Hi:

Only text is imported from incoming emails.  Importing random html elements will break most forums (and likely introduce some security issues.)

Thanks.

Thank you for the reply.

Does SimplePress provide hooks or other ways to readily extend the functionality of post-by-email so that I can let my select group of privileged-to-post-to-the-forum users post emails that are readable when they include inline-responses?

It seems that you’ve already done some of the work required to support html-from-email in your sp-email-post-imap.php file; on these lines:

$body_structure = sp_pbe_body_structure_for_html_part($body_structure);

function sp_pbe_is_mime_subtype_html($a_part) {
	return(sp_pbe_is_mime_subtype_specified($a_part) && sp_pbe_does_mime_subtype_match_given_string($a_part, "HTML"));
}

# tests the key parameters of the part and returns true if it looks like a properly formed HTML part.
function sp_pbe_is_part_a_valid_html_part($a_part) {
	return(sp_pbe_is_primary_body_type_text($a_part) && sp_pbe_is_mime_subtype_html($a_part));
}

# internally used by sp_pbe_body_structure_for_html_part().
function sp_pbe_get_a_default_part($a_structure) {
	$result = $a_structure; # if for some reason we don't find a better part to assign, just use this top-level structure.
	if(is_array($a_structure->parts) && count($a_structure->parts) >= 1) {
		$result = $a_structure->parts[0];
	}
	return $result;
}

# works through the given parts in the structure to find
# a well-formed HTML part. Returns that part if it finds
# it. Returns null otherwise.
# internally used by sp_pbe_body_structure_for_html_part().
function sp_pbe_scan_for_html_part($the_parts_array) {
	$result = null;
	foreach($the_parts_array as $this_part) {
		if(sp_pbe_is_part_a_valid_html_part($this_part)) {
			$result = $this_part; # found our matching part
			break;
		}
	}
	return $result;
}

# works through the given parts in the structure to find
# a well-formed PLAIN part. Returns that part if it finds
# it. Returns null otherwise.
# internally used by sp_pbe_body_structure_for_html_part().
function sp_pbe_scan_for_plain_part($the_parts_array) {
	$result = null;
	foreach($the_parts_array as $this_part) {
		if(sp_pbe_is_part_a_valid_plain_part($this_part)) {
			return $this_part; # found our matching part
			break;
		}
	}
	return $result;
}

# gets the structure of the most appropriate part of a multipart
# message that we can find. We use this 'most appropriate part'
# to identify how the raw content of the message has been encoded.
function sp_pbe_body_structure_for_html_part($a_structure) {
	$parts = $a_structure->parts;

	$result = sp_pbe_scan_for_html_part($parts);
	if(is_null($result)) {
		# if we didn't find an HTML part, now fall back to looking for a PLAIN part
		$result = sp_pbe_scan_for_plain_part($parts);
		if(is_null($result)) {
			# still didn't find a matching part, so fall back to our best default part option
			$result = sp_pbe_get_a_default_part($a_structure);
		}
	}

	return $result;
}






Further, you already allow "arbitrary html" to be posted to the forum directly, and already have privileging structures in place that 
strip, say, iframes from forum posts by users not allowed to post iframes; it seems arbitrary to block html from an 
email source when all other posting options support html... via paid plugin.

Am I wrong in assuming that html from an email would only cause problems if it was poorly formed (eg, no closing tag for an opening tag)? 
To this end, php libraries like simpledom could help by completing un-closed tags and otherwise allowing you guys to 
loop through valid tags for html content.

Hi:

HTML allowed on our forum vs HTML allowed in emails are two different things. 

HTML allowed on our forums can be controlled and filtered at the time the data is entered.  We can control what is allowed and what is not in real-time. 

With email, you never know what you’re going to get because you’re getting it after-the-fact and based on someone else’s rules and whims. 

Between the various email clients that format their messages differently and the numerous servers an email might pass through before it gets to your inbox, there is zero control over what you’re actually getting.  You can easily spend a year writing code for fiddling with stripping and formatting emailed html.  Even the biggest companies haven’t gotten it right – though, of course, they do it better than most.  We’ve done it for other large software projects before so we know the issues – not a week goes by where we didn’t see a new twist on what email clients are sending over and having to write new code to handle it. 

The most reliable thing you can do is extract the text which is what we do for Simple:Press.  The code you posted does make a very very basic attempt at extracting the html but unless it comes in absolutely perfectly, you’re going to get the text portion only. 

Thanks.