*/
+
+#ifndef _GNU_SOURCE
+#define _GNU_SOURCE
+#endif /*_GNU_SOURCE*/
+#include <string.h> /* for strcasestr */
+
+
#include <glib.h>
-#include <string.h>
#include <stdlib.h>
#include <glib/gi18n.h>
#include <regex.h>
#define EMPTY_STRING ""
/*
+ * do the hyperlinkification only for texts < 50 Kb,
+ * as it's quite slow. Without this, e.g. mail with
+ * an uuencoded part (which is not recognized as attachment,
+ * will hang modest
+ */
+#define HYPERLINKIFY_MAX_LENGTH (1024*50)
+
+/*
* we need these regexps to find URLs in plain text e-mails
*/
typedef struct _url_match_pattern_t url_match_pattern_t;
{ "(file|rtsp|http|ftp|https)://[-A-Za-z0-9_$.+!*(),;:@%&=?/~#]+[-A-Za-z0-9_$%&=?/~#]",\
NULL, NULL },\
{ "www\\.[-a-z0-9.]+[-a-z0-9](:[0-9]*)?(/[-A-Za-z0-9_$.+!*(),;:@%&=?/~#]*[^]}\\),?!;:\"]?)?",\
- NULL, "http://" },\
+ NULL, "http://" }, \
{ "ftp\\.[-a-z0-9.]+[-a-z0-9](:[0-9]*)?(/[-A-Za-z0-9_$.+!*(),;:@%&=?/~#]*[^]}\\),?!;:\"]?)?",\
NULL, "ftp://" },\
{ "(voipto|callto|chatto|jabberto|xmpp):[-_a-z@0-9.\\+]+", \
static gchar* cite (const time_t sent_date, const gchar *from);
static void hyperlinkify_plain_text (GString *txt);
static gint cmp_offsets_reverse (const url_match_t *match1, const url_match_t *match2);
-static void chk_partial_match (const url_match_t *match, guint* offset);
static GSList* get_url_matches (GString *txt);
static GString* get_next_line (const char *b, const gsize blen, const gchar * iter);
guint i;
gboolean space_seen = FALSE;
gsize len;
+ guint break_dist = 0; /* distance since last break point */
len = strlen (data);
space_seen = FALSE;
}
+ /* we artificially insert a breakpoint (newline)
+ * after 256, to make sure our lines are not so long
+ * they will DOS the regexping later
+ */
+ if (++break_dist == 256) {
+ g_string_append_c (html, '\n');
+ break_dist = 0;
+ }
+
switch (kar) {
case 0: break; /* ignore embedded \0s */
case '<' : g_string_append (html, "<"); break;
case '&' : g_string_append (html, "&"); break;
case '"' : g_string_append (html, """); break;
case '\'' : g_string_append (html, "'"); break;
- case '\n' : g_string_append (html, "<br>\n"); break;
- case '\t' : g_string_append (html, " "); break; /* note the space at the end*/
+ case '\n' : g_string_append (html, "<br>\n"); break_dist= 0; break;
+ case '\t' : g_string_append (html, " "); break_dist=0; break; /* note the space at the end*/
case ' ':
+ break_dist = 0;
if (space_seen) { /* second space in a row */
g_string_append (html, " ");
space_seen = FALSE;
modest_text_utils_convert_buffer_to_html (html, data);
g_string_append (html, "</body></html>");
- hyperlinkify_plain_text (html);
+
+ if (len <= HYPERLINKIFY_MAX_LENGTH)
+ hyperlinkify_plain_text (html);
return g_string_free (html, FALSE);
}
modest_text_utils_convert_buffer_to_html (html, data);
- hyperlinkify_plain_text (html);
+ if (len < HYPERLINKIFY_MAX_LENGTH)
+ hyperlinkify_plain_text (html);
return g_string_free (html, FALSE);
}
}
-
-/*
- * check if the match is inside an existing match... */
-static void
-chk_partial_match (const url_match_t *match, guint* offset)
-{
- if (*offset >= match->offset && *offset < match->offset + match->len)
- *offset = -1;
-}
-
static GSList*
get_url_matches (GString *txt)
{
for (i = 0; i != pattern_num; ++i) {
offset = 0;
while (1) {
- int test_offset;
+ url_match_t *match;
+ gboolean is_submatch;
+ GSList *cursor;
+
if ((rv = regexec (patterns[i].preg, txt->str + offset, 1, &rm, 0)) != 0) {
g_return_val_if_fail (rv == REG_NOMATCH, NULL); /* this should not happen */
break; /* try next regexp */
}
if (rm.rm_so == -1)
break;
-
- /* FIXME: optimize this */
- /* to avoid partial matches on something that was already found... */
- /* check_partial_match will put -1 in the data ptr if that is the case */
- test_offset = offset + rm.rm_so;
- g_slist_foreach (match_list, (GFunc)chk_partial_match, &test_offset);
- /* make a list of our matches (<offset, len, prefix> tupels)*/
- if (test_offset != -1) {
- url_match_t *match = g_slice_new (url_match_t);
+ is_submatch = FALSE;
+ /* check old matches to see if this has already been matched */
+ cursor = match_list;
+ while (cursor && !is_submatch) {
+ const url_match_t *old_match =
+ (const url_match_t *) cursor->data;
+ guint new_offset = offset + rm.rm_so;
+ is_submatch = (new_offset > old_match->offset &&
+ new_offset < old_match->offset + old_match->len);
+ cursor = g_slist_next (cursor);
+ }
+
+ if (!is_submatch) {
+ /* make a list of our matches (<offset, len, prefix> tupels)*/
+ match = g_slice_new (url_match_t);
match->offset = offset + rm.rm_so;
match->len = rm.rm_eo - rm.rm_so;
match->prefix = patterns[i].prefix;
+ g_warning ("<%d, %d, %s>", match->offset, match->len, match->prefix);
match_list = g_slist_prepend (match_list, match);
}
+
offset += rm.rm_eo;
}
}
return address;
}
+gchar *
+modest_text_utils_get_email_address (const gchar *full_address)
+{
+ const gchar *left, *right;
+
+ if (!full_address)
+ return NULL;
+
+ g_return_val_if_fail (g_utf8_validate (full_address, -1, NULL), NULL);
+
+ left = g_strrstr_len (full_address, strlen(full_address), "<");
+ if (left == NULL)
+ return g_strdup (full_address);
+
+ right = g_strstr_len (left, strlen(left), ">");
+ if (right == NULL)
+ return g_strdup (full_address);
+ return g_strndup (left + 1, right - left - 1);
+}
gint
modest_text_utils_get_subject_prefix_len (const gchar *sub)
modest_text_utils_get_display_date (time_t date)
{
time_t now;
- const guint BUF_SIZE = 64;
+ static const guint BUF_SIZE = 64;
+ static const guint ONE_DAY = 24 * 60 * 60; /* seconds in one day */
gchar date_buf[BUF_SIZE];
- gchar now_buf [BUF_SIZE];
-
+ gchar today_buf [BUF_SIZE];
+
+ modest_text_utils_strftime (date_buf, BUF_SIZE, "%x", date);
+
now = time (NULL);
- /* use the localized dates */
- modest_text_utils_strftime (date_buf, BUF_SIZE, "%x", date);
- modest_text_utils_strftime (now_buf, BUF_SIZE, "%x", now);
-
- /* if this is today, get the time instead of the date */
- if (strcmp (date_buf, now_buf) == 0)
- modest_text_utils_strftime (date_buf, BUF_SIZE, "%X", date);
+ /* we check if the date is within the last 24h, if not, we don't
+ * have to do the extra, expensive strftime, which was very visible
+ * in the profiles.
+ */
+ if (abs(now - date) < ONE_DAY) {
+
+ /* it's within the last 24 hours, but double check */
+ /* use the localized dates */
+ modest_text_utils_strftime (today_buf, BUF_SIZE, "%x", now);
+
+ /* if it's today, use the time instead */
+ if (strcmp (date_buf, today_buf) == 0)
+ modest_text_utils_strftime (date_buf, BUF_SIZE, "%X", date);
+ }
return g_strdup(date_buf);
}
-gboolean
-modest_text_utils_validate_email_address (const gchar *email_address)
+
+gboolean
+modest_text_utils_validate_folder_name (const gchar *folder_name)
+{
+ /* based on http://msdn2.microsoft.com/en-us/library/aa365247.aspx,
+ * with some extras */
+
+ guint len;
+ const gchar **cursor;
+ const gchar *forbidden_chars[] = {
+ "<", ">", ":", "\"", "/", "\\", "|", "?", "*", "^", "%", "$", NULL
+ };
+ const gchar *forbidden_names[] = { /* windows does not like these */
+ "CON", "PRN", "AUX", "NUL", "COM1", "COM2", "COM3", "COM4", "COM5", "COM6",
+ "COM7", "COM8", "COM9", "LPT1", "LPT2", "LPT3", "LPT4", "LPT5", "LPT6", "LPT7", "LPT8", "LPT9",
+ ".", "..", NULL
+ };
+
+ /* cannot be NULL */
+ if (!folder_name)
+ return FALSE;
+
+ /* cannot be empty */
+ len = strlen(folder_name);
+ if (len == 0)
+ return FALSE;
+
+ /* cannot start or end with a space */
+ if (g_ascii_isspace(folder_name[0]) || g_ascii_isspace(folder_name[len - 1]))
+ return FALSE;
+
+ /* cannot contain a forbidden char */
+ for (cursor = forbidden_chars; cursor && *cursor; ++cursor)
+ if (strstr(folder_name, *cursor) != NULL)
+ return FALSE;
+
+ /* cannot contain a forbidden word */
+ if (len <= 4) {
+ for (cursor = forbidden_names; cursor && *cursor; ++cursor) {
+ g_warning ("%s", *cursor);
+ if (g_ascii_strcasecmp (folder_name, *cursor) == 0)
+ return FALSE;
+ }
+ }
+ return TRUE; /* it's valid! */
+}
+
+
+
+gboolean
+modest_text_utils_validate_domain_name (const gchar *domain)
+{
+ gboolean valid = FALSE;
+ regex_t rx;
+ const gchar* domain_regex = "^[a-z0-9]([.]?[a-z0-9-])*[a-z0-9]$";
+
+ if (!domain)
+ return FALSE;
+
+ /* domain name: all alphanum or '-' or '.',
+ * but beginning/ending in alphanum */
+ if (regcomp (&rx, domain_regex, REG_ICASE|REG_EXTENDED|REG_NOSUB)) {
+ g_warning ("BUG: error in regexp");
+ return FALSE;
+ }
+
+ valid = (regexec (&rx, domain, 1, NULL, 0) == 0);
+ regfree (&rx);
+
+ return valid;
+}
+
+
+
+gboolean
+modest_text_utils_validate_email_address (const gchar *email_address, const gchar **invalid_char_position)
{
int count = 0;
const gchar *c = NULL, *domain = NULL;
- static gchar *rfc822_specials = "()<>@,;:\\\"[]";
+ static gchar *rfc822_specials = "()<>@,;:\\\"[]&";
+
+ if (invalid_char_position != NULL)
+ *invalid_char_position = NULL;
/* first we validate the name portion (name@domain) */
for (c = email_address; *c; c++) {
break;
if (*c <= ' ' || *c >= 127)
return FALSE;
- if (strchr(rfc822_specials, *c))
+ if (strchr(rfc822_specials, *c)) {
+ if (invalid_char_position)
+ *invalid_char_position = c;
return FALSE;
+ }
}
if (c == email_address || *(c - 1) == '.')
return FALSE;
return FALSE;
do {
if (*c == '.') {
- if (c == domain || *(c - 1) == '.')
+ if (c == domain || *(c - 1) == '.' || *(c + 1) == '\0')
return FALSE;
count++;
}
if (*c <= ' ' || *c >= 127)
return FALSE;
- if (strchr(rfc822_specials, *c))
+ if (strchr(rfc822_specials, *c)) {
+ if (invalid_char_position)
+ *invalid_char_position = c;
return FALSE;
+ }
} while (*++c);
return (count >= 1) ? TRUE : FALSE;
}
gboolean
-modest_text_utils_validate_recipient (const gchar *recipient)
+modest_text_utils_validate_recipient (const gchar *recipient, const gchar **invalid_char_position)
{
gchar *stripped, *current;
gchar *right_part;
gboolean has_error = FALSE;
- if (modest_text_utils_validate_email_address (recipient))
+ if (modest_text_utils_validate_email_address (recipient, invalid_char_position))
return TRUE;
stripped = g_strdup (recipient);
stripped = g_strstrip (stripped);
address = g_strndup (right_part+1, strlen (right_part) - 2);
g_free (right_part);
- valid = modest_text_utils_validate_email_address (address);
+ valid = modest_text_utils_validate_email_address (address, invalid_char_position);
g_free (address);
return valid;
} else {
(color->blue >> 12) & 0xf, (color->blue >> 8) & 0xf,
(color->blue >> 4) & 0xf, (color->blue) & 0xf);
}
+
+gchar *
+modest_text_utils_text_buffer_get_text (GtkTextBuffer *buffer)
+{
+ GtkTextIter start, end;
+ gchar *slice, *current;
+ GString *result = g_string_new ("");
+
+ g_return_val_if_fail (GTK_IS_TEXT_BUFFER (buffer), NULL);
+
+ gtk_text_buffer_get_start_iter (buffer, &start);
+ gtk_text_buffer_get_end_iter (buffer, &end);
+
+ slice = gtk_text_buffer_get_slice (buffer, &start, &end, FALSE);
+ current = slice;
+
+ while (current && current != '\0') {
+ if (g_utf8_get_char (current) == 0xFFFC) {
+ result = g_string_append_c (result, ' ');
+ current = g_utf8_next_char (current);
+ } else {
+ gchar *next = g_utf8_strchr (current, -1, 0xFFFC);
+ if (next == NULL) {
+ result = g_string_append (result, current);
+ } else {
+ result = g_string_append_len (result, current, next - current);
+ }
+ current = next;
+ }
+ }
+ g_free (slice);
+
+ return g_string_free (result, FALSE);
+
+}