Cross-linked database of Comparative Concepts, extracted from "Morphosyntax: constructions of the world's languages", by William Croft (2022)

You can use this database in the following ways:

• You can explore the interactive glossary, where you can read definitions and see how different CCs relate to each other.
• You can play with our interactive visualization, which shows the relations between CCs as self-organizing graphs. You can filter these graphs to focus on just some CCs if you want.
• You can use the raw database in your own applications. It is stored in YAML, and its structure is described below.

The CC database is stored in a YAML file. Here is an example of a CC entry:

This says the the unique id is sem:less-affected-p, it has the name "less affected P" and the type "sem". It also has an alternative name, "LAP", and is a subtype of sem:p-role. The definition is written in a pseudo-html format which can be parsed into correct HTML by the parsing script. Finally there are two example sentences that are extracted from the definition.

The YAML database consists of a list of entries of the following form:

- Id: cc-id (a string)
  Name: str
  Alias: list of strings (possibly empty)
  Type: one of sem/cxn/inf/str/def
  Definition: pseudo-html string
  Examples: list of examples, either as plain strings or with information about language, gloss, translation, etc.

  SubtypeOf, ConstituentOf, HeadOf,
  AttributeOf, RoleOf,
  ExpressionOf, ModeledOn, RecruitedFrom: these are the different relations, stored as lists of cc-ids

The script ccdb_parser.py parses (and validates) the YAML database and outputs it in different formats. Currently only HTML output is supported.

usage: ccdb_parser.py [-h] [--quiet] [--format {html,karp,fnbr,graph}] [--keep-deleted] cc_database

Parse the comparative concepts database and export it in different formats.

positional arguments:
  cc_database           YAML database of comparative concepts

options:
  -h, --help            show this help message and exit
  --quiet, -q           suppress warnings
  --format FMT, -f FMT  export format (FMT = html, karp, fnbr, graph)
  --keep-deleted, -d    keep deleted terms

There's a Makefile that reads the database and creates the files docs/index.html and docs/cc-graph-data.js. They are used in the interactive glossary and the interactive visualization.