Why GenServer to hold the document?

[alappe]

Dear Andrew,

thanks for starting this package – much appreciated. While converting an old pdf over to your library, I wondered why the document is hold inside a GenServer? I read through it and even converted a local version to return the %Document{} instead of a pid(), all working fine.

I do see a benefit for holding fonts and objects though.

Regards
Andreas

[Hermanverschooten]

@alappe I know this was some time ago, but would you be willing to share your version?

[alappe]

@Hermanverschooten sure, but it will take until next week due to holiday.

[alappe]

@Hermanverschooten Sorry, I forgot it's on github anyway:

https://github.com/zimmer7/elixir-pdf/tree/document_without_process

This is the branch I'm using in one project.

[Hermanverschooten]

@alappe Thanks! Small question though, why did you change the name to PDF?

[alappe]

I can't recall, probably because it worked and that's the usual writing of the format. This complicates merging – sorry for that!

Looking into it, I wasn't really that wrong: https://hexdocs.pm/elixir/1.12/naming-conventions.html#casing

[Hermanverschooten]

Wasn't complaining, just wondering.

[alappe]

Alright :)

[DGollings]

if anyone is reading this, but wants a fork with fewer changes, check out: https://github.com/sephianl/elixir-pdf

it pretty much only applies the first commit that @alappe made in his fork, but to the most recent version of the original library (this one

Thanks @alappe and @andrewtimberlake :)

No idea what the genserver is for, but it was timing out in tests and decided this was the simplest way to fix that.

Usage is simple, change Pdf.build(opts, function) to Pdf.new(opts)