Menu
Community /Pin to ProfileBookmark
@NogDogMay 27.2021 — #What sort of "thing" would you be searching (database, CSV file, JSON file, form input,...)? How many names would that "thing" typically have in it?
Most databases these days -- whether SQL or NoSQL -- have "fuzzy search" mechanisms you could leverage that are likely much more efficient than what we might come up with on the PHP side; so that would be my general preference unless we're talking a pretty small amount of data (for some undefined value of "pretty small"). @NogDogMay 27.2021 — #@kiwis80#1632194 That would be my initial approach -- but then I'm kind of a database-centric developer, so I may be biased. :) Assuming MySQL, the first fairly recent article I found on the FULLTEXT mechanism is https://severalnines.com/database-blog/full-text-searches-mysql-good-bad-and-ugly , and seems like a good quick intro in case it looks attractive to you. (I haven't messed with that stuff in several years now, as at work we're PostgreSQL, and I haven't really had to do fuzzy stuff in it yet.)@NogDogMay 28.2021 — #Hmm...yeah, I think fulltext match is more for finding relevant matches is larger text strings (like text documents). The [soundex function](https://dev.mysql.com/doc/refman/8.0/en/string-functions.html#function_soundex) might be more applicable for this sort of thing.
As to whether it makes sense to do it via MySQL, or use PHP's similar
Person Search
I’ve got a list of people objects, the person has a lastname and firstname field.
I’m looking to do a person look up function.
My plan is to loop through and find possible matches on last and then first name based on the search string.
I’m considering using the similar_text method and return results over 65% for example. I’m not sure on that value yet.
Obvious issue is spelling errors.
Are there any other ideas/options out there
Sign in
to post a comment7 Comments(s) ↴
0
@NogDogMay 27.2021 — #Most databases these days -- whether SQL or NoSQL -- have "fuzzy search" mechanisms you could leverage that are likely much more efficient than what we might come up with on the PHP side; so that would be my general preference unless we're talking a pretty small amount of data (for some undefined value of "pretty small").
0
@NogDogMay 27.2021 — #0
@NogDogMay 28.2021 — #Hmm...yeah, I think fulltext match is more for finding relevant matches is larger text strings (like text documents). The [code=sql]
select * from whatever
where last_name sounds like 'joness' and first_name sounds like 'Thom'
[/code]As to whether it makes sense to do it via MySQL, or use PHP's similar
soundex()