Synchronization of generated preview versions of dictiōnaria and cōdex files with some CDN

[fititnt]

• Currently using:
  • https://github.com/EticaAI/n-data (for non-binary data, e.g. we use .adoc, but not generated .pdf, .epub, etc)
  • https://eticaai.github.io/n-data/

---

We're using GitHub pages to free hosting of the dictionaries, however the binary formats (such as PDF and epub) would take too much space, even for git-tracked repository which is not really intended to have full history. However, even the .adoc formats already are getting bigger to preview. It still wort have then, but we're already in a moment were we should already have some CDN for the internal files while we still waiting to be published on humanitarian channels.

Github limitations to preview

Non-binary (but large files) already not render

One of our text formats, the Cōdex [1603:63:101]: //Dictiōnāria basibus dē rēbus vītālibus necessāriīs//, already does not render on GitHub web interface

---

Example of how it render

For sake of comparison, this is the preview of Cōdex [1603:45:31]: Dictiōnāria de calamitātibus

[fititnt]

The end result actually is better than expected. We're using S3 storage we also use for other files on @EticaAI to store a copy of everything. Wasabi (while not designed to store static sites like AWS do) allow to transparently be used as CDN behind cloudflare.

The explanation of how was done is on the files, but in short we're using https://lsf1603.etica.ai/ as CDN (not as site) similar to what we're doing on https://eticaai.github.io/n-data/. however since the bumber also is part of the prefix, the paths start without 1603.

So "1603_63_101", for, example, can be accessed at https://lsf1603.etica.ai/63/101/1603_63_101.mul-Latn.codex.pdf and https://lsf1603.etica.ai/63/101/1603_63_101.mul-Latn.codex.epub. All other URls are intuitive.

Closing for now.