On 30/05/17 04:06 AM, Carlos E. R. wrote:
In fact, the text files have to be converted to binary, as this:
postmap transport
Not necessarily. Yes you can hash/index the text files. It's an old Berkeley trick, hash/indexing text files. It is one option. Having plain text files, not hashed, is another. There is no *HAVE* about it. As a speed-up trick for less than 200 entries compared to a binary chop search of a sorted file it's like 0.000001 sec vs 0.0000015 sec. BFD. if you want to see what types of lookup mechanism are supported by postfix and its allies and relations and components, the run "postconf -m" I get: # postconf -m btree cidr environ fail hash internal ldap nis pcre proxy regexp sdbm socketmap static tcp texthash unix See the man page "postconf(1)" for details of what each of those are and how they are used. So you have plenty of options beside SQL, the various Berkeley-DB/hashed-text and plain-text files. BFD. Get over it! Given what I said about dependency on external components/systems/subsystems as opposed to just local files or libraries, I'd make two observations: * The Berkeley hash DB or the Btree mechanism are quite fast even for large (thousands, perhaps tens of thousands) entries. * If you're obsessed with SQL then SQLight is both 'local' and fast. I don't have that compiled in to my version, not do I have the Postgress interface compiled in, but they are options if you want them. -- A: Yes. > Q: Are you sure? >> A: Because it reverses the logical flow of conversation. >>> Q: Why is top posting frowned upon? -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org