Here's another simple client program for symbol tables related to indexing. Again, another common function that's easily handled by symbol tables. There's all kinds of cases where we just have a lot of information, maybe on our PC or all over the web, and we want to create an index that allows us to specify. A search key and get all the associated information. And you've, you've used programs like this on, on your computer many instances, most likely. So, more generally, here's the goal. So, we're going to list of files that are specified. And then, and maybe it's all files on the computer, or maybe it's specified in, in some other way. And what we want to do is create an index, so that we can efficiently find all the files that contain a given query. String. so, in this small example. So, our client is going to called file index. And in this small example say we are going to have five text files. And these are just pieces of literature. And so what we want to do is, build and index from that set of text files. And then given a key, a string key, we want it to print out the files that might contain that key. So, for example, the word freedom appears in the Magna Ca, Carta and in Moby Dick and A Tale of Two Cities [COUGH] in all three of them but not in Tom Sawyer and not in Aesop's Fables. The word whale is only in, Moby Dick. The word lamb appears both in Tom Sawyer and Aesop's Fables, and so forth. So we're looking for a piece of information somewhere in a computer, and it'll give us the names of the files that contain that information. so, with the, more [COUGH] complex user interface, this is, very much what the spotlight or find function on your computer is doing. And a programmer might use the same program, to find places where certain, Programming terms are used in a bunch of programs. So, normally we have a bunch of dot java files when we're working on an application. Like these are all the ones associated with this, lecture, so we might build an index from that set of files, and then we might wonder well where do we use the import statements. So we just search for import and it'll tell us, look in those three files. Or, did we use comparators anywhere here? In, this Index will tell us no. So again, a very general capability that's got lots and lots of applications and easy to implement with our symbol table api. And the idea is, that what we're going to do is associate. On keys with values well the key is the string we type what's the value? Well what we're going to use for value is a set of files, the files that contain the query string. So just given that high level description then the implimintation is pretty direct. Here's the implementation of file index using our symbol file implimintation. So now we're going to build a symbol table. That associates string keys with sets of files. So this creates a new symbol table associating string keys with sets of files. So first thing we do, as before, is build the index. So we're going to take, this time we're going to take the list of file names from the command line. And [COUGH] for each file we're going to create an input stream. We're going to read every word in the file if the symbol contains does not contain the word. Will create a new set for that word and put that in the symbol table, and then, we simply get the set associated with the key and add the new word to that set, the new file name to that set. So, this one, for each word in the file. It'll add that file name to the corresponding set for that word. That's building a symbol table associating keys with sets of files. And then processing queries is the same as before, as long as standard in is not empty we take a query. And then simply print out the set of strings associated with that word which is a list of file names. So again, our symbol table API gives a very easy way to implement this file indexing function, which is very widely useful. And similarly, maybe we want to build a, a book index, maybe for a real book, or maybe for an ebook. Or more generally, people want to often process, preprocess text to, maybe a huge amount of text, to support, so called, concordance queries. So what you want to do, is given a word, find all occurrences of that word along with immediate contexts. And context just means a few words before and a few words after. So for example, in tail of two cities. The words cities actually only appears in this one place and this is [COUGH] Context that it appears. With the forwards tongues of the two before and that were blended in after. The word majesty appears in three places, and there's, there's the context. In well, this is, the very special case that you're very familiar with. In web searches. You type in a word, and then you get places that, where that word appears in context. Again, our symbol table API provides for an easy way to implement this. This is a, a. [COUGH] concordance client that that does the job. Now what we want to do is read text, and we want to build an index for each word. So this is a [COUGH] a, a java construct for. Reading all the strings on an input stream, on standard input, and, splitting em by blank space and putting em into array, so now all the words, are, in the an array. On standard input. And then we're going to create a new symbol table that associates strings with sets of integers. And again the string is the key and the sets of integers are going to be the places in the array where the given work appears. So, we go through. All the words, pick out our key, s, and again, if, it's not there yet, we create a new set associated with that s, and, then, afterwards, we go ahead and. Get that set, and put the new index on that set. And then that allows us to process queries, where we take a query, and then get the set of indices associated with that query. And then we left out, left out the code where you print out the nine words that you want. The four to the left and the four to the right. Again, quite useful and familiar functionality and very easy to implement with our simul table client.
