-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On Thursday, July 18, 2002, Thomas F wrote...
IW>> Content-Type: text/html in the headers it works but if it has a IW>> Content-Type: multipart/alternative in the headers and the IW>> Content-Type: text/html is in the message body it doesn't work. > I have a suspicion here. If I see it correctly, you have set it to > "Presence in Anywhere". Try "Presence in Kludges" and "Presence in > Text" and see whether the latter works then. I think not. > My suspicion is that the "lower headers" are not being checked. If I remember correctly, when the content type is set to multipart/alternative, it is attaching a file that is an HTML file. But when the content type is set to text/html the HTML text is just plain text as if you had sat their and written HTML tags in your window. The problem when you try querying for text/html in the body is that TB!s filters stop when they reach the multipart-boundaries, and don't scan the attachment. Take this as an example: =======================8<--------------------------------------------- Content-Type: text/html Content-Transfer-Encoding: 7bit X-Mailer: com.snowball.mail Sender: [EMAIL PROTECTED] X-Sender: [EMAIL PROTECTED] Return-Path: [EMAIL PROTECTED] <HTML> <HEAD> <TITLE>Hi Jonathan</TITLE> =======================8<--------------------------------------------- This is the headers from an email where the HTML code appears to be written straight into the message. As the original post said, this would have been caught by the filter as expected. =======================8<--------------------------------------------- Content-Type: multipart/alternative; boundary="0-1415967510-1026920267=:49617" Sender: [EMAIL PROTECTED] Precedence: bulk Reply-To: [EMAIL PROTECTED] X-MailScanner: Found to be clean X-UIDL: KP-!!G-N"!//k!!QJ'"! =======================8<--------------------------------------------- That is the header for another email, with a multipart/alternative set. That sets a boundary further down in the email (after the main body)... which is as follows: =======================8<--------------------------------------------- - --0-1415967510-1026920267=:49617 Content-Type: text/html; charset=us-ascii =======================8<--------------------------------------------- Now we see the text/html header appearing again. But this header appears in the body. The issue being, TB! stops filtering when it hits the boundary line (or so I've heard). So that would be the reason some mails are sneaking through. Hope that made sense. - -- Jonathan Angliss ([EMAIL PROTECTED]) -----BEGIN PGP SIGNATURE----- Version: 6.5.8ckt iQA/AwUBPTb43SuD6BT4/R9zEQKKAgCfRL9flws3MX/7Bg/eLg+8h0ZKGl4AoNig gSC7FwPrYV96aCyFyTMSC8Zg =uhOG -----END PGP SIGNATURE----- ________________________________________________________ Current Ver: 1.61 FAQ : http://faq.thebat.dutaint.com Unsubscribe: mailto:[EMAIL PROTECTED] Archives : http://tbudl.thebat.dutaint.com Moderators : mailto:[EMAIL PROTECTED] TBTech List: mailto:[EMAIL PROTECTED] Bug Reports: https://www.ritlabs.com/bt/
