technoluddites

1_Data_and_Aesthetic/Code/NYT_ArticleSearch_v2/NYT_ASearch.pde

@@ -0,0 +1,85 @@
+int queryDelay = 1200;
+
+int[] doASearchYears(String q, int startYear, int endYear) {
+  int[] counts = new int[endYear - startYear];
+  int i = startYear;
+  while (i < endYear) {
+    println(i);
+    try {
+      ASResult r = doASearch(q, i + "0101", i + "1231");
+      println(r.hits);
+      counts[i - startYear] = r.hits;
+      delay(queryDelay);
+      i++;
+    } catch (Exception e) {
+     println("FAILED ON " + i + ". IF YOU SEE THIS MESSAGE A BUNCH OF TIMES IN A ROW, TRY AGAIN LATER, OR WITH A DIFFERENT QUERY. \n AND MAKE SURE YOU'VE ENTERED YOUR API KEY!"); 
+    }
+  }
+  return(counts);
+}
+
+ASResult doASearch(String q, String beginDate, String endDate) {
+  ASResult result = new ASResult();
+  result.resultJSON = loadJSONObject(getASURL(q, beginDate, endDate));
+  result.parse();
+  return(result);
+}
+
+String getASURL(String q, String beginDate, String endDate) {
+  return(baseURL + "q=" + java.net.URLEncoder.encode(q) + "&begin_date=" + beginDate + "&end_date=" + endDate + "&api-key=" + apiKey);
+}
+
+class ASResult {
+  JSONObject resultJSON; 
+  ASDoc[] docs;
+  int hits;
+
+  void parse() {
+    JSONObject response = resultJSON.getJSONObject("response");
+    JSONObject meta = response.getJSONObject("meta");
+    hits = meta.getInt("hits");
+
+    JSONArray docArray = response.getJSONArray("docs");
+    int totalDocs = docArray.size();
+    docs = new ASDoc[totalDocs];
+
+    for (int i = 0; i < totalDocs; i++) {
+      JSONObject doco = docArray.getJSONObject(i);
+      ASDoc doc = new ASDoc();
+      doc.docJSON = doco;
+
+      docs[i] = doc;
+      doc.parse();
+    }
+  }
+}
+
+class ASDoc {
+  JSONObject docJSON;
+
+  //** Note - all other fields of the documents can be fished out manually from the docJSON object. ie. docJSON.getString("web_url");
+  //          I've just built in these three for convenience sake
+  String headline;
+  int page;
+  String snippet;
+
+
+  void parse() {
+    headline = docJSON.getJSONObject("headline").getString("main");
+    try {
+      page = int(docJSON.getString("print_page"));
+    } 
+    catch(Exception e) {
+      //Sometimes pages are null. In this case, we'll set it to -1
+      page = -1;
+    }
+
+    try {
+      snippet = docJSON.getString("snippet");
+    } 
+    catch (Exception e) {
+      //Sometimes the snippets are null. In this case we'll set it to "";
+      snippet = "";
+    }
+  }
+}

1_Data_and_Aesthetic/Code/NYT_ArticleSearch_v2/NYT_ArticleSearch_v2.pde

@@ -0,0 +1,72 @@
+/*
+
+ ITP Data Art
+ NYTimes Article Search v2 Simple Example
+
+ **Note - you must put your API key in the first field for this to work!
+
+ Article Search v2 docs: http://developer.nytimes.com/docs/read/article_search_api_v2 
+
+ */
+
+String apiKey = "YOUR_API_KEY_GOES_HERE";
+String baseURL = "http://api.nytimes.com/svc/search/v2/articlesearch.json?";
+
+void setup() {
+  size(1280, 720, P3D);
+  smooth(8);
+  background(255);
+
+  //This function returns a list of integers, counting a search term per year
+  int[] monkeyCounts = doASearchYears("data", 2011, 2015);
+
+  //Which we can draw a bar chart from:
+  for (int i = 0; i < monkeyCounts.length; i++) {
+   fill(0,150);
+   float x = map(i,0, monkeyCounts.length,100, width - 100);
+   float y = height - 100;
+   float w = (width - 200)/monkeyCounts.length;
+   float h = -map(monkeyCounts[i], 0, max(monkeyCounts), 0, height - 200);
+   rect(x, y, w, h);
+   fill(255);
+   pushMatrix();
+   translate(x,y);
+   rotate(PI/2);
+   textSize(10);
+   text(i + 1901, 0, 8);
+   popMatrix();
+  }
+
+  //It's often useful to save data like this (so we don't have to call the API every time once we're visualizing)
+  PrintWriter writer = createWriter("data/monkeyData.csv");
+  for (int i:monkeyCounts) writer.println(i);
+  writer.flush();
+  writer.close();
+
+  /*
+  //There is really a lot more we can do with this.
+  //Here, a search for 'bin laden' on the day he was killed
+  ASResult osama = doASearch("bin laden", "20110502", "20110502");
+  //Find out how many articles
+  println(osama.hits);
+  //Get the headline of the fourth one
+  println(osama.docs[3].headline);
+  //And a snippet from the second one
+  println(osama.docs[2].snippet);
+  //We can get access to any of the fields that are returned (see API docs). For example the URL to the first story:
+  println(osama.docs[0].docJSON.getString("web_url"));
+  //Or, a JSON list of the keywords associated with the story
+  JSONArray keyWords = osama.docs[0].docJSON.getJSONArray("keywords");
+  for(int i =0; i < keyWords.size(); i++) {
+   JSONObject keyWord = keyWords.getJSONObject(i); 
+   println(keyWord.getString("name"), keyWord.getString("value"));
+  }
+  */
+
+}
+
+void draw() {
+}
+
+
+

1_Data_and_Aesthetic/Code/NYT_ArticleSearch_v2/class2example/class2example.pde

@@ -0,0 +1,29 @@
+int[] counts;
+
+void setup() {
+  size(500, 500);
+  loadData();
+}
+void draw() {
+  renderCircles(counts);
+}
+void loadData() {
+  Table myTable = loadTable("monkeyData.csv");
+  counts = new int[myTable.getRowCount()];
+
+//  for (TableRow row:myTable.rows()) {
+//    println(row.getInt(0));
+//  }
+
+  for(int i = 0; i < myTable.getRowCount(); i++){
+    println(myTable.getRow(i).getInt(0));
+    counts[i] = myTable.getRow(i).getInt(0);
+  }
+}
+
+void renderCircles(int[] nums){
+  for(int i = 0; i <nums.length; i++){
+  ellipse( i* 10, 300, nums[i], nums[i]);
+}
+}
+

1_Data_and_Aesthetic/Code/NYT_ArticleSearch_v2/class2example/data/monkeyData.csv

@@ -0,0 +1,4 @@
+7581
+11565
+13898
+11572

1_Data_and_Aesthetic/Code/NYT_ArticleSearch_v2/data/monkeyData.csv

@@ -0,0 +1,4 @@
+7581
+11565
+13898
+11572