Second: in my opinion, using libpcre in not a very wise idea: 1. it's slow, at least slower, that posix regex 2. it's unstable/insecure, just like the whole exim mta (well, it's not a piece of shit, like some other solutions I've seen, but on the other hand, it's not so great)
I strongly oppose to this statement, being a long time Exim user in an ISP-Environment I can say that Exim is imho one of the most stable and reliable OSS-Projects i have ever worked with. I also think that most of Philip Hazels code is very well documented and easy to follow, do you have a more detailed explaination why do you think that Exim and/or PCRE is unstable/insecure ?
Comparing to qmail, exim is a piece of shit. How many bugs in exim were discovered in exim this year?
I don't want to start any kind of MTA-flamewar(esp. on clamav-devel ML) but imho qmail (in the unpatched version) is more or less unusable due to lot's of missing "modern" features. In qmail you might need lot's of external patches and applications to gain those might or might not have the same quality as the original source code.
No bugs were found in qmail since version 1.03, released 1998. Once again, how many bugs were discovered in exim since 1998?
again I have not used qmail for a long time (was tired of patching it to get some features) - but i don't think you can compare exim and qmail featurewise. On the other side I got the impression that qmail is basically unmaintained (featurewise) and basically freecing an applications state is nothing I hope will happen with clamav :-)
PCRE is too used in a wide variety of high profile OSS-projects I don't think they would use it if it's really unstable or insecure.
1. libpcre is about 2 times slower than posix regex, for example.
that might be - can't comment on that.
2. I'm not saying that libpcre is a piece of shit, but it is not completely secure. Quoting Bill G., it's "good enough". And only that. Not perfect. And because of complication level, it won't be perfect. But ClamAV COULD be. At least the engine.
hmm to be honest - clamav has in my experience still a long long way to go before becoming as stable and as reliable as exim. We have yet found a snapshot/stable release of clamd that survived more that a few hours under load. I too think there are much more critical design problems in clamav which should be fixed before (i.e. the network scanning protocol,..)
Please note - this will be my last message on this subject on this ML (as i think it's moving rapidly off-topic) - if you still think strongly abou this - feel free to contact me off-list.
Stefan
------------------------------------------------------- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ _______________________________________________ Clamav-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/clamav-devel
