added period for filtering search results

server/database.py

@@ -1,8 +1,11 @@
+from datetime import datetime, timedelta
+
 import motor.motor_asyncio
 from bson import ObjectId
 from bson.errors import InvalidId
 
 from server.scraper import scrap_from_search
+from server.models.search_result import Period
 
 
 MONGO_DETAILS = 'mongodb://localhost:27017'
@@ -56,37 +59,25 @@ async def retrieve_search_result_by_id(id_: str):
         return
 
 
-async def retrieve_search_results_by_tags(tags: list[str], page: int, limit: int):
+async def retrieve_search_results_by_tags(tags: list[str], page: int, limit: int, period: Period):
     """Find articles by tags
 
     Take search words and check if database contain articles,
-    which have more than :percentage: of words in 'tags' fields matches
+    which have more than 'percentage' of words in 'tags' fields matches
     with words in search query. If database have them, return
     paginated articles and total amount of them.
     :param limit: Page size
     :param page: Number of page
     :param tags: List of search words
+    :param period: Filtering period
     :return: Count and List of articles
     """
-    percentage = 0.75
     tags = list(set(tags))
     filter_expression = {
-        '$expr': {
-            '$function': {
-                'body': """
-                        function(search, document, percentage) {
-                            const searchTags = search;
-                            const documentTags = document;
-                            const intersection = documentTags.filter(tag => searchTags.includes(tag));
-                            return intersection.length >= (searchTags.length * percentage);
-                        }
-                        """,
-                'args': [tags, '$tags', percentage],
-                'lang': 'js'
-            }
-        }
+        **resolve_period_expression(period),
+        **resolve_tags_expression(tags)
     }
-    results = search_results_collection.find(filter_expression).skip((page - 1) * limit).limit(limit)
+    results = search_results_collection.find(filter_expression).sort('date', -1).skip((page - 1) * limit).limit(limit)
     count = await search_results_collection.count_documents(filter_expression)
     return count, [search_results_helper(result) async for result in results]
 
@@ -113,3 +104,41 @@ async def update_content_of_article(id_: str, content: list[list]):
     await search_results_collection.update_one({'_id': ObjectId(id_)}, {"$set": {"content": content}})
     article = await search_results_collection.find_one({'_id': ObjectId(id_)})
     return search_results_helper(article)
+
+
+def resolve_period_expression(period: Period) -> dict:
+    """Create expression based on Period from query"""
+    if period is Period.last_week:
+        end_date = datetime.now()
+        start_date = end_date - timedelta(days=7)
+        end_date = end_date.strftime('%Y-%m-%dT%H:%M:%S.%f')[:-3] + 'Z'
+        start_date = start_date.strftime('%Y-%m-%dT%H:%M:%S.%f')[:-3] + 'Z'
+        return {'date': {'$gte': start_date, '$lt': end_date}}
+    if period is Period.last_month:
+        end_date = datetime.now()
+        start_date = end_date - timedelta(days=30)
+        end_date = end_date.strftime('%Y-%m-%dT%H:%M:%S.%f')[:-3] + 'Z'
+        start_date = start_date.strftime('%Y-%m-%dT%H:%M:%S.%f')[:-3] + 'Z'
+        return {'date': {'$gte': start_date, '$lt': end_date}}
+    return {}
+
+
+def resolve_tags_expression(tags: list[str]) -> dict:
+    """Create expression based on search tags"""
+    percentage = 0.75
+    return {
+        '$expr': {
+            '$function': {
+                'body': """
+                    function(search, document, percentage) {
+                        const searchTags = search;
+                        const documentTags = document;
+                        const intersection = documentTags.filter(tag => searchTags.includes(tag));
+                        return intersection.length >= (searchTags.length * percentage);
+                    }
+                    """,
+                'args': [tags, '$tags', percentage],
+                'lang': 'js'
+            }
+        }
+    }

server/models/search_result.py

@@ -1,4 +1,5 @@
 from datetime import datetime
+from enum import Enum
 
 from pydantic import BaseModel, HttpUrl
 
@@ -21,3 +22,9 @@ class SearchResponseModel(BaseModel):
 
 class ExtendArticleModel(ArticleModel):
     content: list[list] = []
+
+
+class Period(str, Enum):
+    last_week = "last-week"
+    last_month = "last-month"
+    all = "all"

server/routes/search_result.py

@@ -1,7 +1,9 @@
-from fastapi import APIRouter, status, HTTPException
+from typing import Annotated
+
+from fastapi import APIRouter, status, HTTPException, Query
 
 from server.scraper import scrap_content
-from server.models.search_result import SearchResponseModel, ExtendArticleModel
+from server.models.search_result import SearchResponseModel, ExtendArticleModel, Period
 from server.database import (
     update_search_results,
     retrieve_search_result_by_id,
@@ -15,7 +17,10 @@
 
 
 @router.get("/search", status_code=status.HTTP_200_OK, response_model=SearchResponseModel)
-async def get_search_results(find: str | None = None, page: int = 1, limit: int = 5):
+async def get_search_results(find: Annotated[str | None, Query(description='Write search query here')] = None,
+                             page: Annotated[int, Query(ge=1)] = 1,
+                             limit: Annotated[int, Query(ge=1, le=10)] = 5,
+                             period: Period = Period.all):
     """Find articles by search query
 
     Get list of articles which match with search query from database.
@@ -24,10 +29,10 @@ async def get_search_results(find: str | None = None, page: int = 1, limit: int
     articles.
     """
     if find:
-        count, results = await retrieve_search_results_by_tags(find.split(), page, limit)
+        count, results = await retrieve_search_results_by_tags(find.split(), page, limit, period)
         if count < 5:
             await update_search_results(find)
-            count, results = await retrieve_search_results_by_tags(find.split(), page, limit)
+            count, results = await retrieve_search_results_by_tags(find.split(), page, limit, period)
         return {'count': count, 'results': results}
     count, results = await retrieve_newest_search_results(page, limit)
     return {'count': count, 'results': results}

server/scraper.py

@@ -63,5 +63,3 @@ def scrap_content(link: str) -> list[list]:
             image = block.find('img').get('data-pin-media')
             content.append(('image', image))
     return content
-
-