redesign

source_conversion.py

@@ -6,6 +6,7 @@ sent to the browser or reverse proxy.
 """
 
 import codecs
+import hashlib
 import re
 import tempfile
 from dataclasses import dataclass
@@ -26,6 +27,7 @@ class ConversionResult:
     lines_read: int
     lines_emitted: int
     output_size: int
+    content_sha256: str
 
 
 def _get_encoding(content_type: str) -> str:
@@ -37,11 +39,11 @@ def _get_encoding(content_type: str) -> str:
     try:
         encoding = codecs.lookup(encoding).name
     except (LookupError, ValueError) as exc:
-        raise InvalidSourceError("Nieznane kodowanie tekstu z serwera zrodlowego.") from exc
+        raise InvalidSourceError("Unsupported source text encoding.") from exc
     # The reader splits on ASCII newline octets. Multi-byte encodings need a
     # different framing protocol and must not be interpreted as hosts text.
     if encoding in {"utf-16", "utf-16-le", "utf-16-be", "utf-32", "utf-32-le", "utf-32-be"}:
-        raise InvalidSourceError("Nieobslugiwane kodowanie pliku listy.")
+        raise InvalidSourceError("Unsupported source list encoding.")
     return encoding
 
 
@@ -55,10 +57,10 @@ def _bounded_lines(response, *, chunk_size: int, line_limit: int, source_limit:
         if not chunk:
             continue
         if not isinstance(chunk, bytes):
-            raise InvalidSourceError("Zrodlo nie zwrocilo poprawnych danych binarnych.")
+            raise InvalidSourceError("Source returned invalid binary data.")
         total += len(chunk)
         if total > source_limit:
-            raise SourceTooLargeError("Plik zrodlowy przekracza dopuszczalny rozmiar.")
+            raise SourceTooLargeError("Source list exceeds the maximum allowed size.")
         pieces = chunk.split(b"\n")
         for part in pieces[:-1]:
             if skipping or len(pending) + len(part) > line_limit:
@@ -104,7 +106,7 @@ def convert_source(
     if length:
         try:
             if int(length) > source_limit:
-                raise SourceTooLargeError("Plik zrodlowy przekracza dopuszczalny rozmiar.")
+                raise SourceTooLargeError("Source list exceeds the maximum allowed size.")
         except ValueError:
             pass
 
@@ -112,13 +114,15 @@ def convert_source(
     lines_read = 0
     lines_emitted = 0
     written = 0
+    content_hasher = hashlib.sha256()
 
     def append(data: bytes):
         nonlocal written
         if written + len(data) > output_limit:
-            raise SourceTooLargeError("Wynik konwersji przekracza dopuszczalny rozmiar.")
+            raise SourceTooLargeError("Converted output exceeds the maximum allowed size.")
         output.write(data)
         written += len(data)
+        content_hasher.update(data)
 
     try:
         if debug_lines is not None and debug_lines:
@@ -133,37 +137,37 @@ def convert_source(
             lines_read += 1
             if raw is None:
                 if debug_lines is not None and lines_read <= 5:
-                    append(b"# [DEBUG] pominieto zbyt dlugi wiersz\n")
+                    append(b"# [DEBUG] skipped an overlong line\n")
                 continue
             if b"\x00" in raw:
-                raise InvalidSourceError("Zrodlo zawiera dane binarne zamiast listy tekstowej.")
+                raise InvalidSourceError("Source contains binary data instead of a text list.")
             try:
                 line = raw.decode(encoding, errors="strict")
                 if lines_read == 1:
                     line = line.lstrip("\ufeff")
             except UnicodeError as exc:
-                raise InvalidSourceError("Nie mozna zdekodowac tekstu listy (kodowanie).") from exc
+                raise InvalidSourceError("Unable to decode source list text (character encoding).") from exc
             try:
                 converted = convert_line(line, target_ip)
             except (TypeError, ValueError) as exc:
-                raise InvalidSourceError("Nie mozna przetworzyc formatu listy.") from exc
+                raise InvalidSourceError("Unable to process this list format.") from exc
             if converted:
                 lines_emitted += 1
                 append((converted + "\n").encode("utf-8"))
             if debug_lines is not None and lines_read <= 5:
                 preview = line[:200].replace("\r", "\\r").replace("\n", "\\n")
-                append(f"# [DEBUG] podglad {lines_read}: {preview}\n".encode("utf-8"))
+                append(f"# [DEBUG] preview {lines_read}: {preview}\n".encode("utf-8"))
 
         if lines_emitted == 0:
             raise InvalidSourceError(
-                "Nie znaleziono domen w obslugiwanym formacie. Sprawdz adres i zawartosc listy."
+                "No domains found in a supported format. Check the URL and list contents."
             )
         if debug_lines is not None:
             append(
-                f"# [DEBUG] podsumowanie: przeczytano={lines_read}, wyemitowano={lines_emitted}\n".encode("utf-8")
+                f"# [DEBUG] summary: read={lines_read}, emitted={lines_emitted}\n".encode("utf-8")
             )
         output.seek(0)
-        return ConversionResult(output, lines_read, lines_emitted, written)
+        return ConversionResult(output, lines_read, lines_emitted, written, content_hasher.hexdigest())
     except BaseException:
         output.close()
         raise