project-management-search-utils 1.0.9

Django search utilities for Elasticsearch integration with permission-aware filtering

project-management-search-utils

Django search utilities for Elasticsearch integration with permission-aware filtering.

Overview

This package provides high-level search functionality for project management applications using Elasticsearch as the search backend. It includes:

• SearchService: Main search service for querying Elasticsearch
• SearchPermissionMixin: Permission-aware filtering for search results
• SearchIndexManager: Utilities for managing Elasticsearch indexes

Installation

From PyPI (after publishing)

pip install project-management-search-utils

From Source

pip install -e .

Development Installation

pip install -e ".[dev]"

This will also install development dependencies (pytest, black, flake8, isort).

Requirements

• Django >= 5.0
• djangorestframework >= 3.14
• django-haystack >= 3.2
• elasticsearch >= 8.0, < 9.0
• Python >= 3.11

Quick Start

Basic Search

from project_search import SearchService
from django.contrib.auth.models import User

# Create service instance
service = SearchService()

# Get user
user = User.objects.get(username='john')

# Simple search
results = service.search(
    query="mobile app",
    user=user
)

print(f"Found {results['total']} results")
for result in results['results']:
    print(f"  - {result['title']}")

Search with Filters and Pagination

from project_search import SearchService

service = SearchService()

# Search with filters
results = service.search(
    query="mobile app",
    filters={
        "status": "active",
        "health": "healthy"
    },
    page=1,
    page_size=20,
    user=request.user
)

print(f"Total results: {results['total']}")
print(f"Current page: {results['page']}")
print(f"Total pages: {results['total_pages']}")

# Print results
for result in results['results']:
    print(f"- {result['title']} (Status: {result['status']})")

# Print available filters
print("Available filters:")
for filter_name, options in results['facets'].items():
    print(f"\n{filter_name}:")
    for option, count in options.items():
        print(f"  - {option}: {count}")

Autocomplete Suggestions

from project_search import SearchService

service = SearchService()

# Get autocomplete suggestions
suggestions = service.autocomplete(
    query="mo",  # At least 2 characters
    limit=10,
    user=request.user
)

for suggestion in suggestions:
    print(f"- {suggestion['title']} (Type: {suggestion['type']})")

API Reference

SearchService

Main service for querying Elasticsearch.

Methods

search(query, filters=None, page=1, page_size=20, user=None)

Search for projects using Elasticsearch.

Parameters:

• query (str): Search query string
• filters (dict, optional): Filter criteria
  • status: Project status (e.g., 'active', 'archived')
  • health: Project health status (e.g., 'healthy', 'at_risk')
  • owner: Project owner username
• page (int): Page number (1-indexed, default: 1)
• page_size (int): Results per page (default: 20)
• user (User): Django User object for permission filtering

Returns: Dictionary with:

{
    "results": [
        {
            "id": 1,
            "title": "Mobile App",
            "description": "iOS and Android app",
            "owner": {"id": 1, "username": "john"},
            "status": "active",
            "health": "healthy",
            "progress": 75,
            "tags": [
                {"id": 1, "name": "frontend", "color": "#FF5733"},
                {"id": 2, "name": "mobile", "color": "#33FF57"}
            ]
        }
    ],
    "facets": {
        "status": {"active": 45, "archived": 12},
        "health": {"healthy": 40, "at_risk": 17},
        "owner": {"john": 30, "jane": 27}
    },
    "total": 57,
    "page": 1,
    "page_size": 20,
    "total_pages": 3
}

Raises:

• ValueError: If query is empty
• PermissionDenied: If user lacks permission

Example:

results = service.search(
    query="dashboard",
    filters={"status": "active"},
    page=1,
    user=request.user
)

autocomplete(query, limit=10, user=None)

Get autocomplete suggestions.

Parameters:

• query (str): Partial query for prefix matching (minimum 2 characters)
• limit (int): Maximum number of suggestions (default: 10)
• user (User): Django User object for permission filtering

Returns: List of suggestions:

[
    {
        "id": 1,
        "title": "Mobile App",
        "type": "project"
    },
    {
        "id": 2,
        "title": "Mobile First Design",
        "type": "project"
    }
]

Raises:

• ValueError: If query is less than 2 characters

Example:

suggestions = service.autocomplete(
    query="mob",
    limit=5,
    user=request.user
)

SearchPermissionMixin

Provides permission-aware filtering for search results.

Methods

get_accessible_projects(user)

Get all projects accessible to the given user.

Parameters:

• user (User): Django User object

Returns: QuerySet of accessible projects

Raises:

• ValueError: If user is not authenticated

Example:

from project_search import SearchPermissionMixin

mixin = SearchPermissionMixin()
accessible = mixin.get_accessible_projects(user=request.user)

filter_by_permissions(queryset, user)

Filter a queryset to include only accessible projects.

Parameters:

• queryset (QuerySet): QuerySet to filter
• user (User): Django User object

Returns: Filtered QuerySet

Example:

projects = Project.objects.all()
filtered = mixin.filter_by_permissions(projects, user=request.user)

has_project_access(user, project_id)

Check if user has access to a specific project.

Parameters:

• user (User): Django User object
• project_id (int): Project ID

Returns: Boolean indicating access

Example:

if mixin.has_project_access(user=request.user, project_id=42):
    print("User has access")

get_accessible_project_ids(user)

Get list of accessible project IDs (useful for Elasticsearch filtering).

Parameters:

• user (User): Django User object

Returns: List of project IDs

Example:

project_ids = mixin.get_accessible_project_ids(user=request.user)
# Use for filtering Elasticsearch results

SearchIndexManager

Utilities for managing Elasticsearch indexes.

Methods

rebuild_all_indexes()

Rebuild all search indexes from database.

Returns: Dictionary with rebuild statistics

Example:

from project_search import SearchIndexManager

manager = SearchIndexManager()
result = manager.rebuild_all_indexes()
print(f"Indexed {result['total_documents']} documents")

rebuild_index(index_name)

Rebuild a specific index.

Parameters:

• index_name (str): Name of index ('projects', 'milestones', 'activities', or 'tags')

Returns: Dictionary with rebuild information

Example:

result = manager.rebuild_index('projects')

optimize_index(index_name)

Optimize an index for better performance.

Parameters:

• index_name (str): Name of index to optimize

Returns: Dictionary with optimization results

delete_index(index_name)

Delete an index (WARNING: irreversible).

Parameters:

• index_name (str): Name of index to delete

Returns: Dictionary with deletion status

get_index_stats()

Get statistics for all search indexes.

Returns: Dictionary with index statistics including document counts and health status

Example:

stats = manager.get_index_stats()
print(f"Total documents: {stats['total_documents']}")
print(f"Health: {stats['health_status']}")

get_index_mapping(index_name)

Get the mapping (schema) for an index.

Parameters:

• index_name (str): Name of index

Returns: Dictionary with index mapping

reindex_document(doc_id, doc_type)

Reindex a single document.

Parameters:

• doc_id (int): Document ID
• doc_type (str): Document type ('project', 'milestone', 'activity', 'tag')

Returns: Dictionary with reindexing status

clear_cache()

Clear Elasticsearch request cache.

Returns: Dictionary with cache clear status

health_check()

Check the health of Elasticsearch.

Returns: Dictionary with health information and any issues found

Integration with Django

Settings Configuration

Add to your settings.py:

INSTALLED_APPS = [
    # ...
    'haystack',
    'projects',
]

HAYSTACK_CONNECTIONS = {
    'default': {
        'ENGINE': 'haystack.backends.elasticsearch_backend.ElasticsearchEngine',
        'URL': os.getenv('ELASTICSEARCH_URL', 'http://elasticsearch:9200/'),
        'INDEX_NAME': 'projects_index',
    },
}

HAYSTACK_SIGNAL_PROCESSOR = 'haystack.signals.RealtimeSignalProcessor'

Django Admin Integration

Use the SearchPermissionMixin to filter admin querysets:

from django.contrib import admin
from project_search import SearchPermissionMixin
from projects.models import Project

class ProjectAdmin(admin.ModelAdmin, SearchPermissionMixin):
    list_display = ('title', 'owner', 'status')

    def get_queryset(self, request):
        qs = super().get_queryset(request)
        if request.user.is_superuser:
            return qs
        return self.filter_by_permissions(qs, user=request.user)

admin.site.register(Project, ProjectAdmin)

Usage in Views

Class-Based View with SearchService

from django.views.generic import ListView
from project_search import SearchService
from projects.models import Project

class ProjectSearchView(ListView):
    model = Project
    template_name = 'projects/search.html'
    context_object_name = 'results'

    def get_queryset(self):
        service = SearchService()
        query = self.request.GET.get('q', '')

        if not query:
            return Project.objects.none()

        results = service.search(
            query=query,
            user=self.request.user
        )
        return results['results']

    def get_context_data(self, **kwargs):
        context = super().get_context_data(**kwargs)
        service = SearchService()
        query = self.request.GET.get('q', '')

        results = service.search(query=query, user=self.request.user)
        context['facets'] = results['facets']
        context['total'] = results['total']

        return context

REST API View with DRF

from rest_framework.views import APIView
from rest_framework.response import Response
from rest_framework.permissions import IsAuthenticated
from project_search import SearchService

class SearchAPIView(APIView):
    permission_classes = [IsAuthenticated]

    def get(self, request):
        query = request.query_params.get('q', '')
        filters = {
            'status': request.query_params.get('status'),
            'health': request.query_params.get('health'),
        }
        page = int(request.query_params.get('page', 1))

        service = SearchService()
        results = service.search(
            query=query,
            filters={k: v for k, v in filters.items() if v},
            page=page,
            user=request.user
        )

        return Response(results)

Performance Considerations

Elasticsearch Configuration

For production deployments:

1. Index Sharding: Configure appropriate number of shards based on data volume
2. Replicas: Set replica count for high availability
3. Refresh Interval: Adjust for your throughput requirements
4. Bulk Indexing: Use bulk API for initial index creation

Caching

Results are not cached by default. Implement your own caching strategy:

from django.core.cache import cache

def cached_search(query, user):
    cache_key = f"search:{query}:{user.id}"
    results = cache.get(cache_key)

    if results is None:
        service = SearchService()
        results = service.search(query=query, user=user)
        cache.set(cache_key, results, timeout=300)  # 5 minutes

    return results

Query Optimization

1. Limit page size: Don't return excessive results per page
2. Use filters: Narrow results before full-text search
3. Debounce autocomplete: Implement client-side debouncing for suggestions
4. Index optimization: Periodically optimize indexes after bulk operations

Testing

Run tests with:

pytest

Run tests with coverage:

pytest --cov=project_search

Run specific test file:

pytest tests/test_search_service.py

Documentation

For detailed documentation on Elasticsearch integration, see the full project documentation:

• Backend Search Guide
• Elasticsearch Integration
• Search Indexes

Contributing

Contributions are welcome! Please ensure:

1. Code follows PEP 8 style guide (checked with flake8)
2. Code is formatted with black
3. Import statements are sorted with isort
4. All tests pass
5. New features include tests

Format code before submitting:

black project_search/
isort project_search/
flake8 project_search/

License

MIT License - see LICENSE file for details

Support

For issues, questions, or suggestions, please visit: https://github.com/paulkokos/project-management-dashboard/issues

Related Projects

• @paulkokos/search-components: NPM package for React search UI components
• project-management-dashboard: Full project management application

See PACKAGES.md for more information on available packages.