represent all paths internally as PurePath

entangled/code_reader.py

@@ -1,5 +1,5 @@
 from dataclasses import dataclass, field
-from pathlib import Path
+from pathlib import Path, PurePath
 
 import mawk
 import re
@@ -18,9 +18,9 @@ class Frame:
 class CodeReader(mawk.RuleSet):
     """Reads an annotated code file."""
 
-    def __init__(self, path: str, refs: ReferenceMap):
+    def __init__(self, path: PurePath, refs: ReferenceMap):
         self.location = TextLocation(path, 0)
-        self.stack: list[Frame] = [Frame(ReferenceId("#root#", "", -1), "")]
+        self.stack: list[Frame] = [Frame(ReferenceId("#root#", PurePath("-"), -1), "")]
         self.refs: ReferenceMap = refs
 
     @property
@@ -56,7 +56,7 @@ def on_block_begin(self, m: re.Match):
 
         self.stack.append(
             Frame(
-                ReferenceId(m["ref_name"], m["source"], ref_count), m["indent"], content
+                ReferenceId(m["ref_name"], PurePath(m["source"]), ref_count), m["indent"], content
             )
         )
         return []

entangled/commands/stitch.py

@@ -57,7 +57,7 @@ def stitch(*, force: bool = False, show: bool = False):
                 logging.debug("reading `%s`", path)
                 t.update(path)
                 with open(path, "r") as f:
-                    CodeReader(str(path), refs).run(f.read())
+                    CodeReader(path, refs).run(f.read())
 
             for path in input_file_list:
                 t.write(path, stitch_markdown(refs, content[path]), [])

entangled/document.py

@@ -3,6 +3,7 @@
 from collections import defaultdict
 from functools import singledispatchmethod
 from itertools import chain
+from pathlib import PurePath
 
 from .config import Language, AnnotationMethod, config
 from .properties import Property, get_attribute
@@ -17,7 +18,7 @@ def length(iter: Iterable[Any]) -> int:
 @dataclass
 class ReferenceId:
     name: str
-    file: str
+    file: PurePath
     ref_count: int
 
     def __hash__(self):
@@ -71,7 +72,7 @@ def by_name(self, n: str) -> Iterable[CodeBlock]:
 
         return (self.map[r] for r in self.index[n])
 
-    def new_id(self, filename: str, name: str) -> ReferenceId:
+    def new_id(self, filename: PurePath, name: str) -> ReferenceId:
         c = length(filter(lambda r: r.file == filename, self.index[name]))
         return ReferenceId(name, filename, c)
 

entangled/markdown_reader.py

@@ -1,6 +1,6 @@
 from typing import Optional
 from copy import copy
-from pathlib import Path
+from pathlib import Path, PurePath
 
 import re
 import mawk
@@ -20,8 +20,8 @@ class MarkdownLexer(mawk.RuleSet):
     content."""
     def __init__(
         self,
-        filename: str
-    ):
+        filename: Path
+    ): 
         self.location = TextLocation(filename)
         self.raw_content: list[RawContent] = []
         self.inside_codeblock: bool = False
@@ -122,13 +122,13 @@ def read_markdown_file(
     -> tuple[ReferenceMap, list[Content]]:
 
     with open(path, "r") as f:
-        path_str = str(path.resolve().relative_to(Path.cwd()))
-        return read_markdown_string(f.read(), path_str, refs, hooks)
+        rel_path = path.resolve().relative_to(Path.cwd())
+        return read_markdown_string(f.read(), rel_path, refs, hooks)
 
 
 def read_markdown_string(
         text: str,
-        path_str: str = "-",
+        path_str: Path = Path("-"),
         refs: ReferenceMap | None = None,
         hooks: list[HookBase] | None = None) \
         -> tuple[ReferenceMap, list[Content]]:

entangled/tangle.py

@@ -95,7 +95,7 @@ def on_begin(self):
         if self.cb.header is not None:
             result.append(self.cb.header)
         result.append(
-            f"{self.cb.language.comment.open} ~/~ begin <<{self.ref.file}#{self.ref.name}>>[{count}]{self.close_comment}"
+            f"{self.cb.language.comment.open} ~/~ begin <<{self.ref.file.as_posix()}#{self.ref.name}>>[{count}]{self.close_comment}"
         )
         return result
 

entangled/text_location.py

@@ -1,9 +1,10 @@
 from dataclasses import dataclass
+from pathlib import PurePath
 
 
 @dataclass
 class TextLocation:
-    filename: str
+    filename: PurePath
     line_number: int = 0
 
     def __str__(self):

test/test_os_interop.py

@@ -2,7 +2,7 @@
 from entangled.tangle import tangle_ref
 from entangled.code_reader import CodeReader
 
-from pathlib import Path
+from pathlib import Path, PurePath
 import os
 from shutil import copytree, move
 from contextlib import chdir
@@ -13,7 +13,7 @@ def test_tangle_ref(data, tmp_path):
     with chdir(tmp_path / "os_interop"):
         refs, _ = read_markdown_file(Path("doc/index.md"))
         tangled, deps = tangle_ref(refs, "src/euler_number.c")
-        assert deps == {"doc/index.md"}
+        assert deps == {PurePath("doc/index.md")}
         with open("src/euler_number.c", "r") as f:
             assert f.read().strip() == tangled.strip()
 

test/test_tangle.py

@@ -1,7 +1,7 @@
 from entangled.markdown_reader import read_markdown_file
 from entangled.tangle import tangle_ref
 from entangled.code_reader import CodeReader
-from pathlib import Path
+from pathlib import Path, PurePath
 import os
 from shutil import copytree, move
 from contextlib import chdir
@@ -12,7 +12,7 @@ def test_tangle_ref(data, tmp_path):
     with chdir(tmp_path / "hello-world"):
         refs, _ = read_markdown_file(Path("hello-world.md"))
         tangled, deps = tangle_ref(refs, "hello_world.cc")
-        assert deps == {"hello-world.md"}
+        assert deps == {PurePath("hello-world.md")}
         with open("hello_world.cc", "r") as f:
             assert f.read() == tangled