CVSROOT: /cvs/gnome
Module name: nautilus-cd-burner
Changes by: mccann 05/12/20 18:38:19
Modified files:
. : ChangeLog mapping-protocol.c mapping-method.c
mapping-daemon.c
Log message:
2005-12-20 William Jon McCann <[EMAIL PROTECTED]>
* mapping-daemon.c (connection_free, handle_request)
(handle_new_client):
Replace use of GIOChannel with MappingProtocolChannel.
* mapping-method.c (monitor_do_iter_unlocked): Add debugging.
(monitor_setup, vfs_module_init, vfs_module_shutdown):
Replace use of GIOChannel with MappingProtocolChannel.
* mapping-protocol.[ch] (mapping_protocol_data_available)
(mapping_protocol_channel_new, mapping_protocol_channel_free)
(mapping_protocol_channel_ref, mapping_protocol_channel_unref)
(write_all, read_all, decode_int, decode_string, decode_pointer)
(mapping_protocol_request_encode)
(mapping_protocol_request_decode, decode_handshake)
(mapping_protocol_reply_encode, mapping_protocol_reply_decode)
(mapping_protocol_monitor_event_encode)
(mapping_protocol_monitor_event_decode):
The patch to add buffering caused any data in the buffer
after decoding a single reply or event to be lost. This
was because the buffer only existed in the scope of a
single decode function. I've created a new ref counted
structure to wrap GIOChannel and the buffer. Now the
mapping_protocol_data_available() function checks this
buffer in addition to the socket.
URL :
http://cvs.gnome.org/bonsai/cvsquery.cgi?branch=&dir=nautilus-cd-burner&who=mccann&date=explicit&mindate=2005-12-20%2018:37&maxdate=2005-12-20%2018:39
_______________________________________________
cvs-commits-list mailing list
cvs-commits-list@gnome.org
http://mail.gnome.org/mailman/listinfo/cvs-commits-list
