|
|
Sorry, html tags are interpreted in this tickets:
...
Mail with accented chars replaced with html patterns (like "& eacute;" instead of "é") are very-very hard to read ! For example, french word for "summer" is "été" and is displayed "& eacute;t& eacute;" on smartphone...
... |
|
|
Not at all, try this:
In SOGo, send an HTML mail, go to Sent Box, and view generated source code by right-clicking it and select "View source code".
You'll see mail headers (Return-path, Received, User-Agent, from, to, content/type = text/html, charset=utf-8, etc...) and your message "badly" html encoded !
Real HTML compliant documents have to start with "html", "head"..."/head" and "body" tags and ends with "/body" tag.
If viewed outside the entire SOGo web page, browsers (even desktop known browsers) won't switch in HTML mode and will display this mail in text mode, showing HTML codes.
Try to view foreign HTML mail (not composed by SOGo), you'll find three parts:
-mail headers with multipart switch
-1st part follow with text encoded mail body
-2nd part follow with html encoded mail body (watch out html, head and body tags!)
Since SOGo doesn't generate multiparts mails and doesn't compose real html compliant mails, these mails can't be viewed on smartphones/tablets. |
|
|
I agree at least part of the problem is from SOGo. When I send an HTML mail from SOGo on a spamassassin protected server, this check is triggered:
HTML_MIME_NO_HTML_TAG
which, according to this: http://wiki.apache.org/spamassassin/Rules/HTML_MIME_NO_HTML_TAG means that the <html></html> tags are missing (and indeed, looking at the source, we see that those tags are missing)
This check is also triggered: MIME_HTML_ONLY, which can be avoided if the mail offers a plain text version as an alternative/multipart |
|
|
We could use libxml to do this for us quite easily - using code like that:
void walkTree(xmlNode a_node) {
xmlNode cur_node = NULL;
for (cur_node = a_node; cur_node; cur_node = cur_node->next) {
if (cur_node->type == XML_TEXT_NODE) {
printf("%s", cur_node->content);
}
walkTree(cur_node->children);
}
}
void main () {
htmlParserCtxtPtr parser = htmlCreatePushParserCtxt(NULL, NULL, NULL, 0, NULL, 0);
htmlCtxtUseOptions(parser, HTML_PARSE_NOBLANKS | HTML_PARSE_NOERROR | HTML_PARSE_NOWARNING | HTML_PARSE_NONET);
len = ReadFile(data);
htmlParseChunk(parser, data, len, 0);
walkTree(xmlDocGetRootElement(parser->myDoc));
} |
