v1.1.0 Release

[garyan2]

FOR THOSE UPGRADING FROM PREVIOUS VERSIONS OF EPG123: IF YOU DON'T READ ANYTHING ELSE IN THIS POST, READ THE FOLLOWING:

1. FOLLOW THE !UPGRADE_README.PDF FILE TO AVOID LOSING CHANNELS IN YOUR GUIDE, AND
2. DELETE YOUR EXISTING SCHEDULED TASK AND CREATE A NEW ONE WITH EPG123 OR EPG123CLIENT DEPENDING (you'll have to read the manual to find out why)

http://epg123.garyan2.net

The need for v1.1.0.
For most of us, there is no desperate need to upgrade to v1.1.0. I did however find that some, especially those of our international friends, definitely need it. Those that download lineups provided by Schedules Direct that have multiple stations on the same channel number find that only one of those stations is available in the guide. Correcting the problem introduces the need to perform a "transition" operation.

The Problem
The naming convention used since inception of epg123 was to use the lineup id combined with the channel number. It was assumed there wouldn't be multiple stations from the same lineup with the same channel number. Those lineups that did have duplicate channel numbers appeared to be just the same station in either SD or HD. I then found that some lineups have many stations with channel numbers of -1 or multiple differing stations with the same channel number.

The Fix
The channel naming convention obviously had to change to accommodate the above lineup scenarios. The name will now include the station id to make a three variable name.

The Bad
Unfortunately it is not possible for everyone to make the leap from a 2 variable name to a 3 variable name. Those running epg123 with automatch always on would, for the most part, be able to do just that. Those that had to manually match channel lineups would miserably fail the jump. A method to smoothly transition was needed.

The Transition
A transition tool has been created to make the transition smooth for everyone and takes but a minute to perform (7 clicks of the mouse button). The simple transition is to import the EPG123 lineup into the WMC database with both the 2 variable name and the 3 variable name at the same time. They both must be present in the media center database at least 1 time in order to successfully make the transition. In the new epg123Client, it will appear there a 2 of every station that existed before, plus the newly accessible channels. Once both exist in the database, it is safe to complete a guide update with v1.1.0.

The Good
Not only are the 3 variable names now being used, but all available stations in the lineup are now available and no further action is required. There is no need to perform an automatch or any manual matching of the "new" stations.


Fixes
The client now checks to see if a recording is in progress... if there is, then the mxf import will be aborted. There has been some reports that recordings can be interrupted if an import/reindex takes place.

EPG123 now counts the number of channels downloaded during an update and compares to an expected number. If the number of channels during updates is less than 90% of the expected number, then the update will be aborted. This allows a user to investigate the reason for loss of stations from their lineup(s). This avoids losing guide data for the missing stations.

The problem of mismatched md5s between a schedule entry and a program from schedules direct will no longer prevent the mxf file creation. There is actually no longer a comparison intended or otherwise.

Changes/Improvements
The epg123 GUI now automatically logs in to Schedules Direct when it is opened for convenience.

Vast improvement in the main EPG123 GUI. The listviews are much more responsive, quicker to load, and are less likely to experience a null error.

Improved the automatch logic and sealed a gap in which 2 runs would have to be performed to complete an automatch in some circumstances.

There has been significant changes to the Scheduled Task code. The previous option to automatically import guide data and to automatch stations to channels is no longer part of the epg123 configuration. These functions are now controlled by the task created for automated updates. It is possible to perform those actions from the epg123 GUI just like previous versions regardless of whether a task has been created or not.

The epg123Client now has a GUI of its own to be a truly stand-alone client. It handles the Step 1 actions of deleting the ehome directory, sets registry entries, creates the HD PVR/Colossus software tuner file, performs database manual imports and rebuilds, and creates a client only scheduled task. It can also perform station/channel matching by number or call sign, delete channels, manually subscribe and unsubscribe channels, and enable/disable channels in the guide.

The epg123Client now writes to the event log as well.

... additional improvements you may or may not notice.

[Space]

Nice! Congrats on the release!

[stuartm]

Nice, one question. I experienced a problem of tuners being left unavailable on next access if the EPG123 update (probably the mxf import) was run not while a recording is going on but if a channel was being watched live. Is that situation also detected? And what happens if the mxf import is aborted? will it reschedule itself to try again later (e.g. an hour or two)? and keep retrying for some number of retries? Thanks for all your hard work on this.

[choliscott]

I'm also curious on the answer to Stuartm's question, as my setup is always recording something

[garyan2]

Nice, one question. I experienced a problem of tuners being left unavailable on next access if the EPG123 update (probably the mxf import) was run not while a recording is going on but if a channel was being watched live. Is that situation also detected? And what happens if the mxf import is aborted? will it reschedule itself to try again later (e.g. an hour or two)? and keep retrying for some number of retries? Thanks for all your hard work on this.

I couldn't find a reliable way to detect if a tuner was in use, so to answer the first question, not at this time. I'll keep looking for a way to do just that.

Unfortunately, right now if an active recording is detected during update and we abort, that's the end of it. I don't reschedule the task or sit and wait for it to stop. If nothing is done, then it will be the next update time that epg123 runs again. The mxf file is created, so it would be easy enough for it to be imported with the epg123client.

I know these answers aren't what we would all like, but it's what we have right now. Like I said, I'll keep working at it.

[Space]

WMC tells the computer to not go to sleep automatically whenever it is recording or displaying live TV, perhaps this can be used if there is not a more direct way of doing it through the WMC API?

Here is a thread that (eventually) gets to the point of creating a script that uses "powercfg -requests" to determine if a recording is about to start or a recording is in progress:

http://www.thegreenbutton.tv/forums/vie ... f=6&t=5927

EDIT: I just checked and it looks like this method may not be able to tell the difference between watching live TV and watching an existing recorded video, so I guess this is not going to help...

[Sancho]

Nice, one question. I experienced a problem of tuners being left unavailable on next access if the EPG123 update (probably the mxf import) was run not while a recording is going on but if a channel was being watched live. Is that situation also detected? And what happens if the mxf import is aborted? will it reschedule itself to try again later (e.g. an hour or two)? and keep retrying for some number of retries? Thanks for all your hard work on this.

I couldn't find a reliable way to detect if a tuner was in use, so to answer the first question, not at this time. I'll keep looking for a way to do just that.

Unfortunately, right now if an active recording is detected during update and we abort, that's the end of it. I don't reschedule the task or sit and wait for it to stop. If nothing is done, then it will be the next update time that epg123 runs again. The mxf file is created, so it would be easy enough for it to be imported with the epg123client.

I know these answers aren't what we would all like, but it's what we have right now. Like I said, I'll keep working at it.

Upgrade seemed to go smoothly. Thanks for the clear instructions.

I understand the need for some to have the MXF import aborted if there's a recording in progress; was never a problem for me, based on the few times the scheduled update occurred while a recording was in progress. A couple things: could you add an option to not abort, but have the abort option enabled by default? That might solve the problem for those whose machine is always recording, or for whom it's not a problem. Given that the MXF import now will be aborted, but the MXF is created, what's the best method to import it once the machine is free? If I understand correctly, the epg123Client.exe can be used to manually import, so I guess manually importing epg123.mxf would suffice to finish the aborted scheduled update... but maybe I'm missing something. I assume the manual import is a better method than re-running a full mode Save & Execute.

[IT Troll]

Thanks for the new release, I'll take it for a spin at the weekend. Sounds like this will help out the satellite users.

I have yet to read the upgrade guide (which I will study before I do anything) but does the process leave the old two variable station in the database or does it clean them up as part of the transition?

[Sammy2]

Do I have to??

[stuartm]

Do I have to??

Of course not. If you read the first post in this thread Gary explains the changes and why they are there.
And explicitly mentions there is no requirement to do an upgrade.
It's up to you if you want to go this release.

[dkrom]

did the upgrade and now all of my satellite stations are off. If I tune to 2 it tries to go to 1, 4 goes to 4, 7 goes to 5, 20 goes to 7, 31 to 8, 38 to 9, etc ... I followed the utility directions, then ran the epg123.exe to download the data.

[garyan2]

A couple things: could you add an option to not abort, but have the abort option enabled by default? That might solve the problem for those whose machine is always recording, or for whom it's not a problem. Given that the MXF import now will be aborted, but the MXF is created, what's the best method to import it once the machine is free? If I understand correctly, the epg123Client.exe can be used to manually import, so I guess manually importing epg123.mxf would suffice to finish the aborted scheduled update... but maybe I'm missing something. I assume the manual import is a better method than re-running a full mode Save & Execute.

Not in the plans to add an abort/don't abort option. I'll have to think what that would look like. The problem has been reported that recordings were being interrupted by the update. Though I only managed to reproduce the problem once, I don't think I could not address it is some way. The potential for this to be a bigger issue is great so I opted for the conservative approach until something better comes along. Personally, I close all my media centers at the end of the day and have all the clients update at 0500 ... no tuners in use let alone recording something.

The epg123Client would be the best way to manually import the file. It doesn't check for an active recording and will be much quicker than a full scheduled run.

Thanks for the new release, I'll take it for a spin at the weekend. Sounds like this will help out the satellite users.

I have yet to read the upgrade guide (which I will study before I do anything) but does the process leave the old two variable station in the database or does it clean them up as part of the transition?

I'm hoping it will help out the satellite users, but evidently they're pretty secretive about their problems and won't share. I can only assume this was one of the issues.

There is really nothing to clean up going from 2 variable channel name to 3 variable channel name... they are just connectors between the tuner channel and the service (station). Think of it like dancing with your partner, blindfolded, and holding on with your left hand (2 variable name). You must change hands but if you let go with your left, you will lose your partner. So instead, you grab your partner with your right hand (3 variable name) and for a time you are using both hands. You are now free to release your left hand and everything is fine ... you never lost your partner. There was nothing added, nothing taken away, and everything is right with the world. The dance will continue without missing a beat.

Do I have to??

No, but there are some improvements you might be missing out on. I believe this version, taking care of the null exceptions and protecting from loss of channels during an update, is significantly better than v1.0.2.

[garyan2]

did the upgrade and now all of my satellite stations are off. If I tune to 2 it tries to go to 1, 4 goes to 4, 7 goes to 5, 20 goes to 7, 31 to 8, 38 to 9, etc ... I followed the utility directions, then ran the epg123.exe to download the data.

Need to know some information about your setup ... obviously that is not supposed to happen. Is there something special about your setup. Do you have custom channel numbers? Do you use automatch? Does the Schedules Direct channel numbers match your satellite feed?

[garyan2]

@dkrom, could you send me your epg123utility.mxf (compressed) file and the trace.log entry from last update? I should be able to see something in those files.

[DaveInPa]

I upgraded to 1.1.0 today and did the transition. I read the instructions and was expecting it to be complicated. I think it took longer to read the instructions than to upgrade Everything is still working great!

I also have concerns that the mxf load may be aborted. I fetch 21 days of guide twice a day at times when a recording is unlikely, but sometimes happening. So, it's unlikely that my guide will get too far behind.

[royalpain33]

I downloaded the upgrade. And ran the epg123utility.exe as the readme file instructs, but no mxf file was created in the folder. I even tried a couple more times, but still no mxf file.

Any suggestions as to what is wrong with it?

[garyan2]

I downloaded the upgrade. And ran the epg123utility.exe as the readme file instructs, but no mxf file was created in the folder. I even tried a couple more times, but still no mxf file.

Any suggestions as to what is wrong with it?

Is the utility in the same directory as your epg123.cfg file? Also, take a look in your trace.log file ... should say something.

[dkrom]

@dkrom, could you send me your epg123utility.mxf (compressed) file and the trace.log entry from last update? I should be able to see something in those files.

Will do. In response to your other questions - I don't think I have anything special, but I did have to rename all of my OTA channels in WMC or the guide data didn't show up on the MyMediaCenter app. I added -DTV to the channel OTA names, but all of the satellite channels worked at that time. I did use automatch and the SD numbers did match my satellite feed.

[stefanr]

Hi Gary - thank you for the update.

I have followed the instructions in the update guide and everything is working OK except for the scheduled task. As instructed I deleted the existing task using epg123.exe and created a new scheduled task (also using epg123.exe as I only have a single machine).

The scheduled task is created OK, but if I run it I get a return code of 0x0 in Task Sceduler, but in the history I can see and event code 101 - Launch Failure and the message "Task Scheduler failed to start "\epg123_update" task for user "XXXXXX\XXXXXX". Additional Data: Error Value: 2147943767. (I have changed the User information)

If I look at the epg123 trace, I can see that the first task action, epg123.exe -update has run ok, but the second action epg123Client.exe -i "epg123.mxf" -match did not run. I can successfully run the latter comand from a comand prompt. There are no errors in the trace, it just doesn't run.

[IT Troll]

Think of it like dancing with your partner, blindfolded, and holding on with your left hand (2 variable name). You must change hands but if you let go with your left, you will lose your partner. So instead, you grab your partner with your right hand (3 variable name) and for a time you are using both hands. You are now free to release your left hand and everything is fine ... you never lost your partner. There was nothing added, nothing taken away, and everything is right with the world. The dance will continue without missing a beat.

Thanks. Not sure about the analogy. But I know Microsoft have been leading us on a merry dance for years!