PgcEdit 0.5

[jeanl]

mp3boy, don't forget to do that on all the PGCs where you want the PUOs removed!
Jeanl

[2COOL]

@r0lZ

Thanks for the current version! I see you have been very busy. Just wanted to put a request out. Since you have already a macro to find out unused GPRMs, how about implementing in your watch upon start. You can, say, grey out the GPRM numbers on the left to signify that it's unused. If per chance, a new GPRM command gets set inbetween tracing, you could do a refresh. It's just that when I was doing Jump2PGC, I had to manually check for unused GPRM.

By the way, I noticed that if I copied a single command, my command didn't have { }s. But if I did more that one, they are present. It still works with them for single commands with pasting into PgcEdit though. Minor thing.

[r0lZ]

Originally posted by 2COOL
Just wanted to put a request out. Since you have already a macro to find out unused GPRMs, how about implementing in your watch upon start. You can, say, grey out the GPRM numbers on the left to signify that it's unused. If per chance, a new GPRM command gets set inbetween tracing, you could do a refresh. It's just that when I was doing Jump2PGC, I had to manually check for unused GPRM.Hum... The register numbers are already coloured in red when there is a breakpoint on that register. (BTW, you may click on the GPRM number to toggle the breakpoint.) It's difficult to add more colors. Also, the Find Unused GPRMs is a relatively slow routine, especially on DVDs with many commands. I don't want to slow down the startup of the trace.

By the way, I noticed that if I copied a single command, my command didn't have { }s. But if I did more that one, they are present. It still works with them for single commands with pasting into PgcEdit though. Minor thing. Yes, that's normal. A single commands is stored as a simple string, while multiple commands are stored as Tcl lists, with { }s to separate the elements.

[2COOL]

Originally posted by r0lZ
Also, the Find Unused GPRMs is a relatively slow routine, especially on DVDs with many commands. I don't want to slow down the startup of the trace. I've worked with alot of DVDs with lots of commands and it's always been quick for me. Then, perhaps you can have it as an option setting like you did with BOV inquiries.

[2COOL]

@r0lZ

Minor request.

When I use the trace, any PGCs that were accessed were tagged by highlighting them in PGC Selection window. They are still there when I exit out of trace mode. The only way for me to get rid of the highlights would to reload the DVD. I don't want to do that if I had some test edits. Anyway of giving us the option to clearing them without reloading?

BTW, you can disable the dialog of cloning PGC when First Play PGC is selected. I see you have disabled the other relevant menus too. I guess you forgot that one. ;)

[EDIT]Other suggestion...

In the way you did with "calls cross references", how about implementing automated GPRM/SPRM inquries? Say, I wanted to inquire on GPRM 1. A list would be generated on which what GPRM 1 command in which PGC it exists in. I believe I was asking for something like this last year in regards to logging of automated search feature.

[lark]

i support 2cool's idea about GPRM/SPRM inquries.
currently somewhat similar can be done with the search dacility (on the lower right corner). just type e.g. gprm(12) and hit the >> -button to cycle all occurrences (i think this excludes menu & BOV who) and obviously works with one gprm or sprm at the time (unless you use regex)...

regards
t :)

[r0lZ]

Originally posted by 2COOL
@r0lZ

Minor request.

When I use the trace, any PGCs that were accessed were tagged by highlighting them in PGC Selection window. They are still there when I exit out of trace mode. The only way for me to get rid of the highlights would to reload the DVD. I don't want to do that if I had some test edits. Anyway of giving us the option to clearing them without reloading?Although somewhat hidden, this function exists already. Just right click on an empty line of the PGC Selector, and select "Clear 'PGC Played' Highlights".

BTW, you can disable the dialog of cloning PGC when First Play PGC is selected. I see you have disabled the other relevant menus too. I guess you forgot that one. ;)Which dialog? The Clone PGC function doesn't work on FP-PGC.

Other suggestion...

In the way you did with "calls cross references", how about implementing automated GPRM/SPRM inquries? Say, I wanted to inquire on GPRM 1. A list would be generated on which what GPRM 1 command in which PGC it exists in. I believe I was asking for something like this last year in regards to logging of automated search feature. It's possible, but if I do that, I will do it for any search string. This way, you will be able to search for, say, SetTmpPML, too. But I want to finish the Jump2PGC macro first, and I have also some bugs to fix. So, be patient...

[2COOL]

Originally posted by r0lZ
Although somewhat hidden, this function exists already. Just right click on an empty line of the PGC Selector, and select "Clear 'PGC Played' Highlights".Pretty sneaky of you! Sorry if it was brought up before but I have been away for awhile. :(

Which dialog? The Clone PGC function doesn't work on FP-PGC.It shouldn't. I'm referring to Utilities > Clone current PGC.

It's possible, but if I do that, I will do it for any search string. This way, you will be able to search for, say, SetTmpPML, too. But I want to finish the Jump2PGC macro first, and I have also some bugs to fix. So, be patient... Ok.

[CoNS]

Originally posted by r0lZ
But I want to finish the Jump2PGC macro first, and I have also some bugs to fix....and a blanking interface to add in the trace/preview, no?! :D

[r0lZ]

Originally posted by 2COOL
It shouldn't. I'm referring to Utilities > Clone current PGC. Me too. This function do nothing when applied to the FP-PGC. What do you want more?
Note that the Clone PGC function is in the contextual popup menu in the PGC Selector, but is not present on the FP-PGC.
When building the utilities menu (as well as all the main menus and the toolbar), I cannot know what PGC will be selected. Therefore, all functions are always available. It is theorically possible to disable the functions that are not suitable for the currently selected PGC, but I don't do that. It's complicated, and a source of bugs...

[2COOL]

Originally posted by r0lZ
Note that the Clone PGC function is in the contextual popup menu in the PGC Selector, but is not present on the FP-PGC. Try selecting FP PGC and go up to Utilities > Clone current PGC. A dialog pops up. The other utilities don't do it so I figure you had accounted for people like me that would access the utilities this way instead of right clicking in PGC Selector pane.

[r0lZ]

Originally posted by 2COOL
Try selecting FP PGC and go up to Utilities > Clone current PGC. A dialog pops up. The other utilities don't do it so I figure you had accounted for people like me that would access the utilities this way instead of right clicking in PGC Selector pane. Oh, yes! I work usually in expert mode, so I did not notice this dialog. It's not an important bug, as the FP-PGC is not duplicated anyway. I have now removed the dialog in this case. Thanks!

[2COOL]

With so much posts recently in regards to intros creations, I looked at the "Import First-Play Clip" macro and came up with 2 questions.

1. Why do we need to have an unused GPRM? It is possible to use a used GPRM. Here's one of my First Play PGC examples. RED are new commands.

Before

************* pre commands
1 (JumpSS) Jump to VTSM 1, Root menu (TNN 1)


After

************* pre commands
1 if ( gprm(15) == 1 ) then ( Goto line 4)
2 Set gprm(15) =(mov) 1
3 (JumpTT) Jump to Title 2 //Imported Title
4 Set gprm(15) =(mov) 0 //GPRM reset back to default 0
5 NOP
6 (JumpSS) Jump to VTSM 1, Root menu (TNN 1)


2. Why can't we have a value of 255 for an infinite still Title? The Title would display indefinite until the user presses Play.

[northwind]

Is this a known problem?



can't read "::buttons::butinfo(0,errorcode)": no such variable
can't read "::buttons::butinfo(0,errorcode)": no such variable
while executing
"if {[info exists ::menus($cur_vts,$cur_lu,$cur_pgc,numbuttons)] && $::buttons::butinfo($cur_vts,errorcode)==0} {
set cellpostable $::menus($cur_..."
(procedure "simulate_play" line 127)
invoked from within
"simulate_play"
invoked from within
"if {$idx == "-"} {
::main::set_listbox_selection end
set idx [lindex [::main::get_selection_indexes] 0]
bell
} elseif $::tr::now_playin..."
(procedure "step" line 5)
invoked from within
"step "
(procedure "::tr::dvd_trace" line 329)
invoked from within
"::tr::dvd_trace"
(command bound to event)

[jeanl]

do you have the latest version? Grab the latest beta from the videohelp mirror...
jeanl

[2COOL]

@northwind

Not all of us are Tcl/Tk programmers. Exactly, what did you do to get this error?

[northwind]

0.5.5 of course. Ok - I will download the beta.

Sorry - just saw 2cools post...

I just copied the errormessage. No need to go into details if it was something everybody just knew about.

But to your question: Just tracing a dvd 24 season 3. It doesn't matter if I let it do the BOV tracing or not.


Found the beta..

But getting the same error (0.5.6b1)
(btw: 0.4.7 does not give that error)

[blutach]

This is what I have got in a particular menu. DVD is The DInner Game (actually a French movie Le Diner De Cons). I press ctrl-M and there is no background displayed. Pressing image forward also gives nothing.

http://img237.echo.cx/img237/812/untitled7sx1.png (http://www.imageshack.us)

The menu is not large - 14Mb if you want to share it, or I can demux a PGC and email (same th9ing happens on all PGCs).

Regards

[CoNS]

Originally posted by blutach
...a French movie Le Diner De ConsYeah, it's one of my early performances! :p

[jeanl]

Originally posted by blutach
This is what I have got in a particular menu. DVD is The DInner Game (actually a French movie Le Diner De Cons). I press ctrl-M and there is no background displayed. Pressing image forward also gives nothing.

http://img237.echo.cx/img237/812/untitled7sx1.png (http://www.imageshack.us)

The menu is not large - 14Mb if you want to share it, or I can demux a PGC and email (same th9ing happens on all PGCs).

Regards
blutach this looks like the problem another user was having on the forum (can't find the message again, the search utility is simply useless when you have threads that are 20 pages long :devil: ).
There was a bug in the original DVD2AVI code that got carried through to PgcEditPreview. Can you preview the vob? (that should work)...
Jeanl

[lamster]

Originally posted by 2COOL
Why do we need to have an unused GPRM? It is possible to use a used GPRM.
You're assuming that there's no JumpSS-FP command. If there is, then your assumptions regarding what values r15 (in your example) may have when the First Play PGC is entered may not be correct.

(Which ties in rather nicely with r0lZ' comment, "I will do it for any search string." I'd like an easy way to see if anything jumps back to the FP PGC, to know how paranoid I need to be in my programming.)

[2COOL]

Originally posted by lamster
You're assuming that there's no JumpSS-FP command. If there is, then your assumptions regarding what values r15 (in your example) may have when the First Play PGC is entered may not be correct.All GPRMs are set at 0 at DVD insert. I was using any GPRM, since they are at 0, setting them to 1, go back to FP, and setting back to it's original value 0.

(Which ties in rather nicely with r0lZ' comment, "I will do it for any search string." I'd like an easy way to see if anything jumps back to the FP PGC, to know how paranoid I need to be in my programming.) There is. Just right click on First Play PGC and select "go to calling command". If anything pops up in the inquiry, then you'll know. ;)

[lamster]

Originally posted by 2COOL
All GPRMs are set at 0 at DVD insert. I was using any GPRM, since they are at 0, setting them to 1, go back to FP, and setting back to it's original value 0.
It's 0 at DVD insert, but if some other PGC does a "JumpSS-FP (http://www.dvd-replica.com/DVD/jumpss.php?p=1)", which jumps to the First Play PGC, then[list=1]
r15 could have any value at that point, including 1;
the assumption that the value to reset it to is 0 is also invalid.
[/list=1]

[2COOL]

Originally posted by lamster
r15 could have any value at that point, including 1 I know "1" was a bad example. That's why in my Jump2PGC, I had values like 12345, 11111, 22222, ... to select from.

the assumption that the value to reset it to is 0 is also invalid. I like to look at this as a small minor loop. Start at point A, play intro title, and return to point A with original value (0) before executing original FP commands.

[EDIT] OK, now I'm having doubts. I've had a long day of work and just got home. I will research again. ;)

[lark]

Originally posted by jeanl
blutach this looks like the problem another user was having on the forum (can't find the message again, the search utility is simply useless when you have threads that are 20 pages long :devil: ).
There was a bug in the original DVD2AVI code that got carried through to PgcEditPreview. Can you preview the vob? (that should work)...
Jeanl
well, in this case it wasn't 20 pages (not even 20 posts): http://forum.doom9.org/showthread.php?s=&threadid=93771

regards
t :)

[jeanl]

Originally posted by lark
well, in this case it wasn't 20 pages (not even 20 posts): http://forum.doom9.org/showthread.php?s=&threadid=93771

regards
t :)
I do have a tendency to exaggerate! :o
Thanks Lark!
jeanl

[2COOL]

OK, how's this. SPRM 6 (PGC Number) is 0 at default. As also with GRPM(15). As soon as a Title is accessed, sprm(6) gets a value of 1 and will never revert back to 0.

Before

VMG , First-Play PGC

************* pre commands
1 (JumpSS) Jump to VTSM 1, Root menu (TNN 1)

_________________________________________________

After

VMG , First-Play PGC

************* pre commands
1 if ( gprm(15) < sprm(6:PGC number) ) then { Goto line 3 }
2 (JumpTT) Jump to Title 1
3 NOP
4 (JumpSS) Jump to VTSM 1, Root menu (TNN 1)

The only disclaimer here is that if FP is executed the third time around, grpm(15) shouldn't be a 0 or else Title 1 will play again. Of course, we have 15 other gprms to work with and at least one of them should be greater than zero at this point.

or we could also do this

After

VMG , First-Play PGC

************* pre commands
1 if ( gprm(15) != sprm(6:PGC number) ) then { Goto line 3 }
2 (JumpTT) Jump to Title 1
3 NOP
4 (JumpSS) Jump to VTSM 1, Root menu (TNN 1)

[r0lZ]

Originally posted by 2COOL
With so much posts recently in regards to intros creations, I looked at the "Import First-Play Clip" macro and came up with 2 questions.

1. Why do we need to have an unused GPRM? It is possible to use a used GPRM. Here's one of my First Play PGC examples. RED are new commands.

Before

************* pre commands
1 (JumpSS) Jump to VTSM 1, Root menu (TNN 1)


After

************* pre commands
1 if ( gprm(15) == 1 ) then ( Goto line 4)
2 Set gprm(15) =(mov) 1
3 (JumpTT) Jump to Title 2 //Imported Title
4 Set gprm(15) =(mov) 0 //GPRM reset back to default 0
5 NOP
6 (JumpSS) Jump to VTSM 1, Root menu (TNN 1)


2. Why can't we have a value of 255 for an infinite still Title? The Title would display indefinite until the user presses Play. With your method, you reset the GPRM to 0 after the intro clip has been played. Therefore, if the FP-PGC is called again by an original command somewhere in the DVD, the intro clip will be played again, and we don't want that!
Try with a DVDShrink reauthored DVD with 2 or 3 clips: the intro clip will be played before each reauthored title.

[EDIT:] Obsolete post. I've not read the answers before posting. Error! :o

[blutach]

Originally posted by jeanl
blutach this looks like the problem another user was having on the forum (can't find the message again, the search utility is simply useless when you have threads that are 20 pages long :devil: ).
There was a bug in the original DVD2AVI code that got carried through to PgcEditPreview. Can you preview the vob? (that should work)...
Jeanl Yeah, I rememeber that. When I try to preview the menu VOB, either with the preview in 0.5.6B4 or the one you sent me, it hangs PgcEditPreview.exe and processor % goes up to 100%.

Regards

[lark]

Originally posted by blutach
preview in 0.5.6B4
hmm, what's that?

regards
t :)

[r0lZ]

Originally posted by lark
hmm, what's that? Well, I've uploaded Beta 6 at VideoHelp. The new preview, which should fix the problem, is included in this beta (but not referenced in the history).

Please note that the Jump2PGC macro is still in a somewhat experimental state.

[r0lZ]

Originally posted by northwind
Is this a known problem?



can't read "::buttons::butinfo(0,errorcode)": no such variable
can't read "::buttons::butinfo(0,errorcode)": no such variable
while executing
"if {[info exists ::menus($cur_vts,$cur_lu,$cur_pgc,numbuttons)] && $::buttons::butinfo($cur_vts,errorcode)==0} {
set cellpostable $::menus($cur_..."
(procedure "simulate_play" line 127)
invoked from within
"simulate_play"
invoked from within
"if {$idx == "-"} {
::main::set_listbox_selection end
set idx [lindex [::main::get_selection_indexes] 0]
bell
} elseif $::tr::now_playin..."
(procedure "step" line 5)
invoked from within
"step "
(procedure "::tr::dvd_trace" line 329)
invoked from within
"::tr::dvd_trace"
(command bound to event) Will have a look... Thanks!

[lark]

Originally posted by r0lZ
I've uploaded Beta 6 at VideoHelp. The new preview, which should fix the problem, is included in this beta (but not referenced in the history).
ok, thx. i can confirm that menu has background image (for my src).

regards
t :)

[blutach]

Will try B6 and revert. Thanks r0lZ and lark for alerting me to that "old" thread (3 whole days LMAO).

EDIT: Beta 6 doesn't work for me. :(

Regards

[r0lZ]

Originally posted by blutach
EDIT: Beta 6 doesn't work for me. :( So, could you send me a small DVD to test?

[blutach]

I would, but if I put the menu through VobBlanker (just to blank the only cell with audio and get it down to a couple of Mb) or MenuShrink (which shrinks them all), PgcEdit reads all the menu PGCs beautifully!

What was that site to upload the DVD? I can zip it up - it's only about 6Mb once I blanked the title VOBs?

EDIT: I'll try to email it.

Regards

[jeanl]

Originally posted by blutach
I would, but if I put the menu through VobBlanker (just to blank the only cell with audio and get it down to a couple of Mb) or MenuShrink (which shrinks them all), PgcEdit reads all the menu PGCs beautifully!

What was that site to upload the DVD? I can zip it up - it's only about 6Mb once I blanked the title VOBs?

EDIT: I'll try to email it.

Regards

www.yousendit.com

works really well.
jeanl

[jeanl]

Thanks blutach, I got the files. I looked into it and traced the problem back to a part of the code that I don't understand very well, where the original author commented out part of the code saying:

// This code is designed to ignore false (emulated) leading
// SEQUENCE_HEADER_CODEs. But it precludes decoding MPEG1
// because it would trigger this. I'll wait until I see
// how real the issue is before doing anything. So far it is
// just one VOB reported and the notifier no longer even has
// it.

When I re-enable the code that was commented out, things work normally. From what I understand, this particular vob has a "false - emulated leading SEQUENCE_HEADER_CODE", which is what causes DVD2AVI to fail. I'm not sure I know what that means, nor do I see that in the VOB file (but I don't really know where to look, it's hard to figure out where you are in the file with DVD2AVI!). What I know is that this fake sequence header has the wrong picture height/width, which is then responsible for the later failure. The code that I re-enabled skips the fake sequence header and goes to the next one, which has the right data...

I'm going to look into it a bit more to see what's going on. In the meantime, if anybody has any input/hint on that, I would really appreciate it.

jeanl

[jeanl]

This is a good one! Check this out:

http://img105.echo.cx/img105/6341/foo7re.png (http://www.imageshack.us)

Look at the binary value for the VOBU_SRI circled in red. 0x0000001b3! This matches exactly SEQUENCE_HEADER_CODE and DVD2AVI isn't very smart about it! So it's fooled into thinking this is the beginning of a mpeg sequence header, smack in the middle of the VOBU_SRI block! If it wasn't so sad, it would be funny!
:D
Anyway, I'm not too sure what to do about it. I could leave the code that was originally commented out, but then preview will fail on MPEG-1 files. I'm not sure how important that is.
Any opinion?
jeanl
EDIT: The reason why everything works well if MenuShrink or Vobblanker is used to still this guy is because the VOBU_SRI pointers are all reset! AH AH AH! :D

[mpucoder]

When I first heard of this "false (emulated) header" thing I thought it might be something dumb like this. While header emulation IS impossible within the video stream, it is only within the video stream. When interpretting a system stream (mpeg-1) or program stream (mpeg-2) there are two levels of context, the system level, and the elementary stream level. Looks like DVD2AVI tried to take a shortcut and not recognize pack headers or pes headers. If it did the entire NAV pack would get skipped as it is a private stream, not part of the video.

Why does mpeg-1 fail if the code is compiled? Is it possibly because mpeg-1's pack header is different than mpeg-2's? If so there is a simple way to know you are in mpeg-2, and it is even mentioned in the spec. The presence of any extension header means mpeg-2 (it was not used in mpeg-1). For that matter (just looked it up on my site) the pack header for mpeg-2 is recognizable by the value in the byte following the start code.

[jeanl]

Originally posted by mpucoder
When I first heard of this "false (emulated) header" thing I thought it might be something dumb like this. While header emulation IS impossible within the video stream, it is only within the video stream. When interpretting a system stream (mpeg-1) or program stream (mpeg-2) there are two levels of context, the system level, and the elementary stream level. Looks like DVD2AVI tried to take a shortcut and not recognize pack headers or pes headers. If it did the entire NAV pack would get skipped as it is a private stream, not part of the video.
Yes, you're right. I realize this now, but there isn't much I can do about it, short of rewriting a whole lot of code. Maybe you can tell me: If all we're interested in is previewing DVDs, can we be sure that the video will be MPEG-2? If yes, I think I'll re-enable the kludge that the original author put in to avoid the problem, if it does not break anything else!
jeanl

[mpucoder]

That depends on whether you mean mpeg-2 PS or mpeg-2 video. Mpeg-1 video is allowed, but the system level is always mpeg-2.

[jeanl]

mmm, I'm not sure what I mean :)! The original author writes that the kludge to avoid the problem will "trigger" with MPEG-1 video. I'm not sure what that means exactly... Maybe I should try to get my hands on a DVD with MPEG-1 video (not easy, I guess!, unless I make one myself).
jeanl

[r0lZ]

Anyway, I'm glad to see that the problem is not in PgcEdit's code. I have tried to understand what's happening in my code, but everything was correct! :)
Good luck, jeanl!

[jsoto]

The original DVD2AVI author has been recently registered here as jackei and is posting (not too much) in DVD2AVI forum
May be you can ask him by PM or in DVD2AVI forum.

jsoto

[blutach]

Why doesn't the macro allow you to clone the current PGC's behaviour if there is more than one title in the VTS?

Regards

[r0lZ]

It's not related to the number of titles.

To be able to Clone the original PGC behaviour in the 'Import Title's Closing Clip' macro, the original commands are copied to a new PGC in the VMGM domain. Some commands (such as SetSTN, SetNVTMR) are illegal in the VMGM. Some commands (all Links, JumpVTS_PTT...) cannot be used from the VMGM to reach the target PGC.
If there are such commands in the original PGC, the Clone option is disabled, because it's impossible (or too difficult) to do.

[Guest]

IIRC, I'm the one who made that comment.

The mention of MPEG1 was just forward thinking. DVD2AVI/DGIndex currently do not support MPEG1 video, but I plan to add support for it. That comment was to remind me that the test was relevant to that plan.

The current code does an initial scan looking for a sequence header. mpucoder is correct that this initial scan does not take into account the system layer, and that it should do so.

DGIndex has diverged and I'll have to check how it will react to this situation. Is there a VOB you can give me?

[jeanl]

Sure, I'll upload the VOB to your ftp site.

jeanl

[jeanl]

OK, the vob is on your ftp site (gee, that's faast), VTS_01_0.VOB (sorry for the lame name!). Note that you can create the problem yourself by editing the first navpack in Vobedit on any vob file, and putting the right hex value in one of the VOBU_SRI pointers for example (or anywhere else for that matter!).

Let me know if you can think of a more robust way to avoid the problem than what's in the DVD2AVI code... Still, according to your response, DVD2AVI does not support MPEG1, so I shouldn't worry about that part!

Thanks!
Jeanl