X-Git-Url: http://git.maemo.org/git/?p=modest;a=blobdiff_plain;f=src%2Fmodest-text-utils.c;h=fb4da8426d6a96c8a42a6552903a0649aff382ee;hp=6b67214a2cd9a66d95ed41bbb7b5e710cc6ad889;hb=12a672c559d983c4e49a7e4054ee14c0177ecb1c;hpb=fdf3571635d4c34c057c7734785e8e6685036567;ds=sidebyside diff --git a/src/modest-text-utils.c b/src/modest-text-utils.c index 6b67214..fb4da84 100644 --- a/src/modest-text-utils.c +++ b/src/modest-text-utils.c @@ -28,26 +28,32 @@ */ + +#ifndef _GNU_SOURCE +#define _GNU_SOURCE +#endif /*_GNU_SOURCE*/ +#include /* for strcasestr */ + + #include -#include #include #include #include #include #include #include - +#include #ifdef HAVE_CONFIG_H #include #endif /*HAVE_CONFIG_H */ /* defines */ -#define FORWARD_STRING _("-----Forwarded Message-----") -#define FROM_STRING _("From:") -#define SENT_STRING _("Sent:") -#define TO_STRING _("To:") -#define SUBJECT_STRING _("Subject:") +#define FORWARD_STRING _("mcen_ia_editor_original_message") +#define FROM_STRING _("mail_va_from") +#define SENT_STRING _("mcen_fi_message_properties_sent") +#define TO_STRING _("mail_va_to") +#define SUBJECT_STRING _("mail_va_subject") #define EMPTY_STRING "" /* @@ -75,14 +81,40 @@ struct _url_match_t { const gchar* prefix; }; + +/* + * we mark the ampersand with \007 when converting text->html + * because after text->html we do hyperlink detecting, which + * could be screwed up by the ampersand. + * ie. 1<3 ==> 1\007lt;3 + */ +#define MARK_AMP '\007' +#define MARK_AMP_STR "\007" + +/* mark & separately, because they are parts of urls. + * ie. a&b => a\006amp;b, but a>b => a\007gt;b + * + * we need to handle '&' separately, because it can be part of URIs + * (as in href="http://foo.bar?a=1&b=1"), so inside those URIs + * we need to re-replace \006amp; with '&' again, while outside uri's + * it will be '&' + * + * yes, it's messy, but a consequence of doing text->html first, then hyperlinkify + */ +#define MARK_AMP_URI '\006' +#define MARK_AMP_URI_STR "\006" + + +/* note: match MARK_AMP_URI_STR as well, because after txt->html, a '&' will look like $(MARK_AMP_URI_STR)"amp;" */ #define MAIL_VIEWER_URL_MATCH_PATTERNS { \ - { "(file|rtsp|http|ftp|https)://[-A-Za-z0-9_$.+!*(),;:@%&=?/~#]+[-A-Za-z0-9_$%&=?/~#]",\ + { "(feed:|)(file|rtsp|http|ftp|https|mms|mmsh|webcal|feed|rtsp|rdp|lastfm|sip)://[-a-z0-9_$.+!*(),;:@%=\?/~#&" MARK_AMP_URI_STR \ + "]+[-a-z0-9_$%&" MARK_AMP_URI_STR "=?/~#]", \ NULL, NULL },\ - { "www\\.[-a-z0-9.]+[-a-z0-9](:[0-9]*)?(/[-A-Za-z0-9_$.+!*(),;:@%&=?/~#]*[^]}\\),?!;:\"]?)?",\ - NULL, "http://" },\ - { "ftp\\.[-a-z0-9.]+[-a-z0-9](:[0-9]*)?(/[-A-Za-z0-9_$.+!*(),;:@%&=?/~#]*[^]}\\),?!;:\"]?)?",\ + { "www\\.[-a-z0-9_$.+!*(),;:@%=?/~#" MARK_AMP_URI_STR "]+[-a-z0-9_$%" MARK_AMP_URI_STR "=?/~#]",\ + NULL, "http://" }, \ + { "ftp\\.[-a-z0-9_$.+!*(),;:@%=?/~#" MARK_AMP_URI_STR "]+[-a-z0-9_$%" MARK_AMP_URI_STR "=?/~#]",\ NULL, "ftp://" },\ - { "(voipto|callto|chatto|jabberto|xmpp):[-_a-z@0-9.\\+]+", \ + { "(jabberto|voipto|sipto|sip|chatto|skype|xmpp):[-_a-z@0-9.+]+", \ NULL, NULL}, \ { "mailto:[-_a-z0-9.\\+]+@[-_a-z0-9.]+", \ NULL, NULL},\ @@ -90,16 +122,30 @@ struct _url_match_t { NULL, "mailto:"}\ } +const gchar account_title_forbidden_chars[] = { + '\\', '/', ':', '*', '?', '\'', '<', '>', '|', '^' +}; +const gchar folder_name_forbidden_chars[] = { + '<', '>', ':', '\'', '/', '\\', '|', '?', '*', '^', '%', '$', '#', '&' +}; +const gchar user_name_forbidden_chars[] = { + '<', '>' +}; +const guint ACCOUNT_TITLE_FORBIDDEN_CHARS_LENGTH = G_N_ELEMENTS (account_title_forbidden_chars); +const guint FOLDER_NAME_FORBIDDEN_CHARS_LENGTH = G_N_ELEMENTS (folder_name_forbidden_chars); +const guint USER_NAME_FORBIDDEN_CHARS_LENGTH = G_N_ELEMENTS (user_name_forbidden_chars); + /* private */ static gchar* cite (const time_t sent_date, const gchar *from); -static void hyperlinkify_plain_text (GString *txt); +static void hyperlinkify_plain_text (GString *txt, gint offset); static gint cmp_offsets_reverse (const url_match_t *match1, const url_match_t *match2); -static GSList* get_url_matches (GString *txt); +static GSList* get_url_matches (GString *txt, gint offset); static GString* get_next_line (const char *b, const gsize blen, const gchar * iter); static int get_indent_level (const char *l); -static void unquote_line (GString * l); -static void append_quoted (GString * buf, const int indent, const GString * str, +static void unquote_line (GString * l, const gchar *quote_symbol); +static void append_quoted (GString * buf, const gchar *quote_symbol, + const int indent, const GString * str, const int cutpoint); static int get_breakpoint_utf8 (const gchar * s, const gint indent, const gint limit); static int get_breakpoint_ascii (const gchar * s, const gint indent, const gint limit); @@ -117,6 +163,8 @@ static gchar* modest_text_utils_quote_html (const gchar *text, GList *attachments, int limit); static gchar* get_email_from_address (const gchar *address); +static void remove_extra_spaces (gchar *string); + /* ******************************************************************* */ @@ -161,18 +209,21 @@ modest_text_utils_cite (const gchar *text, { gchar *retval; gchar *tmp_sig; - + g_return_val_if_fail (text, NULL); g_return_val_if_fail (content_type, NULL); + + if (!signature) { + tmp_sig = g_strdup (text); + } else { + tmp_sig = g_strconcat (text, "\n", MODEST_TEXT_UTILS_SIGNATURE_MARKER, "\n", signature, NULL); + } - if (!signature) - retval = g_strdup (""); - else if (!strcmp(content_type, "text/html")) { - tmp_sig = g_strconcat ("\n", signature, NULL); - retval = modest_text_utils_convert_to_html_body(tmp_sig); + if (strcmp (content_type, "text/html") == 0) { + retval = modest_text_utils_convert_to_html_body (tmp_sig, -1, TRUE); g_free (tmp_sig); } else { - retval = g_strconcat ("\n", signature, NULL); + retval = tmp_sig; } return retval; @@ -180,10 +231,12 @@ modest_text_utils_cite (const gchar *text, static gchar * forward_cite (const gchar *from, - const gchar *sent, - const gchar *to, - const gchar *subject) + const gchar *sent, + const gchar *to, + const gchar *subject) { + g_return_val_if_fail (sent, NULL); + return g_strdup_printf ("%s\n%s %s\n%s %s\n%s %s\n%s %s\n", FORWARD_STRING, FROM_STRING, (from)?from:"", @@ -229,6 +282,13 @@ modest_text_utils_strftime(char *s, gsize max, const char *fmt, time_t timet) { struct tm tm; + /* To prevent possible problems in strftime that could leave + garbage in the s variable */ + if (s) + s[0] = '\0'; + else + return 0; + /* does not work on old maemo glib: * g_date_set_time_t (&date, timet); */ @@ -237,29 +297,117 @@ modest_text_utils_strftime(char *s, gsize max, const char *fmt, time_t timet) } gchar * -modest_text_utils_derived_subject (const gchar *subject, const gchar *prefix) +modest_text_utils_derived_subject (const gchar *subject, gboolean is_reply) { - gchar *tmp; + gchar *tmp, *subject_dup, *retval, *prefix; + const gchar *untranslated_prefix; + gint prefix_len, untranslated_prefix_len; + gboolean translated_found = FALSE; + gboolean first_time; - g_return_val_if_fail (prefix, NULL); - - if (!subject) - return g_strdup (prefix); + if (!subject || subject[0] == '\0') + subject = _("mail_va_no_subject"); + + subject_dup = g_strdup (subject); + tmp = g_strchug (subject_dup); - tmp = g_strchug (g_strdup (subject)); + prefix = (is_reply) ? _("mail_va_re") : _("mail_va_fw"); + prefix = g_strconcat (prefix, ":", NULL); + prefix_len = g_utf8_strlen (prefix, -1); - if (!strncmp (tmp, prefix, strlen (prefix))) { - return tmp; + untranslated_prefix = (is_reply) ? "Re:" : "Fw:"; + untranslated_prefix_len = 3; + + /* We do not want things like "Re: Re: Re:" or "Fw: Fw:" so + delete the previous ones */ + first_time = TRUE; + do { + if (g_str_has_prefix (tmp, prefix)) { + tmp += prefix_len; + tmp = g_strchug (tmp); + /* Do not consider translated prefixes in the + middle of a Re:Re:..Re: like sequence */ + if (G_UNLIKELY (first_time)) + translated_found = TRUE; + } else if (g_str_has_prefix (tmp, untranslated_prefix)) { + tmp += untranslated_prefix_len; + tmp = g_strchug (tmp); + } else { + gchar *prefix_down, *tmp_down; + + /* We need this to properly check the cases of + some clients adding FW: instead of Fw: for + example */ + prefix_down = g_utf8_strdown (prefix, -1); + tmp_down = g_utf8_strdown (tmp, -1); + if (g_str_has_prefix (tmp_down, prefix_down)) { + tmp += prefix_len; + tmp = g_strchug (tmp); + g_free (prefix_down); + g_free (tmp_down); + } else { + g_free (prefix_down); + g_free (tmp_down); + break; + } + } + first_time = FALSE; + } while (tmp); + + if (!g_strcmp0 (subject, tmp)) { + /* normal case */ + retval = g_strdup_printf ("%s %s", untranslated_prefix, tmp); } else { - g_free (tmp); - return g_strdup_printf ("%s %s", prefix, subject); + if (translated_found) { + /* Found a translated prefix, i.e, "VS:" in Finish */ + retval = g_strdup_printf ("%s %s", prefix, tmp); + } else { + retval = g_strdup_printf ("%s %s", untranslated_prefix, tmp); + } + } + g_free (subject_dup); + g_free (prefix); + + return retval; +} + + +/* Performs a case-insensitive strstr for ASCII strings */ +static const gchar * +ascii_stristr(const gchar *haystack, const gchar *needle) +{ + int needle_len; + int haystack_len; + const gchar *pos; + const gchar *max_pos; + + if (haystack == NULL || needle == NULL) { + return haystack; /* as in strstr */ } + + needle_len = strlen(needle); + + if (needle_len == 0) { + return haystack; /* as in strstr */ + } + + haystack_len = strlen(haystack); + max_pos = haystack + haystack_len - needle_len; + + for (pos = haystack; pos <= max_pos; pos++) { + if (g_ascii_strncasecmp (pos, needle, needle_len) == 0) { + return pos; + } + } + + return NULL; } + gchar* modest_text_utils_remove_address (const gchar *address_list, const gchar *address) { - gchar *dup, *token, *ptr, *result; + gchar *dup, *token, *ptr = NULL, *result; GString *filtered_emails; gchar *email_address; @@ -269,21 +417,21 @@ modest_text_utils_remove_address (const gchar *address_list, const gchar *addres return g_strdup (address_list); email_address = get_email_from_address (address); - + /* search for substring */ - if (!strstr ((const char *) address_list, (const char *) email_address)) { + if (!ascii_stristr ((const char *) address_list, (const char *) email_address)) { g_free (email_address); return g_strdup (address_list); } dup = g_strdup (address_list); filtered_emails = g_string_new (NULL); - + token = strtok_r (dup, ",", &ptr); while (token != NULL) { /* Add to list if not found */ - if (!strstr ((const char *) token, (const char *) email_address)) { + if (!ascii_stristr ((const char *) token, (const char *) email_address)) { if (filtered_emails->len == 0) g_string_append_printf (filtered_emails, "%s", g_strstrip (token)); else @@ -301,48 +449,110 @@ modest_text_utils_remove_address (const gchar *address_list, const gchar *addres return result; } + +gchar* +modest_text_utils_remove_duplicate_addresses (const gchar *address_list) +{ + GSList *addresses, *cursor; + GHashTable *table; + gchar *new_list = NULL; + + g_return_val_if_fail (address_list, NULL); + + table = g_hash_table_new_full (g_str_hash, g_str_equal, g_free, NULL); + addresses = modest_text_utils_split_addresses_list (address_list); + + cursor = addresses; + while (cursor) { + const gchar* address = (const gchar*)cursor->data; + + /* We need only the email to just compare it and not + the full address which would make "a " + different from "a@a.com" */ + const gchar *email = get_email_from_address (address); + + /* ignore the address if already seen */ + if (g_hash_table_lookup (table, email) == 0) { + gchar *tmp; + + /* Include the full address and not only the + email in the returned list */ + if (!new_list) { + tmp = g_strdup (address); + } else { + tmp = g_strjoin (",", new_list, address, NULL); + g_free (new_list); + } + new_list = tmp; + + g_hash_table_insert (table, (gchar*)email, GINT_TO_POINTER(1)); + } + cursor = g_slist_next (cursor); + } + + g_hash_table_unref (table); + g_slist_foreach (addresses, (GFunc)g_free, NULL); + g_slist_free (addresses); + + if (new_list == NULL) + new_list = g_strdup (""); + + return new_list; +} + + static void -modest_text_utils_convert_buffer_to_html (GString *html, const gchar *data) +modest_text_utils_convert_buffer_to_html_start (GString *html, const gchar *data, gssize n) { - guint i; + guint i; gboolean space_seen = FALSE; - gsize len; guint break_dist = 0; /* distance since last break point */ - len = strlen (data); + if (n == -1) + n = strlen (data); /* replace with special html chars where needed*/ - for (i = 0; i != len; ++i) { - char kar = data[i]; + for (i = 0; i != n; ++i) { + guchar kar = data[i]; if (space_seen && kar != ' ') { - g_string_append_c (html, ' '); + g_string_append (html, " "); 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 + * Also, check that kar is ASCII to make sure that we + * don't break a UTF8 char in two */ - if (++break_dist == 256) { + if (++break_dist >= 256 && kar < 127) { 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 '\'' : g_string_append (html, "'"); break; - case '\n' : g_string_append (html, "
\n"); break_dist= 0; break; - case '\t' : g_string_append (html, "    "); break_dist=0; break; /* note the space at the end*/ + case 0: + case MARK_AMP: + case MARK_AMP_URI: + /* this is a temp place holder for '&'; we can only + * set the real '&' after hyperlink translation, otherwise + * we might screw that up */ + break; /* ignore embedded \0s and MARK_AMP */ + case '<' : g_string_append (html, MARK_AMP_STR "lt;"); break; + case '>' : g_string_append (html, MARK_AMP_STR "gt;"); break; + case '&' : g_string_append (html, MARK_AMP_URI_STR "amp;"); break; /* special case */ + case '"' : g_string_append (html, MARK_AMP_STR "quot;"); break; + + /* don't convert ' --> wpeditor will try to re-convert it... */ + //case '\'' : g_string_append (html, "'"); break; + case '\n' : g_string_append (html, "
\n");break_dist= 0; break; + case '\t' : g_string_append (html, MARK_AMP_STR "nbsp;" MARK_AMP_STR "nbsp;" MARK_AMP_STR "nbsp; "); + 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; } else space_seen = TRUE; break; @@ -352,11 +562,25 @@ modest_text_utils_convert_buffer_to_html (GString *html, const gchar *data) } } + +static void +modest_text_utils_convert_buffer_to_html_finish (GString *html) +{ + int i; + /* replace all our MARK_AMPs with real ones */ + for (i = 0; i != html->len; ++i) + if ((html->str)[i] == MARK_AMP || (html->str)[i] == MARK_AMP_URI) + (html->str)[i] = '&'; +} + + gchar* modest_text_utils_convert_to_html (const gchar *data) { GString *html; gsize len; + + g_return_val_if_fail (data, NULL); if (!data) return NULL; @@ -370,142 +594,192 @@ modest_text_utils_convert_to_html (const gchar *data) "" ""); - modest_text_utils_convert_buffer_to_html (html, data); + modest_text_utils_convert_buffer_to_html_start (html, data, -1); g_string_append (html, ""); if (len <= HYPERLINKIFY_MAX_LENGTH) - hyperlinkify_plain_text (html); + hyperlinkify_plain_text (html, 0); + modest_text_utils_convert_buffer_to_html_finish (html); + return g_string_free (html, FALSE); } gchar * -modest_text_utils_convert_to_html_body (const gchar *data) +modest_text_utils_convert_to_html_body (const gchar *data, gssize n, gboolean hyperlinkify) { GString *html; - gsize len; - + + g_return_val_if_fail (data, NULL); + if (!data) return NULL; - len = strlen (data); - html = g_string_sized_new (1.5 * len); /* just a guess... */ + if (n == -1) + n = strlen (data); + html = g_string_sized_new (1.5 * n); /* just a guess... */ - modest_text_utils_convert_buffer_to_html (html, data); + modest_text_utils_convert_buffer_to_html_start (html, data, n); - if (len < HYPERLINKIFY_MAX_LENGTH) - hyperlinkify_plain_text (html); + if (hyperlinkify && (n < HYPERLINKIFY_MAX_LENGTH)) + hyperlinkify_plain_text (html, 0); + modest_text_utils_convert_buffer_to_html_finish (html); + return g_string_free (html, FALSE); } void modest_text_utils_get_addresses_indexes (const gchar *addresses, GSList **start_indexes, GSList **end_indexes) { - gchar *current, *start, *last_blank; - gint start_offset = 0, current_offset = 0; + GString *str; + gchar *start, *cur; + + if (!addresses) + return; - g_return_if_fail (start_indexes != NULL); - g_return_if_fail (end_indexes != NULL); + if (strlen (addresses) == 0) + return; - start = (gchar *) addresses; - current = start; - last_blank = start; + str = g_string_new (""); + start = (gchar*) addresses; + cur = (gchar*) addresses; - while (*current != '\0') { - if ((start == current)&&((*current == ' ')||(*current == ',')||(*current == ';'))) { - start = g_utf8_next_char (start); - start_offset++; - last_blank = current; - } else if ((*current == ',')||(*current == ';')) { + for (cur = start; *cur != '\0'; cur = g_utf8_next_char (cur)) { + if (*cur == ',' || *cur == ';') { gint *start_index, *end_index; - start_index = g_new0(gint, 1); - end_index = g_new0(gint, 1); - *start_index = start_offset; - *end_index = current_offset; + gchar *next_char = g_utf8_next_char (cur); + + if (!g_utf8_strchr (start, (cur - start + 1), g_utf8_get_char ("@")) && + next_char && *next_char != '\n' && *next_char != '\0') + continue; + + start_index = g_new0 (gint, 1); + end_index = g_new0 (gint, 1); + *start_index = g_utf8_pointer_to_offset (addresses, start); + *end_index = g_utf8_pointer_to_offset (addresses, cur);; *start_indexes = g_slist_prepend (*start_indexes, start_index); *end_indexes = g_slist_prepend (*end_indexes, end_index); - start = g_utf8_next_char (current); - start_offset = current_offset + 1; - last_blank = start; - } else if (*current == '"') { - current = g_utf8_next_char (current); - current_offset ++; - while ((*current != '"')&&(*current != '\0')) { - current = g_utf8_next_char (current); - current_offset ++; - } + start = g_utf8_next_char (cur); } - - current = g_utf8_next_char (current); - current_offset ++; } - if (start != current) { - gint *start_index, *end_index; - start_index = g_new0(gint, 1); - end_index = g_new0(gint, 1); - *start_index = start_offset; - *end_index = current_offset; - *start_indexes = g_slist_prepend (*start_indexes, start_index); - *end_indexes = g_slist_prepend (*end_indexes, end_index); + if (start != cur) { + gint *start_index, *end_index; + start_index = g_new0 (gint, 1); + end_index = g_new0 (gint, 1); + *start_index = g_utf8_pointer_to_offset (addresses, start); + *end_index = g_utf8_pointer_to_offset (addresses, cur);; + *start_indexes = g_slist_prepend (*start_indexes, start_index); + *end_indexes = g_slist_prepend (*end_indexes, end_index); } - - *start_indexes = g_slist_reverse (*start_indexes); - *end_indexes = g_slist_reverse (*end_indexes); - return; + if (*start_indexes) + *start_indexes = g_slist_reverse (*start_indexes); + if (*end_indexes) + *end_indexes = g_slist_reverse (*end_indexes); } + GSList * modest_text_utils_split_addresses_list (const gchar *addresses) { - gchar *current, *start, *last_blank; - GSList *result = NULL; - - start = (gchar *) addresses; - current = start; - last_blank = start; - - while (*current != '\0') { - if ((start == current)&&((*current == ' ')||(*current == ',')||(*current == ';'))) { - start = g_utf8_next_char (start); - last_blank = current; - } else if ((*current == ',')||(*current == ';')) { - gchar *new_address = NULL; - new_address = g_strndup (start, current - last_blank); - result = g_slist_prepend (result, new_address); - start = g_utf8_next_char (current); - last_blank = start; - } else if (*current == '\"') { - if (current == start) { - current = g_utf8_next_char (current); - start = g_utf8_next_char (start); - } - while ((*current != '\"')&&(*current != '\0')) - current = g_utf8_next_char (current); + GSList *head; + const gchar *my_addrs = addresses; + const gchar *end; + gchar *addr; + gboolean after_at = FALSE; + + g_return_val_if_fail (addresses, NULL); + + /* skip any space, ',', ';' '\n' at the start */ + while (my_addrs && (my_addrs[0] == ' ' || my_addrs[0] == ',' || + my_addrs[0] == ';' || my_addrs[0] == '\n')) + ++my_addrs; + + /* are we at the end of addresses list? */ + if (!my_addrs[0]) + return NULL; + + /* nope, we are at the start of some address + * now, let's find the end of the address */ + end = my_addrs + 1; + while (end[0] && end[0] != ';' && !(after_at && end[0] == ',')) { + if (end[0] == '\"') { + while (end[0] && end[0] != '\"') + ++end; } - - current = g_utf8_next_char (current); + if (end[0] == '@') { + after_at = TRUE; + } + if ((end[0] && end[0] == '>')&&(end[1] && end[1] == ',')) { + ++end; + break; + } + ++end; } - if (start != current) { - gchar *new_address = NULL; - new_address = g_strndup (start, current - last_blank); - result = g_slist_prepend (result, new_address); + /* we got the address; copy it and remove trailing whitespace */ + addr = g_strndup (my_addrs, end - my_addrs); + g_strchomp (addr); + + remove_extra_spaces (addr); + + head = g_slist_append (NULL, addr); + head->next = modest_text_utils_split_addresses_list (end); /* recurse */ + + return head; +} + +gchar * +modest_text_utils_join_addresses (const gchar *from, + const gchar *to, + const gchar *cc, + const gchar *bcc) +{ + GString *buffer; + gboolean add_separator = FALSE; + + buffer = g_string_new (""); + + if (from && strlen (from)) { + buffer = g_string_append (buffer, from); + add_separator = TRUE; } + if (to && strlen (to)) { + if (add_separator) + buffer = g_string_append (buffer, "; "); + else + add_separator = TRUE; - result = g_slist_reverse (result); - return result; + buffer = g_string_append (buffer, to); + } + if (cc && strlen (cc)) { + if (add_separator) + buffer = g_string_append (buffer, "; "); + else + add_separator = TRUE; + + buffer = g_string_append (buffer, cc); + } + if (bcc && strlen (bcc)) { + if (add_separator) + buffer = g_string_append (buffer, "; "); + else + add_separator = TRUE; + + buffer = g_string_append (buffer, bcc); + } + return g_string_free (buffer, FALSE); } void modest_text_utils_address_range_at_position (const gchar *recipients_list, - gint position, - gint *start, - gint *end) + guint position, + guint *start, + guint *end) { gchar *current = NULL; gint range_start = 0; @@ -513,8 +787,12 @@ modest_text_utils_address_range_at_position (const gchar *recipients_list, gint index; gboolean is_quoted = FALSE; + g_return_if_fail (recipients_list); + g_return_if_fail (position < g_utf8_strlen(recipients_list, -1)); + index = 0; - for (current = (gchar *) recipients_list; *current != '\0'; current = g_utf8_find_next_char (current, NULL)) { + for (current = (gchar *) recipients_list; *current != '\0'; + current = g_utf8_find_next_char (current, NULL)) { gunichar c = g_utf8_get_char (current); if ((c == ',') && (!is_quoted)) { @@ -538,6 +816,30 @@ modest_text_utils_address_range_at_position (const gchar *recipients_list, *end = range_end; } +gchar * +modest_text_utils_address_with_standard_length (const gchar *recipients_list) +{ + gchar ** splitted; + gchar ** current; + GString *buffer = g_string_new (""); + + splitted = g_strsplit (recipients_list, "\n", 0); + current = splitted; + while (*current) { + gchar *line; + if (current != splitted) + buffer = g_string_append_c (buffer, '\n'); + line = g_strndup (*splitted, 1000); + buffer = g_string_append (buffer, line); + g_free (line); + current++; + } + + g_strfreev (splitted); + + return g_string_free (buffer, FALSE); +} + /* ******************************************************************* */ /* ************************* UTILIY FUNCTIONS ************************ */ @@ -582,7 +884,7 @@ get_indent_level (const char *l) /* if we hit the signature marker "-- ", we return -(indent + 1). This * stops reformatting. */ - if (strcmp (l, "-- ") == 0) { + if (strcmp (l, MODEST_TEXT_UTILS_SIGNATURE_MARKER) == 0) { return -1 - indent; } else { return indent; @@ -590,15 +892,17 @@ get_indent_level (const char *l) } static void -unquote_line (GString * l) +unquote_line (GString * l, const gchar *quote_symbol) { gchar *p; + gint quote_len; p = l->str; + quote_len = strlen (quote_symbol); while (p[0]) { - if (p[0] == '>') { - if (p[1] == ' ') { - p++; + if (g_str_has_prefix (p, quote_symbol)) { + if (p[quote_len] == ' ') { + p += quote_len; } } else { break; @@ -609,15 +913,19 @@ unquote_line (GString * l) } static void -append_quoted (GString * buf, int indent, const GString * str, +append_quoted (GString * buf, const gchar *quote_symbol, + int indent, const GString * str, const int cutpoint) { int i; + gchar *quote_concat; indent = indent < 0 ? abs (indent) - 1 : indent; + quote_concat = g_strconcat (quote_symbol, " ", NULL); for (i = 0; i <= indent; i++) { - g_string_append (buf, "> "); + g_string_append (buf, quote_concat); } + g_free (quote_concat); if (cutpoint > 0) { g_string_append_len (buf, str->str, cutpoint); } else { @@ -706,32 +1014,25 @@ quoted_attachments (GList *attachments) } -static gchar * -modest_text_utils_quote_plain_text (const gchar *text, - const gchar *cite, - const gchar *signature, - GList *attachments, - int limit) +static GString * +modest_text_utils_quote_body (GString *output, const gchar *text, + const gchar *quote_symbol, + int limit) { + const gchar *iter; - gint indent, breakpoint, rem_indent = 0; - GString *q, *l, *remaining; gsize len; - gchar *attachments_string = NULL; - - /* remaining will store the rest of the line if we have to break it */ - q = g_string_new ("\n"); - q = g_string_append (q, cite); - q = g_string_append_c (q, '\n'); - remaining = g_string_new (""); + gint indent, breakpoint, rem_indent = 0; + GString *l, *remaining; iter = text; len = strlen(text); + remaining = g_string_new (""); do { l = get_next_line (text, len, iter); iter = iter + l->len + 1; indent = get_indent_level (l->str); - unquote_line (l); + unquote_line (l, quote_symbol); if (remaining->len) { if (l->len && indent == rem_indent) { @@ -739,17 +1040,20 @@ modest_text_utils_quote_plain_text (const gchar *text, g_string_prepend (l, remaining->str); } else { do { + gunichar remaining_first; breakpoint = get_breakpoint (remaining->str, rem_indent, limit); - append_quoted (q, rem_indent, + append_quoted (output, quote_symbol, rem_indent, remaining, breakpoint); g_string_erase (remaining, 0, breakpoint); - if (remaining->str[0] == ' ') { - g_string_erase (remaining, 0, - 1); + remaining_first = g_utf8_get_char_validated (remaining->str, remaining->len); + if (remaining_first != ((gunichar) -1)) { + if (g_unichar_isspace (remaining_first)) { + g_string_erase (remaining, 0, g_utf8_next_char (remaining->str) - remaining->str); + } } } while (remaining->len); } @@ -761,22 +1065,54 @@ modest_text_utils_quote_plain_text (const gchar *text, g_string_erase (remaining, 0, 1); } rem_indent = indent; - append_quoted (q, indent, l, breakpoint); + append_quoted (output, quote_symbol, indent, l, breakpoint); g_string_free (l, TRUE); } while ((iter < text + len) || (remaining->str[0])); - attachments_string = quoted_attachments (attachments); - q = g_string_append (q, attachments_string); - g_free (attachments_string); + return output; +} + +static gchar * +modest_text_utils_quote_plain_text (const gchar *text, + const gchar *cite, + const gchar *signature, + GList *attachments, + int limit) +{ + GString *q; + gchar *attachments_string = NULL; + + q = g_string_new (""); if (signature != NULL) { - q = g_string_append_c (q, '\n'); + g_string_append_printf (q, "\n%s\n", MODEST_TEXT_UTILS_SIGNATURE_MARKER); q = g_string_append (q, signature); } + q = g_string_append (q, "\n"); + q = g_string_append (q, cite); + q = g_string_append_c (q, '\n'); + + q = modest_text_utils_quote_body (q, text, ">", limit); + + attachments_string = quoted_attachments (attachments); + q = g_string_append (q, attachments_string); + g_free (attachments_string); + return g_string_free (q, FALSE); } +static void +quote_html_add_to_gstring (GString *string, + const gchar *text) +{ + if (text && strcmp (text, "")) { + gchar *html_text = modest_text_utils_convert_to_html_body (text, -1, TRUE); + g_string_append_printf (string, "%s
", html_text); + g_free (html_text); + } +} + static gchar* modest_text_utils_quote_html (const gchar *text, const gchar *cite, @@ -784,37 +1120,37 @@ modest_text_utils_quote_html (const gchar *text, GList *attachments, int limit) { - gchar *result = NULL; - gchar *signature_result = NULL; - const gchar *format = \ - "\n" \ - "\n" \ - "\n" \ - "
%s
" \ - "
%s
%s
%s
\n" \ - "\n" \ - "\n"; - gchar *attachments_string = NULL; - gchar *q_attachments_string = NULL; - gchar *q_cite = NULL; - gchar *html_text = NULL; - - if (signature == NULL) - signature_result = g_strdup (""); - else - signature_result = modest_text_utils_convert_to_html_body (signature); + GString *result_string; + + result_string = + g_string_new ( \ + "\n" \ + "\n" \ + "\n
\n"); + + if (text || cite || signature) { + GString *quoted_text; + g_string_append (result_string, "
\n");
+		if (signature) {
+			quote_html_add_to_gstring (result_string, MODEST_TEXT_UTILS_SIGNATURE_MARKER);
+			quote_html_add_to_gstring (result_string, signature);
+		}
+		quote_html_add_to_gstring (result_string, cite);
+		quoted_text = g_string_new ("");
+		quoted_text = modest_text_utils_quote_body (quoted_text, (text) ? text : "", ">", limit);
+		quote_html_add_to_gstring (result_string, quoted_text->str);
+		g_string_free (quoted_text, TRUE);
+		if (attachments) {
+			gchar *attachments_string = quoted_attachments (attachments);
+			quote_html_add_to_gstring (result_string, attachments_string);
+			g_free (attachments_string);
+		}
+		g_string_append (result_string, "
"); + } + g_string_append (result_string, ""); + g_string_append (result_string, ""); - attachments_string = quoted_attachments (attachments); - q_attachments_string = modest_text_utils_convert_to_html_body (attachments_string); - q_cite = modest_text_utils_convert_to_html_body (cite); - html_text = modest_text_utils_convert_to_html_body (text); - result = g_strdup_printf (format, signature_result, q_cite, html_text, q_attachments_string); - g_free (q_cite); - g_free (html_text); - g_free (attachments_string); - g_free (q_attachments_string); - g_free (signature_result); - return result; + return g_string_free (result_string, FALSE); } static gint @@ -823,34 +1159,86 @@ cmp_offsets_reverse (const url_match_t *match1, const url_match_t *match2) return match2->offset - match1->offset; } +static gint url_matches_block = 0; +static url_match_pattern_t patterns[] = MAIL_VIEWER_URL_MATCH_PATTERNS; +static GMutex *url_patterns_mutex = NULL; + + +static gboolean +compile_patterns () +{ + guint i; + const size_t pattern_num = sizeof(patterns)/sizeof(url_match_pattern_t); + for (i = 0; i != pattern_num; ++i) { + patterns[i].preg = g_slice_new0 (regex_t); + + /* this should not happen */ + if (regcomp (patterns[i].preg, patterns[i].regex, + REG_ICASE|REG_EXTENDED|REG_NEWLINE) != 0) { + g_warning ("%s: error in regexp:\n%s\n", __FUNCTION__, patterns[i].regex); + return FALSE; + } + } + return TRUE; +} + +static void +free_patterns () +{ + guint i; + const size_t pattern_num = sizeof(patterns)/sizeof(url_match_pattern_t); + for (i = 0; i != pattern_num; ++i) { + regfree (patterns[i].preg); + g_slice_free (regex_t, patterns[i].preg); + } /* don't free patterns itself -- it's static */ +} + +void +modest_text_utils_hyperlinkify_begin (void) +{ + + if (url_patterns_mutex == NULL) { + url_patterns_mutex = g_mutex_new (); + } + g_mutex_lock (url_patterns_mutex); + if (url_matches_block == 0) + compile_patterns (); + url_matches_block ++; + g_mutex_unlock (url_patterns_mutex); +} + +void +modest_text_utils_hyperlinkify_end (void) +{ + g_mutex_lock (url_patterns_mutex); + url_matches_block--; + if (url_matches_block <= 0) + free_patterns (); + g_mutex_unlock (url_patterns_mutex); +} + static GSList* -get_url_matches (GString *txt) +get_url_matches (GString *txt, gint offset) { regmatch_t rm; - guint rv, i, offset = 0; + guint rv, i, tmp_offset = 0; GSList *match_list = NULL; - static url_match_pattern_t patterns[] = MAIL_VIEWER_URL_MATCH_PATTERNS; const size_t pattern_num = sizeof(patterns)/sizeof(url_match_pattern_t); /* initalize the regexps */ - for (i = 0; i != pattern_num; ++i) { - patterns[i].preg = g_slice_new0 (regex_t); + modest_text_utils_hyperlinkify_begin (); - /* this should not happen */ - g_return_val_if_fail (regcomp (patterns[i].preg, patterns[i].regex, - REG_ICASE|REG_EXTENDED|REG_NEWLINE) == 0, NULL); - } /* find all the matches */ for (i = 0; i != pattern_num; ++i) { - offset = 0; + tmp_offset = offset; while (1) { url_match_t *match; gboolean is_submatch; GSList *cursor; - if ((rv = regexec (patterns[i].preg, txt->str + offset, 1, &rm, 0)) != 0) { + if ((rv = regexec (patterns[i].preg, txt->str + tmp_offset, 1, &rm, 0)) != 0) { g_return_val_if_fail (rv == REG_NOMATCH, NULL); /* this should not happen */ break; /* try next regexp */ } @@ -863,7 +1251,7 @@ get_url_matches (GString *txt) while (cursor && !is_submatch) { const url_match_t *old_match = (const url_match_t *) cursor->data; - guint new_offset = offset + rm.rm_so; + guint new_offset = tmp_offset + rm.rm_so; is_submatch = (new_offset > old_match->offset && new_offset < old_match->offset + old_match->len); cursor = g_slist_next (cursor); @@ -872,21 +1260,16 @@ get_url_matches (GString *txt) if (!is_submatch) { /* make a list of our matches ( tupels)*/ match = g_slice_new (url_match_t); - match->offset = offset + rm.rm_so; + match->offset = tmp_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; + } + tmp_offset += rm.rm_eo; } } - for (i = 0; i != pattern_num; ++i) { - regfree (patterns[i].preg); - g_slice_free (regex_t, patterns[i].preg); - } /* don't free patterns itself -- it's static */ + modest_text_utils_hyperlinkify_end (); /* now sort the list, so the matches are in reverse order of occurence. * that way, we can do the replacements starting from the end, so we don't need @@ -899,11 +1282,34 @@ get_url_matches (GString *txt) +/* replace all occurences of needle in haystack with repl*/ +static gchar* +replace_string (const gchar *haystack, const gchar *needle, gchar repl) +{ + gchar *str, *cursor; + + if (!haystack || !needle || strlen(needle) == 0) + return haystack ? g_strdup(haystack) : NULL; + + str = g_strdup (haystack); + + for (cursor = str; cursor && *cursor; ++cursor) { + if (g_str_has_prefix (cursor, needle)) { + cursor[0] = repl; + memmove (cursor + 1, + cursor + strlen (needle), + strlen (cursor + strlen (needle)) + 1); + } + } + + return str; +} + static void -hyperlinkify_plain_text (GString *txt) +hyperlinkify_plain_text (GString *txt, gint offset) { GSList *cursor; - GSList *match_list = get_url_matches (txt); + GSList *match_list = get_url_matches (txt, offset); /* we will work backwards, so the offsets stay valid */ for (cursor = match_list; cursor; cursor = cursor->next) { @@ -912,10 +1318,17 @@ hyperlinkify_plain_text (GString *txt) gchar *url = g_strndup (txt->str + match->offset, match->len); gchar *repl = NULL; /* replacement */ + /* the string still contains $(MARK_AMP_URI_STR)"amp;" for each + * '&' in the original, because of the text->html conversion. + * in the href-URL (and only there), we must convert that back to + * '&' + */ + gchar *href_url = replace_string (url, MARK_AMP_URI_STR "amp;", '&'); + /* the prefix is NULL: use the one that is already there */ repl = g_strdup_printf ("%s", match->prefix ? match->prefix : EMPTY_STRING, - url, url); + href_url, url); /* replace the old thing with our hyperlink * replacement thing */ @@ -924,6 +1337,7 @@ hyperlinkify_plain_text (GString *txt) g_free (url); g_free (repl); + g_free (href_url); g_slice_free (url_match_t, match); } @@ -931,43 +1345,93 @@ hyperlinkify_plain_text (GString *txt) g_slist_free (match_list); } +void +modest_text_utils_hyperlinkify (GString *string_buffer) +{ + gchar *after_body; + gint offset = 0; + after_body = strstr (string_buffer->str, ""); + if (after_body != NULL) + offset = after_body - string_buffer->str; + hyperlinkify_plain_text (string_buffer, offset); +} -gchar* + +/* for optimization reasons, we change the string in-place */ +void modest_text_utils_get_display_address (gchar *address) { - gchar *cursor; + int i; + + g_return_if_fail (address); if (!address) - return NULL; + return; - g_return_val_if_fail (g_utf8_validate (address, -1, NULL), NULL); + /* should not be needed, and otherwise, we probably won't screw up the address + * more than it already is :) + * g_return_val_if_fail (g_utf8_validate (address, -1, NULL), NULL); + * */ - g_strchug (address); /* remove leading whitespace */ + /* remove leading whitespace */ + if (address[0] == ' ') + g_strchug (address); + + for (i = 0; address[i]; ++i) { + if (address[i] == '<') { + if (G_UNLIKELY(i == 0)) { + break; /* there's nothing else, leave it */ + }else { + address[i] = '\0'; /* terminate the string here */ + break; + } + } + } - /* from display name */ - cursor = g_strstr_len (address, strlen(address), "<"); - if (cursor == address) /* there's nothing else? leave it */ - return address; - if (cursor) - cursor[0]='\0'; + g_strchomp (address); +} - /* remove (bla bla) from display name */ - cursor = g_strstr_len (address, strlen(address), "("); - if (cursor == address) /* there's nothing else? leave it */ - return address; - if (cursor) - cursor[0]='\0'; - g_strchomp (address); /* remove trailing whitespace */ +gchar * +modest_text_utils_get_display_addresses (const gchar *recipients) +{ + gchar *addresses; + GSList *recipient_list; + + addresses = NULL; + recipient_list = modest_text_utils_split_addresses_list (recipients); + if (recipient_list) { + GString *add_string = g_string_sized_new (strlen (recipients)); + GSList *iter = recipient_list; + gboolean first = TRUE; + + while (iter) { + /* Strings are changed in place */ + modest_text_utils_get_display_address ((gchar *) iter->data); + if (G_UNLIKELY (first)) { + g_string_append_printf (add_string, "%s", (gchar *) iter->data); + first = FALSE; + } else { + g_string_append_printf (add_string, ", %s", (gchar *) iter->data); + } + iter = g_slist_next (iter); + } + g_slist_foreach (recipient_list, (GFunc) g_free, NULL); + g_slist_free (recipient_list); + addresses = g_string_free (add_string, FALSE); + } - return address; + return addresses; } + gchar * modest_text_utils_get_email_address (const gchar *full_address) { const gchar *left, *right; + + g_return_val_if_fail (full_address, NULL); if (!full_address) return NULL; @@ -988,97 +1452,238 @@ modest_text_utils_get_email_address (const gchar *full_address) gint modest_text_utils_get_subject_prefix_len (const gchar *sub) { - gint i; - static const gchar* prefix[] = { - "Re:", "RE:", "RV:", "re:" - "Fwd:", "FWD:", "FW:", "fwd:", "Fw:", "fw:", NULL - }; - - if (!sub || (sub[0] != 'R' && sub[0] != 'F' && sub[0] != 'r' && sub[0] != 'f')) /* optimization */ - return 0; + gint prefix_len = 0; - i = 0; + g_return_val_if_fail (sub, 0); + + if (!sub) + return 0; - while (prefix[i]) { - if (g_str_has_prefix(sub, prefix[i])) { - int prefix_len = strlen(prefix[i]); - if (sub[prefix_len] == ' ') - ++prefix_len; /* ignore space after prefix as well */ - return prefix_len; - } - ++i; + /* optimization: "Re", "RE", "re","Fwd", "FWD", "fwd","FW","Fw", "fw" */ + if (sub[0] != 'R' && sub[0] != 'F' && sub[0] != 'r' && sub[0] != 'f') + return 0; + else if (sub[0] && sub[1] != 'e' && sub[1] != 'E' && sub[1] != 'w' && sub[1] != 'W') + return 0; + + prefix_len = 2; + if (sub[2] == 'd') + ++prefix_len; + + /* skip over a [...] block */ + if (sub[prefix_len] == '[') { + int c = prefix_len + 1; + while (sub[c] && sub[c] != ']') + ++c; + if (!sub[c]) + return 0; /* no end to the ']' found */ + else + prefix_len = c + 1; } - return 0; + + /* did we find the ':' ? */ + if (sub[prefix_len] == ':') { + ++prefix_len; + if (sub[prefix_len] == ' ') + ++prefix_len; + prefix_len += modest_text_utils_get_subject_prefix_len (sub + prefix_len); +/* g_warning ("['%s','%s']", sub, (char*) sub + prefix_len); */ + return prefix_len; + } else + return 0; } gint modest_text_utils_utf8_strcmp (const gchar* s1, const gchar *s2, gboolean insensitive) { - gint result = 0; - gchar *n1, *n2; - /* work even when s1 and/or s2 == NULL */ +/* work even when s1 and/or s2 == NULL */ if (G_UNLIKELY(s1 == s2)) return 0; - - /* if it's not case sensitive */ - if (!insensitive) - return strcmp (s1 ? s1 : "", s2 ? s2 : ""); - - n1 = g_utf8_collate_key (s1 ? s1 : "", -1); - n2 = g_utf8_collate_key (s2 ? s2 : "", -1); + if (G_UNLIKELY(!s1)) + return -1; + if (G_UNLIKELY(!s2)) + return 1; - result = strcmp (n1, n2); + /* if it's not case sensitive */ + if (!insensitive) { + + /* optimization: shortcut if first char is ascii */ + if (((s1[0] & 0x80)== 0) && ((s2[0] & 0x80) == 0) && + (s1[0] != s2[0])) + return s1[0] - s2[0]; + + return g_utf8_collate (s1, s2); + + } else { + gint result; + gchar *n1, *n2; - g_free (n1); - g_free (n2); + /* optimization: shortcut if first char is ascii */ + if (((s1[0] & 0x80) == 0) && ((s2[0] & 0x80) == 0) && + (tolower(s1[0]) != tolower (s2[0]))) + return tolower(s1[0]) - tolower(s2[0]); + + n1 = g_utf8_strdown (s1, -1); + n2 = g_utf8_strdown (s2, -1); + + result = g_utf8_collate (n1, n2); + + g_free (n1); + g_free (n2); - return result; + return result; + } } -gchar* +const gchar* modest_text_utils_get_display_date (time_t date) { - time_t now; - static const guint BUF_SIZE = 64; - static const guint ONE_DAY = 24 * 60 * 60; /* seconds in one day */ - gchar date_buf[BUF_SIZE]; - gchar today_buf [BUF_SIZE]; +#define DATE_BUF_SIZE 64 + static gchar date_buf[DATE_BUF_SIZE]; + + /* calculate the # of days since epoch for + * for today and for the date provided + * based on idea from pvanhoof */ + int day = time(NULL) / (24 * 60 * 60); + int date_day = date / (24 * 60 * 60); + + /* if it's today, show the time, if it's not today, show the date instead */ + + /* TODO: take into account the system config for 24/12h */ +#ifdef MODEST_TOOLKIT_HILDON2 + if (day == date_day) /* is the date today? */ + modest_text_utils_strftime (date_buf, DATE_BUF_SIZE, _HL("wdgt_va_24h_time"), date); + else + modest_text_utils_strftime (date_buf, DATE_BUF_SIZE, _HL("wdgt_va_date"), date); +#else + if (day == date_day) /* is the date today? */ + modest_text_utils_strftime (date_buf, DATE_BUF_SIZE, "%X", date); + else + modest_text_utils_strftime (date_buf, DATE_BUF_SIZE, "%x", date); +#endif + + return date_buf; /* this is a static buffer, don't free! */ +} - modest_text_utils_strftime (date_buf, BUF_SIZE, "%x", date); - now = time (NULL); - /* 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); +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; + gint i; + const gchar **cursor = NULL; + const gchar *forbidden_names[] = { /* windows does not like these */ + "CON", "PRN", "AUX", "NUL", ".", "..", "cur", "tmp", "new", + NULL /* cur, tmp, new are reserved for Maildir */ + }; + + /* cannot be NULL */ + if (!folder_name) + return FALSE; + + /* cannot be empty */ + len = strlen(folder_name); + if (len == 0) + return FALSE; + + /* cannot start with a dot, vfat does not seem to like that */ + if (folder_name[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 (i = 0; i < len; i++) + if (modest_text_utils_is_forbidden_char (folder_name[i], FOLDER_NAME_FORBIDDEN_CHARS)) + return FALSE; + + /* Cannot contain Windows port numbers. I'd like to use GRegex + but it's still not available in Maemo. sergio */ + if (!g_ascii_strncasecmp (folder_name, "LPT", 3) || + !g_ascii_strncasecmp (folder_name, "COM", 3)) { + glong val; + gchar *endptr; + + /* We skip the first 3 characters for the + comparison */ + val = strtol(folder_name+3, &endptr, 10); + + /* If the conversion to long succeeded then the string + is not valid for us */ + if (*endptr == '\0') + return FALSE; + else + return TRUE; + } + + /* cannot contain a forbidden word */ + if (len <= 4) { + for (cursor = forbidden_names; cursor && *cursor; ++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-]*[a-z0-9]$"; - /* 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); + g_return_val_if_fail (domain, FALSE); + + if (!domain) + return FALSE; + + memset (&rx, 0, sizeof(regex_t)); /* coverity wants this... */ + + /* 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; } - return g_strdup(date_buf); + 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) +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 = "()<>@,;:\\\"[]&"; - - if (invalid_char_position != NULL) + + if (invalid_char_position) *invalid_char_position = NULL; - + + g_return_val_if_fail (email_address, FALSE); + + /* check that the email adress contains exactly one @ */ + if (!strstr(email_address, "@") || + (strstr(email_address, "@") != g_strrstr(email_address, "@"))) + return FALSE; + /* first we validate the name portion (name@domain) */ for (c = email_address; *c; c++) { if (*c == '\"' && @@ -1119,7 +1724,7 @@ modest_text_utils_validate_email_address (const gchar *email_address, const gcha return FALSE; do { if (*c == '.') { - if (c == domain || *(c - 1) == '.') + if (c == domain || *(c - 1) == '.' || *(c + 1) == '\0') return FALSE; count++; } @@ -1142,8 +1747,14 @@ modest_text_utils_validate_recipient (const gchar *recipient, const gchar **inva gchar *right_part; gboolean has_error = FALSE; + if (invalid_char_position) + *invalid_char_position = NULL; + + g_return_val_if_fail (recipient, FALSE); + if (modest_text_utils_validate_email_address (recipient, invalid_char_position)) return TRUE; + stripped = g_strdup (recipient); stripped = g_strstrip (stripped); current = stripped; @@ -1155,6 +1766,7 @@ modest_text_utils_validate_recipient (const gchar *recipient, const gchar **inva /* quoted string */ if (*current == '\"') { + gchar *last_quote = NULL; current = g_utf8_next_char (current); has_error = TRUE; for (; *current != '\0'; current = g_utf8_next_char (current)) { @@ -1170,9 +1782,11 @@ modest_text_utils_validate_recipient (const gchar *recipient, const gchar **inva } else if (*current == '\"') { has_error = FALSE; current = g_utf8_next_char (current); - break; + last_quote = current; } } + if (last_quote) + current = g_utf8_next_char (last_quote); } else { has_error = TRUE; for (current = stripped ; *current != '\0'; current = g_utf8_next_char (current)) { @@ -1192,6 +1806,9 @@ modest_text_utils_validate_recipient (const gchar *recipient, const gchar **inva g_free (stripped); right_part = g_strstrip (right_part); + if (g_str_has_suffix (right_part, ",") || g_str_has_suffix (right_part, ";")) + right_part [(strlen (right_part) - 1)] = '\0'; + if (g_str_has_prefix (right_part, "<") && g_str_has_suffix (right_part, ">")) { gchar *address; @@ -1217,19 +1834,19 @@ modest_text_utils_get_display_size (guint64 size) const guint GB=1024 * MB; if (size == 0) - return g_strdup_printf(_FM("sfil_li_size_kb"), 0); - if (0 < size && size < KB) - return g_strdup_printf (_FM("sfil_li_size_kb"), 1); + return g_strdup_printf (_FM("sfil_li_size_kb"), (int) 0); + if (0 <= size && size < KB) + return g_strdup_printf (_FM("sfil_li_size_1kb_99kb"), (int) 1); else if (KB <= size && size < 100 * KB) - return g_strdup_printf (_FM("sfil_li_size_1kb_99kb"), size / KB); + return g_strdup_printf (_FM("sfil_li_size_1kb_99kb"), (int) size / KB); else if (100*KB <= size && size < MB) - return g_strdup_printf (_FM("sfil_li_size_100kb_1mb"), (float) size / MB); + return g_strdup_printf (_FM("sfil_li_size_100kb_1mb"), (int) size / KB); else if (MB <= size && size < 10*MB) return g_strdup_printf (_FM("sfil_li_size_1mb_10mb"), (float) size / MB); else if (10*MB <= size && size < GB) - return g_strdup_printf (_FM("sfil_li_size_10mb_1gb"), size / MB); + return g_strdup_printf (_FM("sfil_li_size_10mb_1gb"), (float) size / MB); else - return g_strdup_printf (_FM("sfil_li_size_1gb_or_greater"), (float) size / GB); + return g_strdup_printf (_FM("sfil_li_size_1gb_or_greater"), (float) size / GB); } static gchar * @@ -1246,9 +1863,10 @@ get_email_from_address (const gchar * address) return g_strndup (left_limit + 1, (right_limit - left_limit) - 1); } -gchar * +gchar * modest_text_utils_get_color_string (GdkColor *color) { + g_return_val_if_fail (color, NULL); return g_strdup_printf ("#%x%x%x%x%x%x%x%x%x%x%x%x", (color->red >> 12) & 0xf, (color->red >> 8) & 0xf, @@ -1266,8 +1884,8 @@ modest_text_utils_text_buffer_get_text (GtkTextBuffer *buffer) gchar *slice, *current; GString *result = g_string_new (""); - g_return_val_if_fail (GTK_IS_TEXT_BUFFER (buffer), NULL); - + g_return_val_if_fail (buffer && GTK_IS_TEXT_BUFFER (buffer), NULL); + gtk_text_buffer_get_start_iter (buffer, &start); gtk_text_buffer_get_end_iter (buffer, &end); @@ -1293,3 +1911,335 @@ modest_text_utils_text_buffer_get_text (GtkTextBuffer *buffer) return g_string_free (result, FALSE); } + +gboolean +modest_text_utils_is_forbidden_char (const gchar character, + ModestTextUtilsForbiddenCharType type) +{ + gint i, len; + const gchar *forbidden_chars = NULL; + + /* We need to get the length in the switch because the + compiler needs to know the size at compile time */ + switch (type) { + case ACCOUNT_TITLE_FORBIDDEN_CHARS: + forbidden_chars = account_title_forbidden_chars; + len = G_N_ELEMENTS (account_title_forbidden_chars); + break; + case FOLDER_NAME_FORBIDDEN_CHARS: + forbidden_chars = folder_name_forbidden_chars; + len = G_N_ELEMENTS (folder_name_forbidden_chars); + break; + case USER_NAME_FORBIDDEN_NAMES: + forbidden_chars = user_name_forbidden_chars; + len = G_N_ELEMENTS (user_name_forbidden_chars); + break; + default: + g_return_val_if_reached (TRUE); + } + + for (i = 0; i < len ; i++) + if (forbidden_chars[i] == character) + return TRUE; + + return FALSE; /* it's valid! */ +} + +gchar * +modest_text_utils_label_get_selection (GtkLabel *label) +{ + gint start, end; + gchar *selection; + + if (gtk_label_get_selection_bounds (GTK_LABEL (label), &start, &end)) { + const gchar *start_offset; + const gchar *end_offset; + start_offset = gtk_label_get_text (GTK_LABEL (label)); + start_offset = g_utf8_offset_to_pointer (start_offset, start); + end_offset = gtk_label_get_text (GTK_LABEL (label)); + end_offset = g_utf8_offset_to_pointer (end_offset, end); + selection = g_strndup (start_offset, end_offset - start_offset); + return selection; + } else { + return g_strdup (""); + } +} + +static gboolean +_forward_search_image_char (gunichar ch, + gpointer userdata) +{ + return (ch == 0xFFFC); +} + +gboolean +modest_text_utils_buffer_selection_is_valid (GtkTextBuffer *buffer) +{ + gboolean result; + GtkTextIter start, end; + + g_return_val_if_fail (GTK_IS_TEXT_BUFFER (buffer), FALSE); + + result = gtk_text_buffer_get_has_selection (GTK_TEXT_BUFFER (buffer)); + + /* check there are no images in selection */ + if (result) { + gtk_text_buffer_get_selection_bounds (buffer, &start, &end); + if (gtk_text_iter_get_char (&start)== 0xFFFC) + result = FALSE; + else { + gtk_text_iter_backward_char (&end); + if (gtk_text_iter_forward_find_char (&start, _forward_search_image_char, + NULL, &end)) + result = FALSE; + } + + } + + return result; +} + +static void +remove_quotes (gchar **quotes) +{ + if (g_str_has_prefix (*quotes, "\"") && g_str_has_suffix (*quotes, "\"")) { + gchar *result; + result = g_strndup ((*quotes)+1, strlen (*quotes) - 2); + g_free (*quotes); + *quotes = result; + } +} + +static void +remove_extra_spaces (gchar *string) +{ + gchar *start; + + start = string; + while (start && start[0] != '\0') { + if ((start[0] == ' ') && (start[1] == ' ')) { + g_strchug (start+1); + } + start++; + } +} + +gchar * +modest_text_utils_escape_mnemonics (const gchar *text) +{ + const gchar *p; + GString *result = NULL; + + if (text == NULL) + return NULL; + + result = g_string_new (""); + for (p = text; *p != '\0'; p++) { + if (*p == '_') + result = g_string_append (result, "__"); + else + result = g_string_append_c (result, *p); + } + + return g_string_free (result, FALSE); +} + +gchar * +modest_text_utils_simplify_recipients (const gchar *recipients) +{ + GSList *addresses, *node; + GString *result; + gboolean is_first = TRUE; + + if (recipients == NULL) + return g_strdup (""); + + addresses = modest_text_utils_split_addresses_list (recipients); + result = g_string_new (""); + + for (node = addresses; node != NULL; node = g_slist_next (node)) { + const gchar *address = (const gchar *) node->data; + gchar *left_limit, *right_limit; + + left_limit = strstr (address, "<"); + right_limit = g_strrstr (address, ">"); + + if (is_first) + is_first = FALSE; + else + result = g_string_append (result, ", "); + + if ((left_limit == NULL)||(right_limit == NULL)|| (left_limit > right_limit)) { + result = g_string_append (result, address); + } else { + gchar *name_side; + gchar *email_side; + name_side = g_strndup (address, left_limit - address); + name_side = g_strstrip (name_side); + remove_quotes (&name_side); + email_side = get_email_from_address (address); + if (name_side && email_side && !strcmp (name_side, email_side)) { + result = g_string_append (result, email_side); + } else { + result = g_string_append (result, address); + } + g_free (name_side); + g_free (email_side); + } + + } + g_slist_foreach (addresses, (GFunc)g_free, NULL); + g_slist_free (addresses); + + return g_string_free (result, FALSE); + +} + +GSList * +modest_text_utils_remove_duplicate_addresses_list (GSList *address_list) +{ + GSList *new_list, *iter; + GHashTable *table; + + g_return_val_if_fail (address_list, NULL); + + table = g_hash_table_new_full (g_str_hash, g_str_equal, g_free, NULL); + + new_list = address_list; + iter = address_list; + while (iter) { + const gchar* address = (const gchar*)iter->data; + + /* We need only the email to just compare it and not + the full address which would make "a " + different from "a@a.com" */ + const gchar *email = get_email_from_address (address); + + /* ignore the address if already seen */ + if (g_hash_table_lookup (table, email) == 0) { + g_hash_table_insert (table, (gchar*)email, GINT_TO_POINTER(1)); + iter = g_slist_next (iter); + } else { + GSList *tmp = g_slist_next (iter); + new_list = g_slist_delete_link (new_list, iter); + iter = tmp; + } + } + + g_hash_table_unref (table); + + return new_list; +} + +gchar * +modest_text_utils_get_secure_header (const gchar *value, + const gchar *header) +{ + const gint max_len = 16384; + gchar *new_value = NULL; + gchar *needle = g_strrstr (value, header); + + if (needle && value != needle) + new_value = g_strdup (needle + strlen (header)); + + if (!new_value) + new_value = g_strdup (value); + + /* Do a max length check to prevent DoS attacks caused by huge + malformed headers */ + if (g_utf8_validate (new_value, -1, NULL)) { + if (g_utf8_strlen (new_value, -1) > max_len) { + gchar *tmp = g_malloc0 (max_len * 4); + g_utf8_strncpy (tmp, (const gchar *) new_value, max_len); + g_free (new_value); + new_value = tmp; + } + } else { + if (strlen (new_value) > max_len) { + gchar *tmp = g_malloc0 (max_len); + strncpy (new_value, tmp, max_len); + g_free (new_value); + new_value = tmp; + } + } + + return new_value; +} + +static gboolean +is_quoted (const char *start, const gchar *end) +{ + gchar *c; + + c = (gchar *) start; + while (*c == ' ') + c = g_utf8_next_char (c); + + if (*c == '\0' || *c != '\"') + return FALSE; + + c = (gchar *) end; + while (*c == ' ' && c != start) + c = g_utf8_prev_char (c); + + if (c == start || *c != '\"') + return FALSE; + + return TRUE; +} + + +static void +quote_name_part (GString **str, gchar **cur, gchar **start) +{ + gchar *blank; + gint str_len = *cur - *start; + + while (**start == ' ') { + *start = g_utf8_next_char (*start); + str_len--; + } + + blank = g_utf8_strrchr (*start, str_len, g_utf8_get_char (" ")); + if (blank && (blank != *start)) { + if (is_quoted (*start, blank - 1)) { + *str = g_string_append_len (*str, *start, str_len); + *str = g_string_append (*str, ";"); + *start = g_utf8_next_char (*cur); + } else { + *str = g_string_append_c (*str, '"'); + *str = g_string_append_len (*str, *start, (blank - *start)); + *str = g_string_append_c (*str, '"'); + *str = g_string_append_len (*str, blank, (*cur - blank)); + *str = g_string_append (*str, ";"); + *start = g_utf8_next_char (*cur); + } + } else { + *str = g_string_append_len (*str, *start, str_len); + *str = g_string_append (*str, ";"); + *start = g_utf8_next_char (*cur); + } +} + +gchar * +modest_text_utils_quote_names (const gchar *recipients) +{ + GString *str; + gchar *start, *cur; + + str = g_string_new (""); + start = (gchar*) recipients; + cur = (gchar*) recipients; + + for (cur = start; *cur != '\0'; cur = g_utf8_next_char (cur)) { + if (*cur == ',' || *cur == ';') { + if (!g_utf8_strchr (start, (cur - start + 1), g_utf8_get_char ("@"))) + continue; + quote_name_part (&str, &cur, &start); + } + } + + quote_name_part (&str, &cur, &start); + + return g_string_free (str, FALSE); +}