Docling layout block audit

original.pdf docling.json layout_blocks.json layout_blocks.tsv layout_blocks.jsonl excluded_blocks.html excluded_blocks.tsv final_body_blocks.tsv visual_assets.tsv visual_assets.json

Summary

{
  "blocks": 149,
  "included_in_body": 9,
  "gray_background_blocks": 13,
  "framed_blocks": 7,
  "by_role_guess": {
    "page_header": 13,
    "metadata": 11,
    "title_candidate": 2,
    "affiliation": 7,
    "abstract_candidate": 2,
    "body_heading": 4,
    "body": 9,
    "page_footer": 15,
    "caption": 5,
    "back_matter_heading": 5,
    "body_candidate_excluded": 20,
    "unknown_text": 13,
    "back_matter_text": 1,
    "reference": 42
  },
  "by_docling_label": {
    "page_header": 14,
    "text": 56,
    "section_header": 13,
    "footnote": 4,
    "page_footer": 15,
    "caption": 5,
    "list_item": 42
  },
  "by_body_decision_reason": {
    "document_ui": 1,
    "document_web_address": 1,
    "first_page_front_matter_heading": 2,
    "front_matter_author_line": 1,
    "first_page_author_or_affiliation": 5,
    "implicit_abstract": 1,
    "first_page_summary": 1,
    "body_heading": 2,
    "body": 9,
    "first_page_metadata": 8,
    "docling_page_footer": 3,
    "docling_page_header": 3,
    "outside_body_flow_caption": 2,
    "back_matter_heading": 1,
    "after_back_matter_stop": 109
  },
  "visual_assets": {
    "count": 6,
    "indexable_count": 6,
    "suppressed_count": 0,
    "by_type": {
      "figure": 6
    },
    "by_caption_source": {
      "direct_caption_ref": 5,
      "missing_caption": 1
    },
    "by_duplicate_reason": {
      "": 6
    },
    "missing_caption_count": 1
  }
}

Diff Summary

{
  "parsed_text_blocks": 149,
  "final_body_blocks": 9,
  "excluded_blocks": {
    "count": 140,
    "by_reason": {
      "after_back_matter_stop": 109,
      "first_page_metadata": 8,
      "first_page_author_or_affiliation": 5,
      "docling_page_footer": 3,
      "docling_page_header": 3,
      "body_heading": 2,
      "first_page_front_matter_heading": 2,
      "outside_body_flow_caption": 2,
      "back_matter_heading": 1,
      "document_ui": 1,
      "document_web_address": 1,
      "first_page_summary": 1,
      "front_matter_author_line": 1,
      "implicit_abstract": 1
    },
    "by_role_guess": {
      "reference": 42,
      "body_candidate_excluded": 20,
      "page_footer": 15,
      "page_header": 13,
      "unknown_text": 13,
      "metadata": 11,
      "affiliation": 7,
      "back_matter_heading": 5,
      "caption": 5,
      "body_heading": 4,
      "abstract_candidate": 2,
      "title_candidate": 2,
      "back_matter_text": 1
    },
    "by_risk_level": {
      "low": 115,
      "medium": 25
    },
    "high_risk_count": 0,
    "medium_risk_count": 25
  },
  "char_counts": {
    "parsed_text_chars": 36458,
    "final_body_chars": 7680,
    "excluded_chars": 28778
  }
}

Truncation

{
  "truncated": true,
  "message": "Body extraction stopped at page 3 block #/texts/38: Supporting Information Figure S9). Before the cell is charged to ca. 3.1 V, a solid-solution reaction occurs with negligible variation of cell parameters (Li2MoO3 → Li1.75MoO3 + 0.25Li + ).. 109 following text blocks were excluded as after_back_matter_stop.",
  "stop_trigger": {
    "ref": "#/texts/38",
    "page": 3,
    "layout_order": 39,
    "role_guess": "back_matter_heading",
    "body_decision_reason": "back_matter_heading",
    "text_preview": "Supporting Information Figure S9). Before the cell is charged to ca. 3.1 V, a solid-solution reaction occurs with negligible variation of cell parameters (Li2MoO3 → Li1.75MoO3 + 0.25Li + )."
  },
  "first_truncated_block": {
    "ref": "#/texts/39",
    "page": 3,
    "layout_order": 40,
    "role_guess": "body_candidate_excluded",
    "body_decision_reason": "after_back_matter_stop",
    "text_preview": "As the cell is further charged to ca. 3.6 V, the (003) di ff raction peak shifts to lower angles and gradually becomes broader. Meanwhile, the intensities of the other peaks indexed to Li2MoO3 decrease while those of ne…"
  },
  "truncated_block_count": 109,
  "truncated_pages": [
    3,
    4,
    5,
    6,
    7
  ],
  "by_role_guess": {
    "reference": 42,
    "body_candidate_excluded": 20,
    "unknown_text": 13,
    "page_footer": 10,
    "page_header": 9,
    "back_matter_heading": 4,
    "metadata": 4,
    "caption": 3,
    "body_heading": 2,
    "affiliation": 1,
    "back_matter_text": 1
  }
}

Page Overlays

Green = final body chunks. Orange/purple asset boxes = actual figure/table assets used by ingestion. Cyan dashed text boxes = text blocks consumed by those assets as caption continuations. Red STOP = truncation trigger. Red boxes = blocks after truncation.

Page 1

Page 2

Page 3

Page 4

Page 5

Page 6

Page 7

Visual Assets

这里对齐真实图表资产提取链路。caption_source=embedded_table_cell 表示表注来自 Docling table cell，不会出现在 text block 审计差集里；caption_continuation_used_by_asset 表示某个 text block 已被图表 caption 吸收，不应按普通 metadata 解读。

Blocks