Avro-2785: Update specs on how unions encoded

doc/src/content/xdocs/spec.xml

@@ -546,7 +546,7 @@
 
           <section id="union_encoding">
             <title>Unions</title>
-            <p>A union is encoded by first writing a <code>long</code>
+            <p>A union is encoded by first writing an <code>int</code>
               value indicating the zero-based position within the
               union of the schema of its value.  The value is then
               encoded per the indicated schema within the union.</p>
@@ -560,6 +560,8 @@
                 followed by the serialized string:
                 <source>02 02 61</source></li>
             </ul>
+            <p><em>NOTE</em>: Currently for C/C++ implementtions, the positions are practically an int, but theoretically a long. 
+            In reality, we don't expect unions with 215M members </p>
           </section>
 
           <section id="fixed_encoding">

lang/perl/lib/Avro/BinaryDecoder.pm

@@ -328,7 +328,7 @@ sub skip_union {
     $class->skip($union_schema, $reader);
 }
 
-## 1.3.2 A union is encoded by first writing a long value indicating the
+## 1.3.2 A union is encoded by first writing an int value indicating the
 ## zero-based position within the union of the schema of its value. The value
 ## is then encoded per the indicated schema within the union.
 sub decode_union {

lang/perl/lib/Avro/BinaryEncoder.pm

@@ -234,7 +234,7 @@ sub encode_map {
     $class->encode_long(undef, 0, $cb);
 }
 
-## 1.3.2 A union is encoded by first writing a long value indicating the
+## 1.3.2 A union is encoded by first writing an int value indicating the
 ## zero-based position within the union of the schema of its value. The value
 ## is then encoded per the indicated schema within the union.
 sub encode_union {

lang/py/avro/io.py

@@ -846,7 +846,7 @@ def skip_map(self, writers_schema, decoder):
 
     def read_union(self, writers_schema, readers_schema, decoder):
         """
-        A union is encoded by first writing a long value indicating
+        A union is encoded by first writing an int value indicating
         the zero-based position within the union of the schema of its value.
         The value is then encoded per the indicated schema within the union.
         """
@@ -1096,7 +1096,7 @@ def write_map(self, writers_schema, datum, encoder):
 
     def write_union(self, writers_schema, datum, encoder):
         """
-        A union is encoded by first writing a long value indicating
+        A union is encoded by first writing an int value indicating
         the zero-based position within the union of the schema of its value.
         The value is then encoded per the indicated schema within the union.
         """

lang/py3/avro/io.py

@@ -636,7 +636,7 @@ def skip_map(self, writer_schema, decoder):
 
   def read_union(self, writer_schema, reader_schema, decoder):
     """
-    A union is encoded by first writing a long value indicating
+    A union is encoded by first writing an int value indicating
     the zero-based position within the union of the schema of its value.
     The value is then encoded per the indicated schema within the union.
     """
@@ -866,7 +866,7 @@ def write_map(self, writer_schema, datum, encoder):
 
   def write_union(self, writer_schema, datum, encoder):
     """
-    A union is encoded by first writing a long value indicating
+    A union is encoded by first writing an int value indicating
     the zero-based position within the union of the schema of its value.
     The value is then encoded per the indicated schema within the union.
     """