![]() |
* Martin Bene <martin.bene@xxxxxxxxxxxxx> [071012 14:48]: > Hi Aidan, > > Well, I found the previously used converters based on mime type quite > nice. The content - based classification used by typerules effectively > discards valuable information (i.e the mime-type) present in the email > structure. Fair enough. In my experience, I found that most of the time I was getting everything besides text attached as application/octect-stream if it wasn't a text or multipart. > The current handling based on partial mime doesn't semm optimal - while > the internal conversion can deal with text/plain, the results you get > for text/html parts are certainly not useful :-) > > I think that all parts of a message (text/plain, text/html, > attachments) should be handled equaly; I'm not 100% sure how to best > deal with the headers/metainformation. > > A possible scenario could look like this: > > * create a postscript file using the internal converter containing only > meta-information from the email (headers, sender/recipient information, > whatever); submit as postscript file > > * treat all parts of a multipart/mime message, like attachments, i.e try > converting them based on typerules file and submit them as separate > documents. > > * I would love to have the added flexibility of the mimeconverters back: > for each part of the multipart message, first check if there's a > converter for the specified mime type. Use if it exists and try > converting via typerules if there's no converter. OK, so how about this: 1) Use internal formatter for "message/rfc822" (the main message starts this way) which prints headers, and the "body" if it's not a mime body, all as text (subject to the FormatEnvHeaders). Keep this as 1 document to submit to HylaFAX, with a option to suppress this document. 2) For every mime part, check for a mimeconverter program. The *output* of the mimeconverter program (if there is one) is submitted in place of the original mime part as a document to HylaFAX in the job through the normal machinery. Note that this means the output doesn't have to be Postscript/pdf/etc. It just needs to be something that the regular typerules based submission can convert to PS/PDF/TIFF if it isn't that directly. Having the mimeconverter program output be "empty" would mean that that mime part is not submitted as a document. 3) If no "mimeconverter" is found, handle the part directly. - multipart are recursively handled, as now - message/rfc are handle internally, as now - all others are submitted as separate documents, as now Would this framework be flexible enough for you? It fixes the main problems of faxmail (the squish everything into a single postscript document, and handling application/octet-stream), and gives back the flexibility of handling individual parts with special mechanisms. a. -- Aidan Van Dyk aidan@xxxxxxxx Senior Software Developer +1 215 825-8700 x8103 iFAX Solutions, Inc. http://www.ifax.com/ ____________________ HylaFAX(tm) Users Mailing List _______________________ To subscribe/unsubscribe, click http://lists.hylafax.org/cgi-bin/lsg2.cgi On UNIX: mail -s unsubscribe hylafax-users-request@xxxxxxxxxxx < /dev/null *To learn about commercial HylaFAX(tm) support, mail sales@xxxxxxxxx*