Quote:
Originally Posted by John Wagenseil
It is interesting that Edge, Inkscape and Sumatra have same problems opening the Canon PDF, I wonder if they share some code?
|
Well, most applications rely on libraries to do the serious stuff, and by their own nature libraries are designed to be reused in several applications or even operating systems. For instance, Edge's HTML rendering engine and JavaScript interpreter are two of Microsoft's libraries, which will soon be replaced by their counterparts in Chromium/Google Chrome.
Assuming the wyvern PDF is 100% kosher (yes, the software which
created the file could have introduced a nonconformity which tools like Evince, Okular and Foxit, by accident or design, "forgive"; this happens all the time, e.g., with loose standards like HTML <5), tools like Inkscape have a bug
- when reading the PDF into their own memory representation of the image
- and/or when rendering that representation into screen
Both tasks are likely to performed by libraries, which might be shared by the other buggy tools. For instance, since Evince, Okular and Inkscape all use the poppler library for rendering PDF, and in my system only one version of poppler is installed, it's not likely to be culprit (poppler uses Cairo for raw rendering, but again I have only one version of Cairo here).