commonmeta-py 0.23__py3-none-any.whl → 0.25__py3-none-any.whl

commonmeta/readers/crossref_reader.py

@@ -0,0 +1,409 @@
+"""crossref reader for commonmeta-py"""
+
+from typing import Optional
+import httpx
+from pydash import py_
+
+from ..utils import (
+    dict_to_spdx,
+    normalize_cc_url,
+    normalize_url,
+    normalize_doi,
+    normalize_issn,
+    issn_as_url,
+)
+from ..base_utils import wrap, compact, presence, sanitize, parse_attributes
+from ..author_utils import get_authors
+from ..date_utils import get_date_from_date_parts
+from ..doi_utils import (
+    doi_as_url,
+    doi_from_url,
+    crossref_api_url,
+    crossref_api_query_url,
+    crossref_api_sample_url,
+)
+from ..constants import (
+    CR_TO_CM_TRANSLATIONS,
+    CR_TO_CM_CONTAINER_TRANSLATIONS,
+    CROSSREF_CONTAINER_TYPES,
+    Commonmeta,
+)
+
+
+def get_crossref_list(query: dict, **kwargs) -> list[dict]:
+    """get_crossref list from Crossref API."""
+    url = crossref_api_query_url(query, **kwargs)
+    response = httpx.get(url, timeout=30, **kwargs)
+    if response.status_code != 200:
+        return []
+    return response.json().get("message", {}).get("items", [])
+
+
+def get_crossref(pid: str, **kwargs) -> dict:
+    """get_crossref"""
+    doi = doi_from_url(pid)
+    if doi is None:
+        return {"state": "not_found"}
+    url = crossref_api_url(doi)
+    response = httpx.get(url, timeout=10, **kwargs)
+    if response.status_code != 200:
+        return {"state": "not_found"}
+    return response.json().get("message", {}) | {"via": "crossref"}
+
+
+def read_crossref(data: Optional[dict], **kwargs) -> Commonmeta:
+    """read_crossref"""
+    if data is None:
+        return {"state": "not_found"}
+    meta = data
+    # read_options = ActiveSupport::HashWithIndifferentAccess.
+    # new(options.except(:doi, :id, :url,
+    # :sandbox, :validate, :ra))
+    read_options = kwargs or {}
+
+    doi = meta.get("DOI", None)
+    _id = doi_as_url(doi)
+    _type = CR_TO_CM_TRANSLATIONS.get(meta.get("type", None)) or "Other"
+
+    archive_locations = wrap(meta.get("archive", None))
+
+    if meta.get("author", None):
+        contributors = get_authors(wrap(meta.get("author")), via="crossref")
+    else:
+        contributors = []
+
+    def editor_type(item):
+        item["contributorType"] = "Editor"
+        return item
+
+    editors = [editor_type(i) for i in wrap(meta.get("editor", None))]
+    if editors:
+        contributors += get_authors(editors)
+
+    url = normalize_url(py_.get(meta, "resource.primary.URL"))
+    titles = get_titles(meta)
+    publisher = compact({"name": meta.get("publisher", None)})
+
+    date = compact(
+        {
+            "published": py_.get(meta, "issued.date-time")
+            or get_date_from_date_parts(meta.get("issued", None))
+            or py_.get(meta, "created.date-time")
+        }
+    )
+    identifiers = []
+    identifiers.append(
+        compact(
+            {
+                "identifier": _id,
+                "identifierType": "DOI",
+            }
+        )
+    )
+    license_ = meta.get("license", None)
+    if license_ is not None:
+        license_ = normalize_cc_url(license_[0].get("URL", None))
+        license_ = dict_to_spdx({"url": license_}) if license_ else None
+    issn = get_issn(meta)
+    container = get_container(meta, issn=issn)
+    relations = get_relations(meta.get("relation", None))
+    if issn is not None:
+        relations.append(
+            {
+                "id": issn_as_url(issn),
+                "type": "IsPartOf",
+            }
+        )
+        relations = py_.uniq(relations)
+    references = py_.uniq([get_reference(i) for i in wrap(meta.get("reference", None))])
+    funding_references = from_crossref_funding(wrap(meta.get("funder", None)))
+
+    description = meta.get("abstract", None)
+    if description is not None:
+        descriptions = [{"description": sanitize(description), "type": "Abstract"}]
+    else:
+        descriptions = None
+
+    subjects = py_.uniq(
+        [
+            {"subject": i}
+            for i in wrap(meta.get("subject", None) or meta.get("group-title", None))
+        ]
+    )
+    files = py_.uniq(
+        [
+            get_file(i)
+            for i in wrap(meta.get("link", None))
+            if i["content-type"] != "unspecified"
+        ]
+    )
+
+    return {
+        # required properties
+        "id": _id,
+        "type": _type,
+        # recommended and optional properties
+        "additionalType": None,
+        "archiveLocations": presence(archive_locations),
+        "container": presence(container),
+        "contributors": presence(contributors),
+        "date": presence(date),
+        "descriptions": presence(descriptions),
+        "files": presence(files),
+        "fundingReferences": presence(funding_references),
+        "geoLocations": None,
+        "identifiers": identifiers,
+        "language": meta.get("language", None),
+        "license": license_,
+        "provider": "Crossref",
+        "publisher": presence(publisher),
+        "references": presence(references),
+        "relations": presence(relations),
+        "subjects": presence(subjects),
+        "titles": presence(titles),
+        "url": url,
+        "version": meta.get("version", None),
+    } | read_options
+
+
+def get_titles(meta):
+    """Title information from Crossref metadata."""
+    titles = wrap(parse_attributes(meta.get("title", None)))
+    subtitles = wrap(parse_attributes(meta.get("subtitle", None)))
+    original_language_titles = wrap(
+        parse_attributes(meta.get("original_language_title", None))
+    )
+    language = None
+    return (
+        [{"title": sanitize(i)} for i in titles]
+        + [
+            compact(
+                {
+                    "title": sanitize(i),
+                    "titleType": "Subtitle",
+                }
+            )
+            for i in subtitles
+        ]
+        + [
+            compact(
+                {
+                    "title": sanitize(i),
+                    "titleType": "TranslatedTitle",
+                    "lang": language,
+                }
+            )
+            for i in original_language_titles
+        ]
+    )
+
+
+def get_reference(reference: Optional[dict]) -> Optional[dict]:
+    """Get reference from Crossref reference"""
+    if reference is None or not isinstance(reference, dict):
+        return None
+    doi = reference.get("DOI", None)
+    metadata = {
+        "key": reference.get("key", None),
+        "id": normalize_doi(doi) if doi else None,
+        "contributor": reference.get("author", None),
+        "title": reference.get("article-title", None),
+        "publisher": reference.get("publisher", None),
+        "publicationYear": reference.get("year", None),
+        "volume": reference.get("volume", None),
+        "issue": reference.get("issue", None),
+        "firstPage": reference.get("first-page", None),
+        "lastPage": reference.get("last-page", None),
+        "containerTitle": reference.get("journal-title", None),
+        "edition": None,
+        "unstructured": reference.get("unstructured", None),
+    }
+    return compact(metadata)
+
+
+def get_relations(relations: list) -> list:
+    """Get relations from Crossref"""
+    supported_types = [
+        "IsNewVersionOf",
+        "IsPreviousVersionOf",
+        "IsVersionOf",
+        "HasVersion",
+        "IsPartOf",
+        "HasPart",
+        "IsVariantFormOf",
+        "IsOriginalFormOf",
+        "IsIdenticalTo",
+        "IsTranslationOf",
+        "IsReviewedBy",
+        "Reviews",
+        "HasReview",
+        "IsPreprintOf",
+        "HasPreprint",
+        "IsSupplementTo",
+        "IsSupplementedBy",
+    ]
+
+    if not relations:
+        return []
+
+    def format_relation(key, values):
+        _type = py_.pascal_case(key)
+        if _type not in supported_types:
+            return None
+        rs = []
+        for value in values:
+            if value.get("id-type", None) == "doi":
+                _id = doi_as_url(value.get("id", None))
+            elif value.get("id-type", None) == "issn":
+                _id = issn_as_url(value.get("id", None))
+            else:
+                _id = value.get("id", None)
+
+            rs.append({"type": _type, "id": _id})
+
+        return rs
+
+    return py_.uniq(
+        py_.compact(py_.flatten([format_relation(k, v) for k, v in relations.items()]))
+    )
+
+
+def get_file(file: dict) -> dict:
+    """Get file from Crossref"""
+    return compact(
+        {
+            "url": file.get("URL", None),
+            "mimeType": file.get("content-type", None),
+        }
+    )
+
+
+def get_issn(meta: dict) -> Optional[str]:
+    """Get ISSN from Crossref"""
+    issn = (
+        next(
+            (
+                item
+                for item in wrap(meta.get("issn-type", None))
+                if item["type"] == "electronic"
+            ),
+            None,
+        )
+        or next(
+            (
+                item
+                for item in wrap(meta.get("issn-type", None))
+                if item["type"] == "print"
+            ),
+            None,
+        )
+        or next(
+            (
+                item
+                for item in py_.get(meta, "relation.is-part-of", [])
+                if item["id-type"] == "issn"
+            ),
+            None,
+        )
+        or {}
+    )
+    return (
+        normalize_issn(issn.get("value", None) or issn.get("id", None))
+        if issn
+        else None
+    )
+
+
+def get_container(meta: dict, issn: str) -> dict:
+    """Get container from Crossref"""
+    container_type = CROSSREF_CONTAINER_TYPES.get(meta.get("type", None))
+    container_type = CR_TO_CM_CONTAINER_TRANSLATIONS.get(container_type, None)
+    isbn = (
+        next(
+            (
+                item
+                for item in wrap(meta.get("isbn-type", None))
+                if item["type"] == "electronic"
+            ),
+            None,
+        )
+        or next(
+            (
+                item
+                for item in wrap(meta.get("isbn-type", None))
+                if item["type"] == "print"
+            ),
+            None,
+        )
+        or {}
+    )
+    isbn = isbn["value"] if isbn else None
+    container_title = parse_attributes(meta.get("container-title", None), first=True)
+    volume = meta.get("volume", None)
+    issue = py_.get(meta, "journal-issue.issue")
+    if meta.get("page", None):
+        pages = meta.get("page", None).split("-")
+        first_page = pages[0]
+        last_page = pages[1] if len(pages) > 1 else None
+    else:
+        first_page = None
+        last_page = None
+
+    # TODO: add support for series, location, missing in Crossref JSON
+
+    return compact(
+        {
+            "type": container_type,
+            "identifier": issn or isbn,
+            "identifierType": "ISSN" if issn else "ISBN" if isbn else None,
+            "title": container_title,
+            "volume": volume,
+            "issue": issue,
+            "firstPage": first_page,
+            "lastPage": last_page,
+        }
+    )
+
+
+def from_crossref_funding(funding_references: list) -> list:
+    """Get funding references from Crossref"""
+    formatted_funding_references = []
+    for funding in funding_references:
+        f = compact(
+            {
+                "funderName": funding.get("name", None),
+                "funderIdentifier": doi_as_url(funding["DOI"])
+                if funding.get("DOI", None) is not None
+                else None,
+                "funderIdentifierType": "Crossref Funder ID"
+                if funding.get("DOI", "").startswith("10.13039")
+                else None,
+            }
+        )
+        f = py_.omit(f, "DOI", "doi-asserted-by")
+        if (
+            funding.get("name", None) is not None
+            and funding.get("award", None) is not None
+        ):
+            for award in wrap(funding["award"]):
+                fund_ref = f.copy()
+                fund_ref["awardNumber"] = award
+                formatted_funding_references.append(fund_ref)
+        elif f != {}:
+            formatted_funding_references.append(f)
+    return py_.uniq(formatted_funding_references)
+
+
+def get_random_crossref_id(number: int = 1, **kwargs) -> list:
+    """Get random DOI from Crossref"""
+    number = 20 if number > 20 else number
+    url = crossref_api_sample_url(number, **kwargs)
+    try:
+        response = httpx.get(url, timeout=10)
+        if response.status_code != 200:
+            return []
+
+        items = py_.get(response.json(), "message.items")
+        return [i.get("DOI") for i in items]
+    except (httpx.ReadTimeout, httpx.ConnectError):
+        return []