Fix line ending handling in reverse readline

src/monty/io.py

@@ -11,12 +11,11 @@
 import io
 import mmap
 import os
-import platform
 import subprocess
 import time
 import warnings
 from pathlib import Path
-from typing import TYPE_CHECKING
+from typing import TYPE_CHECKING, cast
 
 try:
     import lzma
@@ -169,7 +168,6 @@ def reverse_readline(
 
     Cases where file would be read forwards and reversed in RAM:
     - If file size is smaller than RAM usage limit (max_mem).
-    - In Windows. TODO: explain reason.
     - For Gzip files, as reverse seeks are not supported.  # TODO: now supported
 
     Files larger than max_mem are read one block each time.
@@ -193,7 +191,7 @@ def reverse_readline(
     """
     # Generate line ending
     l_end: Literal["\r\n", "\n"] = _get_line_ending(m_file)
-    len_l_end: Literal[1, 2] = len(l_end)
+    len_l_end: Literal[1, 2] = cast(Literal[1, 2], len(l_end))
 
     # Check if the file stream is a buffered text stream (text instead of binary)
     is_text: bool = isinstance(m_file, io.TextIOWrapper)
@@ -207,15 +205,7 @@ def reverse_readline(
 
     # If the file size is within desired RAM limit, just reverse it in memory.
     # Gzip files must use this method because there is no way to negative seek.
-    # For windows, we also read the whole file.
-    if (
-        platform.system() == "Windows"  # TODO: platform is not important, len_l_end is
-        or (
-            len_l_end != 1
-        )  # TODO: the following code wouldn't work for "\r\n" as its len is 2
-        or file_size < max_mem
-        or isinstance(m_file, gzip.GzipFile)
-    ):
+    if file_size < max_mem or isinstance(m_file, gzip.GzipFile):
         for line in reversed(m_file.readlines()):
             yield (line if isinstance(line, str) else line.decode("utf-8"))
 
@@ -230,20 +220,21 @@ def reverse_readline(
 
         buffer: str = ""
         m_file.seek(0, 2)
+        eof_pos = m_file.tell()  # need end of file to skip first empty line
 
         while True:
             l_end_pos: int = buffer.rfind(l_end)
-            pt_pos: int = (
-                m_file.tell()
-            )  # pointer position (also size of remaining file to read)
+            # Pointer position (also size of remaining file to read)
+            pt_pos: int = m_file.tell()
 
             # Line ending found within buffer
             if l_end_pos != -1:
                 line = buffer[l_end_pos + len_l_end :]
                 buffer = buffer[:l_end_pos]  # buffer doesn't include l_end
-                if pt_pos != 0 or l_end_pos != 0:  # TODO: why is this condition needed?
-                    line += l_end
-                yield line
+
+                # Skip first match (the last line ending)
+                if l_end_pos != eof_pos:
+                    yield line + l_end
 
             # Line ending not in current buffer, load next block into the buffer
             elif pt_pos > 0:
@@ -254,7 +245,7 @@ def reverse_readline(
                 else:
                     buffer += m_file.read(to_read).decode("utf-8")
 
-                # Move pointer forward  # TODO: why pointer is moved forward again?
+                # Move pointer forward
                 m_file.seek(pt_pos - to_read)
 
                 # Add a l_end to the start of file

tests/test_io.py

@@ -143,11 +143,11 @@ def test_file_with_empty_lines(self, l_end):
 
         with ScratchDir("."):
             # Test text file
-            with open(filename, "w", newline="", encoding="utf-8") as file:
+            with open(filename, "wb") as file:
                 for line in contents:
-                    file.write(line)
+                    file.write(line.encode())
 
-            with zopen(filename, mode="r") as file:
+            with open(filename, mode="r", newline="") as file:
                 revert_contents = tuple(reverse_readline(file))
             assert revert_contents[::-1] == contents
 
@@ -175,14 +175,19 @@ def test_line_ending(self, l_end):
         file_name = "test_file.txt"
 
         with ScratchDir("."):
-            with open(file_name, "w", newline="", encoding="utf-8") as file:
+            with open(file_name, "wb") as file:
                 for line in contents:
-                    file.write(line)
+                    file.write(line.encode())
 
             # Test text mode
             with open(file_name, "r", encoding="utf-8") as file:
                 for idx, line in enumerate(reverse_readline(file)):
-                    assert line == contents[len(contents) - idx - 1]
+                    # Open text in "r" mode would trigger OS
+                    # line ending handing
+                    assert (
+                        line.rstrip(os.linesep) + l_end
+                        == contents[len(contents) - idx - 1]
+                    )
                     assert isinstance(line, str)
 
             # # TODO: Test binary mode