obsidian-plugin/lllin000_PaperForge
1
0
Fork 0
mirror of https://github.com/lllin000/PaperForge.git synced 2026-07-22 06:50:53 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

fix(pr3): remove same-column guard from body_blocks, fix _page_width_for_zone_block arg

Browse source
This commit is contained in:
LLLin000 2026-07-04 22:19:12 +08:00
parent 53a5336b6a
commit 2216ac7d87
2 changed files with 16 additions and 5 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

7
paperforge/worker/ocr_document.py
View file

@ -1591,7 +1591,7 @@ def infer_zones(
_zone_page_width = _page_width_for_zone_block(
ref_heading_block if ref_heading_block else (blocks[0] if blocks else {}),
blocks,
anchors,
(anchors or {}).get("body_family_anchor"),
)
body_blocks = [
@ -1609,10 +1609,7 @@ def infer_zones(
)
and (
(body_end_page is None or int(block.get("page", 0) or 0) <= body_end_page)
or (
_is_above_same_page_reference_heading(block, refs_start_page, ref_heading_top if ref_heading_block else None)
and _is_in_same_reference_column(block, ref_heading_block, _zone_page_width)
)
or _is_above_same_page_reference_heading(block, refs_start_page, ref_heading_top if ref_heading_block else None)
)
and not _is_reference_item_candidate(block)
and not _is_reference_heading_candidate(block)

14
tests/test_ocr_document.py
View file

@ -5601,3 +5601,17 @@ def test_same_page_tail_column_aware() -> None:
# Left-column body below ref heading but in different column -> NOT in tail zone
assert "p3_b1" not in region_bus["tail_nonref_hold_zone"]["block_ids"], \
"Left-column body should not be in tail_nonref_hold_zone despite being below ref heading"
def test_is_in_same_reference_column_cross_column() -> None:
from paperforge.worker.ocr_document import _is_in_same_reference_column
pw = 1200.0
ref_heading = {"bbox": [700, 100, 1100, 140], "page_width": pw}
left_block = {"bbox": [100, 200, 500, 250], "page_width": pw}
right_block = {"bbox": [700, 200, 1100, 250], "page_width": pw}
assert _is_in_same_reference_column(left_block, ref_heading, pw) is False, "Left-column block should not be in same column as right-column ref heading"
assert _is_in_same_reference_column(right_block, ref_heading, pw) is True, "Right-column block should be in same column as right-column ref heading"
single_ref = {"bbox": [50, 100, 1150, 140], "page_width": pw}
single_block = {"bbox": [50, 200, 1150, 250], "page_width": pw}
assert _is_in_same_reference_column(single_block, single_ref, pw) is True, "Full-width blocks on single-column page should return True"
assert _is_in_same_reference_column(left_block, None, pw) is True, "None ref heading should return True (conservative)"