kausalflow/lightkone / components/pdf-filing/versions/1/symbolon.toml · raw

[contract]
id = "lightkone/pdf-filing"
version = 1
title = "Name and place a paper's PDF the way Paperpile does"
description = """
Decides where a new PDF for a paper is stored. Lightkone files PDFs exactly
as Paperpile files them in Google Drive, so a Paperpile "All Papers" folder
can be copied into a library unchanged:

  pdf  = "papers/" + file
  file = "All Papers/<FOLDER>/<NAME>.pdf"      e.g.
         All Papers/HINTON/Hinton et al. 2012 - Improving neural networks.pdf

clean(s), in order: "---" -> "—" (em dash), "--" -> "–" (en dash); each of
: / \\ ? * " < > | -> " - "; whitespace runs -> one space; runs of " - "
-> one " - "; then remove leading and trailing spaces and hyphens.

surname(a): the part of a before the first comma (all of a if none), trimmed.

FOLDER: clean(surname(first author)) upper-cased; "Other" when there is no
first author or that is empty.

NAME:
  title = clean(title), or "Untitled" if that is empty.
  who   = from the authors' surnames, ignoring empty ones:
          none -> title; one -> "A"; two -> "A and B"; more -> "A et al.".
          If who is longer than 40, keep its first 37 and add "...".
  head  = clean(who + " " + year), or clean(who) when there is no year.
  NAME  = head + " - " + title. If NAME is longer than 146, it becomes its
          first 72 with trailing whitespace removed, then " ... ", then its
          last 69 with leading whitespace removed.
Lengths count UTF-16 code units; v1 defines behaviour only for text without
characters outside the Basic Multilingual Plane (no emoji).

Upper-casing follows Unicode full case mapping (ß -> SS).

Taken names: every OTHER paper in the library (other id) takes its pdf, and
"papers/" + each part of its file split on ";" (trimmed). If papers/<file> is
taken, try NAME + " (2).pdf", " (3).pdf", ... until one is free.
"""
regenerability = "disposable"
license = "Proprietary (kausalflow)"
authors = ["kausalflow/lightkone"]

[entrypoint]
function = "run"

[requires]
network = []
credentials = []
state = []
permissions = []

[spec]
seed = 0
max_examples = 200
timeout_seconds = 300