Обсуждение: soundex and metaphone
Hey everyone, I've been working with a couple people who didn't know that soundex and metaphone were included in the distribution as contrib modules. While it's their fault that they didn't check contrib, soundex is pretty common among database systems and I was wondering if there was a reason it is not included as a core function? -Jonah
"Jonah H. Harris" <jharris@tvi.edu> writes: > Hey everyone, > > I've been working with a couple people who didn't know that soundex > and metaphone were included in the distribution as contrib modules. > While it's their fault that they didn't check contrib, soundex is > pretty common among database systems and I was wondering if there was > a reason it is not included as a core function? Because no one's taken ownership of the code and made a case for integrating it into the core backend. -Doug
At a minimum I think we should support soundex in the core. I'm willing to move soundex and metaphone into the backend. Does anyone see a reason not to do so? Douglas McNaught wrote: >"Jonah H. Harris" <jharris@tvi.edu> writes: > > > >>Hey everyone, >> >>I've been working with a couple people who didn't know that soundex >>and metaphone were included in the distribution as contrib modules. >>While it's their fault that they didn't check contrib, soundex is >>pretty common among database systems and I was wondering if there was >>a reason it is not included as a core function? >> >> > >Because no one's taken ownership of the code and made a case for >integrating it into the core backend. > >-Doug > >---------------------------(end of broadcast)--------------------------- >TIP 9: the planner will ignore your desire to choose an index scan if your > joining column's datatypes do not match > >
Jonah H. Harris wrote: > At a minimum I think we should support soundex in the core. I'm > willing to move soundex and metaphone into the backend. Does anyone > see a reason not to do so? Soundex is really only useful for English names with English pronunciation. If we were to adapt a phonetic algorithm into the core, I'd like to see something more general. -- Peter Eisentraut http://developer.postgresql.org/~petere/
Peter, I don't disagree with you that a more generalized function would also be good, just that soundex is common and would be helpful if it were built-in. Peter Eisentraut wrote: >Jonah H. Harris wrote: > > >>At a minimum I think we should support soundex in the core. I'm >>willing to move soundex and metaphone into the backend. Does anyone >>see a reason not to do so? >> >> > >Soundex is really only useful for English names with English >pronunciation. If we were to adapt a phonetic algorithm into the core, >I'd like to see something more general. > > >
Jonah H. Harris wrote: > At a minimum I think we should support soundex in the core. I'm > willing to move soundex and metaphone into the backend. Does anyone > see a reason not to do so? > > I take it you mean apart from the fact that soundex is horribly limited and out of data and probably nobody should be using it? cheers andrew
"Jonah H. Harris" <jharris@tvi.edu> writes: > At a minimum I think we should support soundex in the core. I'm willing > to move soundex and metaphone into the backend. Does anyone see a > reason not to do so? Is it really ready for prime time? For one thing, a quick look shows no evidence of being multibyte-ready. There's a fair amount of cleanup of random private coding conventions (META_MALLOC!?) to be done too. Doug's point is valid: there's some actual work needed here, not just arguing to shove the code from point A to point B. regards, tom lane
Tom Lane wrote: >"Jonah H. Harris" <jharris@tvi.edu> writes: > > >>At a minimum I think we should support soundex in the core. I'm willing >>to move soundex and metaphone into the backend. Does anyone see a >>reason not to do so? >> >> > >Is it really ready for prime time? For one thing, a quick look shows no >evidence of being multibyte-ready. There's a fair amount of cleanup of >random private coding conventions (META_MALLOC!?) to be done too. > >Doug's point is valid: there's some actual work needed here, not just >arguing to shove the code from point A to point B. > > > > Well, META_MALLOC occurs in part of the code that he didn't ask for ... it was inherited from the perl module code that I adapted to do double metaphone. And a minimal wrapper suited my purposes at the time just fine. But the point is well taken nevertheless. cheers andrew
Jonah H. Harris wrote: > I'm willing to move soundex and metaphone into the backend. > Does anyone see a reason not to do so? As a kinda strange reason, I like them in contrib because they demonstrate a nice simple example of how one can write a contrib extension. This module has simple functions that take a string or two and return a string or number. Most of the other contrib modules do tricky stuff with weird types or indexes that make them rather complex to use as a starting point. If they were to be moved out of contrib, I think it'd be really nice if someone add a "hello_world" contrib that demonstrates a bunch of simple operations in C to be used as such a model.