X-Git-Url: https://pintos-os.org/cgi-bin/gitweb.cgi?a=blobdiff_plain;f=src%2Fdata%2Fsys-file-writer.c;h=b5e03eccc336728f6b39a820c1573e0f9fd435a1;hb=5188e51e07df8f0c1dc34ee70d4cb2d3d53a6016;hp=ac5f4c903ed0266a4e50bc87a43c9b29e177cf9e;hpb=d4135192829ae81ec367c3e42289822e2629aeef;p=pspp-builds.git diff --git a/src/data/sys-file-writer.c b/src/data/sys-file-writer.c index ac5f4c90..b5e03ecc 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, 2010 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 @@ -47,6 +47,7 @@ #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" @@ -131,6 +132,10 @@ 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_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); @@ -215,7 +220,7 @@ sfm_open_writer (struct file_handle *fh, struct dictionary *d, &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; } @@ -238,7 +243,7 @@ sfm_open_writer (struct file_handle *fh, struct dictionary *d, 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); @@ -382,7 +387,7 @@ 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); @@ -430,6 +435,7 @@ 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; @@ -460,12 +466,12 @@ write_variable (struct sfm_writer *w, const struct variable *v) /* 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 (var_get_encoding (v), 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); @@ -498,7 +504,7 @@ write_variable (struct sfm_writer *w, const struct variable *v) 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); } @@ -531,7 +537,8 @@ write_value_labels (struct sfm_writer *w, struct variable *v, int idx) for (i = 0; i < n_labels; i++) { const struct val_lab *vl = labels[i]; - char *label = recode_string (var_get_encoding (v), 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)); @@ -552,11 +559,22 @@ write_value_labels (struct sfm_writer *w, struct variable *v, int idx) 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 @@ -572,31 +590,20 @@ 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); } @@ -615,13 +622,13 @@ write_variable_attributes (struct sfm_writer *w, const struct dictionary *d) if (attrset_count (attrs)) { if (n_attrsets++) - ds_put_char (&s, '/'); + ds_put_byte (&s, '/'); ds_put_format (&s, "%s:", var_get_short_name (v, 0)); 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); } @@ -658,7 +665,7 @@ write_mrsets (struct sfm_writer *w, const struct dictionary *dict, if (mrset->cat_source == MRSET_COUNTEDVALUES) ds_put_format (&s, "E %d ", mrset->label_from_var_label ? 11 : 1); else - ds_put_char (&s, 'D'); + ds_put_byte (&s, 'D'); if (mrset->width == 0) counted = xasprintf ("%.0f", mrset->counted.f); @@ -669,17 +676,17 @@ write_mrsets (struct sfm_writer *w, const struct dictionary *dict, free (counted); } else - ds_put_char (&s, 'C'); - ds_put_char (&s, ' '); + ds_put_byte (&s, 'C'); + ds_put_byte (&s, ' '); label = mrset->label && !mrset->label_from_var_label ? mrset->label : ""; ds_put_format (&s, "%zu %s", strlen (label), label); for (j = 0; j < mrset->n_vars; j++) ds_put_format (&s, " %s", var_get_short_name (mrset->vars[j], 0)); - ds_put_char (&s, '\n'); + ds_put_byte (&s, '\n'); } - write_attribute_record (w, &s, 7); + write_utf8_record (w, dict_get_encoding (dict), &s, pre_v14 ? 7 : 19); ds_destroy (&s); } @@ -736,13 +743,7 @@ 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); } @@ -761,16 +762,22 @@ 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; @@ -785,28 +792,37 @@ 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); @@ -824,6 +840,10 @@ write_encoding_record (struct sfm_writer *w, write_int (w, 7); /* Record type. */ write_int (w, 20); /* Record subtype. */ write_int (w, 1); /* Data item (char) size. */ + + /* 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. */ write_int (w, strlen (enc)); /* Number of data items. */ write_string (w, enc, strlen (enc)); } @@ -840,21 +860,12 @@ 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); } @@ -982,7 +993,7 @@ 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)) @@ -1189,9 +1200,9 @@ 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) { @@ -1202,6 +1213,35 @@ 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_int (w, 7); + write_int (w, subtype); + write_int (w, 1); + write_int (w, ss_length (s)); + write_bytes (w, ss_data (s), ss_length (s)); + ss_dealloc (&s); +} + /* Writes SIZE bytes of DATA to W's output file. */ static void write_bytes (struct sfm_writer *w, const void *data, size_t size)