fix: contol word recognition, closes #18

Author: Sven Siegmund

changelog.d/18.fixed.md

@@ -0,0 +1 @@
+Recognize control words with where the parameter's digital sequence is delimited by any character other than an ASCII digit

src/rtfparse/entities.py

@@ -20,11 +20,7 @@
 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
 )
 
 
@@ -33,9 +29,7 @@ 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:
@@ -54,9 +48,7 @@ def probe(
             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()}")
@@ -100,9 +92,7 @@ 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)
@@ -188,9 +178,7 @@ 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/minimal.py

@@ -6,6 +6,7 @@
 """
 
 from pathlib import Path
+
 from rtfparse.parser import Rtf_Parser
 from rtfparse.renderers.html_decapsulator import HTML_Decapsulator
 

src/rtfparse/parser.py

@@ -27,9 +27,7 @@ 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)

src/rtfparse/re_patterns.py

@@ -7,7 +7,10 @@
 
 
 def group(content: bytes) -> bytes:
-    return rb"[" + content + rb"]"
+    if content:
+        return rb"[" + content + rb"]"
+    else:
+        return b""
 
 
 def named_regex_group(name: str, content: bytes) -> bytes:
@@ -54,15 +57,15 @@ def no_capture(content: bytes) -> bytes:
 minus = named_regex_group("minus", rb"-?")
 digit = named_regex_group("digit", minus + group(_digits) + rb"{1,10}")
 hdigit = named_regex_group("hdigit", group(_hdigits))
-# int16 = minus + digit + rb"{1,5}"
 parameter_pattern = named_regex_group("parameter", digit)
 space = named_regex_group("space", rb" ")
 newline = named_regex_group("newline", _newline)
 other = named_regex_group("other", group(rb"^" + _letters + _digits))
+nothing = named_regex_group("nothing", group(rb""))
 
 
 ascii_letter_sequence = named_regex_group("control_name", ascii_letters + parameter_pattern + rb"?")
-delimiter = named_regex_group("delimiter", rb"|".join((space, newline, other, rb"$")))
+delimiter = named_regex_group("delimiter", rb"|".join((space, newline, other, nothing, rb"$")))
 symbol = named_regex_group("symbol", other)
 control_word_pattern = named_regex_group(
     "control_word", rtf_backslash + ascii_letter_sequence + delimiter