DeepTutor: Lifelong Personalized Tutoring. https://deeptutor.info/.
Python ai-agentsai-tutorclawbotcli-tooldeepresearchinteractive-learninglearningmulti-agent-systemsrag
style(reading): format combined extraction and store changes
Bingxi Zhao (Frank)0b3d5be873d8
diff --git a/deeptutor/reading/extract.py b/deeptutor/reading/extract.pyindex 8330abb31..0d41ca4c8 100644--- a/deeptutor/reading/extract.py+++ b/deeptutor/reading/extract.py@@ -238,6 +238,7 @@ def _pdf_pages_with_image_markers()return targeted, _media_for_units(targeted, pdf_images.collection.images)+def _extract_epub(data: bytes, filename: str) -> Extraction:"""Preserve EPUB spine order so browser and assistant locators agree."""from deeptutor.utils.document_extractor import DocumentExtractionError, extract_epub_spinediff --git a/deeptutor/reading/store.py b/deeptutor/reading/store.pyindex b93beb540..3f62706ea 100644--- a/deeptutor/reading/store.py+++ b/deeptutor/reading/store.py@@ -328,7 +328,9 @@ class ReadingStore:"bytes": len(item.data),})- _atomic_write(stage_dir / MEDIA_INDEX_NAME, json.dumps(media_rows, ensure_ascii=False))+ _atomic_write(+ stage_dir / MEDIA_INDEX_NAME, json.dumps(media_rows, ensure_ascii=False)+ )if extraction.render_mode != "text":raw_dir = stage_dir / RAW_DIR@@ -558,9 +560,7 @@ class ReadingStore:for state_dir in (ANNOTATIONS_DIR, POSITIONS_DIR, BOOKMARKS_DIR, REVISIONS_DIR):source_state_dir = material_dir / state_dirif source_state_dir.is_dir():- shutil.copytree(- source_state_dir, stage_dir / state_dir, dirs_exist_ok=True- )+ shutil.copytree(source_state_dir, stage_dir / state_dir, dirs_exist_ok=True)for state_name in (ANNOTATIONS_NAME, POSITION_NAME):source_state = material_dir / state_nameif source_state.is_file():@@ -956,9 +956,7 @@ class ReadingStore:path = self._dir(material_id) / MEDIA_DIR / cleanreturn path if path.is_file() else None- def update_media_captions(- self, material_id: str, captions: Mapping[str, str]- ) -> int:+ def update_media_captions(self, material_id: str, captions: Mapping[str, str]) -> int:"""Write per-image captions into the media index. Returns rows changed.*captions* maps an image name to its text. Names absent from the index@@ -991,6 +989,7 @@ class ReadingStore:if changed:_atomic_write(index_path, json.dumps(rows, ensure_ascii=False))return changed+def render_path(self, material_id: str) -> Path | None:"""Browser-ready EPUB archive; original bytes for all other formats."""manifest = self.manifest(material_id)