[rt #84106] document transform() vs bounding_box()

diff --git a/lib/Imager/Font.pm b/lib/Imager/Font.pm
index c97423c9b1e76754b67ab1a4eb51c0c37b6edcdd..7f449b1f1bccbb639753344af212f0e163b49d88 100644 (file)
--- a/lib/Imager/Font.pm
+++ b/lib/Imager/Font.pm
@@ -4,7 +4,7 @@ use Imager::Color;
 use strict;
 use vars qw($VERSION);
 
-$VERSION = "1.036";
+$VERSION = "1.037";
 
 # the aim here is that we can:
 #  - add file based types in one place: here
@@ -674,6 +674,10 @@ On success returns either the list of bounds, or a bounding box object
 object in scalar context.  Returns an empty list or C<undef> on
 failure and sets an error message readable with C<< Imager->errstr >>.
 
+The transformation matrix set by L</transform()> has no effect on the
+result of this method - the bounds of the untransformed text is
+returned.
+
 =item string()
 
 The $img->string(...) method is now documented in
@@ -815,6 +819,8 @@ Note that the transformation is done in font co-ordinates where y
 increases as you move up, not image co-ordinates where y decreases as
 you move up.
 
+C<transform()> has no effect on the results of L</bounding_box()>.
+
 Returns true on success.  Returns false on failure with the cause
 readable from C<< Imager->errstr >>.