Update comment to reflect availability of random casereaders.

diff --git a/src/data/casefile.c b/src/data/casefile.c
index 9e31d431ce39410b0953515e79d1de5b12be0311..453fbf09da4a5af8f6a9effe85aeb3fe9d4fbad2 100644 (file)
--- a/src/data/casefile.c
+++ b/src/data/casefile.c
@@ -66,10 +66,10 @@
           during the reading phase.  Each casereader has an
           independent position in the casefile.
 
-          Casereaders may only move forward.  They cannot move
-          backward to arbitrary records or seek randomly.
-          Cloning casereaders is possible, but it is not yet
-          implemented.
+          Ordinary casereaders may only move forward.  They
+          cannot move backward to arbitrary records or seek
+          randomly.  Cloning casereaders is possible, but it is
+          not yet implemented.
 
           Use casefile_get_reader() to create a casereader for
           use in phase 2.  This also transitions from phase 1 to
@@ -80,6 +80,12 @@
           a case from a casereader.  Use casereader_destroy() to
           discard a casereader when it is no longer needed.
 
+          "Random" casereaders, which support a seek operation,
+          may also be created.  These should not, generally, be
+          used for statistical procedures, because random access
+          is much slower than sequential access.  They are
+          intended for use by the GUI.
+
        3. Destruction.  This phase is optional.  The casefile is
           also read with casereaders in this phase, but the
           ability to create new casereaders is curtailed.