Handling dashes in ISBN barcode?

[nicorikken]

Thanks for making Libre Barcode. I'm happy it exists, because it will simplify barcode generation. Thanks for that!

I'm looking into adopting this font for ISBN barcodes. The human-readable form of barcodes is separated by dashes, like 978-3-16-148410-0. The placement of dashes depends on the code itself.

From my trials it seems that Libre Barcode doesn't ignore the dashes in a code and explicitly requires the code to be without dashes.

For my automation this requires an additional processing step to remove the dashes before rendering the barcode using Libre Barcode. Certainly something that can be overcome, but it would have been easier if Libre Barcode would accept a code with dashes.

[graphicore]

It seems like that could be added as a GSUB processing step just before entering the rules for the barcode itself. I.e. stripping the dashes down to a legit input string.

Maybe the Bookland number 978 could also be used as a trigger. Looking into the Bookland article, supporting ISSN would also be nice, ist even uses the 2-digit Add-On for issue numbers. Also ISMN

I see cases where the original ISBN number is also printed directly above the barcode and wonder if that would be desirable as well, maybe as an addition behind a Stylistic-Set OpenType-Feature.

[nicorikken]

Sounds like a good approach! For books there is the convention to include an additional barcode part for the book price. I like the suggestion of picking up on https://en.wikipedia.org/wiki/Bookland number.

Yes the human-readable ISBN-code above the barcode is a pleasant convention because it reaffirms the human that the barcode is the ISBN-code and the dashes make it easier to type and for advanced readers have a hint at the ISBN-structure. Not sure if librebarcode could made responsible for putting the dashes in the correct place. I can imagine that the horizontal alignment of the ISBN text could be a bit tricky.

For my use-case for the Ada & Zangemann book translation automation I've now managed a workaround using processing to strip the dashes for the barcode, and keep the dashes for the top barcode. The top text is added using manual placement. Besides the automation required, the placement of the top text and ensuring the correct whitespace size was a bit tricky.