Reputation: 2150
I generated a pdf report with ASP.NET + iTextSharp.
I used several types of fonts in it, each font applied to a word or 2 for art reasons.
So the file is large.
How can I only embed the fonts I actually used? Just like what we do with MS Office Options.
MS Office Word 2007 is like this:
"Embed fonts in the file:
Embed only the characters used in the document(best for reducing file size)
Do not embed common system fonts"
OR I can also accept another kind of solution.
Flatten the whole page to a high-resolution picture.
If the programming is convenient, I actually prefer this solution.
Thanks.
Upvotes: 2
Views: 4464
Reputation: 15868
When creating your BaseFont
instance with embedding enabled, you need to call myBaseFont.setSubset(true)
. Note that with the encoding "Identity-H" (AKA BaseFont.IDENTITY_H
), this happens automatically:
// find all fonts in the usual places across multiple OSs.
// This can be pretty slow if you have a large number fonts, or the fonts
// themselves are Really Big (ArialUnicodeMS: 23mb).
FontFactory.registerDirectories();
// here's one way to do it, using identity-h forces subsetting
Font myFontSubset1 = FontFactory.getFont(fontName1, BaseFont.IDENTITY_H);
// here's another, explicitly enable subsetting for the underlying BaseFont.
Font myFontSubset2 = FontFactory.getFont(fontName2, FontFactory.defaultEncoding, true);
myFontSubset2.getBaseFont().setSubset(true);
//or you can create the BaseFont yourself, with automagic subsetting
BaseFont myFontSubset3 = BaseFont.createFont(fontPath, BaseFont.IDENTITY_H);
// or create it with some other encoding, and enable subsetting.
BaseFont myFontSubset4 = BaseFont.createFont(fontPath, BaseFont.WINANSI, true);
myFontSubset4.setSubset(true);
Note that this is all Java. In C# the first letter of function names are capitalized and setX(newX)
and getX()
become properties.
Upvotes: 2