Full-text search
- Thread starter Bvo
- Start date
In my opinion full-text search is most useful to search for examples (usage), some phrases (including proverbs), etc.
(Looking for rare chars and terms also is very useful, but that's another story).
So it's OK if the search in Pinyin will require tone marks (or may be even no tones if more than one char/word in search?).
Pinyin has many good sides and most useful (at least for me as a beginner) - it's easy and quick for typing.
So in F-TS case it can be used just for quick typing and in this case tone requirement is OK (at least it's better than no Pinyin at all).
PS: and sometimes I don't know/cannot remember how to write some phrase, but do know how it sounds (i.e. can write pinyin with tones, though it's even better if I can search a complex phrase without tones).
(Looking for rare chars and terms also is very useful, but that's another story).
So it's OK if the search in Pinyin will require tone marks (or may be even no tones if more than one char/word in search?).
Pinyin has many good sides and most useful (at least for me as a beginner) - it's easy and quick for typing.
So in F-TS case it can be used just for quick typing and in this case tone requirement is OK (at least it's better than no Pinyin at all).
PS: and sometimes I don't know/cannot remember how to write some phrase, but do know how it sounds (i.e. can write pinyin with tones, though it's even better if I can search a complex phrase without tones).
Last edited:
Well that's true about ABC, but the fields that only have pinyin are generally the sort of fields we wouldn't full-text index anyway - links to other words, e.g.
Another practical issue is that this would add quite a bit of bulk to our dictionary databases, when we already get many many complaints about our app being too big (and are in fact working very hard to make our databases smaller) - absent a LOT of user demand for this feature, it would be tough to justify on that basis.
Another practical issue is that this would add quite a bit of bulk to our dictionary databases, when we already get many many complaints about our app being too big (and are in fact working very hard to make our databases smaller) - absent a LOT of user demand for this feature, it would be tough to justify on that basis.
True, but especially on Android we have a *lot* of users in emerging markets where data is limited + expensive. We could always find a way to spin this off into an optional index file, though - had already been considering doing that with Cantonese support in our basic app since that's a good 5 MB or so worth of data + indexing we could drop.