#include <errno.h>
#include "glib.h"
-
#include "glibintl.h"
+#include "galias.h"
GQuark
g_markup_error_quark (void)
{
- static GQuark error_quark = 0;
-
- if (error_quark == 0)
- error_quark = g_quark_from_static_string ("g-markup-error-quark");
-
- return error_quark;
+ return g_quark_from_static_string ("g-markup-error-quark");
}
typedef enum
(*context->parser->error) (context, error, context->user_data);
}
+static void set_error (GMarkupParseContext *context,
+ GError **error,
+ GMarkupError code,
+ const gchar *format,
+ ...) G_GNUC_PRINTF (4, 5);
+
static void
set_error (GMarkupParseContext *context,
GError **error,
s = g_strdup_vprintf (format, args);
va_end (args);
- tmp_error = g_error_new (G_MARKUP_ERROR,
- code,
- _("Error on line %d char %d: %s"),
- context->line_number,
- context->char_number,
- s);
-
+ tmp_error = g_error_new_literal (G_MARKUP_ERROR, code, s);
g_free (s);
+ g_prefix_error (&tmp_error,
+ _("Error on line %d char %d: "),
+ context->line_number,
+ context->char_number);
+
mark_error (context, tmp_error);
g_propagate_error (error, tmp_error);
}
+static void
+propagate_error (GMarkupParseContext *context,
+ GError **dest,
+ GError *src)
+{
+ if (context->flags & G_MARKUP_PREFIX_ERROR_POSITION)
+ g_prefix_error (&src,
+ _("Error on line %d char %d: "),
+ context->line_number,
+ context->char_number);
+
+ mark_error (context, src);
+
+ g_propagate_error (dest, src);
+}
+
+/* To make these faster, we first use the ascii-only tests, then check
+ * for the usual non-alnum name-end chars, and only then call the
+ * expensive unicode stuff. Nobody uses non-ascii in XML tag/attribute
+ * names, so this is a reasonable hack that virtually always avoids
+ * the guniprop call.
+ */
+#define IS_COMMON_NAME_END_CHAR(c) \
+ ((c) == '=' || (c) == '/' || (c) == '>' || (c) == ' ')
+
static gboolean
-is_name_start_char (gunichar c)
+is_name_start_char (const gchar *p)
{
- if (g_unichar_isalpha (c) ||
- c == '_' ||
- c == ':')
+ if (g_ascii_isalpha (*p) ||
+ (!IS_COMMON_NAME_END_CHAR (*p) &&
+ (*p == '_' ||
+ *p == ':' ||
+ g_unichar_isalpha (g_utf8_get_char (p)))))
return TRUE;
else
return FALSE;
}
static gboolean
-is_name_char (gunichar c)
+is_name_char (const gchar *p)
{
- if (g_unichar_isalnum (c) ||
- c == '.' ||
- c == '-' ||
- c == '_' ||
- c == ':')
+ if (g_ascii_isalnum (*p) ||
+ (!IS_COMMON_NAME_END_CHAR (*p) &&
+ (*p == '.' ||
+ *p == '-' ||
+ *p == '_' ||
+ *p == ':' ||
+ g_unichar_isalpha (g_utf8_get_char (p)))))
return TRUE;
else
return FALSE;
char_str (gunichar c,
gchar *buf)
{
- memset (buf, 0, 7);
+ memset (buf, 0, 8);
g_unichar_to_utf8 (c, buf);
return buf;
}
USTATE_AFTER_CHARREF_HASH
} UnescapeState;
-static gboolean
-unescape_text (GMarkupParseContext *context,
- const gchar *text,
- const gchar *text_end,
- gchar **unescaped,
- GError **error)
+typedef struct
{
-#define MAX_ENT_LEN 5
+ GMarkupParseContext *context;
GString *str;
- const gchar *p;
UnescapeState state;
+ const gchar *text;
+ const gchar *text_end;
+ const gchar *entity_start;
+} UnescapeContext;
+
+static const gchar*
+unescape_text_state_inside_text (UnescapeContext *ucontext,
+ const gchar *p,
+ GError **error)
+{
const gchar *start;
gboolean normalize_attribute;
- str = g_string_new (NULL);
-
- if (context->state == STATE_INSIDE_ATTRIBUTE_VALUE_SQ ||
- context->state == STATE_INSIDE_ATTRIBUTE_VALUE_DQ)
+ if (ucontext->context->state == STATE_INSIDE_ATTRIBUTE_VALUE_SQ ||
+ ucontext->context->state == STATE_INSIDE_ATTRIBUTE_VALUE_DQ)
normalize_attribute = TRUE;
else
normalize_attribute = FALSE;
- state = USTATE_INSIDE_TEXT;
- p = text;
start = p;
+
+ while (p != ucontext->text_end)
+ {
+ if (*p == '&')
+ {
+ break;
+ }
+ else if (normalize_attribute && (*p == '\t' || *p == '\n'))
+ {
+ g_string_append_len (ucontext->str, start, p - start);
+ g_string_append_c (ucontext->str, ' ');
+ p = g_utf8_next_char (p);
+ start = p;
+ }
+ else if (*p == '\r')
+ {
+ g_string_append_len (ucontext->str, start, p - start);
+ g_string_append_c (ucontext->str, normalize_attribute ? ' ' : '\n');
+ p = g_utf8_next_char (p);
+ if (p != ucontext->text_end && *p == '\n')
+ p = g_utf8_next_char (p);
+ start = p;
+ }
+ else
+ p = g_utf8_next_char (p);
+ }
+
+ if (p != start)
+ g_string_append_len (ucontext->str, start, p - start);
+
+ if (p != ucontext->text_end && *p == '&')
+ {
+ p = g_utf8_next_char (p);
+ ucontext->state = USTATE_AFTER_AMPERSAND;
+ }
+
+ return p;
+}
+
+static const gchar*
+unescape_text_state_after_ampersand (UnescapeContext *ucontext,
+ const gchar *p,
+ GError **error)
+{
+ ucontext->entity_start = NULL;
+
+ if (*p == '#')
+ {
+ p = g_utf8_next_char (p);
+
+ ucontext->entity_start = p;
+ ucontext->state = USTATE_AFTER_CHARREF_HASH;
+ }
+ else if (!is_name_start_char (p))
+ {
+ if (*p == ';')
+ {
+ set_unescape_error (ucontext->context, error,
+ p, ucontext->text_end,
+ G_MARKUP_ERROR_PARSE,
+ _("Empty entity '&;' seen; valid "
+ "entities are: & " < > '"));
+ }
+ else
+ {
+ gchar buf[8];
+
+ set_unescape_error (ucontext->context, error,
+ p, ucontext->text_end,
+ G_MARKUP_ERROR_PARSE,
+ _("Character '%s' is not valid at "
+ "the start of an entity name; "
+ "the & character begins an entity; "
+ "if this ampersand isn't supposed "
+ "to be an entity, escape it as "
+ "&"),
+ utf8_str (p, buf));
+ }
+ }
+ else
+ {
+ ucontext->entity_start = p;
+ ucontext->state = USTATE_INSIDE_ENTITY_NAME;
+ }
+
+ return p;
+}
+
+static const gchar*
+unescape_text_state_inside_entity_name (UnescapeContext *ucontext,
+ const gchar *p,
+ GError **error)
+{
+ while (p != ucontext->text_end)
+ {
+ if (*p == ';')
+ break;
+ else if (!is_name_char (p))
+ {
+ gchar ubuf[8];
+
+ set_unescape_error (ucontext->context, error,
+ p, ucontext->text_end,
+ G_MARKUP_ERROR_PARSE,
+ _("Character '%s' is not valid "
+ "inside an entity name"),
+ utf8_str (p, ubuf));
+ break;
+ }
+
+ p = g_utf8_next_char (p);
+ }
+
+ if (ucontext->context->state != STATE_ERROR)
+ {
+ if (p != ucontext->text_end)
+ {
+ gint len = p - ucontext->entity_start;
+
+ /* move to after semicolon */
+ p = g_utf8_next_char (p);
+ ucontext->state = USTATE_INSIDE_TEXT;
+
+ if (strncmp (ucontext->entity_start, "lt", len) == 0)
+ g_string_append_c (ucontext->str, '<');
+ else if (strncmp (ucontext->entity_start, "gt", len) == 0)
+ g_string_append_c (ucontext->str, '>');
+ else if (strncmp (ucontext->entity_start, "amp", len) == 0)
+ g_string_append_c (ucontext->str, '&');
+ else if (strncmp (ucontext->entity_start, "quot", len) == 0)
+ g_string_append_c (ucontext->str, '"');
+ else if (strncmp (ucontext->entity_start, "apos", len) == 0)
+ g_string_append_c (ucontext->str, '\'');
+ else
+ {
+ gchar *name;
+
+ name = g_strndup (ucontext->entity_start, len);
+ set_unescape_error (ucontext->context, error,
+ p, ucontext->text_end,
+ G_MARKUP_ERROR_PARSE,
+ _("Entity name '%s' is not known"),
+ name);
+ g_free (name);
+ }
+ }
+ else
+ {
+ set_unescape_error (ucontext->context, error,
+ /* give line number of the & */
+ ucontext->entity_start, ucontext->text_end,
+ G_MARKUP_ERROR_PARSE,
+ _("Entity did not end with a semicolon; "
+ "most likely you used an ampersand "
+ "character without intending to start "
+ "an entity - escape ampersand as &"));
+ }
+ }
+#undef MAX_ENT_LEN
+
+ return p;
+}
+
+static const gchar*
+unescape_text_state_after_charref_hash (UnescapeContext *ucontext,
+ const gchar *p,
+ GError **error)
+{
+ gboolean is_hex = FALSE;
+ const char *start;
+
+ start = ucontext->entity_start;
+
+ if (*p == 'x')
+ {
+ is_hex = TRUE;
+ p = g_utf8_next_char (p);
+ start = p;
+ }
+
+ while (p != ucontext->text_end && *p != ';')
+ p = g_utf8_next_char (p);
+
+ if (p != ucontext->text_end)
+ {
+ g_assert (*p == ';');
+
+ /* digit is between start and p */
+
+ if (start != p)
+ {
+ gulong l;
+ gchar *end = NULL;
+
+ errno = 0;
+ if (is_hex)
+ l = strtoul (start, &end, 16);
+ else
+ l = strtoul (start, &end, 10);
+
+ if (end != p || errno != 0)
+ {
+ set_unescape_error (ucontext->context, error,
+ start, ucontext->text_end,
+ G_MARKUP_ERROR_PARSE,
+ _("Failed to parse '%-.*s', which "
+ "should have been a digit "
+ "inside a character reference "
+ "(ê for example) - perhaps "
+ "the digit is too large"),
+ p - start, start);
+ }
+ else
+ {
+ /* characters XML permits */
+ if (l == 0x9 ||
+ l == 0xA ||
+ l == 0xD ||
+ (l >= 0x20 && l <= 0xD7FF) ||
+ (l >= 0xE000 && l <= 0xFFFD) ||
+ (l >= 0x10000 && l <= 0x10FFFF))
+ {
+ gchar buf[8];
+ g_string_append (ucontext->str, char_str (l, buf));
+ }
+ else
+ {
+ set_unescape_error (ucontext->context, error,
+ start, ucontext->text_end,
+ G_MARKUP_ERROR_PARSE,
+ _("Character reference '%-.*s' does not "
+ "encode a permitted character"),
+ p - start, start);
+ }
+ }
+
+ /* Move to next state */
+ p = g_utf8_next_char (p); /* past semicolon */
+ ucontext->state = USTATE_INSIDE_TEXT;
+ }
+ else
+ {
+ set_unescape_error (ucontext->context, error,
+ start, ucontext->text_end,
+ G_MARKUP_ERROR_PARSE,
+ _("Empty character reference; "
+ "should include a digit such as "
+ "dž"));
+ }
+ }
+ else
+ {
+ set_unescape_error (ucontext->context, error,
+ start, ucontext->text_end,
+ G_MARKUP_ERROR_PARSE,
+ _("Character reference did not end with a "
+ "semicolon; "
+ "most likely you used an ampersand "
+ "character without intending to start "
+ "an entity - escape ampersand as &"));
+ }
+
+ return p;
+}
+
+static gboolean
+unescape_text (GMarkupParseContext *context,
+ const gchar *text,
+ const gchar *text_end,
+ GString **unescaped,
+ GError **error)
+{
+ UnescapeContext ucontext;
+ const gchar *p;
+
+ ucontext.context = context;
+ ucontext.text = text;
+ ucontext.text_end = text_end;
+ ucontext.entity_start = NULL;
+
+ ucontext.str = g_string_sized_new (text_end - text);
+
+ ucontext.state = USTATE_INSIDE_TEXT;
+ p = text;
+
while (p != text_end && context->state != STATE_ERROR)
{
g_assert (p < text_end);
- switch (state)
+ switch (ucontext.state)
{
case USTATE_INSIDE_TEXT:
{
- while (p != text_end && *p != '&')
- {
- if ((*p == '\t' || *p == '\n') && normalize_attribute)
- {
- g_string_append_len (str, start, p - start);
- g_string_append_c (str, ' ');
- p = g_utf8_next_char (p);
- start = p;
- }
- else if (*p == '\r')
- {
- g_string_append_len (str, start, p - start);
- g_string_append_c (str, normalize_attribute ? ' ' : '\n');
- p = g_utf8_next_char (p);
- if (*p == '\n')
- p = g_utf8_next_char (p);
- start = p;
- }
- else
- p = g_utf8_next_char (p);
- }
-
- if (p != start)
- {
- g_string_append_len (str, start, p - start);
-
- start = NULL;
- }
-
- if (p != text_end && *p == '&')
- {
- p = g_utf8_next_char (p);
- state = USTATE_AFTER_AMPERSAND;
- }
+ p = unescape_text_state_inside_text (&ucontext,
+ p,
+ error);
}
break;
case USTATE_AFTER_AMPERSAND:
{
- if (*p == '#')
- {
- p = g_utf8_next_char (p);
-
- start = p;
- state = USTATE_AFTER_CHARREF_HASH;
- }
- else if (!is_name_start_char (g_utf8_get_char (p)))
- {
- if (*p == ';')
- {
- set_unescape_error (context, error,
- p, text_end,
- G_MARKUP_ERROR_PARSE,
- _("Empty entity '&;' seen; valid "
- "entities are: & " < > '"));
- }
- else
- {
- gchar buf[7];
-
- set_unescape_error (context, error,
- p, text_end,
- G_MARKUP_ERROR_PARSE,
- _("Character '%s' is not valid at "
- "the start of an entity name; "
- "the & character begins an entity; "
- "if this ampersand isn't supposed "
- "to be an entity, escape it as "
- "&"),
- utf8_str (p, buf));
- }
- }
- else
- {
- start = p;
- state = USTATE_INSIDE_ENTITY_NAME;
- }
+ p = unescape_text_state_after_ampersand (&ucontext,
+ p,
+ error);
}
break;
case USTATE_INSIDE_ENTITY_NAME:
{
- gchar buf[MAX_ENT_LEN+1] = {
- '\0', '\0', '\0', '\0', '\0', '\0'
- };
- gchar *dest;
-
- while (p != text_end)
- {
- if (*p == ';')
- break;
- else if (!is_name_char (*p))
- {
- gchar ubuf[7];
-
- set_unescape_error (context, error,
- p, text_end,
- G_MARKUP_ERROR_PARSE,
- _("Character '%s' is not valid "
- "inside an entity name"),
- utf8_str (p, ubuf));
- break;
- }
-
- p = g_utf8_next_char (p);
- }
-
- if (context->state != STATE_ERROR)
- {
- if (p != text_end)
- {
- const gchar *src;
-
- src = start;
- dest = buf;
- while (src != p)
- {
- *dest = *src;
- ++dest;
- ++src;
- }
-
- /* move to after semicolon */
- p = g_utf8_next_char (p);
- start = p;
- state = USTATE_INSIDE_TEXT;
-
- if (strcmp (buf, "lt") == 0)
- g_string_append_c (str, '<');
- else if (strcmp (buf, "gt") == 0)
- g_string_append_c (str, '>');
- else if (strcmp (buf, "amp") == 0)
- g_string_append_c (str, '&');
- else if (strcmp (buf, "quot") == 0)
- g_string_append_c (str, '"');
- else if (strcmp (buf, "apos") == 0)
- g_string_append_c (str, '\'');
- else
- {
- set_unescape_error (context, error,
- p, text_end,
- G_MARKUP_ERROR_PARSE,
- _("Entity name '%s' is not known"),
- buf);
- }
- }
- else
- {
- set_unescape_error (context, error,
- /* give line number of the & */
- start, text_end,
- G_MARKUP_ERROR_PARSE,
- _("Entity did not end with a semicolon; "
- "most likely you used an ampersand "
- "character without intending to start "
- "an entity - escape ampersand as &"));
- }
- }
+ p = unescape_text_state_inside_entity_name (&ucontext,
+ p,
+ error);
}
break;
case USTATE_AFTER_CHARREF_HASH:
{
- gboolean is_hex = FALSE;
- if (*p == 'x')
- {
- is_hex = TRUE;
- p = g_utf8_next_char (p);
- start = p;
- }
-
- while (p != text_end && *p != ';')
- p = g_utf8_next_char (p);
-
- if (p != text_end)
- {
- g_assert (*p == ';');
-
- /* digit is between start and p */
-
- if (start != p)
- {
- gchar *digit = g_strndup (start, p - start);
- gulong l;
- gchar *end = NULL;
- gchar *digit_end = digit + (p - start);
-
- errno = 0;
- if (is_hex)
- l = strtoul (digit, &end, 16);
- else
- l = strtoul (digit, &end, 10);
-
- if (end != digit_end || errno != 0)
- {
- set_unescape_error (context, error,
- start, text_end,
- G_MARKUP_ERROR_PARSE,
- _("Failed to parse '%s', which "
- "should have been a digit "
- "inside a character reference "
- "(ê for example) - perhaps "
- "the digit is too large"),
- digit);
- }
- else
- {
- /* characters XML permits */
- if (l == 0x9 ||
- l == 0xA ||
- l == 0xD ||
- (l >= 0x20 && l <= 0xD7FF) ||
- (l >= 0xE000 && l <= 0xFFFD) ||
- (l >= 0x10000 && l <= 0x10FFFF))
- {
- gchar buf[7];
- g_string_append (str, char_str (l, buf));
- }
- else
- {
- set_unescape_error (context, error,
- start, text_end,
- G_MARKUP_ERROR_PARSE,
- _("Character reference '%s' does not encode a permitted character"),
- digit);
- }
- }
-
- g_free (digit);
-
- /* Move to next state */
- p = g_utf8_next_char (p); /* past semicolon */
- start = p;
- state = USTATE_INSIDE_TEXT;
- }
- else
- {
- set_unescape_error (context, error,
- start, text_end,
- G_MARKUP_ERROR_PARSE,
- _("Empty character reference; "
- "should include a digit such as "
- "dž"));
- }
- }
- else
- {
- set_unescape_error (context, error,
- start, text_end,
- G_MARKUP_ERROR_PARSE,
- _("Character reference did not end with a "
- "semicolon; "
- "most likely you used an ampersand "
- "character without intending to start "
- "an entity - escape ampersand as &"));
- }
+ p = unescape_text_state_after_charref_hash (&ucontext,
+ p,
+ error);
}
break;
if (context->state != STATE_ERROR)
{
- switch (state)
+ switch (ucontext.state)
{
case USTATE_INSIDE_TEXT:
break;
if (context->state == STATE_ERROR)
{
- g_string_free (str, TRUE);
+ g_string_free (ucontext.str, TRUE);
*unescaped = NULL;
return FALSE;
}
else
{
- *unescaped = g_string_free (str, FALSE);
+ *unescaped = ucontext.str;
return TRUE;
}
-
-#undef MAX_ENT_LEN
}
-static gboolean
+static inline gboolean
advance_char (GMarkupParseContext *context)
-{
- g_return_val_if_fail (context->iter != context->current_text_end, FALSE);
-
+{
context->iter = g_utf8_next_char (context->iter);
context->char_number += 1;
if (context->iter == context->current_text_end)
- return FALSE;
-
- if (*context->iter == '\n')
+ {
+ return FALSE;
+ }
+ else if (*context->iter == '\n')
{
context->line_number += 1;
context->char_number = 1;
}
-
+
return TRUE;
}
-static gboolean
+static inline gboolean
xml_isspace (char c)
{
return c == ' ' || c == '\t' || c == '\n' || c == '\r';
{
do
{
- if (!is_name_char (g_utf8_get_char (context->iter)))
+ if (!is_name_char (context->iter))
return;
}
while (advance_char (context));
const gchar *text_end)
{
if (context->partial_chunk == NULL)
- context->partial_chunk = g_string_new (NULL);
+ context->partial_chunk = g_string_sized_new (text_end - text_start);
if (text_start != text_end)
g_string_append_len (context->partial_chunk, text_start,
static void
find_current_text_end (GMarkupParseContext *context)
{
- /* This function must be safe (non-segfaulting) on invalid UTF8 */
+ /* This function must be safe (non-segfaulting) on invalid UTF8.
+ * It assumes the string starts with a character start
+ */
const gchar *end = context->current_text + context->current_text_len;
const gchar *p;
const gchar *next;
g_assert (context->current_text_len > 0);
- p = context->current_text;
- next = g_utf8_find_next_char (p, end);
+ p = g_utf8_find_prev_char (context->current_text, end);
- while (next && *next)
- {
- if (p == next)
- next++;
- p = next;
- next = g_utf8_find_next_char (p, end);
- }
+ g_assert (p != NULL); /* since current_text was a char start */
/* p is now the start of the last character or character portion. */
g_assert (p != end);
set_error (context,
error,
G_MARKUP_ERROR_BAD_UTF8,
- _("Invalid UTF-8 encoded text"));
+ _("Invalid UTF-8 encoded text - overlong sequence"));
}
goto finished;
set_error (context,
error,
G_MARKUP_ERROR_BAD_UTF8,
- _("Invalid UTF-8 encoded text"));
+ _("Invalid UTF-8 encoded text - not a start char"));
goto finished;
}
* we could have a trailing incomplete char)
*/
if (!g_utf8_validate (context->current_text,
- context->current_text_len,
- &first_invalid))
+ context->current_text_len,
+ &first_invalid))
{
gint newlines = 0;
- const gchar *p;
- p = context->current_text;
- while (p != context->current_text_end)
+ const gchar *p, *q;
+ q = p = context->current_text;
+ while (p != first_invalid)
{
if (*p == '\n')
- ++newlines;
+ {
+ ++newlines;
+ q = p + 1;
+ context->char_number = 1;
+ }
++p;
}
context->line_number += newlines;
+ context->char_number += g_utf8_strlen (q, first_invalid - q);
set_error (context,
error,
G_MARKUP_ERROR_BAD_UTF8,
- _("Invalid UTF-8 encoded text"));
+ _("Invalid UTF-8 encoded text - not valid '%s'"),
+ g_strndup (context->current_text,
+ context->current_text_len));
goto finished;
}
context->state = STATE_AFTER_CLOSE_TAG_SLASH;
}
- else if (is_name_start_char (g_utf8_get_char (context->iter)))
+ else if (is_name_start_char (context->iter))
{
context->state = STATE_INSIDE_OPEN_TAG_NAME;
}
else
{
- gchar buf[7];
+ gchar buf[8];
+
set_error (context,
error,
G_MARKUP_ERROR_PARSE,
}
else
{
- gchar buf[7];
+ gchar buf[8];
+
set_error (context,
error,
G_MARKUP_ERROR_PARSE,
}
else
{
- gchar buf[7];
+ gchar buf[8];
+
set_error (context,
error,
G_MARKUP_ERROR_PARSE,
advance_char (context);
context->state = STATE_AFTER_CLOSE_ANGLE;
}
- else if (is_name_start_char (g_utf8_get_char (context->iter)))
+ else if (is_name_start_char (context->iter))
{
context->state = STATE_INSIDE_ATTRIBUTE_NAME;
/* start of attribute name */
}
else
{
- gchar buf[7];
+ gchar buf[8];
+
set_error (context,
error,
G_MARKUP_ERROR_PARSE,
context->attr_values[0] == NULL);
if (tmp_error != NULL)
- {
- mark_error (context, tmp_error);
- g_propagate_error (error, tmp_error);
- }
+ propagate_error (context, error, tmp_error);
}
}
break;
}
else
{
- gchar buf[7];
+ gchar buf[8];
+
set_error (context,
error,
G_MARKUP_ERROR_PARSE,
* with the partial chunk if any; set it for the current
* attribute.
*/
+ GString *unescaped;
+
add_to_partial (context, context->start, context->iter);
g_assert (context->cur_attr >= 0);
context->partial_chunk->str,
context->partial_chunk->str +
context->partial_chunk->len,
- &context->attr_values[context->cur_attr],
+ &unescaped,
error))
{
/* success, advance past quote and set state. */
+ context->attr_values[context->cur_attr] = g_string_free (unescaped, FALSE);
advance_char (context);
context->state = STATE_BETWEEN_ATTRIBUTES;
context->start = NULL;
if (context->iter != context->current_text_end)
{
- gchar *unescaped = NULL;
+ GString *unescaped = NULL;
/* The text has ended at the open angle. Call the text
* callback.
if (context->parser->text)
(*context->parser->text) (context,
- unescaped,
- strlen (unescaped),
+ unescaped->str,
+ unescaped->len,
context->user_data,
&tmp_error);
- g_free (unescaped);
+ g_string_free (unescaped, TRUE);
if (tmp_error == NULL)
{
context->start = context->iter;
}
else
- {
- mark_error (context, tmp_error);
- g_propagate_error (error, tmp_error);
- }
+ propagate_error (context, error, tmp_error);
}
truncate_partial (context);
case STATE_AFTER_CLOSE_TAG_SLASH:
/* Possible next state: INSIDE_CLOSE_TAG_NAME */
- if (is_name_start_char (g_utf8_get_char (context->iter)))
+ if (is_name_start_char (context->iter))
{
context->state = STATE_INSIDE_CLOSE_TAG_NAME;
}
else
{
- gchar buf[7];
+ gchar buf[8];
+
set_error (context,
error,
G_MARKUP_ERROR_PARSE,
if (*context->iter != '>')
{
- gchar buf[7];
+ gchar buf[8];
+
set_error (context,
error,
G_MARKUP_ERROR_PARSE,
context->tag_stack);
if (tmp_error)
- {
- mark_error (context, tmp_error);
- g_propagate_error (error, tmp_error);
- }
+ propagate_error (context, error, tmp_error);
}
g_free (close_name);
if (*context->iter == '<')
context->balance++;
if (*context->iter == '>')
- {
+ {
+ gchar *str;
+ gsize len;
+
context->balance--;
add_to_partial (context, context->start, context->iter);
context->start = context->iter;
- if ((g_str_has_prefix (context->partial_chunk->str, "<?")
- && g_str_has_suffix (context->partial_chunk->str, "?")) ||
- (g_str_has_prefix (context->partial_chunk->str, "<!--")
- && g_str_has_suffix (context->partial_chunk->str, "--")) ||
- (g_str_has_prefix (context->partial_chunk->str, "<![CDATA[")
- && g_str_has_suffix (context->partial_chunk->str, "]]")) ||
- (g_str_has_prefix (context->partial_chunk->str, "<!DOCTYPE")
- && context->balance == 0))
+
+ str = context->partial_chunk->str;
+ len = context->partial_chunk->len;
+
+ if (str[1] == '?' && str[len - 1] == '?')
+ break;
+ if (strncmp (str, "<!--", 4) == 0 &&
+ strcmp (str + len - 2, "--") == 0)
+ break;
+ if (strncmp (str, "<![CDATA[", 9) == 0 &&
+ strcmp (str + len - 2, "]]") == 0)
+ break;
+ if (strncmp (str, "<!DOCTYPE", 9) == 0 &&
+ context->balance == 0)
break;
}
}
/* The passthrough hasn't necessarily ended. Merge with
* partial chunk, leave state unchanged.
*/
- add_to_partial (context, context->start, context->iter);
+ add_to_partial (context, context->start, context->iter);
}
else
{
advance_char (context); /* advance past close angle */
add_to_partial (context, context->start, context->iter);
- if (context->parser->passthrough)
+ if (context->flags & G_MARKUP_TREAT_CDATA_AS_TEXT &&
+ strncmp (context->partial_chunk->str, "<![CDATA[", 9) == 0)
+ {
+ if (context->parser->text)
+ (*context->parser->text) (context,
+ context->partial_chunk->str + 9,
+ context->partial_chunk->len - 12,
+ context->user_data,
+ &tmp_error);
+ }
+ else if (context->parser->passthrough)
(*context->parser->passthrough) (context,
context->partial_chunk->str,
context->partial_chunk->len,
context->start = context->iter; /* could begin text */
}
else
- {
- mark_error (context, tmp_error);
- g_propagate_error (error, tmp_error);
- }
+ propagate_error (context, error, tmp_error);
}
break;
break;
case STATE_INSIDE_ATTRIBUTE_NAME:
+ case STATE_AFTER_ATTRIBUTE_NAME:
set_error (context, error, G_MARKUP_ERROR_PARSE,
_("Document ended unexpectedly inside an attribute name"));
break;
case STATE_AFTER_CLOSE_TAG_SLASH:
case STATE_INSIDE_CLOSE_TAG_NAME:
+ case STATE_AFTER_CLOSE_TAG_NAME:
set_error (context, error, G_MARKUP_ERROR_PARSE,
_("Document ended unexpectedly inside the close tag for "
- "element '%s'"), current_element);
+ "element '%s'"), current_element (context));
break;
case STATE_INSIDE_PASSTHROUGH:
*
* Retrieves the name of the currently open element.
*
+ * If called from the start_element or end_element handlers this will
+ * give the element_name as passed to those functions. For the parent
+ * elements, see g_markup_parse_context_get_element_stack().
+ *
* Since: 2.2
**/
G_CONST_RETURN gchar *
}
/**
+ * g_markup_parse_context_get_element_stack:
+ * @context: a #GMarkupParseContext
+ *
+ * Retrieves the element stack from the internal state of the parser.
+ * The returned #GSList is a list of strings where the first item is
+ * the currently open tag (as would be returned by
+ * g_markup_parse_context_get_element()) and the next item is its
+ * immediate parent.
+ *
+ * This function is intended to be used in the start_element and
+ * end_element handlers where g_markup_parse_context_get_element()
+ * would merely return the name of the element that is being
+ * processed.
+ *
+ * Returns: the element stack, which must not be modified
+ *
+ * Since 2.16
+ **/
+G_CONST_RETURN GSList *
+g_markup_parse_context_get_element_stack (GMarkupParseContext *context)
+{
+ g_return_val_if_fail (context != NULL, NULL);
+
+ return context->tag_stack;
+}
+
+/**
* g_markup_parse_context_get_position:
* @context: a #GMarkupParseContext
* @line_number: return location for a line number, or %NULL
{
const gchar *p;
const gchar *end;
+ gunichar c;
p = text;
end = text + length;
break;
default:
- g_string_append_len (str, p, next - p);
+ c = g_utf8_get_char (p);
+ if ((0x1 <= c && c <= 0x8) ||
+ (0xb <= c && c <= 0xc) ||
+ (0xe <= c && c <= 0x1f) ||
+ (0x7f <= c && c <= 0x84) ||
+ (0x86 <= c && c <= 0x9f))
+ g_string_append_printf (str, "&#x%x;", c);
+ else
+ g_string_append_len (str, p, next - p);
break;
}
/**
* g_markup_escape_text:
* @text: some valid UTF-8 text
- * @length: length of @text in bytes
+ * @length: length of @text in bytes, or -1 if the text is nul-terminated
*
* Escapes text so that the markup parser will parse it verbatim.
* Less than, greater than, ampersand, etc. are replaced with the
* from being processed according to the XML rules for normalization
* of line endings and attribute values.
*
- * Return value: escaped text
+ * Return value: a newly allocated string with the escaped text
**/
gchar*
g_markup_escape_text (const gchar *text,
if (length < 0)
length = strlen (text);
- str = g_string_new (NULL);
+ /* prealloc at least as long as original text */
+ str = g_string_sized_new (length);
append_escaped_text (str, text, length);
return g_string_free (str, FALSE);
G_VA_COPY (args2, args);
output1 = g_strdup_vprintf (format1->str, args);
- va_end (args);
if (!output1)
- goto cleanup;
+ {
+ va_end (args2);
+ goto cleanup;
+ }
output2 = g_strdup_vprintf (format2->str, args2);
va_end (args2);
* output, without having to worry that the strings
* might themselves contain markup.
*
- * <informalexample><programlisting>
+ * |[
* const char *store = "Fortnum & Mason";
* const char *item = "Tea";
* char *output;
* "<item>%s</item>"
* "</purchase>",
* store, item);
- * </programlisting></informalexample>
+ * ]|
*
* Return value: newly allocated result from formatting
* operation. Free with g_free().
return result;
}
+
+static gboolean
+g_markup_parse_boolean (const char *string,
+ gboolean *value)
+{
+ char const * const falses[] = { "false", "f", "no", "n", "0" };
+ char const * const trues[] = { "true", "t", "yes", "y", "1" };
+ int i;
+
+ for (i = 0; i < G_N_ELEMENTS (falses); i++)
+ {
+ if (g_ascii_strcasecmp (string, falses[i]) == 0)
+ {
+ if (value != NULL)
+ *value = FALSE;
+
+ return TRUE;
+ }
+ }
+
+ for (i = 0; i < G_N_ELEMENTS (trues); i++)
+ {
+ if (g_ascii_strcasecmp (string, trues[i]) == 0)
+ {
+ if (value != NULL)
+ *value = TRUE;
+
+ return TRUE;
+ }
+ }
+
+ return FALSE;
+}
+
+/**
+ * GMarkupCollectType:
+ * @G_MARKUP_COLLECT_INVALID: used to terminate the list of attributes
+ * to collect.
+ * @G_MARKUP_COLLECT_STRING: collect the string pointer directly from
+ * the attribute_values[] array. Expects a
+ * parameter of type (const char **). If
+ * %G_MARKUP_COLLECT_OPTIONAL is specified
+ * and the attribute isn't present then the
+ * pointer will be set to %NULL.
+ * @G_MARKUP_COLLECT_STRDUP: as with %G_MARKUP_COLLECT_STRING, but
+ * expects a paramter of type (char **) and
+ * g_strdup()s the returned pointer. The
+ * pointer must be freed with g_free().
+ * @G_MARKUP_COLLECT_BOOLEAN: expects a parameter of type (gboolean *)
+ * and parses the attribute value as a
+ * boolean. Sets %FALSE if the attribute
+ * isn't present. Valid boolean values
+ * consist of (case insensitive) "false",
+ * "f", "no", "n", "0" and "true", "t",
+ * "yes", "y", "1".
+ * @G_MARKUP_COLLECT_TRISTATE: as with %G_MARKUP_COLLECT_BOOLEAN, but
+ * in the case of a missing attribute a
+ * value is set that compares equal to
+ * neither %FALSE nor %TRUE.
+ * G_MARKUP_COLLECT_OPTIONAL is implied.
+ * @G_MARKUP_COLLECT_OPTIONAL: can be bitwise ORed with the other
+ * fields. If present, allows the
+ * attribute not to appear. A default
+ * value is set depending on what value
+ * type is used.
+ *
+ * A mixed enumerated type and flags field. You must specify one type
+ * (string, strdup, boolean, tristate). Additionally, you may
+ * optionally bitwise OR the type with the flag
+ * %G_MARKUP_COLLECT_OPTIONAL.
+ *
+ * It is likely that this enum will be extended in the future to
+ * support other types.
+ **/
+
+/**
+ * g_markup_collect_attributes:
+ * @element_name: the current tag name
+ * @attribute_names: the attribute names
+ * @attribute_values: the attribute values
+ * @error: a pointer to a #GError or %NULL
+ * @first_type: the #GMarkupCollectType of the
+ * first attribute
+ * @first_attr: the name of the first attribute
+ * @...: a pointer to the storage location of the
+ * first attribute (or %NULL), followed by
+ * more types names and pointers, ending
+ * with %G_MARKUP_COLLECT_INVALID.
+ *
+ * Collects the attributes of the element from the
+ * data passed to the #GMarkupParser start_element
+ * function, dealing with common error conditions
+ * and supporting boolean values.
+ *
+ * This utility function is not required to write
+ * a parser but can save a lot of typing.
+ *
+ * The @element_name, @attribute_names,
+ * @attribute_values and @error parameters passed
+ * to the start_element callback should be passed
+ * unmodified to this function.
+ *
+ * Following these arguments is a list of
+ * "supported" attributes to collect. It is an
+ * error to specify multiple attributes with the
+ * same name. If any attribute not in the list
+ * appears in the @attribute_names array then an
+ * unknown attribute error will result.
+ *
+ * The #GMarkupCollectType field allows specifying
+ * the type of collection to perform and if a
+ * given attribute must appear or is optional.
+ *
+ * The attribute name is simply the name of the
+ * attribute to collect.
+ *
+ * The pointer should be of the appropriate type
+ * (see the descriptions under
+ * #GMarkupCollectType) and may be %NULL in case a
+ * particular attribute is to be allowed but
+ * ignored.
+ *
+ * This function deals with issuing errors for missing attributes
+ * (of type %G_MARKUP_ERROR_MISSING_ATTRIBUTE), unknown attributes
+ * (of type %G_MARKUP_ERROR_UNKNOWN_ATTRIBUTE) and duplicate
+ * attributes (of type %G_MARKUP_ERROR_INVALID_CONTENT) as well
+ * as parse errors for boolean-valued attributes (again of type
+ * %G_MARKUP_ERROR_INVALID_CONTENT). In all of these cases %FALSE
+ * will be returned and @error will be set as appropriate.
+ *
+ * Return value: %TRUE if successful
+ *
+ * Since: 2.16
+ **/
+gboolean
+g_markup_collect_attributes (const gchar *element_name,
+ const gchar **attribute_names,
+ const gchar **attribute_values,
+ GError **error,
+ GMarkupCollectType first_type,
+ const gchar *first_attr,
+ ...)
+{
+ GMarkupCollectType type;
+ const gchar *attr;
+ guint64 collected;
+ int written;
+ va_list ap;
+ int i;
+
+ type = first_type;
+ attr = first_attr;
+ collected = 0;
+ written = 0;
+
+ va_start (ap, first_attr);
+ while (type != G_MARKUP_COLLECT_INVALID)
+ {
+ gboolean mandatory;
+ const gchar *value;
+
+ mandatory = !(type & G_MARKUP_COLLECT_OPTIONAL);
+ type &= (G_MARKUP_COLLECT_OPTIONAL - 1);
+
+ /* tristate records a value != TRUE and != FALSE
+ * for the case where the attribute is missing
+ */
+ if (type == G_MARKUP_COLLECT_TRISTATE)
+ mandatory = FALSE;
+
+ for (i = 0; attribute_names[i]; i++)
+ if (i >= 40 || !(collected & (1ull << i)))
+ if (!strcmp (attribute_names[i], attr))
+ break;
+
+ /* ISO C99 only promises that the user can pass up to 127 arguments.
+ * Subtracting the first 4 arguments plus the final NULL and dividing
+ * by 3 arguments per collected attribute, we are left with a maximum
+ * number of supported attributes of (127 - 5) / 3 = 40.
+ *
+ * In reality, nobody is ever going to call us with anywhere close to
+ * 40 attributes to collect, so it is safe to assume that if i > 40
+ * then the user has given some invalid or repeated arguments. These
+ * problems will be caught and reported at the end of the function.
+ *
+ * We know at this point that we have an error, but we don't know
+ * what error it is, so just continue...
+ */
+ if (i < 40)
+ collected |= (1ull << i);
+
+ value = attribute_values[i];
+
+ if (value == NULL && mandatory)
+ {
+ g_set_error (error, G_MARKUP_ERROR,
+ G_MARKUP_ERROR_MISSING_ATTRIBUTE,
+ "element '%s' requires attribute '%s'",
+ element_name, attr);
+
+ va_end (ap);
+ goto failure;
+ }
+
+ switch (type)
+ {
+ case G_MARKUP_COLLECT_STRING:
+ {
+ const char **str_ptr;
+
+ str_ptr = va_arg (ap, const char **);
+
+ if (str_ptr != NULL)
+ *str_ptr = value;
+ }
+ break;
+
+ case G_MARKUP_COLLECT_STRDUP:
+ {
+ char **str_ptr;
+
+ str_ptr = va_arg (ap, char **);
+
+ if (str_ptr != NULL)
+ *str_ptr = g_strdup (value);
+ }
+ break;
+
+ case G_MARKUP_COLLECT_BOOLEAN:
+ case G_MARKUP_COLLECT_TRISTATE:
+ if (value == NULL)
+ {
+ gboolean *bool_ptr;
+
+ bool_ptr = va_arg (ap, gboolean *);
+
+ if (bool_ptr != NULL)
+ {
+ if (type == G_MARKUP_COLLECT_TRISTATE)
+ /* constructivists rejoice!
+ * neither false nor true...
+ */
+ *bool_ptr = -1;
+
+ else /* G_MARKUP_COLLECT_BOOLEAN */
+ *bool_ptr = FALSE;
+ }
+ }
+ else
+ {
+ if (!g_markup_parse_boolean (value, va_arg (ap, gboolean *)))
+ {
+ g_set_error (error, G_MARKUP_ERROR,
+ G_MARKUP_ERROR_INVALID_CONTENT,
+ "element '%s', attribute '%s', value '%s' "
+ "cannot be parsed as a boolean value",
+ element_name, attr, value);
+
+ va_end (ap);
+ goto failure;
+ }
+ }
+
+ break;
+
+ default:
+ g_assert_not_reached ();
+ }
+
+ type = va_arg (ap, GMarkupCollectType);
+ attr = va_arg (ap, const char *);
+ written++;
+ }
+ va_end (ap);
+
+ /* ensure we collected all the arguments */
+ for (i = 0; attribute_names[i]; i++)
+ if ((collected & (1 << i)) == 0)
+ {
+ /* attribute not collected: could be caused by two things.
+ *
+ * 1) it doesn't exist in our list of attributes
+ * 2) it existed but was matched by a duplicate attribute earlier
+ *
+ * find out.
+ */
+ int j;
+
+ for (j = 0; j < i; j++)
+ if (strcmp (attribute_names[i], attribute_names[j]) == 0)
+ /* duplicate! */
+ break;
+
+ /* j is now the first occurance of attribute_names[i] */
+ if (i == j)
+ g_set_error (error, G_MARKUP_ERROR,
+ G_MARKUP_ERROR_UNKNOWN_ATTRIBUTE,
+ "attribute '%s' invalid for element '%s'",
+ attribute_names[i], element_name);
+ else
+ g_set_error (error, G_MARKUP_ERROR,
+ G_MARKUP_ERROR_INVALID_CONTENT,
+ "attribute '%s' given multiple times for element '%s'",
+ attribute_names[i], element_name);
+
+ goto failure;
+ }
+
+ return TRUE;
+
+failure:
+ /* replay the above to free allocations */
+ type = first_type;
+ attr = first_attr;
+
+ va_start (ap, first_attr);
+ while (type != G_MARKUP_COLLECT_INVALID)
+ {
+ gpointer ptr;
+
+ ptr = va_arg (ap, gpointer);
+
+ if (ptr == NULL)
+ continue;
+
+ switch (type & (G_MARKUP_COLLECT_OPTIONAL - 1))
+ {
+ case G_MARKUP_COLLECT_STRDUP:
+ if (written)
+ g_free (*(char **) ptr);
+
+ case G_MARKUP_COLLECT_STRING:
+ *(char **) ptr = NULL;
+ break;
+
+ case G_MARKUP_COLLECT_BOOLEAN:
+ *(gboolean *) ptr = FALSE;
+ break;
+
+ case G_MARKUP_COLLECT_TRISTATE:
+ *(gboolean *) ptr = -1;
+ break;
+ }
+
+ type = va_arg (ap, GMarkupCollectType);
+ attr = va_arg (ap, const char *);
+
+ if (written)
+ written--;
+ }
+ va_end (ap);
+
+ return FALSE;
+}
+
+#define __G_MARKUP_C__
+#include "galiasdef.c"