Loading ...
Sorry, an error occurred while loading the content.
 

Re: [svg-developers] SVG Broken in Mozilla :-(

Expand Messages
  • Marcel Ruff
    Hi as well, if you start complain, i will join to pour my tears. For me the missing SVG adobe plugin (or stable embeded SVG support) in mozilla is a disaster
    Message 1 of 24 , Apr 2, 2002
      Hi as well,

      if you start complain, i will join to pour my tears.

      For me the missing SVG adobe plugin (or stable embeded
      SVG support) in mozilla is a disaster as well.
      It is especially the missing support on Linux which
      makes me cry.

      I believe if Adobe wants to play a major role
      with browser SVG support, they need to support
      the major browsers.
      Their support is an Explorer only niche support
      and this is no answer for an internet solution.

      Building SVG based on Adobes plugin is a risk
      the plugin is not open source, and Adobe does
      what ever it likes to do, which is their good right though
      as nobody pays them.



      What is the situation?
      ----------------------

      We are tied up with IE, mozilla and probably Opera.
      (OS specific there are some more like konqueror etc).

      Microsoft will keep on supplying solutions which are
      not 100% standard conforming for business reasons
      (they want perfect monopoly and money).
      IE is not running on UNIX/Linux etc.

      Mozilla has not the man power to cover it all (i'm
      bound currently to other projects so i can't help).

      The Opera philosophy i don't know.



      Solutions?
      ----------

      Adobe could be an integrator in the SVG area.
      But they would need to take the responsibility
      and not leave developers alone.
      And they should make it open source.

      Probably a donation project sponsoring more developers
      for mozilla could help, but how to organize this?
      I believe there are many companies on earth willing
      to sponsor this, as they are depending on mozilla.
      It is more an organisational task to solve.



      regards,

      Marcel



      Paul Selormey wrote:
      > Hello,
      > I just wish to state that, hadly will an application be written to work with
      > a plugin.
      > Plugins are written to work with the application, and it makes sense to
      > track
      > progresses in the application.
      > In this particular case, the application involved is in beta so naturally it
      > is rather
      > the Adobe guys who should update their part.
      >
      > Best regards,
      > Paul.
      >
      > ----- Original Message -----
      > From: "skeetersrow" <skeetersrow@...>
      > To: <svg-developers@yahoogroups.com>
      > Sent: Tuesday, April 02, 2002 5:49 PM
      > Subject: [svg-developers] SVG Broken in Mozilla :-(
      >
      >
      >
      >>Hi All
      >>
      >>I am sadden to report that Plugin SVG is now broken in ALL Mozilla
      >>based browsers and will reamain broken until Adobe changes its plugin.
      >>Infact if one tries to view a HTML page with embeded SVG that calls
      >>the Adobe plugin, your browser will crash and crash very nasty. I've
      >>had to reinstall twice. Mozilla release 0.9.9 and NS 6X (being old
      >>browsers) are not affected.
      >> This is 'NOT' a late April fools joke. For more information see the
      >>bug at BugZilla that I started.
      >>http://bugzilla.mozilla.org/show_bug.cgi?id=133567
      >>
      >>I have tried to call this to the attention of Mozilla.org and at the
      >>Mozillazine forum but the result has been mostly negative from them,
      >>that I should shut up and not complain-- it is a sad day for me in
      >>that as you all may know, I don't use IE which means SVG with SMIL is
      >>gone for me for awhile.
      >>
      >>Inline SVG in an XML or XHTML doc rendered without a plugin has run up
      >>against a tough nut. Seems that the libart.dll works only on 16bit
      >>resolution; 24 or 32 bit produce false colors.
      >>
      >>So once again:
      >>
      >>To all who have bothered to install
      >>* NPSVG3.dll
      >>* NPSVG3.zip
      >>into your plugin folder in order to view embeded SVG files that call
      >>the Adobe Plugin, don't try to view any pages with such images.
      >>I have had to reinstall Mozilla twice now.
      >>
      >>Mozilla.org has put the ball in Adobe's court and it will stay that
      >>way until they (Adobe) write a new plugin.



      --
      Marcel Ruff
      mailto:ruff@...
      http://www.lake.de/home/lake/swand/
      http://www.xmlBlaster.org
    • Alex Fritze
      ... You are right in saying that this is not a libart problem; it is the price we paid to get svg working on all platforms (Win/Mac/Unix). There is an easy
      Message 2 of 24 , Apr 2, 2002
        >
        > s> Inline SVG in an XML or XHTML doc rendered without a plugin has run up
        > s> against a tough nut. Seems that the libart.dll works only on 16bit
        > s> resolution; 24 or 32 bit produce false colors.
        >
        > I find this hard to believe. I always run my displays at 24 bit and
        > have been using Mozilla SVG from the beginning, and never had any
        > false colors. Furthermore, the internal model of libart is 24 bit:

        You are right in saying that this is not a libart problem; it is the price
        we paid to get svg working on all platforms (Win/Mac/Unix).
        There is an easy fix, but it would mean loosing <foreignObject> support on
        Unix and Mac. Since for the moment <foreignObject> is our only method of
        text output in svg, I've have held back with it so far.
        For more detail see
        [Bug 111152] SVG doesn't render in all pixel formats
        http://bugzilla.mozilla.org/show_bug.cgi?id=111152

        Alex
      • Michael Bierman
        Gee, sounds pretty poor on Mozila s part to cause such a major break for a plugin this late in their cycle. It is even worse that they place responsibility for
        Message 3 of 24 , Apr 2, 2002
          Gee, sounds pretty poor on Mozila's part to cause such a major break for a
          plugin this late in their cycle. It is even worse that they place
          responsibility for it on Adobe.

          Michael

          > -----Original Message-----
          > From: skeetersrow [mailto:skeetersrow@...]
          > Sent: Tuesday, April 02, 2002 12:50 AM
          > To: svg-developers@yahoogroups.com
          > Subject: [svg-developers] SVG Broken in Mozilla :-(
          >
          >
          > Hi All
          >
          > I am sadden to report that Plugin SVG is now broken in ALL Mozilla
          > based browsers and will reamain broken until Adobe changes its plugin.
          > Infact if one tries to view a HTML page with embeded SVG that calls
          > the Adobe plugin, your browser will crash and crash very nasty. I've
          > had to reinstall twice. Mozilla release 0.9.9 and NS 6X (being old
          > browsers) are not affected.
          > This is 'NOT' a late April fools joke. For more information see the
          > bug at BugZilla that I started.
          > http://bugzilla.mozilla.org/show_bug.cgi?id=133567
          >
          > I have tried to call this to the attention of Mozilla.org and at the
          > Mozillazine forum but the result has been mostly negative from them,
          > that I should shut up and not complain-- it is a sad day for me in
          > that as you all may know, I don't use IE which means SVG with SMIL is
          > gone for me for awhile.
          >
          > Inline SVG in an XML or XHTML doc rendered without a plugin has run up
          > against a tough nut. Seems that the libart.dll works only on 16bit
          > resolution; 24 or 32 bit produce false colors.
          >
          > So once again:
          >
          > To all who have bothered to install
          > * NPSVG3.dll
          > * NPSVG3.zip
          > into your plugin folder in order to view embeded SVG files that call
          > the Adobe Plugin, don't try to view any pages with such images.
          > I have had to reinstall Mozilla twice now.
          >
          > Mozilla.org has put the ball in Adobe's court and it will stay that
          > way until they (Adobe) write a new plugin.



          _________________________________________________________
          Do You Yahoo!?
          Get your free @... address at http://mail.yahoo.com
        • skeetersrow
          Hi Alex, I of course stand corrected, sorry about the missinformation about the libart.dll. I thought that you had once said that was being
          Message 4 of 24 , Apr 2, 2002
            Hi Alex,

            I of course stand corrected, sorry about the missinformation about the
            libart.dll.

            I thought that you had once said that <foreignObject> was being used
            incorreclty in Mozilla?

            Chris, here is a test page that you are free to pinch.
            http://www.skeeter-s.com/svg/color-test.xml


            --- In svg-developers@y..., "Alex Fritze" <alex@c...> wrote:
            > >
            > > s> Inline SVG in an XML or XHTML doc rendered without a plugin has
            run up
            > > s> against a tough nut. Seems that the libart.dll works only on 16bit
            > > s> resolution; 24 or 32 bit produce false colors.
            > >
            > > I find this hard to believe. I always run my displays at 24 bit and
            > > have been using Mozilla SVG from the beginning, and never had any
            > > false colors. Furthermore, the internal model of libart is 24 bit:
            >
            > You are right in saying that this is not a libart problem; it is the
            price
            > we paid to get svg working on all platforms (Win/Mac/Unix).
            > There is an easy fix, but it would mean loosing <foreignObject>
            support on
            > Unix and Mac. Since for the moment <foreignObject> is our only method of
            > text output in svg, I've have held back with it so far.
            > For more detail see
            > [Bug 111152] SVG doesn't render in all pixel formats
            > http://bugzilla.mozilla.org/show_bug.cgi?id=111152
            >
            > Alex
          • skeetersrow
            Marcel Did you miss this page at Adobe? http://www.adobe.com/svg/viewer/install/old.html Here is a direct download for a RedHat Linux 7.1
            Message 5 of 24 , Apr 2, 2002
              Marcel

              Did you miss this page at Adobe?
              http://www.adobe.com/svg/viewer/install/old.html
              Here is a direct download for a RedHat Linux 7.1
              http://download.adobe.com/pub/adobe/magic/svgviewer/linux/3.x/3.0x77/en/adobesvg-3.0-linux-i386.tar.gz

              --- In svg-developers@y..., Marcel Ruff <ruff@s...> wrote:
              > Hi as well,
              >
              > if you start complain, i will join to pour my tears.
              >
              > For me the missing SVG adobe plugin (or stable embeded
              > SVG support) in mozilla is a disaster as well.
              > It is especially the missing support on Linux which
              > makes me cry.
              >
              > I believe if Adobe wants to play a major role
              > with browser SVG support, they need to support
              > the major browsers.
              > Their support is an Explorer only niche support
              > and this is no answer for an internet solution.
              >
              > Building SVG based on Adobes plugin is a risk
              > the plugin is not open source, and Adobe does
              > what ever it likes to do, which is their good right though
              > as nobody pays them.
              >
              >
              >
              > What is the situation?
              > ----------------------
              >
              > We are tied up with IE, mozilla and probably Opera.
              > (OS specific there are some more like konqueror etc).
              >
              > Microsoft will keep on supplying solutions which are
              > not 100% standard conforming for business reasons
              > (they want perfect monopoly and money).
              > IE is not running on UNIX/Linux etc.
              >
              > Mozilla has not the man power to cover it all (i'm
              > bound currently to other projects so i can't help).
              >
              > The Opera philosophy i don't know.
              >
              >
              >
              > Solutions?
              > ----------
              >
              > Adobe could be an integrator in the SVG area.
              > But they would need to take the responsibility
              > and not leave developers alone.
              > And they should make it open source.
              >
              > Probably a donation project sponsoring more developers
              > for mozilla could help, but how to organize this?
              > I believe there are many companies on earth willing
              > to sponsor this, as they are depending on mozilla.
              > It is more an organisational task to solve.
              >
              >
              >
              > regards,
              >
              > Marcel
              >
              >
              >
              > Paul Selormey wrote:
              > > Hello,
              > > I just wish to state that, hadly will an application be written to
              work with
              > > a plugin.
              > > Plugins are written to work with the application, and it makes
              sense to
              > > track
              > > progresses in the application.
              > > In this particular case, the application involved is in beta so
              naturally it
              > > is rather
              > > the Adobe guys who should update their part.
              > >
              > > Best regards,
              > > Paul.
              > >
              > > ----- Original Message -----
              > > From: "skeetersrow" <skeetersrow@y...>
              > > To: <svg-developers@y...>
              > > Sent: Tuesday, April 02, 2002 5:49 PM
              > > Subject: [svg-developers] SVG Broken in Mozilla :-(
              > >
              > >
              > >
              > >>Hi All
              > >>
              > >>I am sadden to report that Plugin SVG is now broken in ALL Mozilla
              > >>based browsers and will reamain broken until Adobe changes its plugin.
              > >>Infact if one tries to view a HTML page with embeded SVG that calls
              > >>the Adobe plugin, your browser will crash and crash very nasty. I've
              > >>had to reinstall twice. Mozilla release 0.9.9 and NS 6X (being old
              > >>browsers) are not affected.
              > >> This is 'NOT' a late April fools joke. For more information see the
              > >>bug at BugZilla that I started.
              > >>http://bugzilla.mozilla.org/show_bug.cgi?id=133567
              > >>
              > >>I have tried to call this to the attention of Mozilla.org and at the
              > >>Mozillazine forum but the result has been mostly negative from them,
              > >>that I should shut up and not complain-- it is a sad day for me in
              > >>that as you all may know, I don't use IE which means SVG with SMIL is
              > >>gone for me for awhile.
              > >>
              > >>Inline SVG in an XML or XHTML doc rendered without a plugin has run up
              > >>against a tough nut. Seems that the libart.dll works only on 16bit
              > >>resolution; 24 or 32 bit produce false colors.
              > >>
              > >>So once again:
              > >>
              > >>To all who have bothered to install
              > >>* NPSVG3.dll
              > >>* NPSVG3.zip
              > >>into your plugin folder in order to view embeded SVG files that call
              > >>the Adobe Plugin, don't try to view any pages with such images.
              > >>I have had to reinstall Mozilla twice now.
              > >>
              > >>Mozilla.org has put the ball in Adobe's court and it will stay that
              > >>way until they (Adobe) write a new plugin.
              >
              >
              >
              > --
              > Marcel Ruff
              > mailto:ruff@s...
              > http://www.lake.de/home/lake/swand/
              > http://www.xmlBlaster.org
            • skeetersrow
              Hi Andrew Nothing contradictory here. As of a week or so ago there was a change in coding and I won t go into that here. End effect is old browsers such as the
              Message 6 of 24 , Apr 2, 2002
                Hi Andrew
                Nothing contradictory here. As of a week or so ago there was a change
                in coding and I won't go into that here. End effect is old browsers
                such as the last release and an older yet NS 6.2 are not affected,
                they still use the old API. The up coming Mozilla 1 and any newer NS
                are affected by this change.



                > > Mozilla release 0.9.9 and NS 6X (being old
                > > browsers) are not affected.

                > Steve,
                >
                > Your statements seem contradictory. You state that it is broken in
                "all"
                > Mozilla browsers then say releast 0.9.9 isn't affected.
                >
                > The problem page at the URL you give loaded fine for me in Mozilla
                0.9.9. I
                > noticed that it may be being served as image/svg-xml. It should be
                > image/svg+xml.
                >
                > I don't know correcting that will help to solve any of the other
                problems.
                >
                > Andrew Watt
                >
                >
                > [Non-text portions of this message have been removed]
              • Alex Fritze
                ... It wasn t actually that much of a misinformation. Libart only works in 24bit, but that is by design. We tried to make a special Mozilla version of libart
                Message 7 of 24 , Apr 2, 2002
                  > I of course stand corrected, sorry about the missinformation about the
                  > libart.dll.

                  It wasn't actually that much of a misinformation. Libart only works in
                  24bit, but that is by design. We tried to make a special Mozilla version of
                  libart that works with other pixel formats. Unfortunately this work was
                  never fully completed, but for one reason or another the 'special' libart
                  found its way into the Mozilla code tree. As a result we get all those nasty
                  rendering problems or even crashes :-(
                  I just thought it important to point out that we (the Mozilla SVG
                  developers) are at fault here and not the libart developers.

                  > I thought that you had once said that <foreignObject> was being used
                  > incorreclty in Mozilla?

                  Some of the samples on the croczilla site use <foreignObject>s slightly
                  incorrectly.
                  According to the SVG DTD <foreignObject>s are only allowed in a <switch>
                  element. I skip the <switch> elements at the moment, because <switch> is not
                  implemented yet.
                • Chris Lilley
                  ... AF Some of the samples on the croczilla site use s slightly AF incorrectly. AF According to the SVG DTD s are only allowed
                  Message 8 of 24 , Apr 2, 2002
                    On Tuesday, April 2, 2002, 4:55:20 PM, Alex wrote:

                    >> I thought that you had once said that <foreignObject> was being used
                    >> incorreclty in Mozilla?

                    AF> Some of the samples on the croczilla site use <foreignObject>s slightly
                    AF> incorrectly.
                    AF> According to the SVG DTD <foreignObject>s are only allowed in a <switch>
                    AF> element. I skip the <switch> elements at the moment, because <switch> is not
                    AF> implemented yet.

                    A quick fix would be to implement a 'switch' that renders the first
                    child, or all its children.

                    Since Mozilla implements CSS2 selectors and a little CSS3, it might be
                    possible to do this using CSS - make the first child of a switch
                    element have display:block and the rest, display:none.

                    This would at least allow valid examples to be used. Implementing the
                    test evaluation would be a second step.



                    --
                    Chris mailto:chris@...
                  • skeetersrow
                    ... Hey, while you both are there, is this then okay? It is saved as an XML doc and seems to render nicely in Mozilla.
                    Message 9 of 24 , Apr 2, 2002
                      > A quick fix would be to implement a 'switch' that renders the first
                      > child, or all its children.
                      >
                      > Since Mozilla implements CSS2 selectors and a little CSS3, it might be
                      > possible to do this using CSS - make the first child of a switch
                      > element have display:block and the rest, display:none.
                      >
                      > This would at least allow valid examples to be used. Implementing the
                      > test evaluation would be a second step.
                      >
                      >
                      >
                      > --
                      > Chris mailto:chris@w...


                      Hey, while you both are there, is this then okay? It is saved as an
                      XML doc and seems to render nicely in Mozilla.

                      <?xml version="1.0" standalone="yes"?>
                      <svg width="400px" height="300px" xmlns = 'http://www.w3.org/2000/svg'>
                      <switch>
                      <circle cx="150px" cy="100px" r="75px" style="fill:red; fill-opacity:;
                      stroke:blue; stroke-width:5; stroke-opacity:;" />
                      <ellipse cx="150px" cy="100px" rx="150px" ry="25px"
                      style="fill:yellow; fill-opacity:0.40; stroke:maroon; stroke-width:5;
                      stroke-opacity:;" />
                      <foreignObject x="70px" y="93px" width="100px" height="50px"

                      requiredExtensions="http://example.com/SVGExtensions/EmbeddedXHTML">
                      <html xmlns="http://www.w3.org/1999/xhtml">
                      <body>
                      <p style="font: 14px Arial, Helvetica, sans-serif bold;
                      color: maroon; background-color: silver;">Is this doing it correctly?</p>
                      </body>
                      </html>
                      </foreignObject>
                      </switch>
                      </svg>
                    • Marcel Ruff
                      ... This 3.0 beta has some problems: The Javascript communication of the plugin with the browser is broken (it was fine only with mozilla 0.91). Reloading an
                      Message 10 of 24 , Apr 2, 2002
                        skeetersrow wrote:
                        > Marcel
                        >
                        > Did you miss this page at Adobe?
                        > http://www.adobe.com/svg/viewer/install/old.html
                        > Here is a direct download for a RedHat Linux 7.1
                        > http://download.adobe.com/pub/adobe/magic/svgviewer/linux/3.x/3.0x77/en/adobesvg-3.0-linux-i386.tar.gz

                        This 3.0 beta has some problems:

                        The Javascript communication of the plugin with the browser
                        is broken (it was fine only with mozilla 0.91).

                        Reloading an SVG lets the browser crash.

                        thanks,

                        Marcel

                        --
                        Marcel Ruff
                        mailto:ruff@...
                        http://www.lake.de/home/lake/swand/
                        http://www.xmlBlaster.org
                      • Michael Bierman
                        Actually, I think it would be useful to give a little detail on what this change is since so many people here are adament about ASV support in Mozilla.
                        Message 11 of 24 , Apr 2, 2002
                          Actually, I think it would be useful to give a little detail on what this
                          change is since so many people here are adament about ASV support in
                          Mozilla.

                          Michael

                          > -----Original Message-----
                          > From: skeetersrow [mailto:skeetersrow@...]
                          > Sent: Tuesday, April 02, 2002 6:44 AM
                          > To: svg-developers@yahoogroups.com
                          > Subject: [svg-developers] Re: SVG Broken in Mozilla :-(
                          >
                          >
                          > Hi Andrew
                          > Nothing contradictory here. As of a week or so ago there was a change
                          > in coding and I won't go into that here. End effect is old browsers
                          > such as the last release and an older yet NS 6.2 are not affected,
                          > they still use the old API. The up coming Mozilla 1 and any newer NS
                          > are affected by this change.
                          >
                          >
                          >
                          > > > Mozilla release 0.9.9 and NS 6X (being old
                          > > > browsers) are not affected.
                          >
                          > > Steve,
                          > >
                          > > Your statements seem contradictory. You state that it is broken in
                          > "all"
                          > > Mozilla browsers then say releast 0.9.9 isn't affected.
                          > >
                          > > The problem page at the URL you give loaded fine for me in Mozilla
                          > 0.9.9. I
                          > > noticed that it may be being served as image/svg-xml. It should be
                          > > image/svg+xml.
                          > >
                          > > I don't know correcting that will help to solve any of the other
                          > problems.
                          > >
                          > > Andrew Watt



                          _________________________________________________________
                          Do You Yahoo!?
                          Get your free @... address at http://mail.yahoo.com
                        • skeetersrow
                          Hello Here is the bug # and url http://bugzilla.mozilla.org/show_bug.cgi?id=133567 It has been assigned to aruner@netscape.com (Arun K. Ranganathan) So before
                          Message 12 of 24 , Apr 2, 2002
                            Hello

                            Here is the bug # and url
                            http://bugzilla.mozilla.org/show_bug.cgi?id=133567

                            It has been assigned to aruner@... (Arun K. Ranganathan)

                            So before I spread any more misunderstanding around, the above would
                            be the person to contact.

                            --- In svg-developers@y..., "Michael Bierman" <mbier@y...> wrote:
                            > Actually, I think it would be useful to give a little detail on what
                            this
                            > change is since so many people here are adament about ASV support in
                            > Mozilla.
                            >
                            > Michael
                            >
                            > > -----Original Message-----
                            > > From: skeetersrow [mailto:skeetersrow@y...]
                            > > Sent: Tuesday, April 02, 2002 6:44 AM
                            > > To: svg-developers@y...
                            > > Subject: [svg-developers] Re: SVG Broken in Mozilla :-(
                            > >
                            > >
                            > > Hi Andrew
                            > > Nothing contradictory here. As of a week or so ago there was a change
                            > > in coding and I won't go into that here. End effect is old browsers
                            > > such as the last release and an older yet NS 6.2 are not affected,
                            > > they still use the old API. The up coming Mozilla 1 and any newer NS
                            > > are affected by this change.
                            > >
                            > >
                            > >
                            > > > > Mozilla release 0.9.9 and NS 6X (being old
                            > > > > browsers) are not affected.
                            > >
                            > > > Steve,
                            > > >
                            > > > Your statements seem contradictory. You state that it is broken in
                            > > "all"
                            > > > Mozilla browsers then say releast 0.9.9 isn't affected.
                            > > >
                            > > > The problem page at the URL you give loaded fine for me in Mozilla
                            > > 0.9.9. I
                            > > > noticed that it may be being served as image/svg-xml. It should be
                            > > > image/svg+xml.
                            > > >
                            > > > I don't know correcting that will help to solve any of the other
                            > > problems.
                            > > >
                            > > > Andrew Watt
                            >
                            >
                            >
                            > _________________________________________________________
                            > Do You Yahoo!?
                            > Get your free @... address at http://mail.yahoo.com
                          • Tobias Reif
                            Chris, ... [...] ... Perhaps you want to CC these questions to news://news.mozilla.org/netscape.public.mozilla.svg and alex@croczilla.com , so the SVG In
                            Message 13 of 24 , Apr 3, 2002
                              Chris,


                              > I find this hard to believe. I always run my displays at 24 bit and
                              > have been using Mozilla SVG from the beginning, and never had any
                              > false colors. Furthermore, the internal model of libart is 24 bit:

                              [...]
                              > Are you sure? Is this just with particular SVG features? Do you have a
                              > test file that demonstrates this?


                              Perhaps you want to CC these questions to
                              news://news.mozilla.org/netscape.public.mozilla.svg
                              and
                              alex@...
                              , so the SVG In Mozilla developers can address these issues.

                              Also check the thread "SVG: No green?" 22.03.2002 which might be
                              related; but I'm not sure.

                              Tobi

                              --
                              Neither simple nor complex matters are to be complicated.
                            • Tobias Reif
                              ... Ouch; should have read the rest of the thread first :) Tobi -- Neither simple nor complex matters are to be complicated.
                              Message 14 of 24 , Apr 3, 2002
                                I wrote:

                                > Chris,
                                > Perhaps you want to CC these questions to
                                > news://news.mozilla.org/netscape.public.mozilla.svg
                                > and
                                > alex@...
                                > , so the SVG In Mozilla developers can address these issues.


                                Ouch; should have read the rest of the thread first :)


                                Tobi


                                --
                                Neither simple nor complex matters are to be complicated.
                              • Tobias Reif
                                Michael, ... I heard that the API wasn t frozen when Adobe coded to it. It would be great if a) Mozilla 1.0 will have well-defined, frozen, and well-documented
                                Message 15 of 24 , Apr 3, 2002
                                  Michael,

                                  > Gee, sounds pretty poor on Mozila's part to cause such a major break for a
                                  > plugin this late in their cycle. It is even worse that they place
                                  > responsibility for it on Adobe.


                                  I heard that the API wasn't frozen when Adobe coded to it.

                                  It would be great if
                                  a) Mozilla 1.0 will have well-defined, frozen, and well-documented APIs,
                                  and if
                                  b) The then current version of ASV would work well with Mozilla 1.0(and
                                  the corresponding NN), on Windows, Mac, and Linux, including
                                  interactivity, stability, etc,; just like with Windows IE.
                                  (I avoid IE, since it's insecure. NN6 is my main browser.)

                                  The two parties shouldn't blame it on each other, but work together.

                                  Tobi

                                  --
                                  Neither simple nor complex matters are to be complicated.
                                • sorotokin
                                  ... break for a ... It is true, and that was the main reason why we never claimed official support for Mozilla (and why our installer does not install it for
                                  Message 16 of 24 , Apr 4, 2002
                                    --- In svg-developers@y..., Tobias Reif <tobiasreif@p...> wrote:
                                    > Michael,
                                    >
                                    > > Gee, sounds pretty poor on Mozila's part to cause such a major
                                    break for a
                                    > > plugin this late in their cycle. It is even worse that they place
                                    > > responsibility for it on Adobe.
                                    >
                                    >
                                    > I heard that the API wasn't frozen when Adobe coded to it.

                                    It is true, and that was the main reason why we never claimed
                                    official support for Mozilla (and why our installer does not
                                    install it for NN6.x). No scripting APIs were frozen back then.
                                    Support for Mozilla was always just experimental work.

                                    >
                                    > It would be great if
                                    > a) Mozilla 1.0 will have well-defined, frozen, and well-documented
                                    APIs,

                                    It won't. Mozilla plug-in architecture (to which the
                                    plug-in was coded) was deprecated since 0.9.1 and their
                                    extension for NN4.x plug-ins don't even cover everything
                                    that LiveConnect used to be able to do. So current
                                    official way to write Mozilla plug-ins is too limited
                                    even to expose just what is working in NN4.x.

                                    > and if
                                    > b) The then current version of ASV would work well with Mozilla 1.0
                                    (and
                                    > the corresponding NN), on Windows, Mac, and Linux, including
                                    > interactivity, stability, etc,; just like with Windows IE.

                                    Without plug-in architecture it is problematic...

                                    > (I avoid IE, since it's insecure. NN6 is my main browser.)
                                    >
                                    > The two parties shouldn't blame it on each other, but work together.
                                    >

                                    I agree with that.

                                    Peter

                                    > Tobi
                                    >
                                    > --
                                    > Neither simple nor complex matters are to be complicated.
                                  • Tobias Reif
                                    Peter, ... interactivity, ... Many want/need ASV to work in Mozilla/NN. Perhaps you could find someone at Mozilla who s able to make high level decisions, and
                                    Message 17 of 24 , Apr 4, 2002
                                      Peter,

                                      >> and if b) The then current version of ASV would work well with
                                      >> Mozilla 1.0 (and
                                      >> the corresponding NN), on Windows, Mac, and Linux, including
                                      interactivity,
                                      >> stability, etc,; just like with Windows IE.
                                      >>
                                      > Without plug-in architecture it is problematic...

                                      Many want/need ASV to work in Mozilla/NN. Perhaps you could find someone
                                      at Mozilla who's able to make high level decisions, and describe to
                                      him/her what's required on Mozilla's side, so that creators of plugins
                                      can do their job.

                                      >> The two parties shouldn't blame it on each other, but work
                                      >> together.
                                      >>
                                      > I agree with that.

                                      Great! I hope you'll find a way to work with them, so that Adobe will be
                                      able to fully support Moz&NN6+, on Win%Lin.

                                      Tobi

                                      --
                                      Neither simple nor complex matters are to be complicated.
                                    • Chris Lilley
                                      ... s Hey, while you both are there, is this then okay? Its okay in that it is valid (or would be if there was a DOCTYPE). And I se that Mozilla does indeed
                                      Message 18 of 24 , Apr 6, 2002
                                        On Tuesday, April 2, 2002, 10:46:32 PM, skeetersrow wrote:



                                        >> A quick fix would be to implement a 'switch' that renders the first
                                        >> child, or all its children.
                                        >>
                                        >> Since Mozilla implements CSS2 selectors and a little CSS3, it might be
                                        >> possible to do this using CSS - make the first child of a switch
                                        >> element have display:block and the rest, display:none.
                                        >>
                                        >> This would at least allow valid examples to be used. Implementing the
                                        >> test evaluation would be a second step.

                                        s> Hey, while you both are there, is this then okay?

                                        Its okay in that it is valid (or would be if there was a DOCTYPE). And
                                        I se that Mozilla does indeed implement the first part of my suggested
                                        two-part strategy.

                                        s> It is saved as an
                                        s> XML doc and seems to render nicely in Mozilla.

                                        Nicely and correctly are not the same thing.

                                        s> <?xml version="1.0" standalone="yes"?>
                                        s> <svg width="400px" height="300px" xmlns = 'http://www.w3.org/2000/svg'>
                                        s> <switch>
                                        s> <circle cx="150px" cy="100px" r="75px" style="fill:red; fill-opacity:;
                                        s> stroke:blue; stroke-width:5; stroke-opacity:;" />

                                        This will always be displayed by a compliant viewer, as there is no
                                        test on it so it always evaluates to true.

                                        s> <ellipse cx="150px" cy="100px" rx="150px" ry="25px"
                                        s> style="fill:yellow; fill-opacity:0.40; stroke:maroon; stroke-width:5;
                                        stroke-opacity:;" />

                                        This will never be displayed by a compliant viewer because the element
                                        before it evaluated to true so it is not even evaluated.

                                        s> <foreignObject x="70px" y="93px" width="100px" height="50px"

                                        s> requiredExtensions="http://example.com/SVGExtensions/EmbeddedXHTML">

                                        The test makes no difference, this will never be evaluated.

                                        s> <html xmlns="http://www.w3.org/1999/xhtml">
                                        s> <body>
                                        s> <p style="font: 14px Arial, Helvetica, sans-serif bold;
                                        s> color: maroon; background-color: silver;">Is this doing it correctly?</p>
                                        s> </body>
                                        s> </html>

                                        Since you are using HTML as an Embedding Set you should probably have
                                        just a body element. If you are going to have an html element then it
                                        needs to have head as its first child and head needs a title element
                                        as a achild.

                                        s> </foreignObject>
                                        s> </switch>
                                        s> </svg>

                                        This example would make a lot more sense if you moved the circle and
                                        the ellipse up out of the switch. The correct rendering would then be
                                        the same as the current rendering ;-)


                                        --
                                        Chris mailto:chris@...
                                      • skeetersrow
                                        Hi ... That I can understand. Well how about taking a look at the below. But the suggestion of removing the broke the foreign object. however I noticed
                                        Message 19 of 24 , Apr 8, 2002
                                          Hi
                                          > Nicely and correctly are not the same thing.

                                          That I can understand. Well how about taking a look at the below.
                                          But the suggestion of removing the <html> broke the foreign object.
                                          however I noticed that with it back in and taking out the <body> tag
                                          then the width works, where as before it didn't, put the <body> tag
                                          back in the the width is broken ????

                                          Also clean up the doc and put <switch> where it belongs,

                                          <?xml version="1.0" encoding="UTF-8" standalone="no" ?>
                                          <!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 20010719//EN"
                                          "http://www.w3.org/TR/2001/PR-SVG-20010719/DTD/svg10.dtd">

                                          <svg width="400px" height="300px" xmlns = 'http://www.w3.org/2000/svg'>

                                          <circle cx="150px" cy="100px" r="75px" style="fill:red; fill-opacity:;
                                          stroke:blue; stroke-width:5; stroke-opacity:;" />
                                          <ellipse cx="150px" cy="100px" rx="150px" ry="25px"
                                          style="fill:yellow; fill-opacity:0.40; stroke:maroon; stroke-width:5;
                                          stroke-opacity:;" />

                                          <switch>
                                          <foreignObject x="75px" y="80px" width="150px" height="60px"

                                          requiredExtensions="http://example.com/SVGExtensions/EmbeddedXHTML" >
                                          <html xmlns="http://www.w3.org/1999/xhtml">

                                          <p style="padding:3px 0;text-align: center; font: 13px Arial,
                                          Helvetica, sans-serif bold; color: maroon; background-color:#77ffdd;
                                          -moz-outline: 3px dotted Fuchsia;">Is this doing it correctly?
                                          Or is this silly?</p>

                                          </html>
                                          </foreignObject>
                                          </switch>

                                          </svg>


                                          >
                                          > s> <?xml version="1.0" standalone="yes"?>
                                          > s> <svg width="400px" height="300px" xmlns =
                                          'http://www.w3.org/2000/svg'>
                                          > s> <switch>
                                          > s> <circle cx="150px" cy="100px" r="75px" style="fill:red;
                                          fill-opacity:;
                                          > s> stroke:blue; stroke-width:5; stroke-opacity:;" />
                                          >
                                          > This will always be displayed by a compliant viewer, as there is no
                                          > test on it so it always evaluates to true.
                                          >
                                          > s> <ellipse cx="150px" cy="100px" rx="150px" ry="25px"
                                          > s> style="fill:yellow; fill-opacity:0.40; stroke:maroon; stroke-width:5;
                                          > stroke-opacity:;" />
                                          >
                                          > This will never be displayed by a compliant viewer because the element
                                          > before it evaluated to true so it is not even evaluated.
                                          >
                                          > s> <foreignObject x="70px" y="93px" width="100px" height="50px"
                                          >
                                          > s> requiredExtensions="http://example.com/SVGExtensions/EmbeddedXHTML">
                                          >
                                          > The test makes no difference, this will never be evaluated.
                                          >
                                          > s> <html xmlns="http://www.w3.org/1999/xhtml">
                                          > s> <body>
                                          > s> <p style="font: 14px Arial, Helvetica, sans-serif bold;
                                          > s> color: maroon; background-color: silver;">Is this doing it
                                          correctly?</p>
                                          > s> </body>
                                          > s> </html>
                                          >
                                          > Since you are using HTML as an Embedding Set you should probably have
                                          > just a body element. If you are going to have an html element then it
                                          > needs to have head as its first child and head needs a title element
                                          > as a achild.
                                          >
                                          > s> </foreignObject>
                                          > s> </switch>
                                          > s> </svg>
                                          >
                                          > This example would make a lot more sense if you moved the circle and
                                          > the ellipse up out of the switch. The correct rendering would then be
                                          > the same as the current rendering ;-)
                                          >
                                          >
                                          > --
                                          > Chris mailto:chris@w...
                                        Your message has been successfully submitted and would be delivered to recipients shortly.