */ - mode = S_IRUSR | S_IRGRP | S_IROTH; + mode = 0444; if (opts.create_writeable) - mode |= S_IWUSR | S_IWGRP | S_IWOTH; + mode |= 0222; w->rf = replace_file_start (fh_get_file_name (fh), "wb", mode, &w->file, NULL); if (w->rf == NULL) { - msg (ME, _("Error opening \"%s\" for writing as a system file: %s."), + msg (ME, _("Error opening `%s' for writing as a system file: %s."), fh_get_file_name (fh), strerror (errno)); goto error; } @@ -223,7 +234,7 @@ sfm_open_writer (struct file_handle *fh, struct dictionary *d, /* Write basic variable info. */ short_names_assign (d); for (i = 0; i < dict_get_var_cnt (d); i++) - write_variable (w, dict_get_var (d, i), d); + write_variable (w, dict_get_var (d, i)); /* Write out value labels. */ idx = 0; @@ -231,16 +242,18 @@ sfm_open_writer (struct file_handle *fh, struct dictionary *d, { struct variable *v = dict_get_var (d, i); - write_value_labels (w, v, idx, d); + write_value_labels (w, v, idx); idx += sfm_width_to_octs (var_get_width (v)); } - if (dict_get_documents (d) != NULL) + if (dict_get_document_line_cnt (d) > 0) write_documents (w, d); - write_integer_info_record (w); + write_integer_info_record (w, d); write_float_info_record (w); + write_mrsets (w, d, true); + write_variable_display_parameters (w, d); if (opts.version >= 3) @@ -254,6 +267,8 @@ sfm_open_writer (struct file_handle *fh, struct dictionary *d, write_data_file_attributes (w, d); write_variable_attributes (w, d); + write_mrsets (w, d, false); + write_encoding_record (w, d); /* Write end-of-headers record. */ @@ -261,10 +276,7 @@ sfm_open_writer (struct file_handle *fh, struct dictionary *d, write_int (w, 0); if (write_error (w)) - { - close_writer (w); - return NULL; - } + goto error; return casewriter_create (dict_get_proto (d), &sys_file_casewriter_class, w); @@ -378,7 +390,7 @@ write_header (struct sfm_writer *w, const struct dictionary *d) file_label = dict_get_label (d); if (file_label == NULL) file_label = ""; - write_string (w, file_label, 64); + write_utf8_string (w, dict_get_encoding (d), file_label, 64); /* Padding. */ write_zeros (w, 3); @@ -421,11 +433,12 @@ write_variable_continuation_records (struct sfm_writer *w, int width) /* Write the variable record(s) for variable V to system file W. */ static void -write_variable (struct sfm_writer *w, const struct variable *v, const struct dictionary *dict) +write_variable (struct sfm_writer *w, const struct variable *v) { int width = var_get_width (v); int segment_cnt = sfm_width_to_segments (width); int seg0_width = sfm_segment_alloc_width (width, 0); + const char *encoding = var_get_encoding (v); struct missing_values mv; int i; @@ -456,12 +469,12 @@ write_variable (struct sfm_writer *w, const struct variable *v, const struct dic /* Short name. The full name is in a translation table written separately. */ - write_string (w, var_get_short_name (v, 0), 8); + write_utf8_string (w, encoding, var_get_short_name (v, 0), 8); /* Value label. */ if (var_has_label (v)) { - char *label = recode_string (dict_get_encoding (dict), UTF8, var_get_label (v), -1); + char *label = recode_string (encoding, UTF8, var_get_label (v), -1); size_t label_len = MIN (strlen (label), 255); size_t padded_len = ROUND_UP (label_len, 4); write_int (w, label_len); @@ -494,7 +507,7 @@ write_variable (struct sfm_writer *w, const struct variable *v, const struct dic write_int (w, 0); /* No missing values. */ write_format (w, fmt, seg_width); /* Print format. */ write_format (w, fmt, seg_width); /* Write format. */ - write_string (w, var_get_short_name (v, i), 8); + write_utf8_string (w, encoding, var_get_short_name (v, i), 8); write_variable_continuation_records (w, seg_width); } @@ -508,7 +521,7 @@ write_variable (struct sfm_writer *w, const struct variable *v, const struct dic Value labels for long string variables are written separately, by write_long_string_value_labels. */ static void -write_value_labels (struct sfm_writer *w, struct variable *v, int idx, const struct dictionary *dict) +write_value_labels (struct sfm_writer *w, struct variable *v, int idx) { const struct val_labs *val_labs; const struct val_lab **labels; @@ -527,7 +540,8 @@ write_value_labels (struct sfm_writer *w, struct variable *v, int idx, const str for (i = 0; i < n_labels; i++) { const struct val_lab *vl = labels[i]; - char *label = recode_string (dict_get_encoding (dict), UTF8, val_lab_get_label (vl), -1); + char *label = recode_string (var_get_encoding (v), UTF8, + val_lab_get_escaped_label (vl), -1); uint8_t len = MIN (strlen (label), 255); write_value (w, val_lab_get_value (vl), var_get_width (v)); @@ -548,11 +562,22 @@ write_value_labels (struct sfm_writer *w, struct variable *v, int idx, const str static void write_documents (struct sfm_writer *w, const struct dictionary *d) { - size_t line_cnt = dict_get_document_line_cnt (d); + const struct string_array *docs = dict_get_documents (d); + const char *enc = dict_get_encoding (d); + size_t i; write_int (w, 6); /* Record type. */ - write_int (w, line_cnt); - write_bytes (w, dict_get_documents (d), line_cnt * DOC_LINE_LENGTH); + write_int (w, docs->n); + for (i = 0; i < docs->n; i++) + { + char *s = recode_string (enc, "UTF-8", docs->strings[i], -1); + size_t s_len = strlen (s); + size_t write_len = MIN (s_len, DOC_LINE_LENGTH); + + write_bytes (w, s, write_len); + write_spaces (w, DOC_LINE_LENGTH - write_len); + free (s); + } } static void @@ -568,31 +593,20 @@ put_attrset (struct string *string, const struct attrset *attrs) size_t j; ds_put_cstr (string, attribute_get_name (attr)); - ds_put_char (string, '('); + ds_put_byte (string, '('); for (j = 0; j < n_values; j++) ds_put_format (string, "'%s'\n", attribute_get_value (attr, j)); - ds_put_char (string, ')'); + ds_put_byte (string, ')'); } } -static void -write_attribute_record (struct sfm_writer *w, const struct string *content, - int subtype) -{ - write_int (w, 7); - write_int (w, subtype); - write_int (w, 1); - write_int (w, ds_length (content)); - write_bytes (w, ds_data (content), ds_length (content)); -} - static void write_data_file_attributes (struct sfm_writer *w, const struct dictionary *d) { struct string s = DS_EMPTY_INITIALIZER; put_attrset (&s, dict_get_attributes (d)); - write_attribute_record (w, &s, 17); + write_utf8_record (w, dict_get_encoding (d), &s, 17); ds_destroy (&s); } @@ -611,13 +625,88 @@ write_variable_attributes (struct sfm_writer *w, const struct dictionary *d) if (attrset_count (attrs)) { if (n_attrsets++) - ds_put_char (&s, '/'); - ds_put_format (&s, "%s:", var_get_short_name (v, 0)); + ds_put_byte (&s, '/'); + ds_put_format (&s, "%s:", var_get_name (v)); put_attrset (&s, attrs); } } - if (n_attrsets) - write_attribute_record (w, &s, 18); + if (n_attrsets) + write_utf8_record (w, dict_get_encoding (d), &s, 18); + ds_destroy (&s); +} + +/* Write multiple response sets. If PRE_V14 is true, writes sets supported by + SPSS before release 14, otherwise writes sets supported only by later + versions. */ +static void +write_mrsets (struct sfm_writer *w, const struct dictionary *dict, + bool pre_v14) +{ + const char *encoding = dict_get_encoding (dict); + struct string s = DS_EMPTY_INITIALIZER; + size_t n_mrsets; + size_t i; + + n_mrsets = dict_get_n_mrsets (dict); + if (n_mrsets == 0) + return; + + for (i = 0; i < n_mrsets; i++) + { + const struct mrset *mrset = dict_get_mrset (dict, i); + char *name; + size_t j; + + if ((mrset->type != MRSET_MD || mrset->cat_source != MRSET_COUNTEDVALUES) + != pre_v14) + continue; + + name = recode_string (encoding, "UTF-8", mrset->name, -1); + ds_put_format (&s, "%s=", name); + free (name); + + if (mrset->type == MRSET_MD) + { + char *counted; + + if (mrset->cat_source == MRSET_COUNTEDVALUES) + ds_put_format (&s, "E %d ", mrset->label_from_var_label ? 11 : 1); + else + ds_put_byte (&s, 'D'); + + if (mrset->width == 0) + counted = xasprintf ("%.0f", mrset->counted.f); + else + counted = xmemdup0 (value_str (&mrset->counted, mrset->width), + mrset->width); + ds_put_format (&s, "%zu %s", strlen (counted), counted); + free (counted); + } + else + ds_put_byte (&s, 'C'); + ds_put_byte (&s, ' '); + + if (mrset->label && !mrset->label_from_var_label) + { + char *label = recode_string (encoding, "UTF-8", mrset->label, -1); + ds_put_format (&s, "%zu %s", strlen (label), label); + free (label); + } + else + ds_put_cstr (&s, "0 "); + + for (j = 0; j < mrset->n_vars; j++) + { + const char *short_name_utf8 = var_get_short_name (mrset->vars[j], 0); + char *short_name = recode_string (encoding, "UTF-8", + short_name_utf8, -1); + ds_put_format (&s, " %s", short_name); + free (short_name); + } + ds_put_byte (&s, '\n'); + } + + write_string_record (w, ds_ss (&s), pre_v14 ? 7 : 19); ds_destroy (&s); } @@ -674,13 +763,7 @@ write_vls_length_table (struct sfm_writer *w, var_get_short_name (v, 0), var_get_width (v), 0); } if (!ds_is_empty (&map)) - { - write_int (w, 7); /* Record type. */ - write_int (w, 14); /* Record subtype. */ - write_int (w, 1); /* Data item (char) size. */ - write_int (w, ds_length (&map)); /* Number of data items. */ - write_bytes (w, ds_data (&map), ds_length (&map)); - } + write_utf8_record (w, dict_get_encoding (dict), &map, 14); ds_destroy (&map); } @@ -699,16 +782,22 @@ write_long_string_value_labels (struct sfm_writer *w, { struct variable *var = dict_get_var (dict, i); const struct val_labs *val_labs = var_get_value_labels (var); + const char *encoding = var_get_encoding (var); int width = var_get_width (var); const struct val_lab *val_lab; if (val_labs_count (val_labs) == 0 || width < 9) continue; - size += 12 + strlen (var_get_name (var)); + size += 12; + size += recode_string_len (encoding, "UTF-8", var_get_name (var), -1); for (val_lab = val_labs_first (val_labs); val_lab != NULL; val_lab = val_labs_next (val_labs, val_lab)) - size += 8 + width + strlen (val_lab_get_label (val_lab)); + { + size += 8 + width; + size += recode_string_len (encoding, "UTF-8", + val_lab_get_escaped_label (val_lab), -1); + } } if (size == 0) return; @@ -723,28 +812,37 @@ write_long_string_value_labels (struct sfm_writer *w, { struct variable *var = dict_get_var (dict, i); const struct val_labs *val_labs = var_get_value_labels (var); - const char *var_name = var_get_name (var); + const char *encoding = var_get_encoding (var); int width = var_get_width (var); const struct val_lab *val_lab; + char *var_name; if (val_labs_count (val_labs) == 0 || width < 9) continue; + var_name = recode_string (encoding, "UTF-8", var_get_name (var), -1); write_int (w, strlen (var_name)); write_bytes (w, var_name, strlen (var_name)); + free (var_name); + write_int (w, width); write_int (w, val_labs_count (val_labs)); for (val_lab = val_labs_first (val_labs); val_lab != NULL; val_lab = val_labs_next (val_labs, val_lab)) { - const char *label = val_lab_get_label (val_lab); - size_t label_length = strlen (label); + char *label; + size_t len; write_int (w, width); write_bytes (w, value_str (val_lab_get_value (val_lab), width), width); - write_int (w, label_length); - write_bytes (w, label, label_length); + + label = recode_string (var_get_encoding (var), "UTF-8", + val_lab_get_escaped_label (val_lab), -1); + len = strlen (label); + write_int (w, len); + write_bytes (w, label, len); + free (label); } } assert (ftello (w->file) == start + size); @@ -754,19 +852,18 @@ static void write_encoding_record (struct sfm_writer *w, const struct dictionary *d) { - const char *enc = dict_get_encoding (d); - - if ( NULL == enc) - return; - - write_int (w, 7); /* Record type. */ - write_int (w, 20); /* Record subtype. */ - write_int (w, 1); /* Data item (char) size. */ - write_int (w, strlen (enc)); /* Number of data items. */ - write_string (w, enc, strlen (enc)); + /* IANA says "...character set names may be up to 40 characters taken + from the printable characters of US-ASCII," so character set names + don't need to be recoded to be in UTF-8. + + We convert encoding names to uppercase because SPSS writes encoding + names in uppercase. */ + char *encoding = xstrdup (dict_get_encoding (d)); + str_uppercase (encoding); + write_string_record (w, ss_cstr (encoding), 20); + free (encoding); } - /* Writes the long variable name table. */ static void write_longvar_table (struct sfm_writer *w, const struct dictionary *dict) @@ -778,30 +875,23 @@ write_longvar_table (struct sfm_writer *w, const struct dictionary *dict) for (i = 0; i < dict_get_var_cnt (dict); i++) { struct variable *v = dict_get_var (dict, i); - char *longname = recode_string (dict_get_encoding (dict), UTF8, var_get_name (v), -1); - if (i) - ds_put_char (&map, '\t'); + ds_put_byte (&map, '\t'); ds_put_format (&map, "%s=%s", - var_get_short_name (v, 0), longname); - free (longname); + var_get_short_name (v, 0), var_get_name (v)); } - - write_int (w, 7); /* Record type. */ - write_int (w, 13); /* Record subtype. */ - write_int (w, 1); /* Data item (char) size. */ - write_int (w, ds_length (&map)); /* Number of data items. */ - write_bytes (w, ds_data (&map), ds_length (&map)); - + write_utf8_record (w, dict_get_encoding (dict), &map, 13); ds_destroy (&map); } /* Write integer information record. */ static void -write_integer_info_record (struct sfm_writer *w) +write_integer_info_record (struct sfm_writer *w, + const struct dictionary *d) { int version_component[3]; int float_format; + int codepage; /* Parse the version string. */ memset (version_component, 0, sizeof version_component); @@ -819,6 +909,16 @@ write_integer_info_record (struct sfm_writer *w) else abort (); + /* Choose codepage. */ + codepage = sys_get_codepage_from_encoding (dict_get_encoding (d)); + if (codepage == 0) + { + /* Default to "7-bit ASCII" if the codepage number is unknown, because + many files use this codepage number regardless of their actual + encoding. */ + codepage = 2; + } + /* Write record. */ write_int (w, 7); /* Record type. */ write_int (w, 3); /* Record subtype. */ @@ -831,7 +931,7 @@ write_integer_info_record (struct sfm_writer *w) write_int (w, float_format); write_int (w, 1); /* Compression code. */ write_int (w, INTEGER_NATIVE == INTEGER_MSB_FIRST ? 1 : 2); - write_int (w, 2); /* 7-bit ASCII. */ + write_int (w, codepage); } /* Write floating-point information record. */ @@ -881,7 +981,7 @@ sys_file_casewriter_destroy (struct casewriter *writer, void *w_) } /* Returns true if an I/O error has occurred on WRITER, false otherwise. */ -bool +static bool write_error (const struct sfm_writer *writer) { return ferror (writer->file); @@ -889,7 +989,7 @@ write_error (const struct sfm_writer *writer) /* Closes a system file after we're done with it. Returns true if successful, false if an I/O error occurred. */ -bool +static bool close_writer (struct sfm_writer *w) { bool ok; @@ -910,7 +1010,7 @@ close_writer (struct sfm_writer *w) /* Seek back to the beginning and update the number of cases. This is just a courtesy to later readers, so there's no need to check return values or report errors. */ - if (ok && w->case_cnt <= INT32_MAX && !fseek (w->file, 80, SEEK_SET)) + if (ok && w->case_cnt <= INT32_MAX && !fseeko (w->file, 80, SEEK_SET)) { write_int (w, w->case_cnt); clearerr (w->file); @@ -920,7 +1020,7 @@ close_writer (struct sfm_writer *w) ok = false; if (!ok) - msg (ME, _("An I/O error occurred writing system file \"%s\"."), + msg (ME, _("An I/O error occurred writing system file `%s'."), fh_get_file_name (w->fh)); if (ok ? !replace_file_commit (w->rf) : !replace_file_abort (w->rf)) @@ -1127,9 +1227,9 @@ write_value (struct sfm_writer *w, const union value *value, int width) } } -/* Writes null-terminated STRING in a field of the given WIDTH to - W. If STRING is longer than WIDTH, it is truncated; if WIDTH - is narrowed, it is padded on the right with spaces. */ +/* Writes null-terminated STRING in a field of the given WIDTH to W. If STRING + is longer than WIDTH, it is truncated; if STRING is shorter than WIDTH, it + is padded on the right with spaces. */ static void write_string (struct sfm_writer *w, const char *string, size_t width) { @@ -1140,6 +1240,44 @@ write_string (struct sfm_writer *w, const char *string, size_t width) putc (' ', w->file); } +/* Recodes null-terminated UTF-8 encoded STRING into ENCODING, and writes the + recoded version in a field of the given WIDTH to W. The string is truncated + or padded on the right with spaces to exactly WIDTH bytes. */ +static void +write_utf8_string (struct sfm_writer *w, const char *encoding, + const char *string, size_t width) +{ + char *s = recode_string (encoding, "UTF-8", string, -1); + write_string (w, s, width); + free (s); +} + +/* Writes a record with type 7, subtype SUBTYPE that contains CONTENT recoded + from UTF-8 encoded into ENCODING. */ +static void +write_utf8_record (struct sfm_writer *w, const char *encoding, + const struct string *content, int subtype) +{ + struct substring s; + + s = recode_substring_pool (encoding, "UTF-8", ds_ss (content), NULL); + write_string_record (w, s, subtype); + ss_dealloc (&s); +} + +/* Writes a record with type 7, subtype SUBTYPE that contains the string + CONTENT. */ +static void +write_string_record (struct sfm_writer *w, + const struct substring content, int subtype) +{ + write_int (w, 7); + write_int (w, subtype); + write_int (w, 1); + write_int (w, ss_length (content)); + write_bytes (w, ss_data (content), ss_length (content)); +} + /* Writes SIZE bytes of DATA to W's output file. */ static void write_bytes (struct sfm_writer *w, const void *data, size_t size)