path: root/mediagoblin/tools/metadata.py

# GNU MediaGoblin -- federated, autonomous media hosting
# Copyright (C) 2011, 2012 MediaGoblin contributors.  See AUTHORS.
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU Affero General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU Affero General Public License for more details.
#
# You should have received a copy of the GNU Affero General Public License
# along with this program.  If not, see <http://www.gnu.org/licenses/>.


import os
import copy
import json
import re
from pkg_resources import resource_filename

import dateutil.parser
from pyld import jsonld
from jsonschema import validate, FormatChecker, draft4_format_checker

from mediagoblin.tools.pluginapi import hook_handle



########################################################
## Set up the MediaGoblin format checker for json-schema
########################################################

URL_REGEX = re.compile(
    r'^[a-z]+://([^/:]+|([0-9]{1,3}\.){3}[0-9]{1,3})(:[0-9]+)?(\/.*)?$',
    re.IGNORECASE)

def is_uri(instance):
    """
    jsonschema uri validator
    """
    if not isinstance(instance, str):
        return True

    return URL_REGEX.match(instance)

def is_datetime(instance):
    """
    Is a date or datetime readable string.
    """
    if not isinstance(instance, str):
        return True

    return dateutil.parser.parse(instance)


class DefaultChecker(FormatChecker):
    """
    Default MediaGoblin format checker... extended to include a few extra things
    """
    checkers = copy.deepcopy(draft4_format_checker.checkers)


DefaultChecker.checkers["uri"] = (is_uri, ())
DefaultChecker.checkers["date-time"] = (is_datetime, (ValueError, TypeError))
DEFAULT_CHECKER = DefaultChecker()

# Crappy default schema, checks for things we deem important

DEFAULT_SCHEMA = {
    "$schema": "http://json-schema.org/schema#",

    "type": "object",
    "properties": {
        "license": {
            "format": "uri",
            "type": "string",
        },
        "dcterms:created": {
            "format": "date-time",
            "type": "string",
        },
        "dc:created": {
            "format": "date-time",
            "type": "string",
        }
    },
}


def load_resource(package, resource_path):
    """
    Load a resource, return it as a string.

    Args:
    - package: package or module name.  Eg "mediagoblin.media_types.audio"
    - resource_path: path to get to this resource, a list of
      directories and finally a filename.  Will be joined with
      os.path.sep.
    """
    filename = resource_filename(package, os.path.sep.join(resource_path))
    return open(filename, encoding="utf-8").read()

def load_resource_json(package, resource_path):
    """
    Load a resource json file, return a dictionary.

    Args:
    - package: package or module name.  Eg "mediagoblin.media_types.audio"
    - resource_path: path to get to this resource, a list of
      directories and finally a filename.  Will be joined with
      os.path.sep.
    """
    return json.loads(load_resource(package, resource_path))


##################################
## Load the MediaGoblin core files
##################################


BUILTIN_CONTEXTS = {
    "http://www.w3.org/2013/json-ld-context/rdfa11": load_resource(
        "mediagoblin", ["static", "metadata", "rdfa11.jsonld"])}


_CONTEXT_CACHE = {}

def load_context(url):
    """
    A self-aware document loader.  For those contexts MediaGoblin
    stores internally, load them from disk.
    """
    if url in _CONTEXT_CACHE:
        return _CONTEXT_CACHE[url]

    # See if it's one of our basic ones
    document = BUILTIN_CONTEXTS.get(url, None)

    # No?  See if we have an internal schema for this
    if document is None:
        document = hook_handle(("context_url_data", url))

    # Okay, if we've gotten a document by now... let's package it up
    if document is not None:
        document = {'contextUrl': None,
                    'documentUrl': url,
                    'document': document}

    # Otherwise, use jsonld.load_document
    else:
        document = jsonld.load_document(url)

    # cache
    _CONTEXT_CACHE[url] = document
    return document


DEFAULT_CONTEXT = "http://www.w3.org/2013/json-ld-context/rdfa11"

def compact_json(metadata, context=DEFAULT_CONTEXT):
    """
    Compact json with supplied context.

    Note: Free floating" nodes are removed (eg a key just named
    "bazzzzzz" which isn't specified in the context... something like
    bazzzzzz:blerp will stay though.  This is jsonld.compact behavior.
    """
    compacted = jsonld.compact(
        metadata, context,
        options={
            "documentLoader": load_context,
            # This allows for things like "license" and etc to be preserved
            "expandContext": context,
            "keepFreeFloatingNodes": False})

    return compacted


def compact_and_validate(metadata, context=DEFAULT_CONTEXT,
                         schema=DEFAULT_SCHEMA):
    """
    compact json with supplied context, check against schema for errors

    raises an exception (jsonschema.exceptions.ValidationError) if
    there's an error.

    Note: Free floating" nodes are removed (eg a key just named
    "bazzzzzz" which isn't specified in the context... something like
    bazzzzzz:blerp will stay though.  This is jsonld.compact behavior.

    You may wish to do this validation yourself... this is just for convenience.
    """
    compacted = compact_json(metadata, context)
    validate(metadata, schema, format_checker=DEFAULT_CHECKER)

    return compacted


def expand_json(metadata, context=DEFAULT_CONTEXT):
    """
    Expand json, but be sure to use our documentLoader.

    By default this expands with DEFAULT_CONTEXT, but if you do not need this,
    you can safely set this to None.

    # @@: Is the above a good idea?  Maybe it should be set to None by
    #   default.
    """
    options = {
        "documentLoader": load_context}
    if context is not None:
        options["expandContext"] = context
    return jsonld.expand(metadata, options=options)


def rdfa_to_readable(rdfa_predicate):
    readable = rdfa_predicate.split(":")[1].capitalize()
    return readable