From b43b56ed04a94c4157262160d4f8b1bb18a7d0c3 Mon Sep 17 00:00:00 2001
From: Ray <mailtangyu@gmail.com>
Date: Fri, 24 Jul 2026 03:32:27 +0800
Subject: [PATCH] chore: trim non-essential comments

---
 pageindex/__init__.py      | 4 ----
 pageindex/page_index.py    | 6 +-----
 pageindex/page_index_md.py | 9 +--------
 pageindex/tokens.py        | 5 +----
 pageindex/utils.py         | 5 +----
 5 files changed, 4 insertions(+), 25 deletions(-)

diff --git a/pageindex/__init__.py b/pageindex/__init__.py
index ab4350d..1b64ea9 100644
--- a/pageindex/__init__.py
+++ b/pageindex/__init__.py
@@ -9,8 +9,6 @@ _chatgpt_key = _os.getenv("CHATGPT_API_KEY")
 if not _os.getenv("OPENAI_API_KEY") and _chatgpt_key:
     _os.environ["OPENAI_API_KEY"] = _chatgpt_key
 
-# Legacy (pre-SDK) exports. Cheap to import eagerly: litellm/PyPDF2 load
-# inside the functions that use them, not at module import.
 from .index.page_index import *  # noqa: E402
 from .index.page_index_md import md_to_tree
 from .retrieve import get_document, get_document_structure, get_page_content
@@ -71,8 +69,6 @@ __all__ = [
 
 
 def __getattr__(name):
-    # `pageindex.utils` / `pageindex.page_index_md` attribute access without an
-    # explicit submodule import.
     if name in ("utils", "page_index_md"):
         import importlib
         return importlib.import_module(f".{name}", __name__)
diff --git a/pageindex/page_index.py b/pageindex/page_index.py
index 8cace07..b42f98a 100644
--- a/pageindex/page_index.py
+++ b/pageindex/page_index.py
@@ -1,8 +1,4 @@
-# pageindex/page_index.py
-# Compatibility shim. The PDF indexing pipeline now lives in
-# pageindex/index/page_index.py (the single source of truth). This module
-# re-exports it so legacy imports (`from pageindex.page_index import ...`,
-# `from pageindex import page_index`) keep working.
+# pageindex/page_index.py — re-exports from index/page_index.py
 import sys
 import types
 from .index.page_index import *  # noqa: F401,F403,E402
diff --git a/pageindex/page_index_md.py b/pageindex/page_index_md.py
index d570b06..44713f3 100644
--- a/pageindex/page_index_md.py
+++ b/pageindex/page_index_md.py
@@ -1,9 +1,2 @@
-# pageindex/page_index_md.py
-# Compatibility shim. The Markdown indexing pipeline now lives in
-# pageindex/index/page_index_md.py (the single source of truth). This module
-# re-exports it so legacy imports keep working.
-#
-# The canonical md_to_tree coerces legacy 'yes'/'no' string flags itself (a
-# bare 'no' would otherwise be truthy) — this shim used to duplicate that
-# coercion in its own wrapper; now it just re-exports the canonical function.
+# pageindex/page_index_md.py — re-exports from index/page_index_md.py
 from .index.page_index_md import *  # noqa: F401,F403,E402
diff --git a/pageindex/tokens.py b/pageindex/tokens.py
index ab4386c..a47648c 100644
--- a/pageindex/tokens.py
+++ b/pageindex/tokens.py
@@ -1,8 +1,5 @@
 # pageindex/tokens.py
-# Leaf utility so both the parser and index layers can count tokens without
-# the parser reaching back into pageindex.index (a reverse/horizontal
-# dependency). Depends only on litellm (imported lazily — it's several
-# seconds and a network fetch, and cloud-only paths never need it).
+# Shared by parser and index layers (avoids a reverse dependency).
 
 
 def count_tokens(text, model=None):
diff --git a/pageindex/utils.py b/pageindex/utils.py
index ef5e793..c8d1773 100644
--- a/pageindex/utils.py
+++ b/pageindex/utils.py
@@ -1,7 +1,4 @@
-# pageindex/utils.py
-# Compatibility shim. The indexing utilities now live in pageindex/index/utils.py,
-# which is the single source of truth. This module re-exports them so legacy
-# imports (`from pageindex.utils import ...`) keep working.
+# pageindex/utils.py — re-exports from index/utils.py
 from .index.utils import *  # noqa: F401,F403,E402
 # Legacy 0.2.x alias. index.utils keeps it private (_config) so its star-export
 # can't shadow the pageindex.config submodule; re-expose it only in this shim.