[exim] Forced expansion failure in acl ‘set’ modifier

2017-04-20 Thread Stefan Unterweger
Hi! According to the specification (§ 43.26), when ACL ‘condition’s are evaluated, if the expansion is forced to fail, the whole condition is ignored, as if it were not written. But what would happen if the argument of a ‘set’ modifier is forced to fail (instead of just failing on its own)? Is

Re: [exim] Removing all headers with the same prefix

2017-04-19 Thread Stefan Unterweger
* Jeremy Harris on Tue, Apr 18, 2017 at 10:54:58PM +0100: > It'll be a multistep process, probably. Treat ACLs as a programming > language. > > […] > > But start with doing it as a sequence of > > set acl_m_foo = I’m already doing something very similar to implement a home-grown

[exim] Removing all headers with the same prefix

2017-04-18 Thread Stefan Unterweger
Hi, I want to scrub certain kinds of headers for all incoming messages so that I can use them for myself—and trust them. In particular, I want to get rid of all headers which start with ‘Spam-’, no matter what comes after that prefix (be it ‘Spam-Score’, ‘Spam-Whitelisted’ or whatever my