reader_comdirect (tbc)

src/reader.py

@@ -1,4 +1,6 @@
 import csv
+import re
+from enum import Enum, auto
 from pathlib import Path
 
 from src.models import B4wExport
@@ -12,3 +14,47 @@ def reader_raw(path: Path):
 def reader_b4w(path: Path):
     raw = reader_raw(path)
     return list(map(B4wExport.model_validate, raw))
+
+
+def reader_comdirect(path: Path):
+    PATTERN_ACCOUNT_NAME = r'^"Umsätze ([^"]+)";"Zeitraum: .*$'
+    PATTERN_HEADER = r'"([^"]+)"'
+    PATTERN_DATA = r'"([^"]+)"'
+
+    class EXPECT_LINE(Enum):
+        ACCOUNT_NAME = auto()
+        BLANK = auto()
+        HEADER = auto()
+        DATA = auto()
+
+    with path.open(encoding="latin1") as fid:
+        lines = fid.readlines()
+
+    expect_next = EXPECT_LINE.ACCOUNT_NAME
+    account_name = None
+    result = {}
+    for line in lines:
+        if expect_next == EXPECT_LINE.ACCOUNT_NAME:
+            account_name = re.findall(PATTERN_ACCOUNT_NAME, line)
+            if account_name:
+                account_name = account_name[0]
+                result[account_name] = {}
+                expect_next = EXPECT_LINE.BLANK
+        elif expect_next == EXPECT_LINE.BLANK:
+            if line == "\n":
+                expect_next = EXPECT_LINE.HEADER
+        elif expect_next == EXPECT_LINE.HEADER:
+            header = re.findall(PATTERN_HEADER, line)
+            if header:
+                result[account_name]["data"] = []
+                expect_next = EXPECT_LINE.DATA
+        elif expect_next == EXPECT_LINE.DATA:
+            data = re.findall(PATTERN_DATA, line)
+            if data:
+                result[account_name]["data"].append(
+                    {column: value for column, value in zip(header, data)}
+                )
+                expect_next = EXPECT_LINE.ACCOUNT_NAME
+        else:
+            expect_next = EXPECT_LINE.ACCOUNT_NAME
+    return result