more AI optimizations, reworked logger

src/logic/SRU.py

@@ -1,28 +1,17 @@
 import re
-import sys
 import xml.etree.ElementTree as ET
 from dataclasses import dataclass, field
-from datetime import datetime
 from enum import Enum
-from typing import Dict, Iterable, List, Optional, Tuple
+from typing import Dict, Iterable, List, Optional, Tuple, Union
 
-import loguru
 import requests
 from requests.adapters import HTTPAdapter
 
-from src import LOG_DIR
+# centralized logging used via src.shared.logging
 from src.logic.dataclass import BookData
+from src.shared.logging import log
 
-log = loguru.logger
-log.remove()
-log.add(sys.stdout, level="INFO")
-log.add(f"{LOG_DIR}/application.log", rotation="1 MB", retention="10 days")
-
-log.add(
-    f"{LOG_DIR}/{datetime.now().strftime('%Y-%m-%d')}.log",
-    rotation="1 day",
-    retention="1 month",
-)
+log  # ensure imported logger is referenced
 
 
 # -----------------------
@@ -186,7 +175,9 @@ def parse_echoed_request(root: ET.Element) -> Optional[EchoedSearchRequest]:
     )
 
 
-def parse_search_retrieve_response(xml_str: str) -> SearchRetrieveResponse:
+def parse_search_retrieve_response(
+    xml_str: Union[str, bytes],
+) -> SearchRetrieveResponse:
     root = ET.fromstring(xml_str)
 
     # Root is zs:searchRetrieveResponse
@@ -598,12 +589,12 @@ class Api:
             "Accept-Charset": "latin1,utf-8;q=0.7,*;q=0.3",
         }
         # 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}")
-        # extract top-level response (decode to text for the XML parser)
-        response = parse_search_retrieve_response(response.text)
-        return response.records
+        resp = self._session.get(url, headers=headers, timeout=(3.05, 60))
+        if resp.status_code != 200:
+            raise Exception(f"Error fetching data from SWB: {resp.status_code}")
+        # Parse using raw bytes (original behavior) to preserve encoding edge cases
+        sr = parse_search_retrieve_response(resp.content)
+        return sr.records
 
     def getBooks(self, query_args: Iterable[str]) -> List[BookData]:
         records: List[Record] = self.get(query_args)

src/logic/semester.py

@@ -18,16 +18,8 @@ from __future__ import annotations
 
 import datetime
 import re
-import sys
 
-import loguru
-
-from src import LOG_DIR
-
-log = loguru.logger
-log.remove()
-log.add(sys.stdout, level="INFO")
-log.add(f"{LOG_DIR}/application.log", rotation="1 MB", retention="10 days")
+from src.shared.logging import log
 
 
 class Semester:
@@ -124,21 +116,22 @@ class Semester:
     # ------------------------------------------------------------------
     # Comparison helpers
     # ------------------------------------------------------------------
-    def isPastSemester(self, other: "Semester") -> bool:
-        if self.year < other.year:
+    def isPastSemester(self, current: "Semester") -> bool:
+        log.debug(f"Comparing {self} < {current}")
+        if self.year < current.year:
             return True
-        if self.year == other.year:
+        if self.year == current.year:
             return (
-                self.semester == "WiSe" and other.semester == "SoSe"
+                self.semester == "WiSe" and current.semester == "SoSe"
             )  # WiSe before next SoSe
         return False
 
-    def isFutureSemester(self, other: "Semester") -> bool:
-        if self.year > other.year:
+    def isFutureSemester(self, current: "Semester") -> bool:
+        if self.year > current.year:
             return True
-        if self.year == other.year:
+        if self.year == current.year:
             return (
-                self.semester == "SoSe" and other.semester == "WiSe"
+                self.semester == "SoSe" and current.semester == "WiSe"
             )  # SoSe after WiSe of same year
         return False
 

src/logic/webrequest.py

@@ -1,23 +1,16 @@
-import sys
 from typing import Any, Optional, Union
 
-import loguru
 import requests
 from bs4 import BeautifulSoup
 
 # import sleep_and_retry decorator to retry requests
 from ratelimit import limits, sleep_and_retry
 
-from src import LOG_DIR
 from src.logic.dataclass import BookData
+from src.shared.logging import log
 from src.transformers import ARRAYData, BibTeXData, COinSData, RDSData, RISData
 from src.transformers.transformers import RDS_AVAIL_DATA, RDS_GENERIC_DATA
 
-log = loguru.logger
-log.remove()
-log.add(sys.stdout, level="INFO")
-log.add(f"{LOG_DIR}/application.log", rotation="1 MB", retention="10 days")
-
 # logger.add(sys.stderr, format="{time} {level} {message}", level="INFO")
 
 

src/logic/wordparser.py

@@ -1,20 +1,13 @@
-import sys
 import zipfile
 from typing import Any
 
 import fitz  # PyMuPDF
-import loguru
 import pandas as pd
 from bs4 import BeautifulSoup
 from docx import Document
 
-from src import LOG_DIR
 from src.logic.dataclass import Book, SemapDocument
-
-log = loguru.logger
-log.remove()
-log.add(sys.stdout, level="INFO")
-log.add(f"{LOG_DIR}/application.log", rotation="1 MB", retention="10 days")
+from src.shared.logging import log
 
 
 def word_docx_to_csv(path: str) -> list[pd.DataFrame]:
@@ -50,7 +43,6 @@ def get_fach(path: str) -> str:
     soup = BeautifulSoup(xml_data, "xml")
     # text we need is in <w:p w14:paraId="12456A32" ... > -> w:r -> w:t
     paragraphs = soup.find_all("w:p")
-    names = []
     for para in paragraphs:
         para_id = para.get("w14:paraId")
         if para_id == "12456A32":