consume Unicode BOM (0xFEFF) in createReader() and Table parser

core/src/processing/core/PApplet.java

@@ -6744,7 +6744,20 @@ public class PApplet implements PConstants {
   static public BufferedReader createReader(InputStream input) {
     InputStreamReader isr =
       new InputStreamReader(input, StandardCharsets.UTF_8);
-    return new BufferedReader(isr);
+
+    BufferedReader reader = new BufferedReader(isr);
+    // consume the Unicode BOM (byte order marker) if present
+    try {
+      reader.mark(1);
+      int c = reader.read();
+      // if not the BOM, back up to the beginning again
+      if (c != '\uFEFF') {
+        reader.reset();
+      }
+    } catch (IOException e) {
+      e.printStackTrace();
+    }
+    return reader;
   }
 
 

core/src/processing/data/Table.java

@@ -378,6 +378,15 @@ public class Table {
     } else {
       InputStreamReader isr = new InputStreamReader(input, encoding);
       BufferedReader reader = new BufferedReader(isr);
+
+      // strip out the Unicode BOM, if present
+      reader.mark(1);
+      int c = reader.read();
+      // if not the BOM, back up to the beginning again
+      if (c != '\uFEFF') {
+        reader.reset();
+      }
+
       /*
        if (awfulCSV) {
         parseAwfulCSV(reader, header);
@@ -4039,6 +4048,7 @@ public class Table {
 
   // . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
 
+
   /**
    * @webref table:method
    * @brief Trims whitespace from values

core/todo.txt

@@ -2,7 +2,11 @@
 X return null for PApplet.trim(null)
 X StringDict(TableRow) constructor
 X allow lone double quotes in the midst of csv strings
-_ make trim() work on column titles as well
+X make trim() work on column titles as well
+_ add trimRows() and trimColumns()
+_   would you ever use one w/o the other?
+_   should trim() just handle it?
+X consume Unicode BOM (0xFEFF) in createReader() and Table parser
 
 
 _ no prompt shows with selectInput() on 10.11 and 10.12