https://bugs.freedesktop.org/show_bug.cgi?id=44135

Michael Meeks <[email protected]> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|REOPENED                    |NEEDINFO

--- Comment #32 from Michael Meeks <[email protected]> ---
> > So INVALID
>
> Luckily reopened. I agree with other reporters here that this
> level of arrogance is a bit disturbing.

Ho hum - the complaints:useful-work ratio is a bit disturbing here.

It seems the problem is that modern & capable PDF renderers can render the
generated PDF acceptably; older and/or more broken ones have problems - AFAICs.

This may well not be a bug in LibreOffice - the question is: can we do much
about it - can we detect and generate PDF differently ? What is the difference
in the PDF.

Anyone wanting to complain would do better to dig through the PDF spec. and
find a good way to pretty-print that PDF, and then work out which commands are
different in it.

I had a go using a recent xpdf-poplar; interestingly the raw X backend of this
fails to render the PDF, but the cairo / SVG output renders it fine: as does
the evince viewer.

Using the pdf2svg utility - on before + after and diffing them, it seems like
we have:

 <clipPath id="clip1">
-  <path d="M 0 0 L 793.800781 0 L 793.800781 595 L 0 595 Z "/>
-</clipPath>
-<clipPath id="clip2">
-  <path d="M 230.398438 208.699219 L 540 208.699219 L 540 403.300781 L
230.398438 403.300781 Z "/>
-</clipPath>
-<clipPath id="clip3">
-  <path d="M 129.601562 72 L 266.398438 72 L 266.398438 151.199219 L
129.601562 151.199219 Z "/>
+  <path d="M 0 0 L 793.601562 0 L 793.601562 595 L 0 595 Z "/>
 </clipPath>
+<filter id="alpha" filterUnits="objectBoundingBox" x="0%" y="0%" width="100%"
height="100%">
+  <feColorMatrix type="matrix" in="SourceGraphic" values="0 0 0 0 1 0 0 0 0 1
0 0 0 0 1 0 0 0 1 0"/>
+</filter>
+<mask id="mask0">
+  <g filter="url(#alpha)">
+<rect x="0" y="0" width="794" height="595"
style="fill:rgb(0%,0%,0%);fill-opacity:0.7;stroke:none;"/>
+  </g>
+</mask>
+<image id="image6" width="311" height="196"
xlink:href="data:image/png;base64,... <snip huge base64 image >
/>
+<mask id="mask1">
+  <g filter="url(#alpha)">
+<rect x="0" y="0" width="794" height="595"
style="fill:rgb(0%,0%,0%);fill-opacity:0.5;stroke:none;"/>
+  </g>
+</mask>
+<image id="image10" width="138" height="81"
xlink:href="data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAIoAAABRCAYAAAANfj6IAAAABmJLR0QA/wD/AP+gvaeTAAAA+0lEQVR4nO3SsQ3CUBTAwBdETcWMZCYWpIINkMIK7n6Q7iZw4RkIttcxj5l5rg7h1PbL6gL+g1FIjEJiFBKjkBiFxCgkRiExColRSIxCYhQSo5AYhcQoJEYhMQqJUUiMQmIUEqOQGIXEKCRGITEKiVFIjEJiFBKjkBiFxCgkRiExColRSIxCYhQSo5AYhcQoJEYhMQqJUUiMQmIUEqOQGIXEKCRGITEKiVFIjEJiFBKjkBiFxCgkRiExColRSIxCYhQSo5AYhcQoJEYhMQqJUUiMQmIUEqOQGIXEKCRGITEKiVFIjEJiFBKjkBiFZPscc/vO3FeHcF7XmfcPhSYH5NAXBk0AAAAASUVORK5CYII="/>
...
-<g style="fill:rgb(0%,0%,0%);fill-opacity:1;">
+<g style="fill:rgb(12.155151%,10.978699%,10.586548%);fill-opacity:1;">
   <use xlink:href="#glyph0-1" x="39.7" y="554.8"/>
 </g>
-<g style="fill:rgb(0%,0%,0%);fill-opacity:1;">
+<g style="fill:rgb(12.155151%,10.978699%,10.586548%);fill-opacity:1;">
   <use xlink:href="#glyph0-1" x="395.5" y="554.8"/>
 </g>
 <path
style="fill-rule:evenodd;fill:rgb(81.175232%,90.586853%,89.802551%);fill-opacity:1;stroke-width:0.12;stroke-linecap:butt;stroke-linejoin:round;stroke:rgb(50.195312%,50.195312%,50.195312%);stroke-opacity:1;stroke-miterlimit:10;"
d="M 367.199219 292.601562 L 165.601562 292.601562 L 165.601562 559 L
568.800781 559 L 568.800781 292.601562 Z " transform="matrix(1,0,0,-1,0,595)"/>
-<g style="fill:rgb(0%,0%,0%);fill-opacity:1;">
-  <use xlink:href="#glyph1-1" x="328.1" y="175.5"/>
-  <use xlink:href="#glyph1-2" x="338.108" y="175.5"/>
-  <use xlink:href="#glyph1-3" x="342.104" y="175.5"/>
-  <use xlink:href="#glyph1-4" x="352.112" y="175.5"/>
-</g>
-<g style="fill:rgb(0%,0%,0%);fill-opacity:1;">
-  <use xlink:href="#glyph1-5" x="362.192" y="175.5"/>
-</g>
-<g style="fill:rgb(0%,0%,0%);fill-opacity:1;">
-  <use xlink:href="#glyph1-6" x="367.268" y="175.5"/>
-  <use xlink:href="#glyph1-7" x="372.254" y="175.5"/>
-  <use xlink:href="#glyph1-8" x="382.262" y="175.5"/>
+<g style="fill:rgb(12.155151%,10.978699%,10.586548%);fill-opacity:1;">
+  <use xlink:href="#glyph1-1" x="328" y="175.5"/>
+  <use xlink:href="#glyph1-2" x="338.008" y="175.5"/>
+</g>
+<g style="fill:rgb(12.155151%,10.978699%,10.586548%);fill-opacity:1;">
+  <use xlink:href="#glyph1-3" x="342.094" y="175.5"/>
 </g>
etc.

which (I hope) maps reasonably well to the actual PDF generated - certainly the
later PDF is larger which presumably reflects this alpha-mask / image.

So - anyhow - this is a problem that an intelligent user can solve with only a
few hours of research I expect: and here is the question:

What is the difference in the PDF - and what are the human-readable PDF
elements that changed between the versions ?

-- 
You are receiving this mail because:
You are the assignee for the bug.
_______________________________________________
Libreoffice-bugs mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/libreoffice-bugs

Reply via email to