Details
struct GScanner
struct GScanner
{
/* unused fields */
gpointer user_data;
guint max_parse_errors;
/* g_scanner_error() increments this field */
guint parse_errors;
/* name of input stream, featured by the default message handler */
const gchar *input_name;
/* data pointer for derived structures */
gpointer derived_data;
/* link into the scanner configuration */
GScannerConfig *config;
/* fields filled in after g_scanner_get_next_token() */
GTokenType token;
GTokenValue value;
guint line;
guint position;
/* fields filled in after g_scanner_peek_next_token() */
GTokenType next_token;
GTokenValue next_value;
guint next_line;
guint next_position;
/* to be considered private */
GHashTable *symbol_table;
gint input_fd;
const gchar *text;
const gchar *text_end;
gchar *buffer;
guint scope_id;
/* handler function for _warn and _error */
GScannerMsgFunc msg_handler;
}; |
The data structure representing a lexical scanner.
You should set input_name after creating the scanner, since it is used
by the default message handler when displaying warnings and errors.
If you are scanning a file, the file name would be a good choice.
The user_data and
derived_data fields are not used.
If you need to associate extra data with the scanner you can place them here.
If you want to use your own message handler you can set the
msg_handler field. The type of the message
handler function is declared by GScannerMsgFunc.
g_scanner_new ()
Creates a new GScanner.
The config_templ structure specifies the initial settings of the scanner,
which are copied into the GScanner config field.
If you pass NULL then the default settings are used.
(See g_scanner_config_template in gscanner.c for the defaults.)
struct GScannerConfig
struct GScannerConfig
{
/* Character sets
*/
gchar *cset_skip_characters; /* default: " \t\n" */
gchar *cset_identifier_first;
gchar *cset_identifier_nth;
gchar *cpair_comment_single; /* default: "#\n" */
/* Should symbol lookup work case sensitive?
*/
guint case_sensitive : 1;
/* Boolean values to be adjusted "on the fly"
* to configure scanning behaviour.
*/
guint skip_comment_multi : 1; /* C like comment */
guint skip_comment_single : 1; /* single line comment */
guint scan_comment_multi : 1; /* scan multi line comments? */
guint scan_identifier : 1;
guint scan_identifier_1char : 1;
guint scan_identifier_NULL : 1;
guint scan_symbols : 1;
guint scan_binary : 1;
guint scan_octal : 1;
guint scan_float : 1;
guint scan_hex : 1; /* `0x0ff0' */
guint scan_hex_dollar : 1; /* `$0ff0' */
guint scan_string_sq : 1; /* string: 'anything' */
guint scan_string_dq : 1; /* string: "\\-escapes!\n" */
guint numbers_2_int : 1; /* bin, octal, hex => int */
guint int_2_float : 1; /* int => G_TOKEN_FLOAT? */
guint identifier_2_string : 1;
guint char_2_token : 1; /* return G_TOKEN_CHAR? */
guint symbol_2_token : 1;
guint scope_0_fallback : 1; /* try scope 0 on lookups? */
}; |
Specifies the GScanner settings.
cset_skip_characters specifies which characters
should be skipped by the scanner (the default is the whitespace characters:
space, tab, carriage-return and line-feed).
cset_identifier_first specifies the characters
which can start identifiers.
(the default is G_CSET_a_2_z, "_", and G_CSET_A_2_Z).
cset_identifier_nth specifies the characters
which can be used in identifiers, after the first character.
The default is G_CSET_a_2_z, "_0123456789", G_CSET_A_2_Z, G_CSET_LATINS,
G_CSET_LATINC.
cpair_comment_single specifies the characters
at the start and end of single-line comments. The default is "#\n" which
means that single-line comments start with a '#' and continue until a '\n'
(end of line).
case_sensitive specifies if symbols are
case sensitive.
The rest of the fields are flags which turn features on or off.
FIXME: should describe these.
g_scanner_input_file ()
void g_scanner_input_file (GScanner *scanner,
gint input_fd); |
Prepares to scan a file.
g_scanner_sync_file_offset ()
void g_scanner_sync_file_offset (GScanner *scanner); |
g_scanner_stat_mode ()
gint g_scanner_stat_mode (const gchar *filename); |
Gets the file attributes.
This is the st_mode field from the
stat structure. See the stat()
documentation.
g_scanner_input_text ()
Prepares to scan a text buffer.
g_scanner_peek_next_token ()
Gets the next token, without removing it from the input stream.
The token data is placed in the
next_token,
next_value,
next_line, and
next_position fields of the GScanner structure.
g_scanner_get_next_token ()
Gets the next token, removing it from the input stream.
The token data is placed in the
token,
value,
line, and
position fields of the GScanner structure.
g_scanner_cur_line ()
Gets the current line in the input stream (counting from 1).
g_scanner_cur_position ()
Gets the current position in the current line (counting from 0).
g_scanner_cur_token ()
Gets the current token type.
This is simply the token field in the GScanner
structure.
g_scanner_cur_value ()
Gets the current token value.
This is simply the value field in the GScanner
structure.
g_scanner_eof ()
Returns TRUE if the scanner has reached the end of the file or text buffer.
g_scanner_set_scope ()
Sets the current scope.
g_scanner_scope_add_symbol ()
Adds a symbol to the given scope.
g_scanner_scope_foreach_symbol ()
g_scanner_scope_lookup_symbol ()
g_scanner_scope_remove_symbol ()
void g_scanner_scope_remove_symbol (GScanner *scanner,
guint scope_id,
const gchar *symbol); |
g_scanner_freeze_symbol_table ()
void g_scanner_freeze_symbol_table (GScanner *scanner); |
g_scanner_thaw_symbol_table ()
void g_scanner_thaw_symbol_table (GScanner *scanner); |
g_scanner_lookup_symbol ()
g_scanner_warn ()
void g_scanner_warn (GScanner *scanner,
const gchar *format,
...); |
Outputs a warning message, via the GScanner message handler.
g_scanner_error ()
void g_scanner_error (GScanner *scanner,
const gchar *format,
...); |
Outputs an error message, via the GScanner message handler.
g_scanner_unexp_token ()
Outputs a message resulting from an unexpected token in the input stream.
FIXME: I don't understand the arguments here.
g_scanner_destroy ()
void g_scanner_destroy (GScanner *scanner); |
Frees all memory used by the GScanner.
enum GTokenType
typedef enum
{
G_TOKEN_EOF = 0,
G_TOKEN_LEFT_PAREN = '(',
G_TOKEN_RIGHT_PAREN = ')',
G_TOKEN_LEFT_CURLY = '{',
G_TOKEN_RIGHT_CURLY = '}',
G_TOKEN_LEFT_BRACE = '[',
G_TOKEN_RIGHT_BRACE = ']',
G_TOKEN_EQUAL_SIGN = '=',
G_TOKEN_COMMA = ',',
G_TOKEN_NONE = 256,
G_TOKEN_ERROR,
G_TOKEN_CHAR,
G_TOKEN_BINARY,
G_TOKEN_OCTAL,
G_TOKEN_INT,
G_TOKEN_HEX,
G_TOKEN_FLOAT,
G_TOKEN_STRING,
G_TOKEN_SYMBOL,
G_TOKEN_IDENTIFIER,
G_TOKEN_IDENTIFIER_NULL,
G_TOKEN_COMMENT_SINGLE,
G_TOKEN_COMMENT_MULTI,
G_TOKEN_LAST
} GTokenType; |
The possible types of token returned from each g_scanner_get_next_token() call.
union GTokenValue
union GTokenValue
{
gpointer v_symbol;
gchar *v_identifier;
gulong v_binary;
gulong v_octal;
gulong v_int;
gdouble v_float;
gulong v_hex;
gchar *v_string;
gchar *v_comment;
guchar v_char;
guint v_error;
}; |
A union holding the value of the token.
enum GErrorType
typedef enum
{
G_ERR_UNKNOWN,
G_ERR_UNEXP_EOF,
G_ERR_UNEXP_EOF_IN_STRING,
G_ERR_UNEXP_EOF_IN_COMMENT,
G_ERR_NON_DIGIT_IN_CONST,
G_ERR_DIGIT_RADIX,
G_ERR_FLOAT_RADIX,
G_ERR_FLOAT_MALFORMED
} GErrorType; |
The possible errors, used in the v_error field
of GTokenValue, when the token is a G_TOKEN_ERROR.
G_CSET_a_2_z
#define G_CSET_a_2_z "abcdefghijklmnopqrstuvwxyz" |
The set of lower-case ASCII alphabet characters.
Used for specifying valid identifier characters in GScannerConfig.
G_CSET_A_2_Z
#define G_CSET_A_2_Z "ABCDEFGHIJKLMNOPQRSTUVWXYZ" |
The set of upper-case ASCII alphabet characters.
Used for specifying valid identifier characters in GScannerConfig.
G_CSET_LATINC
Part of the set of extended characters in the Latin character sets.
FIXME: lower case?
Used for specifying valid identifier characters in GScannerConfig.
G_CSET_LATINS
Part of the set of extended characters in the Latin character sets.
FIXME: upper case?
Used for specifying valid identifier characters in GScannerConfig.
g_scanner_add_symbol()
#define g_scanner_add_symbol( scanner, symbol, value ) |
Adds a symbol to the default scope.
Deprecated in favour of g_scanner_scope_add_symbol().
g_scanner_remove_symbol()
#define g_scanner_remove_symbol( scanner, symbol ) |
Removes a symbol from the default scope.
Deprecated in favour of g_scanner_scope_remove_symbol().
g_scanner_foreach_symbol()
#define g_scanner_foreach_symbol( scanner, func, data ) |
Calls a function for each symbol in the default scope.
Deprecated in favour of g_scanner_scope_foreach_symbol().