Re: Including an EPS

[Jean Abou Samra]

Le 17/12/2022 à 21:30, Mike Kilmer a écrit :

Thanks, Jean.

With upgrade, compilation still failed, but it was in fact the eps file and/or 
its format that compiler was choking on.

Here you can see the (truncated) diff btwn the (first) one that DID work (from 
my vector editor) and the one that pyqrcode generated:

diff --git a/longdarkroad.eps b/longdarkroad.eps
index 9050b61..2842f16 100644

[...]
+%%BoundingBox: 0 0 152.5 152.5
[...]

According to the PostScript language Document Structuring Conventions
Specification version 3.0,

http://www.lprng.com/RESOURCES/ADOBE/5001.DSC_Spec.pdf

Section 3.1 "Conforming Documents":

"""
Consider the following incorrect example:
%%BoundingBox 43.22 50.45 100.60 143.49

the %%BoundingBox: comment. Abbreviations for comments are not acceptable.
Second, floating point arguments are used instead of the integer arguments
this comment requires
"""

Therefore, it seems to me that your EPS file invalid. Although
LilyPond could emit a more helpful error in this rare case,
the tool you are using should be fixed to emit valid EPS.

Are you sure about the example Python code you gave? For me, it
outputs a PNG image in an ".eps" file, and even with format="eps"
passed to .save(), the EPS output looks different. Furthermore,
the %%Creator embedded comment refers to the pyqrcode library, but
your code uses the qrcode library, which is a different one.

Lastly: are you aware of https://lsr.di.unimi.it/LSR/Item?id=1153?

I find this method more convenient than including EPS images, but I let you
judge for yourself.

Best,
Jean

This comment is incorrect on two counts. First, there is a colon missing from

OpenPGP_signature
Description: OpenPGP digital signature