add SemapDocument and Book dataclasses, improve word document parsing

src/logic/wordparser.py

@@ -1,12 +1,93 @@
 import pandas as pd
 from docx import Document
+from dataclasses import dataclass
 
 letters = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz"
 
 
-def word_docx_to_csv(path) -> pd.DataFrame:
+@dataclass
+class Book:
+    author: str = None
+    year: str = None
+    edition: str = None
+    title: str = None
+    location: str = None
+    publisher: str = None
+    signature: str = None
+    internal_notes: str = None
+
+    @property
+    def has_signature(self) -> bool:
+        return self.signature is not None and self.signature != ""
+
+    @property
+    def is_empty(self) -> bool:
+        return all(
+            [
+                self.author == "",
+                self.year == "",
+                self.edition == "",
+                self.title == "",
+                self.location == "",
+                self.publisher == "",
+                self.signature == "",
+                self.internal_notes == "",
+            ]
+        )
+
+    def from_dict(self, data: dict):
+        for key, value in data.items():
+            if value == "\u2002\u2002\u2002\u2002\u2002":
+                value = ""
+
+            if key == "Autorenname(n):Nachname, Vorname":
+                self.author = value
+            elif key == "Jahr/Auflage":
+                self.year = value.split("/")[0] if "/" in value else value
+                self.edition = value.split("/")[1] if "/" in value else ""
+            elif key == "Titel":
+                self.title = value
+            elif key == "Ort und Verlag":
+                self.location = value.split(",")[0] if "," in value else value
+                self.publisher = value.split(",")[1] if "," in value else ""
+            elif key == "Standnummer":
+                self.signature = value
+            elif key == "Interne Vermerke":
+                self.internal_notes = value
+
+
+@dataclass
+class SemapDocument:
+    subject: str = None
+    phoneNumber: int = None
+    mail: str = None
+    title: str = None
+    semester: str = None
+    books: list[Book] = None
+
+    @property
+    def renameSemester(self) -> None:
+        if self.semester is not None:
+            if "sommersemester" in self.semester.lower():
+                year = self.semester.split(" ")[-1]
+                self.semester = f"SoSe {year}"
+            elif "wintersemester" in self.semester.lower():
+                year = self.semester.split(" ")[-1]
+                self.semester = f"WiSe {year}"
+
+    @property
+    def signatures(self) -> list[str]:
+        if self.books is not None:
+            return [book.signature for book in self.books if book.has_signature]
+        return []
+
+
+def word_docx_to_csv(path: str) -> list[pd.DataFrame]:
     doc = Document(path)
     tables = doc.tables
+    print("Tables: ", len(tables))
+
+    # print content of all tables
 
     m_data = []
     for table in tables:
@@ -24,8 +105,9 @@ def word_docx_to_csv(path) -> pd.DataFrame:
 
         m_data.append(df)
 
-    df = m_data[2]
-    return df
+    # for df[0, 1]: merge i and i+1 as key, value
+
+    return m_data
 
 
 def makeDict():
@@ -122,6 +204,46 @@ def elsa_word_to_csv(path):
     return tuple_to_dict(data, doctype), doctype
 
 
+def word_to_semap(word_path: str) -> SemapDocument:
+    semap = SemapDocument()
+    df = word_docx_to_csv(word_path)
+    apparatdata = df[0]
+    apparatdata = apparatdata.to_dict()
+    keys = list(apparatdata.keys())
+    appdata = {keys[i]: keys[i + 1] for i in range(0, len(keys), 2)}
+    print(appdata.keys())
+    semap.phoneNumber = appdata["Telefon:"]
+    semap.subject = appdata["Ihr Fach:"]
+    semap.mail = appdata["Mailadresse:"]
+    apparatdata = df[1]
+    apparatdata = apparatdata.to_dict()
+    keys = list(apparatdata.keys())
+    appdata = {keys[i]: keys[i + 1] for i in range(0, len(keys), 2)}
+    semap.title = appdata["Veranstaltung:"]
+    semap.semester = appdata["Semester:"]
+    semap.renameSemester
+    books = df[2]
+    booklist = []
+    for i in range(len(books)):
+        if books.iloc[i].isnull().all():
+            continue
+        data = books.iloc[i].to_dict()
+        book = Book()
+        book.from_dict(data)
+        if book.is_empty:
+            continue
+        elif not book.has_signature:
+            continue
+        else:
+            booklist.append(book)
+
+    semap.books = booklist
+
+    return semap
+
+
 if __name__ == "__main__":
-    else_df = elsa_word_to_csv("C:/Users/aky547/Desktop/Antrag ELSA Schweitzer.docx")
-    # print(else_df)
+    else_df = word_to_semap(
+        "C:/Users/aky547/Desktop/SA 80 titelmeldung_SoSe2025 Burth.docx"
+    )
+    print(else_df)