dvdauthor-developer

Re: [Dvdauthor-developer] Feel Like Doing A New Release

[Elio B. <ebl...@us...>]

Il 01/11/2016 04:45, Lawrence D'Oliveiro ha scritto:
> Anybody got any patches/issues they’ve been sitting on? Speak now, or
> ... it’ll be too late.

I remember this suggestion I applied in an old self-build:

https://sourceforge.net/p/dvdauthor/mailman/dvdauthor-developer/thread/135346.47898.qm%40web111920.mail.gq1.yahoo.com/#msg27660464

The source code of subfont.c is still remarking the replacing macro could be an improvement and in 
fact, I remember the new rendering was nicer.
Anyone else's impressions ?
Elio

[Dvdauthor-developer] Feel Like Doing A New Release

[Lawrence D'O. <ld...@ge...>]

I know, it’s been too long since the last regular release. There have
only been a few minor patches since then, but I guess it’s time to
make them official.

Anybody got any patches/issues they’ve been sitting on? Speak now, or
... it’ll be too late.

[Dvdauthor-developer] creating DVD from something other than 4:3/16:9

[Rick C. <ric...@gm...>]

Hello,

Normally using ffmpeg with -target ntsc-dvd or -target pal-dvd then following up with dvdauthor works great. But on something other than 4:3/16:9 for example if the aspect ratio is 2.40:1 then the resulting DVD has its aspect ratio changed and no longer looks good. What is the solution for this?

rc

Re: [Dvdauthor-developer] Errors in UDF format while sending ISO to DVD replication

[Lawrence D'O. <ld...@ge...>]

[Probably best to continue this discussion on dvdauthor-users rather
than dvdauthor-developer.]

On Tue, 28 Jul 2015 17:15:24 -0300, Jonatã Bolzan wrote:

> - Video Manager IFO and BUP files in same ECC block
> 
> Talking with mkisofs developers, it seems that the second error is
> related with the files generated by dvdauthor, and they recommended me
> to ask about this with you.

In what way is it related to the files, exactly? dvdauthor only
generates the contents of the files, and according to what is known of
the spec, the two files are supposed to be identical, so dvdauthor
makes them identical. So what else is dvdauthor supposed to be doing?

> I read the mkisofs man and it says that mkisofs with -dvd-video will
> fill needed gaps. But in the case of an "empty" menu, or a VOB that is
> so little that the sum of IFO + VOB is less than 32K, mkisofs will not
> take action.

Why not?

> Well, mkisofs developers say that this is a kind of issue that should
> be fixed by the dvd authoring software.

How?

Have you tried genisoimage as an alternative to mkisofs?

[Dvdauthor-developer] Errors in UDF format while sending ISO to DVD replication

[Jonatã B. <jo...@jo...>]

Hi there!

We are developing an interface for our dvd authoring software using
dvdauthor. We are able to create the ISO file and we would like to
send it to a replication company. They pointed out that there are some
errors in our ISO file:

- Volume Descriptor Sequences are not equivalent
- Video Manager IFO and BUP files in same ECC block

Talking with mkisofs developers, it seems that the second error is
related with the files generated by dvdauthor, and they recommended me
to ask about this with you.

As you can note, the following information on this email is the same
as in this thread (sorry by the link, sf seems to have some problem by
now):
http://sourceforge.net/p/cdrtools/mailman/cdrtools-developers/?viewmonth=201507&style=threaded

Well, the error description is the following:

"The DVD-Video specifications require that the Video Manager Information
file (VIDEO_TS.IFO) and its backup (VIDEO_TS.BUP) NOT be recorded in
the same ECC block.

Background
For redundancy purposes, the DVD-Video specifications require a backup
file for the Video Manager Information file. If this file should
become corrupt, a DVD-Player can locate the backup file and use it
instead. If the two files overlap within the same ECC Block, it is
possible that if the ECC fails, it will cause both files to be
unreadable.

Possible Cause
This problem is caused by the authoring system that created the image.
It is the responsibility of the authoring system to ensure that these
files do not overlap within the same ECC Block.

Possible Solution
This problem can only be corrected by an authoring system."

So, I search about this. IFO and BUP can not be on the same ECC
sector, that is 32k large (16 x 2048). The behaviour is described
here:

http://club.myce.com/f62/video_ts-ifo-video_ts-bup-same-ecc-block-98278/#post621399

I read the mkisofs man and it says that mkisofs with -dvd-video will
fill needed gaps. But in the case of an "empty" menu, or a VOB that is
so little that the sum of IFO + VOB is less than 32K, mkisofs will not
take action. In fact, this is an issue with some version of Nero too:

http://download.videohelp.com/r0lZ/pgcedit/third_party/blutach/Burning%20With%20PgcEdit.htm

At the same time, it seems that this kind of issue was already discussed:

http://osdir.com/ml/audio.cd-record/2002-12/msg00015.html

Well, mkisofs developers say that this is a kind of issue that should
be fixed by the dvd authoring software.

What do you think? I would like to help testing this.

Thanks for your attention.

--
Jonatã Bolzan Loss

Re: [Dvdauthor-developer] [PATCH] configure.ac: fix static build

[Lawrence D'O. <ld...@ge...>]

On Sun, 15 Feb 2015 22:00:51 +0100, Romain Naour wrote:

> Add LT_INIT macro to handle --enable-shared option.
> 
> If enable-shared=no then add --static flag to resolve FREETYPE_LIBS.
> 
> Otherwise, dvdauthor forget to link witk -lbz2 when linking with
> freetype2.
> 
> Fixes:
> http://autobuild.buildroot.net/results/1d8/1d83390a3dd9f6bb595e9fc7b321500b4dc533a8/
> 
> Signed-off-by: Romain Naour <rom...@op...>

Accepted.

[Dvdauthor-developer] [PATCH] configure.ac: fix static build

[Romain N. <rom...@op...>]

Add LT_INIT macro to handle --enable-shared option.

If enable-shared=no then add --static flag to resolve FREETYPE_LIBS.

Otherwise, dvdauthor forget to link witk -lbz2 when linking with freetype2.

Fixes:
http://autobuild.buildroot.net/results/1d8/1d83390a3dd9f6bb595e9fc7b321500b4dc533a8/

Signed-off-by: Romain Naour <rom...@op...>
---
 configure.ac | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/configure.ac b/configure.ac
index 68301e3..c110087 100644
--- a/configure.ac
+++ b/configure.ac
@@ -4,6 +4,7 @@ AC_CONFIG_HEADERS(src/config.h)
 AC_CONFIG_AUX_DIR(autotools)
 
 AM_INIT_AUTOMAKE
+LT_INIT
 
 AC_PROG_INSTALL
 
@@ -39,6 +40,11 @@ PKG_CHECK_MODULES(LIBPNG, [libpng])
 AC_SUBST(LIBPNG_CFLAGS)
 AC_SUBST(LIBPNG_LIBS)
 
+config_static=''
+if test "$enable_shared" = 'no'; then
+    config_static='--static'
+fi
+
 usemagick=0
 
 AC_CHECK_PROGS(MAGICKCONFIG, [MagickCore-config])
@@ -81,7 +87,7 @@ AC_SUBST(FRIBIDI_LIBS)
 AC_CHECK_PROGS(FREETYPECONFIG, [freetype-config])
 if test -n "$FREETYPECONFIG"; then
     FREETYPE_CPPFLAGS="`$FREETYPECONFIG --cflags`"
-    FREETYPE_LIBS="`$FREETYPECONFIG --libs`"
+    FREETYPE_LIBS="`$FREETYPECONFIG --libs $config_static`"
     AC_DEFINE(HAVE_FREETYPE, 1, [Whether FreeType is available])
 
     ac_save_CPPFLAGS="$CPPFLAGS"
-- 
1.9.3

Re: [Dvdauthor-developer] dvdunauthor two fixes

[Lawrence D'O. <ld...@ge...>]

On Tue, 02 Sep 2014 21:30:50 +0200, Elio Blanca wrote:

> Il 26/08/2014 11:04, Lawrence D'Oliveiro ha scritto:
>
>> While we’re at it, get rid of the unwieldy use of times(2) and substitute
>> time(2) instead. Whole-second accuracy is all we need.
>
> Maybe it's ok the way it is now.

OK, accepted. Thanks.

Re: [Dvdauthor-developer] dvdunauthor two fixes

[Elio B. <ebl...@us...>]

Il 26/08/2014 11:04, Lawrence D'Oliveiro ha scritto:
> Who said anything about checking errno?
Ok, got it.



> The answer is to use floating point for the calculation. While we’re at
> it, get rid of the unwieldy use of times(2) and substitute time(2)
> instead. Whole-second accuracy is all we need.
Maybe it's ok the way it is now.

Re: [Dvdauthor-developer] dvdunauthor two fixes

[Lawrence D'O. <ld...@ge...>]

On Mon, 25 Aug 2014 21:13:01 +0200, Elio Blanca wrote:

> Checking errno at first is not my preferred way to go.

Who said anything about checking errno?

> Of course, the value into `rmn' is small enough to be ok into a
> signed int but the issue lies in intermediate calculation,
> overflowing very fast (the multiplication comes from the `rmn'
> formula).

The answer is to use floating point for the calculation. While we’re at
it, get rid of the unwieldy use of times(2) and substitute time(2)
instead. Whole-second accuracy is all we need.

Re: [Dvdauthor-developer] dvdunauthor two fixes

[Elio B. <ebl...@us...>]

Some clarification is needed.


Il 24/08/2014 03:23, Lawrence D'Oliveiro ha scritto:
> According to the times(2) man page <http://linux.die.net/man/2/times>,
> the only possible error it can return is EFAULT, which means the
> address of the buffer passed is invalid, which is a program bug.

Checking errno at first is not my preferred way to go.
According to the times man page both here http://linux.die.net/man/2/times and here 
https://www.gnu.org/software/libc/manual/html_node/Processor-Time.html
it is said that
*On error, (clock_t) -1 is returned*
and
*times returns (clock_t)(-1) to indicate failure*
Checking errno is another story, but the very first step would be checking the return value, which 
may be -1 if something goes wrong. Dvdunauthor should skip any calculations when the `now' value 
isn't reliable.


Il 24/08/2014 03:23, Lawrence D'Oliveiro ha scritto:
> As for the likelihood of overflowing the 32-bit integer calculation, on
> my system sysconf(_SC_CLK_TCK) returns 100. I believe on some systems
> it may return 1000, but I’m not aware of any where the number may be
> higher. Even at 1000 ticks per second, signed 32-bit arithmetic isn’t
> going to overflow for 49 days, which would seem to be ample time to
> decode a few gigabytes of DVD-Video.


As an example, look at this short dvdunauthor log (enriched with helpful fprintf):

[cut]
totalsect=2284995 numsect=80384 now=1720990742 start=1720989770
(totalsect - numsect) * (now - start) = 2142881892
STAT: [0] VOB 1, Cell 1 (4%, 4:26 remain)
totalsect=2284995 numsect=80896 now=1720990744 start=1720989770
(totalsect - numsect) * (now - start) = 2146792426
STAT: [0] VOB 1, Cell 1 (4%, 4:25 remain)
totalsect=2284995 numsect=81408 now=1720990745 start=1720989770
(totalsect - numsect) * (now - start) = -2146469971
STAT: [0] VOB 1, Cell 1 (4%, -4:-23 remain)
totalsect=2284995 numsect=81920 now=1720990747 start=1720989770
(totalsect - numsect) * (now - start) = -2142563021
STAT: [0] VOB 1, Cell 1 (4%, -4:-21 remain)
totalsect=2284995 numsect=82432 now=1720990748 start=1720989770
(totalsect - numsect) * (now - start) = -2140860682
STAT: [0] VOB 1, Cell 1 (4%, -4:-19 remain)
totalsect=2284995 numsect=82944 now=1720990750 start=1720989770
(totalsect - numsect) * (now - start) = -2136957316
STAT: [0] VOB 1, Cell 1 (4%, -4:-17 remain)
totalsect=2284995 numsect=83456 now=1720990752 start=1720989770
(totalsect - numsect) * (now - start) = -2133055998
STAT: [0] VOB 1, Cell 1 (4%, -4:-15 remain)
totalsect=2284995 numsect=83968 now=1720990753 start=1720989770
(totalsect - numsect) * (now - start) = -2131357755
STAT: [0] VOB 1, Cell 1 (4%, -4:-13 remain)
totalsect=2284995 numsect=84480 now=1720990755 start=1720989770
(totalsect - numsect) * (now - start) = -2127460021
STAT: [0] VOB 1, Cell 1 (4%, -4:-11 remain)
totalsect=2284995 numsect=84992 now=1720990757 start=1720989770
(totalsect - numsect) * (now - start) = -2123564335
STAT: [0] VOB 1, Cell 1 (4%, -4:-9 remain)
totalsect=2284995 numsect=85504 now=1720990760 start=1720989770
(totalsect - numsect) * (now - start) = -2117471206
STAT: [0] VOB 1, Cell 1 (4%, -4:-7 remain)
[cut]


Of course, the value into `rmn' is small enough to be ok into a signed int but the issue lies in 
intermediate calculation, overflowing very fast (the multiplication comes from the `rmn' formula).
At the beginning of the job, dvdunauthor has (totalsect-numsect) which is a big value, then it 
slowly decrease. At the same time, (now-start) starts with small values and grows as time goes on. 
This originates many unpredictable integer overflows and they bring those ugly negative numbers 
reported as remaining time.
The formula is well written, as it assures the smallest approximations, but the range used in 
intermediate calculations IMHO is not adequate.

Re: [Dvdauthor-developer] dvdunauthor two fixes

[Lawrence D'O. <ld...@ge...>]

On Sat, 23 Aug 2014 01:19:03 +0200, Elio Blanca wrote:

> This patch contains two fixes for dvdunauthor.
> The first one is a check on return value from the times() function.
> The second one is a widened range for computation of remaining time,
> in order to prevent dvdunauthor from printing negative numbers as
> remaining time.
> 
> diff -ur dvdauthor-master/src/dvdunauthor.c
> dvdauthor-master-1/src/dvdunauthor.c ---
> dvdauthor-master/src/dvdunauthor.c	2014-07-20
> 03:42:06.000000000 +0200 +++
> dvdauthor-master-1/src/dvdunauthor.c	2014-08-23
> 01:08:32.380673829 +0200 @@ -1158,9 +1158,9 @@ if (rl > BIGBLOCKSECT)
> rl = BIGBLOCKSECT; now = times(&unused_tms);
> -            if (now-start > 3 * clkpsec && numsect > 0)
> +            if (now != (clock_t)-1 && now-start > 3 * clkpsec &&
> numsect > 0) {
> -                const int rmn = (totalsect - numsect) * (now -
> start) / (numsect * clkpsec);
> +                const int rmn = (int)(((long long)totalsect -
> numsect) * ((long long)now - start) / (numsect * clkpsec)); /*
> estimate of time remaining */ fprintf
>                    (

According to the times(2) man page <http://linux.die.net/man/2/times>,
the only possible error it can return is EFAULT, which means the
address of the buffer passed is invalid, which is a program bug.

As for the likelihood of overflowing the 32-bit integer calculation, on
my system sysconf(_SC_CLK_TCK) returns 100. I believe on some systems
it may return 1000, but I’m not aware of any where the number may be
higher. Even at 1000 ticks per second, signed 32-bit arithmetic isn’t
going to overflow for 49 days, which would seem to be ample time to
decode a few gigabytes of DVD-Video.

[Dvdauthor-developer] dvdunauthor two fixes

[Elio B. <ebl...@us...>]

This patch contains two fixes for dvdunauthor.
The first one is a check on return value from the times() function.
The second one is a widened range for computation of remaining time, in order to prevent dvdunauthor 
from printing negative numbers as remaining time.
Elio

Re: [Dvdauthor-developer] [PATCH] compat.h needs stuff from config.h so include it there

[Lawrence D'O. <ld...@ge...>]

On Sat, 19 Jul 2014 21:06:58 +0300, Ville Skyttä wrote:

> This was brought up by dvdvml.* not including it and thus causing a
> build failure on Fedora (22) devel. While at it, uninclude config.h
> from files where it was included just for compat.h reasons.

Committed. Thanks.

[Dvdauthor-developer] [PATCH] compat.h needs stuff from config.h so include it there

[Ville S. <vil...@ik...>]

This was brought up by dvdvml.* not including it and thus causing a
build failure on Fedora (22) devel. While at it, uninclude config.h
from files where it was included just for compat.h reasons.
---
 src/compat.h        | 2 ++
 src/conffile.c      | 1 -
 src/dvdauthor.c     | 2 --
 src/dvdcompile.c    | 2 --
 src/dvdpgc.c        | 1 -
 src/dvdunauthor.c   | 2 --
 src/dvduncompile.c  | 1 -
 src/dvdvob.c        | 2 --
 src/readxml.c       | 2 --
 src/spuunmux.c      | 1 -
 src/subgen-encode.c | 2 --
 11 files changed, 2 insertions(+), 16 deletions(-)

diff --git a/src/compat.h b/src/compat.h
index 126dcd7..93de081 100644
--- a/src/compat.h
+++ b/src/compat.h
@@ -1,6 +1,8 @@
 // basic headers
 #define _GNU_SOURCE /* really just for strndup */
 
+#include "config.h"
+
 #ifdef HAVE_STDBOOL_H
 # include <stdbool.h>
 #else
diff --git a/src/conffile.c b/src/conffile.c
index cab661a..36ded07 100644
--- a/src/conffile.c
+++ b/src/conffile.c
@@ -21,7 +21,6 @@
  * MA 02110-1301 USA.
  */
 
-#include "config.h"
 #include "compat.h"
 #include <errno.h>
 #include "conffile.h"
diff --git a/src/dvdauthor.c b/src/dvdauthor.c
index d591d68..4860258 100644
--- a/src/dvdauthor.c
+++ b/src/dvdauthor.c
@@ -20,8 +20,6 @@
  * MA 02110-1301 USA.
  */
 
-#include "config.h"
-
 #include "compat.h"
 
 #include <sys/types.h>
diff --git a/src/dvdcompile.c b/src/dvdcompile.c
index fd232b6..47f13e5 100644
--- a/src/dvdcompile.c
+++ b/src/dvdcompile.c
@@ -20,8 +20,6 @@
  * MA 02110-1301 USA.
  */
 
-#include "config.h"
-
 #include "compat.h"
 #include <assert.h>
 
diff --git a/src/dvdpgc.c b/src/dvdpgc.c
index e0adc61..cf6eb04 100644
--- a/src/dvdpgc.c
+++ b/src/dvdpgc.c
@@ -20,7 +20,6 @@
  * MA 02110-1301 USA.
  */
 
-#include "config.h"
 #include "compat.h"
 #include <errno.h>
 #include <assert.h>
diff --git a/src/dvdunauthor.c b/src/dvdunauthor.c
index ccd422f..f988cdf 100644
--- a/src/dvdunauthor.c
+++ b/src/dvdunauthor.c
@@ -23,8 +23,6 @@
  * MA 02110-1301 USA.
  */
 
-#include "config.h"
-
 #include "compat.h"
 
 #include <ctype.h>
diff --git a/src/dvduncompile.c b/src/dvduncompile.c
index 0c5eaf1..a064011 100644
--- a/src/dvduncompile.c
+++ b/src/dvduncompile.c
@@ -22,7 +22,6 @@
  * MA 02110-1301 USA.
  */
 
-#include "config.h"
 #include "compat.h"
 
 #include <stdio.h>
diff --git a/src/dvdvob.c b/src/dvdvob.c
index 2d7e2b8..2146689 100644
--- a/src/dvdvob.c
+++ b/src/dvdvob.c
@@ -20,8 +20,6 @@
  * MA 02110-1301 USA.
  */
 
-#include "config.h"
-
 #include "compat.h"
 
 #include <assert.h>
diff --git a/src/readxml.c b/src/readxml.c
index 61ff0a1..3b83f07 100644
--- a/src/readxml.c
+++ b/src/readxml.c
@@ -20,8 +20,6 @@
  * MA 02110-1301 USA.
  */
 
-#include "config.h"
-
 #include "compat.h"
 
 #include <assert.h>
diff --git a/src/spuunmux.c b/src/spuunmux.c
index 1981e41..94fcb10 100644
--- a/src/spuunmux.c
+++ b/src/spuunmux.c
@@ -31,7 +31,6 @@
  * MA 02110-1301 USA.
  */
 
-#include "config.h"
 #include "compat.h"
 
 #include <fcntl.h>
diff --git a/src/subgen-encode.c b/src/subgen-encode.c
index 53ed4b2..962e902 100644
--- a/src/subgen-encode.c
+++ b/src/subgen-encode.c
@@ -21,8 +21,6 @@
  * MA 02110-1301 USA.
  */
 
-#include "config.h"
-
 #include "compat.h"
 
 #include <assert.h>
-- 
1.9.3

Re: [Dvdauthor-developer] [PATCH] Spelling fixes.

[Lawrence D'O. <ld...@ge...>]

On Thu,  7 Mar 2013 18:40:18 +0200, Ville Skyttä wrote:

> ---
>  ChangeLog          | 6 +++---
>  doc/dvdauthor.sgml | 2 +-
>  src/dvdauthor.c    | 4 ++--
>  src/dvdcli.c       | 2 +-
>  src/dvdvob.c       | 2 +-
>  src/subreader.c    | 2 +-
>  src/subrender.c    | 2 +-
>  7 files changed, 10 insertions(+), 10 deletions(-)

Done.

[Dvdauthor-developer] [PATCH] Spelling fixes.

[Ville S. <vil...@ik...>]

---
 ChangeLog          | 6 +++---
 doc/dvdauthor.sgml | 2 +-
 src/dvdauthor.c    | 4 ++--
 src/dvdcli.c       | 2 +-
 src/dvdvob.c       | 2 +-
 src/subreader.c    | 2 +-
 src/subrender.c    | 2 +-
 7 files changed, 10 insertions(+), 10 deletions(-)

diff --git a/ChangeLog b/ChangeLog
index c6aeac4..1a3fc37 100644
--- a/ChangeLog
+++ b/ChangeLog
@@ -108,7 +108,7 @@
 	ensure a row of a subtitle takes less than 1440 bits
 	fix warnings
 	fix 64-bit support (iconv)
-	fix video and sector pts values to be more consistant
+	fix video and sector pts values to be more consistent
 	fix ImageMagick 6.x.x alpha detection
 	improved dvdunauthor by Ralf Engels
 	full sized pts/scr support (for 13+ hours DVD's)
@@ -124,7 +124,7 @@
 	better audio type detection logic
 	clear seamless flag on audio discontinuity
 	only allow <titles> to exist once per titleset
-	use 'name' attribute for buttons and actions in spumux (to be consistant with dvdauthor)
+	use 'name' attribute for buttons and actions in spumux (to be consistent with dvdauthor)
 	dynamically allocate PGC buffer
 	allow up to 2730 titles
 	allow reading from filedescriptors for VOBs, palette, and XML
@@ -174,7 +174,7 @@
 	include XML schemas for XML verification
 
 0.6.7: November 24, 2003
-	autodetect io.h presense for Cygwin
+	autodetect io.h presence for Cygwin
 	have spumux deal with incomplete reads/writes to/from pipes
 	have spumux work with odd sized subtitles
 	allow XML comments in dvd code
diff --git a/doc/dvdauthor.sgml b/doc/dvdauthor.sgml
index 7b10b20..2828609 100644
--- a/doc/dvdauthor.sgml
+++ b/doc/dvdauthor.sgml
@@ -109,7 +109,7 @@ XML control file instead.</para>
 <variablelist>
 
 <varlistentry><term><literal>-j</literal></term><term><literal>--jumppad</literal></term>
-<listitem><para>Enables the creation of jumppads, which allow greater flexibility in choosing jump/call desinations.</para></listitem></varlistentry>
+<listitem><para>Enables the creation of jumppads, which allow greater flexibility in choosing jump/call destinations.</para></listitem></varlistentry>
 
 <varlistentry><term><literal>-g</literal></term><term><literal>--allgprm</literal></term>
 <listitem><para>Enable the use of all 16 general purpose registers.  Prohibits the use of jumppad and some complex expressions that require temporary registers.</para></listitem></varlistentry>
diff --git a/src/dvdauthor.c b/src/dvdauthor.c
index 907300e..d591d68 100644
--- a/src/dvdauthor.c
+++ b/src/dvdauthor.c
@@ -1073,7 +1073,7 @@ static char *makevtsdir(const char *s)
     return strdup(fbuf);
 }
 
-// jumppad requires the existance of a menu to operate
+// jumppad requires the existence of a menu to operate
 // if no languages exist, create an english one
 static void jp_force_menu(struct menugroup *mg, vtypes type)
   {
@@ -1447,7 +1447,7 @@ static void validatesummary(struct pgcgroup *va)
 
             for( j=0; j<8; j++ )
                 if( va->allentries & p->entries & (1<<j) )
-                    fprintf(stderr,"ERR:  Multiple definitions for entry %s, 2nd occurance in PGC #%d\n",entries[j],i);
+                    fprintf(stderr,"ERR:  Multiple definitions for entry %s, 2nd occurrence in PGC #%d\n",entries[j],i);
             err = true;
         }
         if (va->pstype != VTYPE_VTS && (p->entries & ~allowedentries) != 0)
diff --git a/src/dvdcli.c b/src/dvdcli.c
index 8114f10..b9a49bf 100644
--- a/src/dvdcli.c
+++ b/src/dvdcli.c
@@ -311,7 +311,7 @@ static void usage()
             "\n\t" LONGOPT("--fpc or ") "-F CMD sets the commands to be executed when the disc is first\n"
             "\t    inserted.\n"
             "\n\t" LONGOPT("--jumppad or ") "-j enables the creation of jumppads, which allow greater\n"
-            "\t    flexibility in choosing jump/call desinations.\n"
+            "\t    flexibility in choosing jump/call destinations.\n"
             "\n\t" LONGOPT("--allgprm or ") "-g enables the use of all 16 general purpose registers.\n"
             "\n\t" LONGOPT("--help or ") "-h displays this screen.\n"
         );
diff --git a/src/dvdvob.c b/src/dvdvob.c
index b1cc82a..1cd41ce 100644
--- a/src/dvdvob.c
+++ b/src/dvdvob.c
@@ -1866,7 +1866,7 @@ noshow:
                                 "WARN: Using SCR instead.\n"
                           );
                         firstaudiopts = readscr(vi->sectdata + 4) + 4 * 147;
-                          // 147 is roughly the minumum pts that must transpire between packets;
+                          // 147 is roughly the minimum pts that must transpire between packets;
                           // we give a couple packets of buffer to allow the dvd player to
                           // process the data
                       } /*if*/
diff --git a/src/subreader.c b/src/subreader.c
index c7b0182..9a19b93 100644
--- a/src/subreader.c
+++ b/src/subreader.c
@@ -2278,7 +2278,7 @@ sub_data *sub_read_file(const char *filename, float movie_fps)
                         // equal to sub_lines. we store in fragment_position the
                         // position of the shortest group, in fragment_length its
                         // length, and in blank_lines_avail the length of the group currently
-                        // examinated
+                        // examined
                         for (i = 0; i < lines_to_add; ++i)
                           {
                             if (placeholder[subs_done][i] == -1)
diff --git a/src/subrender.c b/src/subrender.c
index 50e7044..0be02d0 100644
--- a/src/subrender.c
+++ b/src/subrender.c
@@ -814,7 +814,7 @@ inline static void vo_update_text_sub
                 prev_line_end = chindex;
                 y += vo_font->height;
               } /*for*/
-            /* Here you could retreive the buffers*/
+            /* Here you could retrieve the buffers*/
           } /*if*/
       }
   } /*vo_update_text_sub*/
-- 
1.7.11.7

Re: [Dvdauthor-developer] Bitwise "and" comparison operator generates parse error...

[Micah Q. <mic...@gm...>]

I've fixed this issue in my local copy of the source code, but again, I'm
not sure how you'd like me to submit the patch.  Should I email you an
updated patch?  Are you interested in fixing this particular bug?


On Mon, Dec 17, 2012 at 11:30 PM, Micah Quinn <mic...@gm...> wrote:

> No, sorry.  This code executed within a <pre> block encapsulated in a
> CDATA section generates the following error:
>
> ERR:  Parse error 'syntax error, unexpected CLOSEPAREN_TOK' on token ')'
>
> Also, to my previously submitted patch, I've noticed a bug in the
> negatecompare function.  It causes the bitwise and logic to be the reverse
> of what it should be.  I'm looking now at how to identify a bitwise and
> operator and not negate the if statement.
>
>
> On Tue, Dec 18, 2012 at 5:08 AM, Andrew Church <ac...@ac...>wrote:
>
>> >> Perhaps I am missing something.  What is the proper way in DVDAuthor
>> >> script to do a bitwise comparison between a register and a static
>> value?
>> >>  The equivalent in 'C' would be the following:
>> >>
>> >> if ( g0 & 2 )
>> >>     printf("Bit one is set\n");
>> >> if ( g0 & 4 )
>> >>     printf("Bit two is set\n");
>> >> if ( g0 & 8 )
>> >>     printf("Bit three is set\n");
>>
>> Somehwat belated, but I presume the answer to your question is:
>>
>> if ((g0 & 1) != 0) g1 = 1;
>> if ((g0 & 2) != 0) g2 = 1;
>> if ((g0 & 4) != 0) g3 = 1;
>>
>>   --Andrew Church
>>     ac...@ac...
>>     http://achurch.org/
>>
>>
>> ------------------------------------------------------------------------------
>> LogMeIn Rescue: Anywhere, Anytime Remote support for IT. Free Trial
>> Remotely access PCs and mobile devices and provide instant support
>> Improve your efficiency, and focus on delivering more value-add services
>> Discover what IT Professionals Know. Rescue delivers
>> http://p.sf.net/sfu/logmein_12329d2d
>> _______________________________________________
>> Dvdauthor-developer mailing list
>> Dvd...@li...
>> https://lists.sourceforge.net/lists/listinfo/dvdauthor-developer
>>
>
>
>
> --
> Micah Quinn
>



-- 
Micah Quinn

Re: [Dvdauthor-developer] Bitwise "and" comparison operator generates parse error...

[Micah Q. <mic...@gm...>]

No, sorry.  This code executed within a <pre> block encapsulated in a CDATA
section generates the following error:

ERR:  Parse error 'syntax error, unexpected CLOSEPAREN_TOK' on token ')'

Also, to my previously submitted patch, I've noticed a bug in the
negatecompare function.  It causes the bitwise and logic to be the reverse
of what it should be.  I'm looking now at how to identify a bitwise and
operator and not negate the if statement.


On Tue, Dec 18, 2012 at 5:08 AM, Andrew Church <ac...@ac...> wrote:

> >> Perhaps I am missing something.  What is the proper way in DVDAuthor
> >> script to do a bitwise comparison between a register and a static value?
> >>  The equivalent in 'C' would be the following:
> >>
> >> if ( g0 & 2 )
> >>     printf("Bit one is set\n");
> >> if ( g0 & 4 )
> >>     printf("Bit two is set\n");
> >> if ( g0 & 8 )
> >>     printf("Bit three is set\n");
>
> Somehwat belated, but I presume the answer to your question is:
>
> if ((g0 & 1) != 0) g1 = 1;
> if ((g0 & 2) != 0) g2 = 1;
> if ((g0 & 4) != 0) g3 = 1;
>
>   --Andrew Church
>     ac...@ac...
>     http://achurch.org/
>
>
> ------------------------------------------------------------------------------
> LogMeIn Rescue: Anywhere, Anytime Remote support for IT. Free Trial
> Remotely access PCs and mobile devices and provide instant support
> Improve your efficiency, and focus on delivering more value-add services
> Discover what IT Professionals Know. Rescue delivers
> http://p.sf.net/sfu/logmein_12329d2d
> _______________________________________________
> Dvdauthor-developer mailing list
> Dvd...@li...
> https://lists.sourceforge.net/lists/listinfo/dvdauthor-developer
>



-- 
Micah Quinn

Re: [Dvdauthor-developer] Bitwise "and" comparison operator generates parse error...

[<ac...@ac...>]

>> Perhaps I am missing something.  What is the proper way in DVDAuthor
>> script to do a bitwise comparison between a register and a static value?
>>  The equivalent in 'C' would be the following:
>>
>> if ( g0 & 2 )
>>     printf("Bit one is set\n");
>> if ( g0 & 4 )
>>     printf("Bit two is set\n");
>> if ( g0 & 8 )
>>     printf("Bit three is set\n");

Somehwat belated, but I presume the answer to your question is:

if ((g0 & 1) != 0) g1 = 1;
if ((g0 & 2) != 0) g2 = 1;
if ((g0 & 4) != 0) g3 = 1;

  --Andrew Church
    ac...@ac...
    http://achurch.org/

Re: [Dvdauthor-developer] Bitwise "and" comparison operator generates parse error...

[Micah Q. <mic...@gm...>]

Perhaps I am missing something.  What is the proper way in DVDAuthor script
to do a bitwise comparison between a register and a static value?  The
equivalent in 'C' would be the following:

if ( g0 & 2 )
    printf("Bit one is set\n");
if ( g0 & 4 )
    printf("Bit two is set\n");
if ( g0 & 8 )
    printf("Bit three is set\n");
...

I suspect there is a real bug here because reviewing dvdcompile.c shows
that it would not be possible to generate the bitwise comparison byte code.

I'd like to submit my patch for review because it might make things a bit
clearer, but I'm getting a 403 error when attempting a git push.  What is
the proper way to submit proposed patches to the dvdauthor github?

On Fri, Dec 7, 2012 at 5:36 PM, Lawrence D'Oliveiro <ld...@ge...
> wrote:

> On Fri, 7 Dec 2012 16:33:25 -0600, Micah Quinn wrote:
>
> > I now have a small patch that fixes this, but I cannot find an easy
> > way to submit my changes for approval.  Where can I submit them?
>
> For maximum flexibility, get yourself a GitHub account. Or you can
> submit an occasional patch direct to this list.
>
> But I’m not sure there is really a problem here: you realize that,
> unlike C, the dvdauthor interaction language does not allow interchange
> of integers and booleans?
>
>
> ------------------------------------------------------------------------------
> LogMeIn Rescue: Anywhere, Anytime Remote support for IT. Free Trial
> Remotely access PCs and mobile devices and provide instant support
> Improve your efficiency, and focus on delivering more value-add services
> Discover what IT Professionals Know. Rescue delivers
> http://p.sf.net/sfu/logmein_12329d2d
> _______________________________________________
> Dvdauthor-developer mailing list
> Dvd...@li...
> https://lists.sourceforge.net/lists/listinfo/dvdauthor-developer
>



-- 
Micah Quinn

Re: [Dvdauthor-developer] Bitwise "and" comparison operator generates parse error...

[Lawrence D'O. <ld...@ge...>]

On Fri, 7 Dec 2012 16:33:25 -0600, Micah Quinn wrote:

> I now have a small patch that fixes this, but I cannot find an easy
> way to submit my changes for approval.  Where can I submit them?

For maximum flexibility, get yourself a GitHub account. Or you can
submit an occasional patch direct to this list.

But I’m not sure there is really a problem here: you realize that,
unlike C, the dvdauthor interaction language does not allow interchange
of integers and booleans?

[Dvdauthor-developer] Bitwise "and" comparison operator generates parse error...

[Micah Q. <mic...@gm...>]

Sorry if this does not thread properly on the mailing list.

I now have a small patch that fixes this, but I cannot find an easy way to
submit my changes for approval.  Where can I submit them?

-- 
Micah Quinn

[Dvdauthor-developer] Bitwise "and" comparison operator generates parse error...

[Micah Q. <mic...@gm...>]

Hello,

First, thanks for a very cool tool.  I've learned a great deal about DVD
creation from the source code.

I've recently tried to use the bitwise comparison operator "and" in one of
my scripts, but the dvdauthor parser chokes on it with the following error:

   "ERR:  Parse error 'syntax error, unexpected CLOSEPAREN_TOK' on token
')'"

The chunk of DVD VM script code in question is as follows:

   if ( g0 and 1 ) g1 = 11 ;

I've reviewed the yacc code in dvdvmy.y and found the following:

1.  "ifstatement" consumes the above script line
2.  _AND_TOK is "and" and only takes "boolexpr" for the lh/rh operands
3.  "boolexpr" does not allow for "regornum" and thus the parser error

I've tried changing "boolexpr" such that _AND_TOK takes an "expression" on
the rh/lh side.  I also remoevd the conflicting handler in "expression" for
_AND_TOK.  This allows me to overcome the parser syntax error.  I added a
case for VM_AND in "compilebool" in "dvdcompile.c".  However, "compilebool"
does not appear to handle VM_VALs correctly.  I've only spent a few hours
digging into this, but I was hoping somebody with more knowledge on the VM
script could assist.  I believe once these parser issues are sorted out,
there will still be some bytecode issues to solve.  I've compiled xine-lib
with GPRM tracing, so I can debug DVD script fairly easily.

Let me know how I can help.

P.S.  I'm using the latest from github.

-- 
Micah Quinn

Re: [Dvdauthor-developer] [PATCH] Fix comparison of boolean with int other than 0/1, kudos to cppcheck.

[<ac...@ac...>]

>> In the former however as (va->vd.vformat ||
>> default_video_format) will only ever result in value of 0 or 1 (a
>> boolean) it'll never be == VF_PAL (which is 2), so the returned value
>> will always be VR_NTSC.
>
>I didn't realize that. How about this for a fix:

The previous fix you posted is fine -- if(intval) or if(enumval) are
equivalent to comparing for inequality with zero (this is defined by the
C spec), so the code will do what it looks like it should do.  It's just
a matter of style whether you use explicit comparisons or not.  FWIW,
I'd probably do without them here since zero (and VF_NONE, which is also
zero) is just used to mean "absence of a value".

  --Andrew Church
    ac...@ac...
    http://achurch.org/