This repository is used by Relaton-CIE to represent known publications.
Data is stored in Relaton YAML format.
This repository contains a crawler script which runs daly. The script parses site https://www.techstreet.com/cie/searches/31156444 and store documents into the data directory.
Two search indexes are published alongside the data:
index-v1.yaml/index-v1.zip-
Legacy flat index (
:id:is the raw docidentifier string). Kept unchanged for existing consumers. index-v2.yaml/index-v2.zip-
Structured index whose
:id:rows arePubid::Cie::Identifiermaps (_type: pubid:cie:…). Consumed by the pubid-aware Relaton-CIE read side.
index-v2 is rebuilt from the local data/*.yaml files (no crawl needed) with:
bundle exec ruby gen_index.rbgen_index.rb parses each primary docidentifier to a Pubid::Cie::Identifier
using the same parse + round-trip guard as the producer, writes index-v2.yaml,
sanity-loads it (a full-index round-trip), and packages index-v2.zip. Any id
that pubid cannot parse or that will not round-trip through the index is skipped
and reported. index-v1.* is left untouched.