From b071da73c15c165537303e948d749a91080afd11 Mon Sep 17 00:00:00 2001
From: Eric Huang <121286342+erichuang0324@users.noreply.github.com>
Date: Mon, 9 Jun 2025 16:59:43 +0800
Subject: [PATCH] Add simple MLA citation generator

---
 README.md                            | 19 +++++++
 gradlew                              |  0
 src/main/java/CitationGenerator.java | 79 ++++++++++++++++++++++++++++
 src/main/java/mainMla.java           | 25 ++-------
 4 files changed, 103 insertions(+), 20 deletions(-)
 create mode 100644 README.md
 mode change 100644 => 100755 gradlew
 create mode 100644 src/main/java/CitationGenerator.java

diff --git a/README.md b/README.md
new file mode 100644
index 0000000..18b6a15
--- /dev/null
+++ b/README.md
@@ -0,0 +1,19 @@
+# MLA Citation Machine
+
+This project provides a small command line program that generates a basic MLA style citation for a web page.
+
+## Usage
+
+Run the application and provide a URL through standard input. The program will download the page, attempt to detect author, title, website name and publication date from common HTML meta tags, and print a citation string.
+
+```
+$ echo "https://example.com" | ./gradlew run
+```
+
+The output will resemble:
+
+```
+Author Name. "Article Title." Website Name, 2024-03-14, https://example.com
+```
+
+The extraction heuristics are simple and may not work for every site but should handle many common news pages.
diff --git a/gradlew b/gradlew
old mode 100644
new mode 100755
diff --git a/src/main/java/CitationGenerator.java b/src/main/java/CitationGenerator.java
new file mode 100644
index 0000000..b94b4f0
--- /dev/null
+++ b/src/main/java/CitationGenerator.java
@@ -0,0 +1,79 @@
+import org.jsoup.Jsoup;
+import org.jsoup.nodes.Document;
+import org.jsoup.nodes.Element;
+
+import java.net.URL;
+
+public class CitationGenerator {
+    public static String generateMlaCitation(String url) throws Exception {
+        Document doc = Jsoup.connect(url)
+                .userAgent("Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/137.0.0.0 Safari/537.36")
+                .referrer("http://www.google.com")
+                .header("Accept-Language", "en-US,en;q=0.9")
+                .timeout(10 * 1000)
+                .followRedirects(true)
+                .get();
+
+        String author = null;
+        String[] authorSelectors = {
+                "meta[name=author]",
+                "meta[property*=author]",
+                "meta[name*=author]"
+        };
+        for (String selector : authorSelectors) {
+            Element element = doc.selectFirst(selector);
+            if (element != null) {
+                String content = element.attr("content");
+                if (!content.isEmpty()) {
+                    author = content;
+                    break;
+                }
+            }
+        }
+
+        String title = null;
+        Element titleMeta = doc.selectFirst("meta[property=og:title]");
+        if (titleMeta != null && !titleMeta.attr("content").isEmpty()) {
+            title = titleMeta.attr("content");
+        } else {
+            title = doc.title();
+        }
+
+        String website = null;
+        Element siteMeta = doc.selectFirst("meta[property=og:site_name]");
+        if (siteMeta != null && !siteMeta.attr("content").isEmpty()) {
+            website = siteMeta.attr("content");
+        } else {
+            website = new URL(url).getHost();
+        }
+
+        String date = null;
+        String[] dateSelectors = {
+                "meta[property*=published]",
+                "meta[name*=pubdate]",
+                "meta[name*=date]"
+        };
+        for (String selector : dateSelectors) {
+            Element element = doc.selectFirst(selector);
+            if (element != null) {
+                String content = element.attr("content");
+                if (!content.isEmpty()) {
+                    date = content;
+                    break;
+                }
+            }
+        }
+
+        StringBuilder citation = new StringBuilder();
+        if (author != null) {
+            citation.append(author).append(". ");
+        }
+        citation.append("\"").append(title).append(".\" ");
+        citation.append(website);
+        if (date != null) {
+            citation.append(", ").append(date);
+        }
+        citation.append(", ").append(url);
+        return citation.toString();
+    }
+}
diff --git a/src/main/java/mainMla.java b/src/main/java/mainMla.java
index 385d25b..31ebcdc 100644
--- a/src/main/java/mainMla.java
+++ b/src/main/java/mainMla.java
@@ -1,28 +1,13 @@
-import org.jsoup.Jsoup;
-import org.jsoup.nodes.Document;
-import org.jsoup.nodes.Element;
-
 import java.util.Scanner;
+
 public class mainMla {
     public static void main(String[] args) {
         Scanner sc = new Scanner(System.in);
         String input = sc.nextLine();
-        try{
-            Document doc = Jsoup.connect(input)
-                    .userAgent("Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/137.0.0.0 Safari/537.36 Edg/137.0.0.0")
-                    .referrer("http://www.google.com")
-                    .header("Accept-Language", "en-US,en;q=0.9")
-                    .timeout(10*1000)
-                    .followRedirects(true)
-                    .get();
-            System.out.println(doc.getElementsByTag("meta"));
-            Element authorData = doc.selectFirst("meta[property=cXenseParse:author]");
-            System.out.println(authorData.attr("content"));
-
-
-
-        }
-        catch(Exception e){
+        try {
+            String citation = CitationGenerator.generateMlaCitation(input);
+            System.out.println(citation);
+        } catch (Exception e) {
             System.out.println(e.getMessage());
         }
     }