Merge pull request #13 from fleetingbytes/develop

Develop

Author: fleetingbytes

CHANGELOG.md

@@ -2,6 +2,14 @@
 
 <!-- towncrier release notes start -->
 
+## 0.8.1 (2023-08-07)
+
+
+### Bugfixes
+
+- Interpret ANSI encoding as CP1252, improve error handling [#11](https://github.com/fleetingbytes/rtfparse/issues/11)
+
+
 ## 0.8.0 (2023-06-29)
 
 

src/rtfparse/__about__.py

@@ -1,4 +1,4 @@
 #!/usr/bin/env python
 
 
-__version__ = "0.8.0"
+__version__ = "0.8.1"

src/rtfparse/entities.py

@@ -20,7 +20,11 @@
 MAX_CW_LETTERS = 32  # As specified in RTF Spec
 INTEGER_MAGNITUDE = 32  # As specified in RTF Spec
 PLAIN_TEXT = CONTROL_WORD = (
-    BACKSLASH + MAX_CW_LETTERS + MINUS + len(str((1 << INTEGER_MAGNITUDE) // 2)) + DELIMITER
+    BACKSLASH
+    + MAX_CW_LETTERS
+    + MINUS
+    + len(str((1 << INTEGER_MAGNITUDE) // 2))
+    + DELIMITER
 )
 
 
@@ -29,7 +33,9 @@ def __init__(self) -> None:
         self.text = ""
 
     @classmethod
-    def probe(cls, pattern: re_patterns.Bytes_Regex, file: io.BufferedReader) -> Bytestring_Type:
+    def probe(
+        cls, pattern: re_patterns.Bytes_Regex, file: io.BufferedReader
+    ) -> Bytestring_Type:
         logger.debug(f"Probing file at position {file.tell()}")
         original_position = file.tell()
         while True:
@@ -48,7 +54,9 @@ def probe(cls, pattern: re_patterns.Bytes_Regex, file: io.BufferedReader) -> Byt
             elif match := re_patterns.plain_text.match(probed):
                 result = Bytestring_Type.PLAIN_TEXT
             else:
-                logger.debug(f"This does not match anything, it's probably a newline, moving on")
+                logger.debug(
+                    f"This does not match anything, it's probably a newline, moving on"
+                )
                 original_position += 1
                 file.seek(original_position)
                 logger.debug(f"Probe moved to position {file.tell()}")
@@ -92,7 +100,9 @@ def __init__(self, encoding: str, file: io.BufferedReader) -> None:
             file.seek(target_position)
             # handle \binN:
             if self.control_name == "bin":
-                self.bindata = file.read(utils.twos_complement(self.parameter, INTEGER_MAGNITUDE))
+                self.bindata = file.read(
+                    utils.twos_complement(self.parameter, INTEGER_MAGNITUDE)
+                )
         else:
             logger.warning(f"Missing Control Word")
             file.seek(self.start_position)
@@ -178,7 +188,9 @@ def __init__(self, encoding: str, file: io.BufferedReader) -> None:
                 logger.debug(f"Returned to position {file.tell()}")
         else:
             logger.warning(
-                utils.warn(f"Expected a group but found no group start. Creating unknown group")
+                utils.warn(
+                    f"Expected a group but found no group start. Creating unknown group"
+                )
             )
             file.seek(self.start_position)
         while True:

src/rtfparse/parser.py

@@ -5,6 +5,7 @@
 import logging
 import pathlib
 import re
+from argparse import Namespace
 
 # Typing
 from typing import Optional, Union
@@ -26,7 +27,9 @@ def __init__(
         self.rtf_file = rtf_file
         if not (self.rtf_path or self.rtf_file):
             raise ValueError("Need `rtf_path` or `rtf_file` argument")
-        self.ENCODING_PROBE = 48  # look for encoding information in the first 48 bytes of the file
+        self.ENCODING_PROBE = (
+            48  # look for encoding information in the first 48 bytes of the file
+        )
 
     def read_encoding(self, file: Union[io.BufferedReader, io.BytesIO]) -> str:
         probed = file.read(self.ENCODING_PROBE)
@@ -52,17 +55,22 @@ def read_encoding(self, file: Union[io.BufferedReader, io.BytesIO]) -> str:
             # if any item is a Control_Word which has a parameter, we assume that this is the parameter of \ansicpg, and that corresponds to the codepage we are looking for
             if item.parameter:
                 param = item.parameter
+            else:
+                param = None
         if param:
             if param == 65001:
                 logger.warning(
-                    "Found encoding 65001, but often this is actually cp1252, so I'm overriding it"
+                    "Found encoding '65001', but often this is actually 'cp1252', so I'm taking that"
                 )
                 encoding = "cp1252"
             else:
                 encoding = f"cp{param}"
         else:
             if names[0].control_name == "ansi":
-                encoding = "ansi"
+                logger.warning(
+                    "Found encoding 'ansi', but often this is actually 'cp1252', so I'm taking that"
+                )
+                encoding = "cp1252"
             elif names[0].control_name == "mac":
                 encoding = "mac_roman"
             elif names[0].control_name == "pc":
@@ -87,6 +95,8 @@ def parse_file(self) -> entities.Group:
             self.parsed = entities.Group(encoding, file)
         except Exception as err:
             logger.exception(err)
+            self.parsed == Namespace()
+            self.parsed.structure = list()
         finally:
             if self.rtf_path is not None:
                 logger.debug(f"Closing {parsed_object}")