summaryrefslogtreecommitdiff
path: root/io.c
diff options
context:
space:
mode:
authorakr <akr@b2dd03c8-39d4-4d8f-98ff-823fe69b080e>2009-02-07 14:50:58 +0000
committerakr <akr@b2dd03c8-39d4-4d8f-98ff-823fe69b080e>2009-02-07 14:50:58 +0000
commit1babcafb031f7bd434099a531482a01860bd9271 (patch)
tree7733f6429826ef05cc53765412bc4941b3f6cc07 /io.c
parentdc20aac2ce20360ddb605a127e3ab583fa6c1540 (diff)
update rdoc.
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_1_8@22113 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Diffstat (limited to 'io.c')
-rw-r--r--io.c5
1 files changed, 3 insertions, 2 deletions
diff --git a/io.c b/io.c
index b789e9f78d..116d3aa8f6 100644
--- a/io.c
+++ b/io.c
@@ -1351,8 +1351,9 @@ io_getpartial(int argc, VALUE *argv, VALUE io, int nonblock)
* The later means that readpartial is nonblocking-flag insensitive.
* It blocks on the situation IO#sysread causes Errno::EAGAIN as if the fd is blocking mode.
*
- * Since readpartial bypass internal stdio buffering mechanism,
- * IO#pos may return wrong value after readpartial.
+ * Since readpartial uses read system call directly,
+ * the stdio buffering mechanism may confuse the file offset.
+ * So IO#pos may return wrong value after readpartial.
* This problem is fixed by Ruby 1.9 which doesn't use stdio.
*/