Roadmap for MythTv Player 0.8

Mikkel (Administrator) 2009-09-01, 09:32 #16

User title: Developer
Member since Oct 2006 · 222 posts · Location: Copenhagen, Denmark
Group memberships: Administrators, Members

Show profile ·

Link to this post

In reply to post ID 761

Sounds critical! Linux, Do you know in which protocol version this changed? From http://www.mythtv.org/wiki/Protocol it seems like protocol version 46 is a candidate. Before releasing 0.6.0 I think we should verify all protocol version bumps to make sure the player does not mess up anything.

Can the FILETRANSFER command really delete recordings?

This post was edited on 2009-09-01, 09:38 by Mikkel.

linux 2009-09-01, 19:50 #17

Member since Aug 2009 · 22 posts
Group memberships: Members

Show profile ·

Link to this post

Hi Mikkel,

Officially ANN FILETRANSFER changes in protocol version 46. But in the source code it changed in Rev. 20297 [1].
I don't know what the write mode in FileTransfer really does. But I had these problems:

1.) In the LiveTv mode the files writtes on disk were damaged mpeg-files.

2.) Played recordings were written while playing (the file's modified date changed). Sometimes files were damaged or truncated.

The problem was repeatable (with patch/without patch).

(In 0.6 this critical bug doesn't exist (the first line of the command is already correct))

[1] http://svn.mythtv.org/trac/changeset/20297/trunk/mythtv/pr…

This post was edited on 2009-09-01, 19:52 by linux.

Lomion

(Moderator) 2009-09-01, 20:55 #18

Member since Jan 2009 · 39 posts · Location: Vienna, Austria
Group memberships: Global Moderators, Members

Show profile ·

Link to this post

What about handling it like this:

m_cmd.clear();
m_cmd.add("ANN FileTransfer MythTvPlayer");
m_cmd.add(filename);
if(m_fileConnection.getProtocolVersion()>=45)
{
m_cmd.add("Default"); // Storagegroup
m_cmd.add("0"); // writemode off
}
m_cmd.add("1"); // usereadahead true (using ringbuffer in Mythtv for fileaccess)
m_cmd.add("-1"); // retries (no idea what this does but its default -1)

It was working when I tried it with 0.22.
Can somebody verify it with 0.19?