Added solution for 1242 Leet Code multi threaded.

anxious-coder-lhs · anxious-coder-lhs · commit 4d085cbaeb18 · 2020-10-07T13:46:55.000+05:30
diff --git a/README.md b/README.md
@@ -48,6 +48,7 @@ Happy to accept any contributions to this in any langugage.
 13. [Leet Code 70 Climbing Stairs](https://leetcode.com/problems/climbing-stairs/) | [Solution DP Top Down](./level_easy/LeetCode_70_Climbing_Stairs_Easy.ts) | [Solution DP Bottoms Up](level_easy/LeetCode_70_Climbing_Stairs_1_Easy.ts)
 14. [Leet Code 200 Number of Islands](https://leetcode.com/problems/number-of-islands/) | [Solution 1 DFS](./level_medium/LeetCode_200_NumberOfIslands.ts) | [Solution BFS](./level_medium/LeetCode_200_NumberOfIslands_1.ts) | [Solution Union Find](./level_medium/LeetCode_200_NumberOfIslands_2.ts)
 15. [Leet Code 362 Design Hit Counter](https://leetcode.com/problems/design-hit-counter/) | [Solution 1](./level_medium/LeetCode_362_Hit_Counter_1.ts) | [Solution 2](./level_medium/LeetCode_362_Hit_Counter_Medium.ts)
+16. [LeetCode Web Crawler 1242](https://leetcode.com/problems/web-crawler-multithreaded/) | [Solution 1 ](./level_medium/LeetCode_WebCrawler_MultiThreaded_1242.java)
 
 
 ## Hard
diff --git a/level_medium/LeetCode_WebCrawler_MultiThreaded_1242.java b/level_medium/LeetCode_WebCrawler_MultiThreaded_1242.java
@@ -0,0 +1,47 @@
+
+/**
+ * // This is the HtmlParser's API interface.
+ * // You should not implement it, or speculate about its implementation
+ * interface HtmlParser {
+ *     public List<String> getUrls(String url) {}
+ * }
+ */
+class Solution {
+    /**
+     * Solution is going to rely on the parallel streams to spin up parallel immutable tasks in a multithreaded fashion. This is ideal as it
+     * would use a default thread pool on its own. Alternatively, a thread pool can also be created for the same.
+     * 
+     * We use a DFS mechanism for crawling all the web pages. For the DFS function to operate, we get all the pages of the url and recursively
+     * call the operation to recursively traverse all the child pages while maintaining a set to see if the url has already been crawled before.
+     * 
+     * It uses the Java Streams to collect all the items in the collection and return it by recursively computing the rest of the items and then
+     * crawling it. This is a complete immutable functional approach, alternatively, it can also be done recursively, by using an additional
+     * storage for the resposnes and using it for the results.
+     */
+    public List<String> crawl(String startUrl, HtmlParser htmlParser) {
+        final String hostName = this.getHostName(startUrl);
+        final Set<String> visitedUrls = Collections.synchronizedSet(new HashSet());
+        return this.crawlUrl(startUrl, htmlParser, hostName, visitedUrls)
+            .collect(Collectors.toList());
+    }
+    
+    private Stream<String> crawlUrl(String startUrl, 
+                                  HtmlParser htmlParser, 
+                                  String domainName,
+                                  Set<String> visitedUrls) {
+        
+        visitedUrls.add(startUrl);
+        final Stream<String> childStream = htmlParser.getUrlsQ(startUrl)
+            .parallelStream()
+            .filter(url -> getHostName(url).equals(domainName))
+            .filter(url -> !visitedUrls.contains(url))
+            .flatMap(url -> crawlUrl(url, htmlParser, domainName, visitedUrls));
+            
+        return Stream.concat(Stream.of(startUrl), childStream);
+    }
+    
+    private String getHostName(String url) {
+        final int index = url.indexOf("/", 7); // Finding the indexes for the url post protocol.
+        return (index == -1) ? url : url.substring(0, index);
+    }
+}
