magic-pdf 0.7.1__py3-none-any.whl → 0.8.1__py3-none-any.whl

magic_pdf/pre_proc/ocr_detect_all_bboxes.py

@@ -133,6 +133,7 @@ def remove_need_drop_blocks(all_bboxes, discarded_blocks):
 
 
 def remove_overlaps_min_blocks(all_bboxes):
+    #  重叠block，小的不能直接删除，需要和大的那个合并成一个更大的。
     #  删除重叠blocks中较小的那些
     need_remove = []
     for block1 in all_bboxes:
@@ -142,9 +143,17 @@ def remove_overlaps_min_blocks(all_bboxes):
                 block2_bbox = block2[:4]
                 overlap_box = get_minbox_if_overlap_by_ratio(block1_bbox, block2_bbox, 0.8)
                 if overlap_box is not None:
-                    bbox_to_remove = next((block for block in all_bboxes if block[:4] == overlap_box), None)
-                    if bbox_to_remove is not None and bbox_to_remove not in need_remove:
-                        need_remove.append(bbox_to_remove)
+                    block_to_remove = next((block for block in all_bboxes if block[:4] == overlap_box), None)
+                    if block_to_remove is not None and block_to_remove not in need_remove:
+                        large_block = block1 if block1 != block_to_remove else block2
+                        x1, y1, x2, y2 = large_block[:4]
+                        sx1, sy1, sx2, sy2 = block_to_remove[:4]
+                        x1 = min(x1, sx1)
+                        y1 = min(y1, sy1)
+                        x2 = max(x2, sx2)
+                        y2 = max(y2, sy2)
+                        large_block[:4] = [x1, y1, x2, y2]
+                        need_remove.append(block_to_remove)
 
     if len(need_remove) > 0:
         for block in need_remove:

magic_pdf/pre_proc/ocr_dict_merge.py

@@ -1,18 +1,15 @@
-from loguru import logger
-
-from magic_pdf.libs.boxbase import __is_overlaps_y_exceeds_threshold, get_minbox_if_overlap_by_ratio, \
-    calculate_overlap_area_in_bbox1_area_ratio, _is_in_or_part_overlap_with_area_ratio
+from magic_pdf.libs.boxbase import (__is_overlaps_y_exceeds_threshold,
+                                    _is_in_or_part_overlap_with_area_ratio,
+                                    calculate_overlap_area_in_bbox1_area_ratio)
 from magic_pdf.libs.drop_tag import DropTag
-from magic_pdf.libs.ocr_content_type import ContentType, BlockType
-from magic_pdf.pre_proc.ocr_span_list_modify import modify_y_axis, modify_inline_equation
-from magic_pdf.pre_proc.remove_bbox_overlap import remove_overlap_between_bbox_for_span
+from magic_pdf.libs.ocr_content_type import BlockType, ContentType
 
 
 # 将每一个line中的span从左到右排序
 def line_sort_spans_by_left_to_right(lines):
     line_objects = []
     for line in lines:
-        # 按照x0坐标排序
+        #  按照x0坐标排序
         line.sort(key=lambda span: span['bbox'][0])
         line_bbox = [
             min(span['bbox'][0] for span in line),  # x0
@@ -21,8 +18,8 @@ def line_sort_spans_by_left_to_right(lines):
             max(span['bbox'][3] for span in line),  # y1
         ]
         line_objects.append({
-            "bbox": line_bbox,
-            "spans": line,
+            'bbox': line_bbox,
+            'spans': line,
         })
     return line_objects
 
@@ -39,16 +36,21 @@ def merge_spans_to_line(spans):
         for span in spans[1:]:
             # 如果当前的span类型为"interline_equation" 或者 当前行中已经有"interline_equation"
             # image和table类型，同上
-            if span['type'] in [ContentType.InterlineEquation, ContentType.Image, ContentType.Table] or any(
-                    s['type'] in [ContentType.InterlineEquation, ContentType.Image, ContentType.Table] for s in
-                    current_line):
+            if span['type'] in [
+                    ContentType.InterlineEquation, ContentType.Image,
+                    ContentType.Table
+            ] or any(s['type'] in [
+                    ContentType.InterlineEquation, ContentType.Image,
+                    ContentType.Table
+            ] for s in current_line):
                 # 则开始新行
                 lines.append(current_line)
                 current_line = [span]
                 continue
 
             # 如果当前的span与当前行的最后一个span在y轴上重叠，则添加到当前行
-            if __is_overlaps_y_exceeds_threshold(span['bbox'], current_line[-1]['bbox']):
+            if __is_overlaps_y_exceeds_threshold(span['bbox'],
+                                                 current_line[-1]['bbox']):
                 current_line.append(span)
             else:
                 # 否则，开始新行
@@ -71,7 +73,8 @@ def merge_spans_to_line_by_layout(spans, layout_bboxes):
         # 遍历spans,将每个span放入对应的layout中
         layout_sapns = []
         for span in spans:
-            if calculate_overlap_area_in_bbox1_area_ratio(span['bbox'], layout_bbox) > 0.6:
+            if calculate_overlap_area_in_bbox1_area_ratio(
+                    span['bbox'], layout_bbox) > 0.6:
                 layout_sapns.append(span)
         # 如果layout_sapns不为空，则放入new_spans中
         if len(layout_sapns) > 0:
@@ -99,12 +102,10 @@ def merge_lines_to_block(lines):
     # 目前不做block拼接,先做个结构,每个block中只有一个line,block的bbox就是line的bbox
     blocks = []
     for line in lines:
-        blocks.append(
-            {
-                "bbox": line["bbox"],
-                "lines": [line],
-            }
-        )
+        blocks.append({
+            'bbox': line['bbox'],
+            'lines': [line],
+        })
     return blocks
 
 
@@ -121,7 +122,8 @@ def sort_blocks_by_layout(all_bboxes, layout_bboxes):
             if block[7] == BlockType.Footnote:
                 continue
             block_bbox = block[:4]
-            if calculate_overlap_area_in_bbox1_area_ratio(block_bbox, layout_bbox) > 0.8:
+            if calculate_overlap_area_in_bbox1_area_ratio(
+                    block_bbox, layout_bbox) > 0.8:
                 layout_blocks.append(block)
 
         # 如果layout_blocks不为空，则放入new_blocks中
@@ -134,7 +136,8 @@ def sort_blocks_by_layout(all_bboxes, layout_bboxes):
     # 如果new_blocks不为空，则对new_blocks中每个block进行排序
     if len(new_blocks) > 0:
         for bboxes_in_layout_block in new_blocks:
-            bboxes_in_layout_block.sort(key=lambda x: x[1])  # 一个layout内部的box，按照y0自上而下排序
+            bboxes_in_layout_block.sort(
+                key=lambda x: x[1])  # 一个layout内部的box，按照y0自上而下排序
             sort_blocks.extend(bboxes_in_layout_block)
 
     # sort_blocks中已经包含了当前页面所有最终留下的block，且已经排好了顺序
@@ -142,9 +145,7 @@ def sort_blocks_by_layout(all_bboxes, layout_bboxes):
 
 
 def fill_spans_in_blocks(blocks, spans, radio):
-    '''
-    将allspans中的span按位置关系，放入blocks中
-    '''
+    """将allspans中的span按位置关系，放入blocks中."""
     block_with_spans = []
     for block in blocks:
         block_type = block[7]
@@ -156,17 +157,15 @@ def fill_spans_in_blocks(blocks, spans, radio):
         block_spans = []
         for span in spans:
             span_bbox = span['bbox']
-            if calculate_overlap_area_in_bbox1_area_ratio(span_bbox, block_bbox) > radio:
+            if calculate_overlap_area_in_bbox1_area_ratio(
+                    span_bbox, block_bbox) > radio:
                 block_spans.append(span)
-
         '''行内公式调整, 高度调整至与同行文字高度一致(优先左侧, 其次右侧)'''
         # displayed_list = []
         # text_inline_lines = []
         # modify_y_axis(block_spans, displayed_list, text_inline_lines)
-
         '''模型识别错误的行间公式, type类型转换成行内公式'''
         # block_spans = modify_inline_equation(block_spans, displayed_list, text_inline_lines)
-
         '''bbox去除粘连'''  # 去粘连会影响span的bbox，导致后续fill的时候出错
         # block_spans = remove_overlap_between_bbox_for_span(block_spans)
 
@@ -182,12 +181,9 @@ def fill_spans_in_blocks(blocks, spans, radio):
 
 
 def fix_block_spans(block_with_spans, img_blocks, table_blocks):
-    '''
-    1、img_block和table_block因为包含caption和footnote的关系，存在block的嵌套关系
-        需要将caption和footnote的text_span放入相应img_block和table_block内的
-        caption_block和footnote_block中
-    2、同时需要删除block中的spans字段
-    '''
+    """1、img_block和table_block因为包含caption和footnote的关系，存在block的嵌套关系
+    需要将caption和footnote的text_span放入相应img_block和table_block内的
+    caption_block和footnote_block中 2、同时需要删除block中的spans字段."""
     fix_blocks = []
     for block in block_with_spans:
         block_type = block['type']
@@ -218,16 +214,13 @@ def merge_spans_to_block(spans: list, block_bbox: list, block_type: str):
     block_spans = []
     # 如果有img_caption，则将img_block中的text_spans放入img_caption_block中
     for span in spans:
-        if calculate_overlap_area_in_bbox1_area_ratio(span['bbox'], block_bbox) > 0.6:
+        if calculate_overlap_area_in_bbox1_area_ratio(span['bbox'],
+                                                      block_bbox) > 0.6:
             block_spans.append(span)
     block_lines = merge_spans_to_line(block_spans)
     # 对line中的span进行排序
     sort_block_lines = line_sort_spans_by_left_to_right(block_lines)
-    block = {
-        'bbox': block_bbox,
-        'type': block_type,
-        'lines': sort_block_lines
-    }
+    block = {'bbox': block_bbox, 'type': block_type, 'lines': sort_block_lines}
     return block, block_spans
 
 
@@ -237,11 +230,7 @@ def make_body_block(span: dict, block_bbox: list, block_type: str):
         'bbox': block_bbox,
         'spans': [span],
     }
-    body_block = {
-        'bbox': block_bbox,
-        'type': block_type,
-        'lines': [body_line]
-    }
+    body_block = {'bbox': block_bbox, 'type': block_type, 'lines': [body_line]}
     return body_block
 
 
@@ -249,13 +238,16 @@ def fix_image_block(block, img_blocks):
     block['blocks'] = []
     # 遍历img_blocks,找到与当前block匹配的img_block
     for img_block in img_blocks:
-        if _is_in_or_part_overlap_with_area_ratio(block['bbox'], img_block['bbox'], 0.95):
+        if _is_in_or_part_overlap_with_area_ratio(block['bbox'],
+                                                  img_block['bbox'], 0.95):
 
             # 创建img_body_block
             for span in block['spans']:
-                if span['type'] == ContentType.Image and img_block['img_body_bbox'] == span['bbox']:
+                if span['type'] == ContentType.Image and img_block[
+                        'img_body_bbox'] == span['bbox']:
                     # 创建img_body_block
-                    img_body_block = make_body_block(span, img_block['img_body_bbox'], BlockType.ImageBody)
+                    img_body_block = make_body_block(
+                        span, img_block['img_body_bbox'], BlockType.ImageBody)
                     block['blocks'].append(img_body_block)
 
                     # 从spans中移除img_body_block中已经放入的span
@@ -265,10 +257,15 @@ def fix_image_block(block, img_blocks):
             # 根据list长度，判断img_block中是否有img_caption
             if img_block['img_caption_bbox'] is not None:
                 img_caption_block, img_caption_spans = merge_spans_to_block(
-                    block['spans'], img_block['img_caption_bbox'], BlockType.ImageCaption
-                )
+                    block['spans'], img_block['img_caption_bbox'],
+                    BlockType.ImageCaption)
                 block['blocks'].append(img_caption_block)
 
+            if img_block['img_footnote_bbox'] is not None:
+                img_footnote_block, img_footnote_spans = merge_spans_to_block(
+                    block['spans'], img_block['img_footnote_bbox'],
+                    BlockType.ImageFootnote)
+                block['blocks'].append(img_footnote_block)
             break
     del block['spans']
     return block
@@ -278,13 +275,17 @@ def fix_table_block(block, table_blocks):
     block['blocks'] = []
     # 遍历table_blocks,找到与当前block匹配的table_block
     for table_block in table_blocks:
-        if _is_in_or_part_overlap_with_area_ratio(block['bbox'], table_block['bbox'], 0.95):
+        if _is_in_or_part_overlap_with_area_ratio(block['bbox'],
+                                                  table_block['bbox'], 0.95):
 
             # 创建table_body_block
             for span in block['spans']:
-                if span['type'] == ContentType.Table and table_block['table_body_bbox'] == span['bbox']:
+                if span['type'] == ContentType.Table and table_block[
+                        'table_body_bbox'] == span['bbox']:
                     # 创建table_body_block
-                    table_body_block = make_body_block(span, table_block['table_body_bbox'], BlockType.TableBody)
+                    table_body_block = make_body_block(
+                        span, table_block['table_body_bbox'],
+                        BlockType.TableBody)
                     block['blocks'].append(table_body_block)
 
                     # 从spans中移除img_body_block中已经放入的span
@@ -294,8 +295,8 @@ def fix_table_block(block, table_blocks):
             # 根据list长度，判断table_block中是否有caption
             if table_block['table_caption_bbox'] is not None:
                 table_caption_block, table_caption_spans = merge_spans_to_block(
-                    block['spans'], table_block['table_caption_bbox'], BlockType.TableCaption
-                )
+                    block['spans'], table_block['table_caption_bbox'],
+                    BlockType.TableCaption)
                 block['blocks'].append(table_caption_block)
 
                 # 如果table_caption_block_spans不为空
@@ -307,8 +308,8 @@ def fix_table_block(block, table_blocks):
             # 根据list长度，判断table_block中是否有table_note
             if table_block['table_footnote_bbox'] is not None:
                 table_footnote_block, table_footnote_spans = merge_spans_to_block(
-                    block['spans'], table_block['table_footnote_bbox'], BlockType.TableFootnote
-                )
+                    block['spans'], table_block['table_footnote_bbox'],
+                    BlockType.TableFootnote)
                 block['blocks'].append(table_footnote_block)
 
             break

magic_pdf/tools/cli.py

@@ -1,53 +1,77 @@
 import os
+from pathlib import Path
+
 import click
 from loguru import logger
-from pathlib import Path
 
-from magic_pdf.rw.DiskReaderWriter import DiskReaderWriter
-from magic_pdf.rw.AbsReaderWriter import AbsReaderWriter
 import magic_pdf.model as model_config
-from magic_pdf.tools.common import parse_pdf_methods, do_parse
 from magic_pdf.libs.version import __version__
+from magic_pdf.rw.AbsReaderWriter import AbsReaderWriter
+from magic_pdf.rw.DiskReaderWriter import DiskReaderWriter
+from magic_pdf.tools.common import do_parse, parse_pdf_methods
 
 
 @click.command()
-@click.version_option(__version__, "--version", "-v", help="display the version and exit")
+@click.version_option(__version__,
+                      '--version',
+                      '-v',
+                      help='display the version and exit')
 @click.option(
-    "-p",
-    "--path",
-    "path",
+    '-p',
+    '--path',
+    'path',
     type=click.Path(exists=True),
     required=True,
-    help="local pdf filepath or directory",
+    help='local pdf filepath or directory',
 )
 @click.option(
-    "-o",
-    "--output-dir",
-    "output_dir",
-    type=str,
-    help="output local directory",
-    default="",
+    '-o',
+    '--output-dir',
+    'output_dir',
+    type=click.Path(),
+    required=True,
+    help='output local directory',
 )
 @click.option(
-    "-m",
-    "--method",
-    "method",
+    '-m',
+    '--method',
+    'method',
     type=parse_pdf_methods,
-    help="""the method for parsing pdf. 
+    help="""the method for parsing pdf.
 ocr: using ocr technique to extract information from pdf.
 txt: suitable for the text-based pdf only and outperform ocr.
 auto: automatically choose the best method for parsing pdf from ocr and txt.
 without method specified, auto will be used by default.""",
-    default="auto",
+    default='auto',
+)
+@click.option(
+    '-d',
+    '--debug',
+    'debug_able',
+    type=bool,
+    help='Enables detailed debugging information during the execution of the CLI commands.',
+    default=False,
+)
+@click.option(
+    '-s',
+    '--start',
+    'start_page_id',
+    type=int,
+    help='The starting page for PDF parsing, beginning from 0.',
+    default=0,
+)
+@click.option(
+    '-e',
+    '--end',
+    'end_page_id',
+    type=int,
+    help='The ending page for PDF parsing, beginning from 0.',
+    default=None,
 )
-def cli(path, output_dir, method):
+def cli(path, output_dir, method, debug_able, start_page_id, end_page_id):
     model_config.__use_inside_model__ = True
-    model_config.__model_mode__ = "full"
-    if output_dir == "":
-        if os.path.isdir(path):
-            output_dir = os.path.join(path, "output")
-        else:
-            output_dir = os.path.join(os.path.dirname(path), "output")
+    model_config.__model_mode__ = 'full'
+    os.makedirs(output_dir, exist_ok=True)
 
     def read_fn(path):
         disk_rw = DiskReaderWriter(os.path.dirname(path))
@@ -63,17 +87,20 @@ def cli(path, output_dir, method):
                 pdf_data,
                 [],
                 method,
+                debug_able,
+                start_page_id=start_page_id,
+                end_page_id=end_page_id,
             )
 
         except Exception as e:
             logger.exception(e)
 
     if os.path.isdir(path):
-        for doc_path in Path(path).glob("*.pdf"):
+        for doc_path in Path(path).glob('*.pdf'):
             parse_doc(doc_path)
     else:
         parse_doc(path)
 
 
-if __name__ == "__main__":
+if __name__ == '__main__':
     cli()

magic_pdf/tools/cli_dev.py

@@ -1,35 +1,32 @@
-import os
 import json as json_parse
-import click
+import os
 from pathlib import Path
-from magic_pdf.libs.path_utils import (
-    parse_s3path,
-    parse_s3_range_params,
-    remove_non_official_s3_args,
-)
-from magic_pdf.libs.config_reader import (
-    get_s3_config,
-)
-from magic_pdf.rw.S3ReaderWriter import S3ReaderWriter
-from magic_pdf.rw.DiskReaderWriter import DiskReaderWriter
-from magic_pdf.rw.AbsReaderWriter import AbsReaderWriter
+
+import click
+
 import magic_pdf.model as model_config
-from magic_pdf.tools.common import parse_pdf_methods, do_parse
+from magic_pdf.libs.config_reader import get_s3_config
+from magic_pdf.libs.path_utils import (parse_s3_range_params, parse_s3path,
+                                       remove_non_official_s3_args)
 from magic_pdf.libs.version import __version__
+from magic_pdf.rw.AbsReaderWriter import AbsReaderWriter
+from magic_pdf.rw.DiskReaderWriter import DiskReaderWriter
+from magic_pdf.rw.S3ReaderWriter import S3ReaderWriter
+from magic_pdf.tools.common import do_parse, parse_pdf_methods
 
 
 def read_s3_path(s3path):
     bucket, key = parse_s3path(s3path)
 
     s3_ak, s3_sk, s3_endpoint = get_s3_config(bucket)
-    s3_rw = S3ReaderWriter(
-        s3_ak, s3_sk, s3_endpoint, "auto", remove_non_official_s3_args(s3path)
-    )
+    s3_rw = S3ReaderWriter(s3_ak, s3_sk, s3_endpoint, 'auto',
+                           remove_non_official_s3_args(s3path))
     may_range_params = parse_s3_range_params(s3path)
     if may_range_params is None or 2 != len(may_range_params):
         byte_start, byte_end = 0, None
     else:
-        byte_start, byte_end = int(may_range_params[0]), int(may_range_params[1])
+        byte_start, byte_end = int(may_range_params[0]), int(
+            may_range_params[1])
     return s3_rw.read_offset(
         remove_non_official_s3_args(s3path),
         byte_start,
@@ -38,51 +35,47 @@ def read_s3_path(s3path):
 
 
 @click.group()
-@click.version_option(__version__, "--version", "-v", help="显示版本信息")
+@click.version_option(__version__, '--version', '-v', help='显示版本信息')
 def cli():
     pass
 
 
 @cli.command()
 @click.option(
-    "-j",
-    "--jsonl",
-    "jsonl",
+    '-j',
+    '--jsonl',
+    'jsonl',
     type=str,
-    help="输入 jsonl 路径，本地或者 s3 上的文件",
+    help='输入 jsonl 路径，本地或者 s3 上的文件',
     required=True,
 )
 @click.option(
-    "-m",
-    "--method",
-    "method",
+    '-m',
+    '--method',
+    'method',
     type=parse_pdf_methods,
-    help="指定解析方法。txt: 文本型 pdf 解析方法， ocr: 光学识别解析 pdf, auto: 程序智能选择解析方法",
-    default="auto",
+    help='指定解析方法。txt: 文本型 pdf 解析方法， ocr: 光学识别解析 pdf, auto: 程序智能选择解析方法',
+    default='auto',
 )
 @click.option(
-    "-o",
-    "--output-dir",
-    "output_dir",
-    type=str,
-    help="输出到本地目录",
-    default="",
+    '-o',
+    '--output-dir',
+    'output_dir',
+    type=click.Path(),
+    required=True,
+    help='输出到本地目录',
 )
 def jsonl(jsonl, method, output_dir):
     model_config.__use_inside_model__ = False
-    if jsonl.startswith("s3://"):
-        jso = json_parse.loads(read_s3_path(jsonl).decode("utf-8"))
-        full_jsonl_path = "."
+    if jsonl.startswith('s3://'):
+        jso = json_parse.loads(read_s3_path(jsonl).decode('utf-8'))
     else:
-        full_jsonl_path = os.path.realpath(jsonl)
         with open(jsonl) as f:
             jso = json_parse.loads(f.readline())
-
-    if output_dir == "":
-        output_dir = os.path.join(os.path.dirname(full_jsonl_path), "output")
-    s3_file_path = jso.get("file_location")
+    os.makedirs(output_dir, exist_ok=True)
+    s3_file_path = jso.get('file_location')
     if s3_file_path is None:
-        s3_file_path = jso.get("path")
+        s3_file_path = jso.get('path')
     pdf_file_name = Path(s3_file_path).stem
     pdf_data = read_s3_path(s3_file_path)
 
@@ -91,8 +84,9 @@ def jsonl(jsonl, method, output_dir):
         output_dir,
         pdf_file_name,
         pdf_data,
-        jso["doc_layout_result"],
+        jso['doc_layout_result'],
         method,
+        False,
         f_dump_content_list=True,
         f_draw_model_bbox=True,
     )
@@ -100,43 +94,45 @@ def jsonl(jsonl, method, output_dir):
 
 @cli.command()
 @click.option(
-    "-p",
-    "--pdf",
-    "pdf",
+    '-p',
+    '--pdf',
+    'pdf',
     type=click.Path(exists=True),
     required=True,
-    help="本地 PDF 文件",
+    help='本地 PDF 文件',
 )
 @click.option(
-    "-j",
-    "--json",
-    "json_data",
+    '-j',
+    '--json',
+    'json_data',
     type=click.Path(exists=True),
     required=True,
-    help="本地模型推理出的 json 数据",
-)
-@click.option(
-    "-o", "--output-dir", "output_dir", type=str, help="本地输出目录", default=""
+    help='本地模型推理出的 json 数据',
 )
+@click.option('-o',
+              '--output-dir',
+              'output_dir',
+              type=click.Path(),
+              required=True,
+              help='本地输出目录')
 @click.option(
-    "-m",
-    "--method",
-    "method",
+    '-m',
+    '--method',
+    'method',
     type=parse_pdf_methods,
-    help="指定解析方法。txt: 文本型 pdf 解析方法， ocr: 光学识别解析 pdf, auto: 程序智能选择解析方法",
-    default="auto",
+    help='指定解析方法。txt: 文本型 pdf 解析方法， ocr: 光学识别解析 pdf, auto: 程序智能选择解析方法',
+    default='auto',
 )
 def pdf(pdf, json_data, output_dir, method):
     model_config.__use_inside_model__ = False
     full_pdf_path = os.path.realpath(pdf)
-    if output_dir == "":
-        output_dir = os.path.join(os.path.dirname(full_pdf_path), "output")
+    os.makedirs(output_dir, exist_ok=True)
 
     def read_fn(path):
         disk_rw = DiskReaderWriter(os.path.dirname(path))
         return disk_rw.read(os.path.basename(path), AbsReaderWriter.MODE_BIN)
 
-    model_json_list = json_parse.loads(read_fn(json_data).decode("utf-8"))
+    model_json_list = json_parse.loads(read_fn(json_data).decode('utf-8'))
 
     file_name = str(Path(full_pdf_path).stem)
     pdf_data = read_fn(full_pdf_path)
@@ -146,10 +142,11 @@ def pdf(pdf, json_data, output_dir, method):
         pdf_data,
         model_json_list,
         method,
+        False,
         f_dump_content_list=True,
         f_draw_model_bbox=True,
     )
 
 
-if __name__ == "__main__":
+if __name__ == '__main__':
     cli()