Issue #7545: improve documentation of the `buffering` argument in io.open().

1 files changed, 14 insertions, 4 deletions

diff --git a/Modules/_io/_iomodule.c b/Modules/_io/_iomodule.c
index a2a6430..ca359bc 100644
--- a/Modules/_io/_iomodule.c
+++ b/Modules/_io/_iomodule.c
@@ -219,10 +219,20 @@ PyDoc_STRVAR(open_doc,
 "returned as strings, the bytes having been first decoded using a\n"
 "platform-dependent encoding or using the specified encoding if given.\n"
 "\n"
-"buffering is an optional integer used to set the buffering policy. By\n"
-"default full buffering is on. Pass 0 to switch buffering off (only\n"
-"allowed in binary mode), 1 to set line buffering, and an integer > 1\n"
-"for full buffering.\n"
+"buffering is an optional integer used to set the buffering policy.\n"
+"Pass 0 to switch buffering off (only allowed in binary mode), 1 to select\n"
+"line buffering (only usable in text mode), and an integer > 1 to indicate\n"
+"the size of a fixed-size chunk buffer.  When no buffering argument is\n"
+"given, the default buffering policy works as follows:\n"
+"\n"
+"* Binary files are buffered in fixed-size chunks; the size of the buffer\n"
+"  is chosen using a heuristic trying to determine the underlying device's\n"
+"  \"block size\" and falling back on `io.DEFAULT_BUFFER_SIZE`.\n"
+"  On many systems, the buffer will typically be 4096 or 8192 bytes long.\n"
+"\n"
+"* \"Interactive\" text files (files for which isatty() returns True)\n"
+"  use line buffering.  Other text files use the policy described above\n"
+"  for binary files.\n"
 "\n"
 "encoding is the name of the encoding used to decode or encode the\n"
 "file. This should only be used in text mode. The default encoding is\n"