Cart
[Home] [By Thread] [By Date] [Recent Entries]
Jeff Greif wrote: "If you don't care where in the text you're searching (whether it's in the title or the excerpt), and you're never going to be searching for the strings 'title' or 'excerpt', then you could just use grep. XML databases and search engines aren't necessary, nor would they be as fast, most likely, unless they indexed the documents in some way (Patricia tree or inverted file) based on the text content. You'd be more likely to find this in a text indexing system not specialized to XML." Jeff, For a handful of documents, this may well be true. It's limited in that there is no syntactic context that filter by, but it will work. But it's an untruth that an XML database wouldn't be necessary or fast for managing many of these documents. Regarding indexing and speed, that's primarily what XML databases do, otherwise the word database wouldn't really apply. The types of indexing that they support depends on the implementation, but full-text inverted indexing is definitely something that most of them do.
|
