Source code for eucrim.search.views

# SPDX-FileCopyrightText: 2024 Thomas Breitner <t.breitner@csl.mpg.de>
#
# SPDX-License-Identifier: EUPL-1.2

from django.core.paginator import EmptyPage, PageNotAnInteger, Paginator
from django.db.models import Count
from django.shortcuts import render

from wagtail.models import Page
from django.conf import settings

from django.contrib.contenttypes.models import ContentType

from .forms import SearchForm, CONTENT_TYPE_CHOICES
from .utils import parse


# Number of results per page in search results
SEARCH_PAGE_SIZE = 10
# How many authors to show by default in the sidebar before "Show more"
SEARCH_AUTHOR_VISIBLE_LIMIT = getattr(settings, "SEARCH_AUTHOR_VISIBLE_LIMIT", 8)

MODEL_LABELS = {
    "articlepage": "Article",
    "newspage": "News",
    "eventpage": "Event",
    "profilepage": "Profile",
    "issuepage": "Issue",
    "procedurepage": "Case",
    "standardpage": "Page",
}




[docs]
def get_content_type_facets(queryset, request, selected_content_types):
    """
    Compute content type facets using Django ORM aggregation.
    This works on any QuerySet, not just search results.
    """
    # Get counts per content type from the queryset
    facet_counts = (
        queryset.values("content_type_id")
        .annotate(count=Count("id"))
        .order_by("-count")
    )

    # Build a map of content_type_id -> count
    count_map = {str(item["content_type_id"]): item["count"] for item in facet_counts}

    # Get all content types we care about
    ct_models = [choice[0] for choice in CONTENT_TYPE_CHOICES]
    cts = ContentType.objects.filter(model__in=ct_models)
    ct_by_model = {ct.model: ct for ct in cts}

    # Build base params for link generation
    base_params = request.GET.copy()
    base_params.pop("page", None)

    facet_list = []
    for model_name, label in CONTENT_TYPE_CHOICES:
        ct = ct_by_model.get(model_name)
        if ct is None:
            continue

        count = count_map.get(str(ct.id), 0)

        # Omit zero-count facets — don't show options that have no results
        if count == 0:
            continue

        # Build link for this facet
        params_for_facet = base_params.copy()
        params_for_facet.setlist("content_types", [model_name])
        link = "?" + params_for_facet.urlencode()

        facet_list.append(
            {
                "content_type_model": model_name,
                "label": label,
                "count": count,
                "link": link,
                "selected": model_name in selected_content_types,
            }
        )

    return facet_list






[docs]
def get_author_facets(queryset, request):
    """
    Compute author facets from ArticlePageAuthor and NewsPage.authors relationships.
    """
    author_facets = []
    try:
        # Import models here to avoid circular imports
        from eucrim.article.models import ArticlePageAuthor
        from eucrim.news.models import NewsPage

        # Aggregate from ArticlePageAuthor (articles) using DB-side filters
        article_authors = (
            ArticlePageAuthor.objects.filter(page__in=queryset)
            .values("author_id", "author__first_name", "author__last_name")
            .annotate(count=Count("author_id"))
            .order_by("-count")[:20]
        )

        # Aggregate from NewsPage.authors M2M using DB-side filters
        news_authors_qs = (
            NewsPage.authors.through.objects.filter(newspage__in=queryset)
            .values(
                "profilepage_id",
                "profilepage__first_name",
                "profilepage__last_name",
            )
            .annotate(count=Count("profilepage_id"))
        )

        # Combine and deduplicate authors
        author_map = {}
        for a in article_authors:
            author_id = a["author_id"]
            if author_id:
                name = f"{a['author__first_name']} {a['author__last_name']}".strip()
                author_map[author_id] = author_map.get(
                    author_id, {"name": name, "count": 0, "id": author_id}
                )
                author_map[author_id]["count"] += a["count"]

        for a in news_authors_qs:
            author_id = a["profilepage_id"]
            if author_id:
                name = f"{a['profilepage__first_name']} {a['profilepage__last_name']}".strip()
                author_map[author_id] = author_map.get(
                    author_id, {"name": name, "count": 0, "id": author_id}
                )
                author_map[author_id]["count"] += a["count"]

        # Sort by count and take top 20, omitting zero-count entries
        author_facets = sorted(
            author_map.values(), key=lambda x: x["count"], reverse=True
        )
        author_facets = [a for a in author_facets if a.get("count", 0) > 0][:20]

    except Exception:
        pass

    return author_facets