#!/usr/bin/env python3

import requests
import json
import sys
import base64

url = "http://localhost:8088/api/v1/flow/0000/service/text-load"

############################################################################

text = open("docs/README.cats", "rb").read()

# Some random identifiers.  The doc ID is important, as extracted knowledge
# is linked back to this identifier
org_id = "https://trustgraph.ai/org/3c35111a-f8ce-54b2-4dd6-c673f8bf0d09"
doc_id = "https://trustgraph.ai/doc/4faa45c1-f91a-a96a-d44f-2e57b9813db8"
pub_id = "https://trustgraph.ai/pubev/a847d950-a281-4099-aaab-c5e35333ff61"

# Organization metadata
org_facts = [
    [org_id, "http://www.w3.org/1999/02/22-rdf-syntax-ns#type",
     "https://schema.org/Organization"],
    [org_id, "http://www.w3.org/2000/01/rdf-schema#label", "trustgraph.ai"],
    [org_id, "https://schema.org/name", "trustgraph.ai"]
]

# Publication metadata.  Note how it links to the Organization
pub_facts = [
    [pub_id, "http://www.w3.org/1999/02/22-rdf-syntax-ns#type",
     "https://schema.org/PublicationEvent"],
    [pub_id, "https://schema.org/description", "Uploading to Github"],
    [pub_id, "https://schema.org/endDate", "2024-10-23"],
    [pub_id, "https://schema.org/publishedBy", org_id],
    [pub_id, "https://schema.org/startDate", "2024-10-23"]
]

# Document metadata.  Note how it links to the publication event
doc_facts = [
    [doc_id, "http://www.w3.org/1999/02/22-rdf-syntax-ns#type",
     "https://schema.org/DigitalDocument"],
    [doc_id, "http://www.w3.org/2000/01/rdf-schema#label", "Mark's cats"],
    [doc_id, "https://schema.org/copyrightHolder", "trustgraph.ai"],
    [doc_id, "https://schema.org/copyrightNotice", "Public domain"],
    [doc_id, "https://schema.org/copyrightYear", "2024"],
    [doc_id, "https://schema.org/description",
     "This document describes Mark's cats"],
    [doc_id, "https://schema.org/keywords", "animals"],
    [doc_id, "https://schema.org/keywords", "cats"],
    [doc_id, "https://schema.org/keywords", "home-life"],
    [doc_id, "https://schema.org/name", "Mark's cats"],
    [doc_id, "https://schema.org/publication", pub_id],
    [doc_id, "https://schema.org/url", "https://example.com"]
]

def to_value(x):
    if x.startswith("https://"):
        return { "v": x, "e": True }
    if x.startswith("http://"):
        return { "v": x, "e": True }
    return { "v": x, "e": False }

# Convert the above metadata into the right form
metadata = [
    { "s": to_value(t[0]), "p": to_value(t[1]), "o": to_value(t[2]) }
    for t in org_facts + pub_facts + doc_facts
]

input = {
    
    # Document identifer.  Knowledge derived by TrustGraph is linked to this
    # identifier, so the additional metadata specified above is linked to the
    # derived knowledge and users of the knowledge graph could see
    # information about the source of knowledge
    "id": doc_id,
    
    # Additional metadata in the form of RDF triples
    "metadata": metadata,
    
    # Text character set.  Default is UTF-8
    "charset": "utf-8",
    
    # The PDF document, is presented as a base64 encoded document.
    "text": base64.b64encode(text).decode("utf-8")
    
}

resp = requests.post(url, json=input)

resp = resp.json()

if "error" in resp:
    print(f"Error: {resp['error']}")
    sys.exit(1)

print(resp)