Fixes #1338 .
In principle it tries assign unique PS font names to Type 1 fonts, and to err on the safe side. Thus we risk to embed a font twice under different names, if it does not have the same pdf ID.
However, and this is the issue with Type 1 fonts, we rely on heuristics to replace the font name. However, these are the same heuristics FoFiType1::parse uses in the first place to determine the embedded font name, which gives me some confidence that this has worked out quite well in the past. On the other hand, if sb. has some good files to test, please test.
Todo:
-
make clang-format happy, although some things look ugly -
replace std::regex::multiline