Ayase

Revision as of 05:05, 22 August 2019 by Antonizoon (talk | contribs) (Created page with "= Ayase Imageboard Archival Standard (Ayase) = The Ayase Imageboard Archival Standard was produced by the Bibliotheca Anonoma to handle the ever growing operations of Desuarc...")
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Ayase Imageboard Archival Standard (Ayase)

The Ayase Imageboard Archival Standard was produced by the Bibliotheca Anonoma to handle the ever growing operations of Desuarchive and RebeccaBlackTech, by completely discarding FoolFuuka/Asagi and starting over with new paradigms and modern languages and technologies. It is made upof

Reference Implementation

  • Operating System: CentOS/RHEL 8
  • Database: PostgreSQL
  • Scraper: Ena or Hydrus (.NET C#)
  • Middleware: Ayase (Python PyPy)
  • Frontends: 4chan X, Clover, iphone app

Specifications

Files

  • All files are to be named by sha256sum and file extension. This was chosen for the broad availability of hardware extensions in Intel/AMD/ARM for the purpose and its use by 8chan/vichan.
    • This is to mitigate the ongoing issue of md5sum collisions becoming ever easier for users to do. However, migration t
    • The trustworthiness of the sha256sum is strong up to the dawn of quantum computing it seems. It is chosen over sha256sums
    • While there is a question of whether the ones generating the first sha256sum can be trusted, md5sums also will have this same issue as md5sum collisions can be generated and injected into the archive anytime. At least the fact that the file has the same sha256sum and md5sum at archival time should at least vouch for the fact that the archiver did not change it after a file was scraped.
    • Even without hardware acceleration chips, sha256sums will still be quicker to generate than files can be read from disk, even SSDs. Even if a mass conversion of filenames has to be done, it would only take 4 days to deal with 30TB, though SSD caching might be necessary.
  • They are to be stored in double nested folders.

Time

  • Ayase requires time to be stored in PostgreSQL datetimes, which also store timezones.
  • Only UTC should be used as the timezone for newly scraped data. The timezone support is not an excuse to store in other timezones.
  • The timezone support is only meant for compatibility purposes with prior Asagi data, given that they store time as US time (maybe Eastern) due to their past HTML scraping. Future scrapes are strongly advised not to replicate this behavior, local time should be up to the frontend to determine.

PostgreSQL JSONB Schema

if we GET json from the 4chan API, and always serve the same json to the user, why deconstruct and reconstruct into post focused sql records every time?

JSONB is different from text blobs of JSON too, its more like a NoSQL database and can be indexed. PostgreSQL is a better NoSQL than MongoDB as The Guardian has found.

Elasticsearch Engine

A seperate elastic search engine kept in sync with, but independent from the sql server, will replace Sphinxsearch which queries the mysql db