name2unicode() should follow the Adobe Glyph List Specification

[pietermarsman]

Fixes #177
Fixes #261
Fixes #204

[pietermarsman]

Just an update of what I've been doing:

• I've been testing name2unicode() with the samples/nonfree/f1040nr.pdf.
• For this pdf it is called from EncodingDb.get_encoding() and from Type1FontHeaderParser.get_encoding(). This latter causes a problem, because it has a font that maps cid's to glyph-names in the form of \H[0-9]+.
• This form is not a valid glyph-name according to Adobe. The current name2unicode() extracts the number part and returns the corresponding unicode character. This is wrong (for this document), since it returns chinese characters for an american document.
• I am trying to figure out how to the \H[0-9]+ should be mapped to valid characters.
• My best guess now is that there is some information in the encrypted part of the embedded font program (see PDF Reference 1.6, page 437)

[tataganesh]

Some small changes.

[tataganesh]

Also is this an extension of #204?

[pietermarsman]

Thanks for the feedback. No time left today to fix it, so I will look into this tomorrow.

[pietermarsman]

Also is this an extension of #204?

Yes, this is also implements the behaviour described by @jbarlow83, and more.

[robinalexandre]

Do you know when this PR and all the others about name2unicode and Cmap will be merged ?

I think we are a lot waiting for it and it seems like checks have passed 👍

Thanks to all