e3395ae8cf
Some checks failed
CI / go-lint (push) Has been skipped
CI / python-lint (push) Has been skipped
CI / nodejs-lint (push) Has been skipped
CI / test-go-school (push) Successful in 28s
CI / test-go-edu-search (push) Successful in 29s
CI / test-python-klausur (push) Failing after 1m57s
CI / test-python-agent-core (push) Successful in 16s
CI / test-nodejs-website (push) Successful in 41s
1. Filter words inside image_overlays (removes OCR from images) 2. Ghost filter: only remove single-char border artifacts, not multi-char like (= which is real content 3. Skip first-row header detection for zones with image_overlays (merged geometry creates artificial gaps) Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
444 lines
20 KiB
Python
444 lines
20 KiB
Python
"""
|
|
Tests for grid_editor_api zone merging, heading detection, and ghost filtering.
|
|
|
|
Covers:
|
|
- _merge_content_zones_across_boxes: zone merging logic
|
|
- _detect_heading_rows_by_color: heading detection by color + height
|
|
- _filter_border_ghosts: single-char ghost detection
|
|
- _detect_header_rows: skip_first_row_header flag
|
|
"""
|
|
|
|
import sys
|
|
sys.path.insert(0, '/app')
|
|
|
|
import pytest
|
|
from cv_vocab_types import PageZone, DetectedBox
|
|
from grid_editor_api import (
|
|
_merge_content_zones_across_boxes,
|
|
_filter_border_ghosts,
|
|
_detect_header_rows,
|
|
_detect_heading_rows_by_color,
|
|
)
|
|
|
|
|
|
# ---------------------------------------------------------------------------
|
|
# _merge_content_zones_across_boxes
|
|
# ---------------------------------------------------------------------------
|
|
|
|
class TestMergeContentZonesAcrossBoxes:
|
|
"""Test zone merging across box zones."""
|
|
|
|
def test_no_merge_when_less_than_3_zones(self):
|
|
"""Fewer than 3 zones → no merge possible."""
|
|
zones = [
|
|
PageZone(index=0, zone_type="content", y=0, height=100, x=0, width=500),
|
|
PageZone(index=1, zone_type="box", y=100, height=50, x=50, width=400,
|
|
box=DetectedBox(x=50, y=100, width=400, height=50, confidence=0.9)),
|
|
]
|
|
result = _merge_content_zones_across_boxes(zones, 0, 500)
|
|
assert len(result) == 2
|
|
assert result[0].zone_type == "content"
|
|
assert result[1].zone_type == "box"
|
|
|
|
def test_merge_content_box_content(self):
|
|
"""[content, box, content] → [merged_content with overlay]."""
|
|
zones = [
|
|
PageZone(index=0, zone_type="content", y=0, height=100, x=0, width=500),
|
|
PageZone(index=1, zone_type="box", y=100, height=50, x=50, width=400,
|
|
box=DetectedBox(x=50, y=100, width=400, height=50, confidence=0.9)),
|
|
PageZone(index=2, zone_type="content", y=150, height=200, x=0, width=500),
|
|
]
|
|
result = _merge_content_zones_across_boxes(zones, 0, 500)
|
|
assert len(result) == 1
|
|
merged = result[0]
|
|
assert merged.zone_type == "content"
|
|
assert merged.y == 0
|
|
assert merged.height == 350 # 0 to 350
|
|
assert len(merged.image_overlays) == 1
|
|
assert merged.image_overlays[0]["y"] == 100
|
|
assert merged.image_overlays[0]["height"] == 50
|
|
|
|
def test_box_at_start_not_merged(self):
|
|
"""Box at the start (not between contents) stays separate."""
|
|
zones = [
|
|
PageZone(index=0, zone_type="box", y=0, height=50, x=50, width=400,
|
|
box=DetectedBox(x=50, y=0, width=400, height=50, confidence=0.9)),
|
|
PageZone(index=1, zone_type="content", y=50, height=100, x=0, width=500),
|
|
PageZone(index=2, zone_type="box", y=150, height=50, x=50, width=400,
|
|
box=DetectedBox(x=50, y=150, width=400, height=50, confidence=0.9)),
|
|
PageZone(index=3, zone_type="content", y=200, height=200, x=0, width=500),
|
|
]
|
|
result = _merge_content_zones_across_boxes(zones, 0, 500)
|
|
# Box at start stays, then content+box+content merges
|
|
assert len(result) == 2
|
|
assert result[0].zone_type == "box"
|
|
assert result[1].zone_type == "content"
|
|
assert len(result[1].image_overlays) == 1
|
|
|
|
def test_consecutive_boxes_not_merged(self):
|
|
"""[content, box, box, content] → no merge (consecutive boxes rare in practice)."""
|
|
zones = [
|
|
PageZone(index=0, zone_type="content", y=0, height=100, x=0, width=500),
|
|
PageZone(index=1, zone_type="box", y=100, height=50, x=50, width=400,
|
|
box=DetectedBox(x=50, y=100, width=400, height=50, confidence=0.9)),
|
|
PageZone(index=2, zone_type="box", y=150, height=30, x=60, width=380,
|
|
box=DetectedBox(x=60, y=150, width=380, height=30, confidence=0.8)),
|
|
PageZone(index=3, zone_type="content", y=180, height=200, x=0, width=500),
|
|
]
|
|
result = _merge_content_zones_across_boxes(zones, 0, 500)
|
|
# Two consecutive boxes: the algorithm only merges [content, box, content]
|
|
# pairs, so consecutive boxes break the pattern.
|
|
assert len(result) == 4
|
|
|
|
def test_zone_reindexing(self):
|
|
"""Zone indices are re-numbered after merging."""
|
|
zones = [
|
|
PageZone(index=0, zone_type="content", y=0, height=100, x=0, width=500),
|
|
PageZone(index=1, zone_type="box", y=100, height=50, x=50, width=400,
|
|
box=DetectedBox(x=50, y=100, width=400, height=50, confidence=0.9)),
|
|
PageZone(index=2, zone_type="content", y=150, height=200, x=0, width=500),
|
|
]
|
|
result = _merge_content_zones_across_boxes(zones, 0, 500)
|
|
assert result[0].index == 0
|
|
|
|
def test_no_boxes_passthrough(self):
|
|
"""All-content zones pass through unchanged."""
|
|
zones = [
|
|
PageZone(index=0, zone_type="content", y=0, height=100, x=0, width=500),
|
|
PageZone(index=1, zone_type="content", y=100, height=100, x=0, width=500),
|
|
]
|
|
result = _merge_content_zones_across_boxes(zones, 0, 500)
|
|
assert len(result) == 2
|
|
|
|
def test_typical_vocab_page_pattern(self):
|
|
"""Typical pattern: [box(VOCABULARY), content, box(image), content]
|
|
→ box stays, content+box+content merges."""
|
|
zones = [
|
|
PageZone(index=0, zone_type="box", y=10, height=40, x=50, width=400,
|
|
box=DetectedBox(x=50, y=10, width=400, height=40, confidence=0.95)),
|
|
PageZone(index=1, zone_type="content", y=60, height=50, x=0, width=500),
|
|
PageZone(index=2, zone_type="box", y=120, height=85, x=50, width=400,
|
|
box=DetectedBox(x=50, y=120, width=400, height=85, confidence=0.8)),
|
|
PageZone(index=3, zone_type="content", y=210, height=500, x=0, width=500),
|
|
]
|
|
result = _merge_content_zones_across_boxes(zones, 0, 500)
|
|
assert len(result) == 2
|
|
assert result[0].zone_type == "box" # VOCABULARY header box stays
|
|
assert result[1].zone_type == "content" # merged content zone
|
|
assert result[1].y == 60
|
|
assert result[1].height == 710 - 60 # 60 to 710
|
|
assert len(result[1].image_overlays) == 1
|
|
assert result[1].image_overlays[0]["y"] == 120
|
|
# Check reindexing
|
|
assert result[0].index == 0
|
|
assert result[1].index == 1
|
|
|
|
|
|
# ---------------------------------------------------------------------------
|
|
# _detect_heading_rows_by_color
|
|
# ---------------------------------------------------------------------------
|
|
|
|
class TestDetectHeadingRowsByColor:
|
|
"""Test heading detection by color + height."""
|
|
|
|
def _make_word_box(self, text, left, top, width, height, color="black"):
|
|
return {
|
|
"text": text,
|
|
"left": left,
|
|
"top": top,
|
|
"width": width,
|
|
"height": height,
|
|
"color_name": color,
|
|
"conf": 90,
|
|
}
|
|
|
|
def _make_zone(self, cells, rows, columns, zone_index=0,
|
|
bbox_x=0, bbox_y=0, bbox_w=800, bbox_h=1000):
|
|
return {
|
|
"zone_index": zone_index,
|
|
"zone_type": "content",
|
|
"bbox_px": {"x": bbox_x, "y": bbox_y, "w": bbox_w, "h": bbox_h},
|
|
"cells": cells,
|
|
"rows": rows,
|
|
"columns": columns,
|
|
}
|
|
|
|
def test_blue_heading_detected(self):
|
|
"""Row with all blue words + taller height → heading."""
|
|
# Normal rows: height ~20
|
|
normal_cells = []
|
|
for ri in range(5):
|
|
normal_cells.append({
|
|
"cell_id": f"Z0_R{ri:02d}_C0",
|
|
"zone_index": 0,
|
|
"row_index": ri,
|
|
"col_index": 0,
|
|
"col_type": "column_1",
|
|
"text": f"word_{ri}",
|
|
"word_boxes": [
|
|
self._make_word_box(f"word_{ri}", 10, 100 + ri * 30, 80, 20),
|
|
],
|
|
})
|
|
normal_cells.append({
|
|
"cell_id": f"Z0_R{ri:02d}_C1",
|
|
"zone_index": 0,
|
|
"row_index": ri,
|
|
"col_index": 1,
|
|
"col_type": "column_2",
|
|
"text": f"translation_{ri}",
|
|
"word_boxes": [
|
|
self._make_word_box(f"translation_{ri}", 300, 100 + ri * 30, 100, 20),
|
|
],
|
|
})
|
|
|
|
# Heading row (index 2): blue, taller (height 25)
|
|
heading_ri = 2
|
|
for c in normal_cells:
|
|
if c["row_index"] == heading_ri:
|
|
for wb in c["word_boxes"]:
|
|
wb["color_name"] = "blue"
|
|
wb["height"] = 25 # > 1.2 * 20 = 24
|
|
|
|
rows = [
|
|
{"index": ri, "y_min_px": 100 + ri * 30, "y_max_px": 120 + ri * 30, "is_header": False}
|
|
for ri in range(5)
|
|
]
|
|
columns = [
|
|
{"index": 0, "label": "column_1"},
|
|
{"index": 1, "label": "column_2"},
|
|
]
|
|
|
|
zones_data = [self._make_zone(normal_cells, rows, columns)]
|
|
count = _detect_heading_rows_by_color(zones_data, 800, 1000)
|
|
|
|
assert count == 1
|
|
# Check that row 2 is now marked as header
|
|
assert rows[2]["is_header"] is True
|
|
# Check that the heading cell was created
|
|
heading_cells = [c for c in zones_data[0]["cells"] if c["row_index"] == heading_ri]
|
|
assert len(heading_cells) == 1
|
|
assert heading_cells[0]["col_type"] == "heading"
|
|
assert "word_2" in heading_cells[0]["text"]
|
|
assert "translation_2" in heading_cells[0]["text"]
|
|
|
|
def test_black_row_not_heading(self):
|
|
"""Row with black words → not a heading, even if tall."""
|
|
cells = [
|
|
{
|
|
"cell_id": "Z0_R00_C0", "zone_index": 0, "row_index": 0,
|
|
"col_index": 0, "col_type": "column_1", "text": "hello",
|
|
"word_boxes": [self._make_word_box("hello", 10, 100, 80, 25, "black")],
|
|
},
|
|
{
|
|
"cell_id": "Z0_R00_C1", "zone_index": 0, "row_index": 0,
|
|
"col_index": 1, "col_type": "column_2", "text": "world",
|
|
"word_boxes": [self._make_word_box("world", 300, 100, 80, 25, "black")],
|
|
},
|
|
{
|
|
"cell_id": "Z0_R01_C0", "zone_index": 0, "row_index": 1,
|
|
"col_index": 0, "col_type": "column_1", "text": "foo",
|
|
"word_boxes": [self._make_word_box("foo", 10, 130, 80, 20, "black")],
|
|
},
|
|
{
|
|
"cell_id": "Z0_R01_C1", "zone_index": 0, "row_index": 1,
|
|
"col_index": 1, "col_type": "column_2", "text": "bar",
|
|
"word_boxes": [self._make_word_box("bar", 300, 130, 80, 20, "black")],
|
|
},
|
|
]
|
|
rows = [
|
|
{"index": 0, "y_min_px": 100, "y_max_px": 125, "is_header": False},
|
|
{"index": 1, "y_min_px": 130, "y_max_px": 150, "is_header": False},
|
|
]
|
|
columns = [{"index": 0, "label": "column_1"}, {"index": 1, "label": "column_2"}]
|
|
zones_data = [self._make_zone(cells, rows, columns)]
|
|
count = _detect_heading_rows_by_color(zones_data, 800, 1000)
|
|
assert count == 0
|
|
|
|
def test_mixed_color_row_not_heading(self):
|
|
"""Row with some blue and some black words → not a heading."""
|
|
cells = [
|
|
{
|
|
"cell_id": "Z0_R00_C0", "zone_index": 0, "row_index": 0,
|
|
"col_index": 0, "col_type": "column_1", "text": "Unit",
|
|
"word_boxes": [self._make_word_box("Unit", 10, 100, 60, 25, "blue")],
|
|
},
|
|
{
|
|
"cell_id": "Z0_R00_C1", "zone_index": 0, "row_index": 0,
|
|
"col_index": 1, "col_type": "column_2", "text": "normal",
|
|
"word_boxes": [self._make_word_box("normal", 300, 100, 80, 25, "black")],
|
|
},
|
|
{
|
|
"cell_id": "Z0_R01_C0", "zone_index": 0, "row_index": 1,
|
|
"col_index": 0, "col_type": "column_1", "text": "foo",
|
|
"word_boxes": [self._make_word_box("foo", 10, 130, 80, 20, "black")],
|
|
},
|
|
{
|
|
"cell_id": "Z0_R01_C1", "zone_index": 0, "row_index": 1,
|
|
"col_index": 1, "col_type": "column_2", "text": "bar",
|
|
"word_boxes": [self._make_word_box("bar", 300, 130, 80, 20, "black")],
|
|
},
|
|
]
|
|
rows = [
|
|
{"index": 0, "y_min_px": 100, "y_max_px": 125, "is_header": False},
|
|
{"index": 1, "y_min_px": 130, "y_max_px": 150, "is_header": False},
|
|
]
|
|
columns = [{"index": 0, "label": "column_1"}, {"index": 1, "label": "column_2"}]
|
|
zones_data = [self._make_zone(cells, rows, columns)]
|
|
count = _detect_heading_rows_by_color(zones_data, 800, 1000)
|
|
assert count == 0
|
|
|
|
def test_colored_but_not_tall_not_heading(self):
|
|
"""Row with all blue words but normal height → not a heading."""
|
|
cells = [
|
|
{
|
|
"cell_id": "Z0_R00_C0", "zone_index": 0, "row_index": 0,
|
|
"col_index": 0, "col_type": "column_1", "text": "Unit",
|
|
"word_boxes": [self._make_word_box("Unit", 10, 100, 60, 20, "blue")],
|
|
},
|
|
{
|
|
"cell_id": "Z0_R00_C1", "zone_index": 0, "row_index": 0,
|
|
"col_index": 1, "col_type": "column_2", "text": "four",
|
|
"word_boxes": [self._make_word_box("four", 300, 100, 60, 20, "blue")],
|
|
},
|
|
{
|
|
"cell_id": "Z0_R01_C0", "zone_index": 0, "row_index": 1,
|
|
"col_index": 0, "col_type": "column_1", "text": "foo",
|
|
"word_boxes": [self._make_word_box("foo", 10, 130, 80, 20, "black")],
|
|
},
|
|
{
|
|
"cell_id": "Z0_R01_C1", "zone_index": 0, "row_index": 1,
|
|
"col_index": 1, "col_type": "column_2", "text": "bar",
|
|
"word_boxes": [self._make_word_box("bar", 300, 130, 80, 20, "black")],
|
|
},
|
|
]
|
|
rows = [
|
|
{"index": 0, "y_min_px": 100, "y_max_px": 120, "is_header": False},
|
|
{"index": 1, "y_min_px": 130, "y_max_px": 150, "is_header": False},
|
|
]
|
|
columns = [{"index": 0, "label": "column_1"}, {"index": 1, "label": "column_2"}]
|
|
zones_data = [self._make_zone(cells, rows, columns)]
|
|
count = _detect_heading_rows_by_color(zones_data, 800, 1000)
|
|
assert count == 0
|
|
|
|
def test_single_column_zone_skipped(self):
|
|
"""Zones with < 2 columns are skipped."""
|
|
cells = [
|
|
{
|
|
"cell_id": "Z0_R00_C0", "zone_index": 0, "row_index": 0,
|
|
"col_index": 0, "col_type": "column_1", "text": "Unit",
|
|
"word_boxes": [self._make_word_box("Unit", 10, 100, 60, 25, "blue")],
|
|
},
|
|
]
|
|
rows = [{"index": 0, "y_min_px": 100, "y_max_px": 125, "is_header": False}]
|
|
columns = [{"index": 0, "label": "column_1"}]
|
|
zones_data = [self._make_zone(cells, rows, columns)]
|
|
count = _detect_heading_rows_by_color(zones_data, 800, 1000)
|
|
assert count == 0
|
|
|
|
def test_already_header_skipped(self):
|
|
"""Rows already marked is_header are not re-detected."""
|
|
cells = [
|
|
{
|
|
"cell_id": "Z0_R00_C0", "zone_index": 0, "row_index": 0,
|
|
"col_index": 0, "col_type": "spanning_header", "text": "Header",
|
|
"word_boxes": [self._make_word_box("Header", 10, 100, 60, 25, "blue")],
|
|
},
|
|
{
|
|
"cell_id": "Z0_R01_C0", "zone_index": 0, "row_index": 1,
|
|
"col_index": 0, "col_type": "column_1", "text": "foo",
|
|
"word_boxes": [self._make_word_box("foo", 10, 130, 80, 20, "black")],
|
|
},
|
|
{
|
|
"cell_id": "Z0_R01_C1", "zone_index": 0, "row_index": 1,
|
|
"col_index": 1, "col_type": "column_2", "text": "bar",
|
|
"word_boxes": [self._make_word_box("bar", 300, 130, 80, 20, "black")],
|
|
},
|
|
]
|
|
rows = [
|
|
{"index": 0, "y_min_px": 100, "y_max_px": 125, "is_header": True},
|
|
{"index": 1, "y_min_px": 130, "y_max_px": 150, "is_header": False},
|
|
]
|
|
columns = [{"index": 0, "label": "column_1"}, {"index": 1, "label": "column_2"}]
|
|
zones_data = [self._make_zone(cells, rows, columns)]
|
|
count = _detect_heading_rows_by_color(zones_data, 800, 1000)
|
|
assert count == 0
|
|
|
|
|
|
# ---------------------------------------------------------------------------
|
|
# _filter_border_ghosts (Fix 2: single-char only)
|
|
# ---------------------------------------------------------------------------
|
|
|
|
class TestFilterBorderGhosts:
|
|
"""Test that ghost filtering only removes single-char words."""
|
|
|
|
def test_single_char_ghost_removed(self):
|
|
"""Single '|' on a box border → filtered as ghost."""
|
|
box = DetectedBox(x=100, y=100, width=200, height=150, confidence=0.9, border_thickness=3)
|
|
words = [
|
|
{"text": "|", "left": 98, "top": 200, "width": 5, "height": 20},
|
|
{"text": "hello", "left": 150, "top": 150, "width": 80, "height": 20},
|
|
]
|
|
filtered, count = _filter_border_ghosts(words, [box])
|
|
assert count == 1
|
|
assert len(filtered) == 1
|
|
assert filtered[0]["text"] == "hello"
|
|
|
|
def test_multi_char_ghost_kept(self):
|
|
"""Multi-char '(=' on a box border → NOT filtered (real content)."""
|
|
box = DetectedBox(x=648, y=129, width=338, height=125, confidence=0.7, border_thickness=0)
|
|
words = [
|
|
{"text": "(=", "left": 644, "top": 294, "width": 16, "height": 17},
|
|
{"text": "I", "left": 665, "top": 294, "width": 9, "height": 18},
|
|
]
|
|
filtered, count = _filter_border_ghosts(words, [box])
|
|
assert count == 0
|
|
assert len(filtered) == 2
|
|
|
|
def test_single_paren_on_border_removed(self):
|
|
"""Single ')' on border → filtered."""
|
|
box = DetectedBox(x=100, y=100, width=200, height=150, confidence=0.9, border_thickness=2)
|
|
words = [
|
|
{"text": ")", "left": 299, "top": 200, "width": 4, "height": 7},
|
|
]
|
|
filtered, count = _filter_border_ghosts(words, [box])
|
|
assert count == 1
|
|
assert len(filtered) == 0
|
|
|
|
|
|
# ---------------------------------------------------------------------------
|
|
# _detect_header_rows (Fix 3: skip_first_row_header)
|
|
# ---------------------------------------------------------------------------
|
|
|
|
class TestDetectHeaderRowsSkipFlag:
|
|
"""Test skip_first_row_header flag."""
|
|
|
|
def test_first_row_detected_without_flag(self):
|
|
"""Without flag, first row with big gap → header."""
|
|
rows = [
|
|
{"y_min": 100, "y_max": 120, "index": 0},
|
|
{"y_min": 160, "y_max": 180, "index": 1},
|
|
{"y_min": 185, "y_max": 205, "index": 2},
|
|
]
|
|
words = [
|
|
{"height": 20, "top": 105, "left": 10, "width": 80},
|
|
{"height": 20, "top": 165, "left": 10, "width": 80},
|
|
{"height": 20, "top": 190, "left": 10, "width": 80},
|
|
]
|
|
headers = _detect_header_rows(rows, words, 0)
|
|
assert 0 in headers
|
|
|
|
def test_first_row_skipped_with_flag(self):
|
|
"""With skip flag, first row NOT detected even with big gap."""
|
|
rows = [
|
|
{"y_min": 100, "y_max": 120, "index": 0},
|
|
{"y_min": 160, "y_max": 180, "index": 1},
|
|
{"y_min": 185, "y_max": 205, "index": 2},
|
|
]
|
|
words = [
|
|
{"height": 20, "top": 105, "left": 10, "width": 80},
|
|
{"height": 20, "top": 165, "left": 10, "width": 80},
|
|
{"height": 20, "top": 190, "left": 10, "width": 80},
|
|
]
|
|
headers = _detect_header_rows(rows, words, 0, skip_first_row_header=True)
|
|
assert 0 not in headers
|