commit AI suggested performance enhancements

src/backend/threads_availchecker.py

@@ -1,5 +1,4 @@
 import sys
-import time
 
 # from src.transformers import RDS_AVAIL_DATA
 import loguru
@@ -23,7 +22,11 @@ class AvailChecker(QThread):
     updateProgress = Signal(int, int)
 
     def __init__(
-        self, links: list = None, appnumber: int = None, parent=None, books=list[dict]
+        self,
+        links: list[str] | None = None,
+        appnumber: int | None = None,
+        parent=None,
+        books: list[dict] | None = None,
     ):
         if links is None:
             links = []
@@ -38,11 +41,13 @@ class AvailChecker(QThread):
         )
         self.links = links
         self.appnumber = appnumber
-        self.books = books
+        self.books = books or []
         log.info(
             f"Started worker with appnumber: {self.appnumber} and links: {self.links} and {len(self.books)} books..."
         )
-        time.sleep(2)
+        # Pre-create reusable request and transformer to avoid per-item overhead
+        self._request = WebRequest().set_apparat(self.appnumber)
+        self._rds_transformer = BibTextTransformer("RDS")
 
     def run(self):
         self.db = Database()
@@ -50,9 +55,8 @@ class AvailChecker(QThread):
         count = 0
         for link in self.links:
             log.info("Processing entry: " + str(link))
-            data = WebRequest().set_apparat(self.appnumber).get_ppn(link).get_data()
+            data = self._request.get_ppn(link).get_data()
-            transformer = BibTextTransformer("RDS")
+            rds = self._rds_transformer.get_data(data).return_data("rds_availability")
-            rds = transformer.get_data(data).return_data("rds_availability")
 
             book_id = None
             if not rds or not rds.items:

src/logic/SRU.py

@@ -1,3 +1,4 @@
+import re
 import sys
 import xml.etree.ElementTree as ET
 from dataclasses import dataclass, field
@@ -7,6 +8,7 @@ from typing import Dict, Iterable, List, Optional, Tuple
 
 import loguru
 import requests
+from requests.adapters import HTTPAdapter
 
 from src import LOG_DIR
 from src.logic.dataclass import BookData
@@ -556,8 +558,23 @@ class Api:
         self.site = site
         self.url = url
         self.prefix = prefix
+        # Reuse TCP connections across requests for better performance
+        self._session = requests.Session()
+        # Slightly larger connection pool for concurrent calls
+        adapter = HTTPAdapter(pool_connections=10, pool_maxsize=20)
+        self._session.mount("http://", adapter)
+        self._session.mount("https://", adapter)
+
+    def close(self):
+        try:
+            self._session.close()
+        except Exception:
             pass
 
+    def __del__(self):
+        # Best-effort cleanup
+        self.close()
+
     def get(self, query_args: Iterable[str]) -> List[Record]:
         # if any query_arg ends with =, remove it
         if self.site == "DNB":
@@ -580,18 +597,18 @@ class Api:
             "Accept": "application/xml",
             "Accept-Charset": "latin1,utf-8;q=0.7,*;q=0.3",
         }
-        response = requests.get(url, headers=headers)
+        # Use persistent session and set timeouts to avoid hanging
+        response = self._session.get(url, headers=headers, timeout=(3.05, 20))
         if response.status_code != 200:
             raise Exception(f"Error fetching data from SWB: {response.status_code}")
-        data = response.content
+        # extract top-level response (decode to text for the XML parser)
-
+        response = parse_search_retrieve_response(response.text)
-        # extract top-level response
-        response = parse_search_retrieve_response(data)
         return response.records
 
     def getBooks(self, query_args: Iterable[str]) -> List[BookData]:
         records: List[Record] = self.get(query_args)
-        print(f"{self.site} found {len(records)} records")
+        # Avoid printing on hot paths; rely on logger if needed
+        log.debug(f"{self.site} found {len(records)} records for args={query_args}")
         books: List[BookData] = []
         # extract title from query_args if present
         title = None
@@ -611,7 +628,8 @@ class Api:
         return books
 
     def getLinkForBook(self, book: BookData) -> str:
-        results = self.getBooks()
+        # Not implemented: depends on catalog front-end; return empty string for now
+        return ""
 
 
 class SWB(Api):

src/ui/dialogs/progress.py

@@ -32,6 +32,7 @@ class CheckThread(QtCore.QThread):
             range(len(self.items)),
             unit_scale=True,
         )
+        swb_client = SWB()
         for i in tqdm_object:
             book: BookData = self.items[i]
             author = (
@@ -43,7 +44,7 @@ class CheckThread(QtCore.QThread):
             # remove trailing punctuation from title
             title = book.title.rstrip(" .:,;!?")
             response: list[BookData] = []
-            response = SWB().getBooks(
+            response = swb_client.getBooks(
                 [
                     "pica.bib=20735",
                     f"pica.tit={title.split(':')[0].strip()}",
@@ -88,4 +89,6 @@ class ProgressDialog(QDialog):
         layout.addWidget(self.start_button)
 
     def start(self):
+        # Start logic is managed externally; keep method for UI wiring
+        pass