I created a batch file to start icepdf-viewer jar with enabled finest level logging.
Unfortunately I wasn't able to enable it for our own application (we have a quite complex logging configuration on our own which I wasn't able to extend so far), but I was able to do so directly for viewer ri of OS version 6.2.1 (which we're currently using). So I tried to enably fine level logging as you suggested.
COPY AND PASTE TEXT FROM A PDF PRO
The icepdf-pro-jars aren't included in the OS version of 6.3-snapshot (they're obviously jars of the pro version), so I downloaded ICEpdf-pro-6.2.5 trial, included it in our application and the bug is gone!!! So it seems to be a bug only of OS version. Now we have real progress!! Including the bcp-jars into the classpath doesn't make any difference. On my system the logging is: FINER: Match Found for: Arial-BoldMT:Bold Substituting arial-boldmt:C:\WINDOWS\Fonts\arialbd.ttf 8:51:47 AM .fonts.FontManager getInstance FINE: Font Substitution: Found system font: Arial-BoldMT for named font Arial-BoldMT 8:51:47 AM .fonts.FontManager findFont FINER: Match Found for: ArialMT:Plain Substituting arialmt:C:\WINDOWS\Fonts\arial.ttf 8:51:47 AM .fonts.FontManager getInstance FINE: Font Substitution: Found system font: ArialMT for named font ArialMT For the 6.3-snapshot can you make sure you have at least the following jars on your class path: Core Rendering and viewer jars icepdf-core.jar icepdf-pro.jar icepdf-pro-intl.jar icepdf-viewer.jar bcpkix-jdk15on.jar bcprov-ext-jdk15on.jar bcprov-jdk15on.jar Also can you enabled fine level logging so that we can see which font is being substituted. I'm still not able to reproduce this locally, I get quite different results as attached earlier. In order to start the viewer ri I had to alter the manifest file of the "icepdf-viewer-6.3.0-SNAPSHOT.jar" (to include the classpath with the "icepdf-core-6.3.0-SNAPSHOT.jar") but that shouldn't make a difference, shouldn't it? Filename Did I make any mistake building the viewer ri? I checked out the OS repository and ran the gradle scripts (core-awt and viewer-awt). And when I mark the text in the framed box (below "Sachverhalt"), there are still line breaks where they shouldn't be (as in the originally attached text file). "Schles wi g-Holstein" instead of "Schleswig-Holstein"). The result still has blanks which are not part of the original text (i.e.
COPY AND PASTE TEXT FROM A PDF PDF
I marked the whole header of the attached pdf document (up to the word "Sachverhalt"), copied the marked text (CTRL-C) and pasted it in the attached text file "test.txt".
I build 6.3-snapshot again today and tried again with the viewer ri. Text file with contents copies and pasted from the attached pdf document via ICEpdf 6.2.1 viewer. We're using ICEpdf version 6.2.1 but it's the same with version 6.3.0-SNAPSHOT. It's the same with the stand alone ICEpdf viewer. When I copy text inside this document via ICEpdf and paste it into a text file, there are spaces between the letters of some words (see attached file "iText-Pdf.txt"). The attached pdf document "iText-Pdf.pdf" has been created this way. We are creating pdf documents via iText by directly drawing the document text into a graphics context obtained via the iText class PdfContentByte (we are currently using iText version 2.1.7 but I also checked on iText 5.5.9).
0 kommentar(er)
