/* PSPP - a program for statistical analysis.
- Copyright (C) 2007, 2009, 2010, 2011, 2012 Free Software Foundation, Inc.
+ Copyright (C) 2007, 2009, 2010, 2011, 2012, 2013, 2016 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
const struct dictionary *dict; /*Dictionary of destination */
enum data_parser_type type; /* Type of data to parse. */
int skip_records; /* Records to skip before first real data. */
- casenumber max_cases; /* Max number of cases to read. */
- int percent_cases; /* Approximate percent of cases to read. */
struct field *fields; /* Fields to parse. */
size_t field_cnt; /* Number of fields. */
parser->type = DP_FIXED;
parser->skip_records = 0;
- parser->max_cases = -1;
- parser->percent_cases = 100;
parser->fields = NULL;
parser->field_cnt = 0;
parser->skip_records = initial_records_to_skip;
}
-/* Sets the maximum number of cases parsed by PARSER to
- MAX_CASES. The default is -1, meaning no limit. */
-void
-data_parser_set_case_limit (struct data_parser *parser, casenumber max_cases)
-{
- parser->max_cases = max_cases;
-}
-
-/* Sets the percentage of cases that PARSER should read from the
- input file to PERCENT_CASES. By default, all cases are
- read. */
-void
-data_parser_set_case_percent (struct data_parser *parser, int percent_cases)
-{
- assert (percent_cases >= 0 && percent_cases <= 100);
- parser->percent_cases = percent_cases;
-}
-
/* Returns true if PARSER is configured to allow cases to span
multiple records. */
bool
}
/* Limit cases. */
- if (parser->max_cases != -1 && parser->max_cases-- == 0)
- return false;
- if (parser->percent_cases < 100
- && dfm_get_percent_read (reader) >= parser->percent_cases)
- return false;
-
if (parser->type == DP_DELIMITED)
{
if (parser->span)
int *first_column, int *last_column, struct string *tmp,
struct substring *field)
{
+ size_t length_before_separators;
struct substring line, p;
+ bool quoted;
if (dfm_eof (reader))
return false;
}
*first_column = dfm_column_start (reader);
- if (ss_find_byte (parser->quotes, ss_first (p)) != SIZE_MAX)
+ quoted = ss_find_byte (parser->quotes, ss_first (p)) != SIZE_MAX;
+ if (quoted)
{
/* Quoted field. */
int quote = ss_get_byte (&p);
*field = ds_ss (tmp);
}
*last_column = *first_column + (ss_length (line) - ss_length (p));
-
- /* Skip trailing soft separator and a single hard separator
- if present. */
- if (!ss_is_empty (p))
- {
- size_t n_seps = ss_ltrim (&p, parser->soft_seps);
- if (!ss_is_empty (p)
- && ss_find_byte (parser->hard_seps, ss_first (p)) != SIZE_MAX)
- {
- ss_advance (&p, 1);
- n_seps++;
- }
- if (!n_seps)
- msg (DW, _("Missing delimiter following quoted string."));
- }
}
else
{
/* Regular field. */
ss_get_bytes (&p, ss_cspan (p, ds_ss (&parser->any_sep)), field);
*last_column = *first_column + ss_length (*field);
+ }
- if (!ss_ltrim (&p, parser->soft_seps) || ss_is_empty (p)
- || ss_find_byte (parser->hard_seps, p.string[0]) != SIZE_MAX)
- {
- /* Advance past a trailing hard separator,
- regardless of whether one actually existed. If
- we "skip" a delimiter that was not actually
- there, then we will return end-of-line on our
- next call, which is what we want. */
- dfm_forward_columns (reader, 1);
- }
+ /* Skip trailing soft separator and a single hard separator if present. */
+ length_before_separators = ss_length (p);
+ ss_ltrim (&p, parser->soft_seps);
+ if (!ss_is_empty (p)
+ && ss_find_byte (parser->hard_seps, ss_first (p)) != SIZE_MAX)
+ {
+ ss_advance (&p, 1);
+ ss_ltrim (&p, parser->soft_seps);
}
+ if (ss_is_empty (p))
+ dfm_forward_columns (reader, 1);
+ else if (quoted && length_before_separators == ss_length (p))
+ msg (DW, _("Missing delimiter following quoted string."));
dfm_forward_columns (reader, ss_length (line) - ss_length (p));
return true;
parse_delimited_span (const struct data_parser *parser,
struct dfm_reader *reader, struct ccase *c)
{
- const char *input_encoding = dfm_reader_get_encoding (reader);
const char *output_encoding = dict_get_encoding (parser->dict);
struct string tmp = DS_EMPTY_INITIALIZER;
struct field *f;
}
}
+ const char *input_encoding = dfm_reader_get_encoding (reader);
error = data_in (s, input_encoding, f->format.type,
case_data_rw_idx (c, f->case_idx),
fmt_var_width (&f->format), output_encoding);
parse_delimited_no_span (const struct data_parser *parser,
struct dfm_reader *reader, struct ccase *c)
{
- const char *input_encoding = dfm_reader_get_encoding (reader);
const char *output_encoding = dict_get_encoding (parser->dict);
struct string tmp = DS_EMPTY_INITIALIZER;
struct substring s;
goto exit;
}
+ const char *input_encoding = dfm_reader_get_encoding (reader);
error = data_in (s, input_encoding, f->format.type,
case_data_rw_idx (c, f->case_idx),
fmt_var_width (&f->format), output_encoding);