Merge pull request #7 from opendatalab/dev

Dev

requirements.txt

@@ -8,7 +8,6 @@ pdfminer.six==20231228
 pydantic>=2.7.2,<2.8.0
 PyMuPDF>=1.24.9
 scikit-learn>=1.0.2
-wordninja>=2.0.0
 torch>=2.2.2,<=2.3.1
 transformers
 # The requirements.txt must ensure that only necessary external dependencies are introduced. If there are new dependencies to add, please contact the project administrator.

setup.py

@@ -45,6 +45,7 @@ if __name__ == '__main__':
                      "paddlepaddle==2.6.1;platform_system=='Windows' or platform_system=='Darwin'",  # windows版本3.0.0b1效率下降，需锁定2.6.1
                      "pypandoc",  # 表格解析latex转html
                      "struct-eqtable==0.1.0",  # 表格解析
+                     "doclayout_yolo==0.0.2",  # doclayout_yolo
                      "detectron2"
                      ],
         },

tests/test_data/assets/jsonl/test_01.jsonl

+{"track_id":"e8824f5a-9fcb-4ee5-b2d4-6bf2c67019dc","path":"s3://sci-hub/enbook-scimag/78800000/libgen.scimag78872000-78872999/10.1017/cbo9780511770425.012.pdf","file_type":"pdf","content_type":"application/pdf","content_length":80078,"title":"German Idealism and the Concept of Punishment || Conclusion","remark":{"file_id":"scihub_78800000/libgen.scimag78872000-78872999.zip_10.1017/cbo9780511770425.012","file_source_type":"paper","original_file_id":"10.1017/cbo9780511770425.012","file_name":"10.1017/cbo9780511770425.012.pdf","author":"Merle, Jean-Christophe"}}

tests/test_data/assets/jsonl/test_02.jsonl

+{"track_id":"e8824f5a-9fcb-4ee5-b2d4-6bf2c67019dc","path":"tests/test_data/assets/pdfs/test_02.pdf","file_type":"pdf","content_type":"application/pdf","content_length":80078,"title":"German Idealism and the Concept of Punishment || Conclusion","remark":{"file_id":"scihub_78800000/libgen.scimag78872000-78872999.zip_10.1017/cbo9780511770425.012","file_source_type":"paper","original_file_id":"10.1017/cbo9780511770425.012","file_name":"10.1017/cbo9780511770425.012.pdf","author":"Merle, Jean-Christophe"}}

tests/test_data/data_reader_writer/test_filebase.py

+import os
+import shutil
+
+from magic_pdf.data.data_reader_writer import (FileBasedDataReader,
+                                               FileBasedDataWriter)
+
+
+def test_filebased_reader_writer():
+
+    unitest_dir = '/tmp/magic_pdf/unittest/data/filebased_reader_writer'
+    sub_dir = os.path.join(unitest_dir, 'sub')
+    abs_fn = os.path.join(unitest_dir, 'abspath.txt')
+
+    os.makedirs(sub_dir, exist_ok=True)
+
+    writer = FileBasedDataWriter(sub_dir)
+    reader = FileBasedDataReader(sub_dir)
+
+    writer.write('test.txt', b'hello world')
+    assert reader.read('test.txt') == b'hello world'
+
+    writer.write(abs_fn, b'hello world')
+    assert reader.read(abs_fn) == b'hello world'
+    shutil.rmtree(unitest_dir)

tests/test_data/data_reader_writer/test_multi_bucket_s3.py

+import json
+import os
+
+import fitz
+import pytest
+
+from magic_pdf.data.data_reader_writer import (MultiBucketS3DataReader,
+                                               MultiBucketS3DataWriter)
+from magic_pdf.data.schemas import S3Config
+
+
+@pytest.mark.skipif(
+    os.getenv('S3_ACCESS_KEY_2', None) is None, reason='need s3 config!'
+)
+def test_multi_bucket_s3_reader_writer():
+    """test multi bucket s3 reader writer must config s3 config in the
+    environment export S3_BUCKET=xxx export S3_ACCESS_KEY=xxx export
+    S3_SECRET_KEY=xxx export S3_ENDPOINT=xxx.
+
+    export S3_BUCKET_2=xxx export S3_ACCESS_KEY_2=xxx export S3_SECRET_KEY_2=xxx export S3_ENDPOINT_2=xxx
+    """
+    bucket = os.getenv('S3_BUCKET', '')
+    ak = os.getenv('S3_ACCESS_KEY', '')
+    sk = os.getenv('S3_SECRET_KEY', '')
+    endpoint_url = os.getenv('S3_ENDPOINT', '')
+
+    bucket_2 = os.getenv('S3_BUCKET_2', '')
+    ak_2 = os.getenv('S3_ACCESS_KEY_2', '')
+    sk_2 = os.getenv('S3_SECRET_KEY_2', '')
+    endpoint_url_2 = os.getenv('S3_ENDPOINT_2', '')
+
+    s3configs = [
+        S3Config(
+            bucket_name=bucket, access_key=ak, secret_key=sk, endpoint_url=endpoint_url
+        ),
+        S3Config(
+            bucket_name=bucket_2,
+            access_key=ak_2,
+            secret_key=sk_2,
+            endpoint_url=endpoint_url_2,
+        ),
+    ]
+
+    reader = MultiBucketS3DataReader(default_bucket=bucket, s3_configs=s3configs)
+    writer = MultiBucketS3DataWriter(default_bucket=bucket, s3_configs=s3configs)
+
+    bits = reader.read('meta-index/scihub/v001/scihub/part-66210c190659-000026.jsonl')
+
+    assert bits == reader.read(
+        f's3://{bucket}/meta-index/scihub/v001/scihub/part-66210c190659-000026.jsonl'
+    )
+
+    bits = reader.read(
+        f's3://{bucket_2}/enbook-scimag/78800000/libgen.scimag78872000-78872999/10.1017/cbo9780511770425.012.pdf'
+    )
+    docs = fitz.open('pdf', bits)
+    assert len(docs) == 10
+
+    bits = reader.read(
+        'meta-index/scihub/v001/scihub/part-66210c190659-000026.jsonl?bytes=566,713'
+    )
+    assert bits == reader.read_at(
+        'meta-index/scihub/v001/scihub/part-66210c190659-000026.jsonl', 566, 713
+    )
+    assert len(json.loads(bits)) > 0
+
+    writer.write_string(
+        'unittest/data/data_reader_writer/multi_bucket_s3_data/test01.txt', 'abc'
+    )
+
+    assert 'abc'.encode() == reader.read(
+        'unittest/data/data_reader_writer/multi_bucket_s3_data/test01.txt'
+    )
+
+    writer.write(
+        'unittest/data/data_reader_writer/multi_bucket_s3_data/test02.txt',
+        '123'.encode(),
+    )
+
+    assert '123'.encode() == reader.read(
+        'unittest/data/data_reader_writer/multi_bucket_s3_data/test02.txt'
+    )

tests/test_data/data_reader_writer/test_s3.py

+import json
+import os
+
+import pytest
+
+from magic_pdf.data.data_reader_writer import S3DataReader, S3DataWriter
+
+
+@pytest.mark.skipif(
+    os.getenv('S3_ACCESS_KEY', None) is None, reason='need s3 config!'
+)
+def test_multi_bucket_s3_reader_writer():
+    """test multi bucket s3 reader writer must config s3 config in the
+    environment export S3_BUCKET=xxx export S3_ACCESS_KEY=xxx export
+    S3_SECRET_KEY=xxx export S3_ENDPOINT=xxx."""
+    bucket = os.getenv('S3_BUCKET', '')
+    ak = os.getenv('S3_ACCESS_KEY', '')
+    sk = os.getenv('S3_SECRET_KEY', '')
+    endpoint_url = os.getenv('S3_ENDPOINT', '')
+
+    reader = S3DataReader(bucket=bucket, ak=ak, sk=sk, endpoint_url=endpoint_url)
+    writer = S3DataWriter(bucket=bucket, ak=ak, sk=sk, endpoint_url=endpoint_url)
+
+    bits = reader.read('meta-index/scihub/v001/scihub/part-66210c190659-000026.jsonl')
+
+    assert bits == reader.read(
+        f's3://{bucket}/meta-index/scihub/v001/scihub/part-66210c190659-000026.jsonl'
+    )
+
+    bits = reader.read(
+        'meta-index/scihub/v001/scihub/part-66210c190659-000026.jsonl?bytes=566,713'
+    )
+    assert bits == reader.read_at(
+        'meta-index/scihub/v001/scihub/part-66210c190659-000026.jsonl', 566, 713
+    )
+    assert len(json.loads(bits)) > 0
+
+    writer.write_string(
+        'unittest/data/data_reader_writer/multi_bucket_s3_data/test01.txt', 'abc'
+    )
+
+    assert 'abc'.encode() == reader.read(
+        'unittest/data/data_reader_writer/multi_bucket_s3_data/test01.txt'
+    )
+
+    writer.write(
+        f'{bucket}/unittest/data/data_reader_writer/multi_bucket_s3_data/test02.txt',
+        '123'.encode(),
+    )
+
+    assert '123'.encode() == reader.read(
+        'unittest/data/data_reader_writer/multi_bucket_s3_data/test02.txt'
+    )

tests/test_data/io/test_s3.py

+import json
+import os
+
+import pytest
+
+from magic_pdf.data.io.s3 import S3Reader, S3Writer
+
+
+@pytest.mark.skipif(
+    os.getenv('S3_ACCESS_KEY', None) is None, reason='s3 config not found'
+)
+def test_s3_reader():
+    """test s3 reader.
+
+    must config s3 config in the environment export S3_BUCKET=xxx export S3_ACCESS_KEY=xxx export S3_SECRET_KEY=xxx
+    export S3_ENDPOINT=xxx
+    """
+
+    bucket = os.getenv('S3_BUCKET', '')
+    ak = os.getenv('S3_ACCESS_KEY', '')
+    sk = os.getenv('S3_SECRET_KEY', '')
+    endpoint_url = os.getenv('S3_ENDPOINT', '')
+    reader = S3Reader(bucket=bucket, ak=ak, sk=sk, endpoint_url=endpoint_url)
+    bits = reader.read(
+        'meta-index/scihub/v001/scihub/part-66210c190659-000026.jsonl'
+    )
+    assert len(bits) > 0
+
+    bits = reader.read_at(
+        'meta-index/scihub/v001/scihub/part-66210c190659-000026.jsonl',
+        566,
+        713,
+    )
+    assert len(json.loads(bits)) > 0
+
+
+@pytest.mark.skipif(
+    os.getenv('S3_ACCESS_KEY', None) is None, reason='s3 config not found'
+)
+def test_s3_writer():
+    """test s3 reader.
+
+    must config s3 config in the environment export S3_BUCKET=xxx export S3_ACCESS_KEY=xxx export S3_SECRET_KEY=xxx
+    export S3_ENDPOINT=xxx
+    """
+    bucket = os.getenv('S3_BUCKET', '')
+    ak = os.getenv('S3_ACCESS_KEY', '')
+    sk = os.getenv('S3_SECRET_KEY', '')
+    endpoint_url = os.getenv('S3_ENDPOINT', '')
+    writer = S3Writer(bucket=bucket, ak=ak, sk=sk, endpoint_url=endpoint_url)
+    test_fn = 'unittest/io/test.jsonl'
+    writer.write(test_fn, '123'.encode())
+    reader = S3Reader(bucket=bucket, ak=ak, sk=sk, endpoint_url=endpoint_url)
+    bits = reader.read(test_fn)
+    assert bits.decode() == '123'

tests/test_data/test_dataset.py

+
+from magic_pdf.data.dataset import ImageDataset, PymuDocDataset
+
+
+def test_pymudataset():
+    with open('tests/test_data/assets/pdfs/test_01.pdf', 'rb') as f:
+        bits = f.read()
+    datasets = PymuDocDataset(bits)
+    assert len(datasets) > 0
+    assert datasets.get_page(0).get_page_info().h > 100
+
+
+def test_imagedataset():
+    with open('tests/test_data/assets/pngs/test_01.png', 'rb') as f:
+        bits = f.read()
+    datasets = ImageDataset(bits)
+    assert len(datasets) == 1
+    assert datasets.get_page(0).get_page_info().w > 100

tests/test_data/test_read_api.py

+import os
+
+import pytest
+
+from magic_pdf.data.data_reader_writer import MultiBucketS3DataReader
+from magic_pdf.data.read_api import (read_jsonl, read_local_images,
+                                     read_local_pdfs)
+from magic_pdf.data.schemas import S3Config
+
+
+def test_read_local_pdfs():
+    datasets = read_local_pdfs('tests/test_data/assets/pdfs')
+    assert len(datasets) == 2
+    assert len(datasets[0]) > 0
+    assert len(datasets[1]) > 0
+
+    assert datasets[0].get_page(0).get_page_info().w > 0
+    assert datasets[0].get_page(0).get_page_info().h > 0
+
+
+def test_read_local_images():
+    datasets = read_local_images('tests/test_data/assets/pngs', suffixes=['png'])
+    assert len(datasets) == 2
+    assert len(datasets[0]) == 1
+    assert len(datasets[1]) == 1
+
+    assert datasets[0].get_page(0).get_page_info().w > 0
+    assert datasets[0].get_page(0).get_page_info().h > 0
+
+
+@pytest.mark.skipif(
+    os.getenv('S3_ACCESS_KEY_2', None) is None, reason='need s3 config!'
+)
+def test_read_json():
+    """test multi bucket s3 reader writer must config s3 config in the
+    environment export S3_BUCKET=xxx export S3_ACCESS_KEY=xxx export
+    S3_SECRET_KEY=xxx export S3_ENDPOINT=xxx.
+
+    export S3_BUCKET_2=xxx export S3_ACCESS_KEY_2=xxx export S3_SECRET_KEY_2=xxx export S3_ENDPOINT_2=xxx
+    """
+    bucket = os.getenv('S3_BUCKET', '')
+    ak = os.getenv('S3_ACCESS_KEY', '')
+    sk = os.getenv('S3_SECRET_KEY', '')
+    endpoint_url = os.getenv('S3_ENDPOINT', '')
+
+    bucket_2 = os.getenv('S3_BUCKET_2', '')
+    ak_2 = os.getenv('S3_ACCESS_KEY_2', '')
+    sk_2 = os.getenv('S3_SECRET_KEY_2', '')
+    endpoint_url_2 = os.getenv('S3_ENDPOINT_2', '')
+
+    s3configs = [
+        S3Config(
+            bucket_name=bucket, access_key=ak, secret_key=sk, endpoint_url=endpoint_url
+        ),
+        S3Config(
+            bucket_name=bucket_2,
+            access_key=ak_2,
+            secret_key=sk_2,
+            endpoint_url=endpoint_url_2,
+        ),
+    ]
+
+    reader = MultiBucketS3DataReader(bucket, s3configs)
+
+    datasets = read_jsonl(
+        f's3://{bucket}/meta-index/scihub/v001/scihub/part-66210c190659-000026.jsonl',
+        reader,
+    )
+    assert len(datasets) > 0
+    assert len(datasets[0]) == 10
+
+    datasets = read_jsonl('tests/test_data/assets/jsonl/test_01.jsonl', reader)
+    assert len(datasets) == 1
+    assert len(datasets[0]) == 10
+
+    datasets = read_jsonl('tests/test_data/assets/jsonl/test_02.jsonl')
+    assert len(datasets) == 1
+    assert len(datasets[0]) == 1

tests/test_model/assets/test_01.model.json

+[
+    {
+        "layout_dets": [
+            {
+                "category_id": 3,
+                "poly": [
+                    776.7277221679688,
+                    688.448974609375,
+                    1242.224365234375,
+                    688.448974609375,
+                    1242.224365234375,
+                    1182.0628662109375,
+                    776.7277221679688,
+                    1182.0628662109375
+                ],
+                "score": 0.999997079372406
+            },
+            {
+                "category_id": 3,
+                "poly": [
+                    775.9269409179688,
+                    1389.754638671875,
+                    1243.672119140625,
+                    1389.754638671875,
+                    1243.672119140625,
+                    1859.716064453125,
+                    775.9269409179688,
+                    1859.716064453125
+                ],
+                "score": 0.9999949932098389
+            },
+            {
+                "category_id": 1,
+                "poly": [
+                    752.11572265625,
+                    1939.3634033203125,
+                    1430.1146240234375,
+                    1939.3634033203125,
+                    1430.1146240234375,
+                    2041.1771240234375,
+                    752.11572265625,
+                    2041.1771240234375
+                ],
+                "score": 0.999975323677063
+            },
+            {
+                "category_id": 3,
+                "poly": [
+                    46.55152893066406,
+                    686.12939453125,
+                    638.8861083984375,
+                    686.12939453125,
+                    638.8861083984375,
+                    1803.419189453125,
+                    46.55152893066406,
+                    1803.419189453125
+                ],
+                "score": 0.999961256980896
+            },
+            {
+                "category_id": 3,
+                "poly": [
+                    33.684722900390625,
+                    150.77980041503906,
+                    1238.0679931640625,
+                    150.77980041503906,
+                    1238.0679931640625,
+                    524.98291015625,
+                    33.684722900390625,
+                    524.98291015625
+                ],
+                "score": 0.9999504089355469
+            },
+            {
+                "category_id": 1,
+                "poly": [
+                    24.685693740844727,
+                    1875.9998779296875,
+                    703.5064697265625,
+                    1875.9998779296875,
+                    703.5064697265625,
+                    2050.7431640625,
+                    24.685693740844727,
+                    2050.7431640625
+                ],
+                "score": 0.9999105334281921
+            },
+            {
+                "category_id": 1,
+                "poly": [
+                    750.97705078125,
+                    1252.206787109375,
+                    1430.0809326171875,
+                    1252.206787109375,
+                    1430.0809326171875,
+                    1357.2947998046875,
+                    750.97705078125,
+                    1357.2947998046875
+                ],
+                "score": 0.999853789806366
+            },
+            {
+                "category_id": 4,
+                "poly": [
+                    904.842041015625,
+                    1213.027099609375,
+                    1273.5655517578125,
+                    1213.027099609375,
+                    1273.5655517578125,
+                    1242.717529296875,
+                    904.842041015625,
+                    1242.717529296875
+                ],
+                "score": 0.9995817542076111
+            },
+            {
+                "category_id": 4,
+                "poly": [
+                    905.3208618164062,
+                    1898.5325927734375,
+                    1273.1282958984375,
+                    1898.5325927734375,
+                    1273.1282958984375,
+                    1928.9906005859375,
+                    905.3208618164062,
+                    1928.9906005859375
+                ],
+                "score": 0.9986443519592285
+            },
+            {
+                "category_id": 4,
+                "poly": [
+                    372.0135498046875,
+                    556.02685546875,
+                    1084.9647216796875,
+                    556.02685546875,
+                    1084.9647216796875,
+                    586.6792602539062,
+                    372.0135498046875,
+                    586.6792602539062
+                ],
+                "score": 0.9985352754592896
+            },
+            {
+                "category_id": 2,
+                "poly": [
+                    1350.63671875,
+                    79.77919006347656,
+                    1379.6220703125,
+                    79.77919006347656,
+                    1379.6220703125,
+                    99.83788299560547,
+                    1350.63671875,
+                    99.83788299560547
+                ],
+                "score": 0.9973036646842957
+            },
+            {
+                "category_id": 4,
+                "poly": [
+                    203.2659912109375,
+                    597.2034912109375,
+                    1251.0240478515625,
+                    597.2034912109375,
+                    1251.0240478515625,
+                    657.985595703125,
+                    203.2659912109375,
+                    657.985595703125
+                ],
+                "score": 0.9622809886932373
+            },
+            {
+                "category_id": 0,
+                "poly": [
+                    70.87332916259766,
+                    1834.5714111328125,
+                    657.8504638671875,
+                    1834.5714111328125,
+                    657.8504638671875,
+                    1865.07373046875,
+                    70.87332916259766,
+                    1865.07373046875
+                ],
+                "score": 0.8580453395843506
+            },
+            {
+                "category_id": 1,
+                "poly": [
+                    189.0360870361328,
+                    597.2406616210938,
+                    1252.3204345703125,
+                    597.2406616210938,
+                    1252.3204345703125,
+                    658.4781494140625,
+                    189.0360870361328,
+                    658.4781494140625
+                ],
+                "score": 0.3083903193473816
+            },
+            {
+                "category_id": 13,
+                "poly": [
+                    1190,
+                    1980,
+                    1206,
+                    1980,
+                    1206,
+                    1997,
+                    1190,
+                    1997
+                ],
+                "score": 0.51,
+                "latex": ":"
+            },
+            {
+                "category_id": 13,
+                "poly": [
+                    1219,
+                    1331,
+                    1235,
+                    1331,
+                    1235,
+                    1348,
+                    1219,
+                    1348
+                ],
+                "score": 0.49,
+                "latex": ":"
+            },
+            {
+                "category_id": 13,
+                "poly": [
+                    798,
+                    2016,
+                    813,
+                    2016,
+                    813,
+                    2033,
+                    798,
+                    2033
+                ],
+                "score": 0.41,
+                "latex": ":"
+            },
+            {
+                "category_id": 13,
+                "poly": [
+                    135,
+                    1991,
+                    148,
+                    1991,
+                    148,
+                    2006,
+                    135,
+                    2006
+                ],
+                "score": 0.39,
+                "latex": ":"
+            },
+            {
+                "category_id": 13,
+                "poly": [
+                    400,
+                    1916,
+                    416,
+                    1916,
+                    416,
+                    1933,
+                    400,
+                    1933
+                ],
+                "score": 0.38,
+                "latex": ":"
+            },
+            {
+                "category_id": 13,
+                "poly": [
+                    1148,
+                    1944,
+                    1162,
+                    1944,
+                    1162,
+                    1961,
+                    1148,
+                    1961
+                ],
+                "score": 0.31,
+                "latex": ":"
+            },
+            {
+                "category_id": 15,
+                "poly": [
+                    798.0,
+                    1943.0,
+                    1147.0,
+                    1943.0,
+                    1147.0,
+                    1968.0,
+                    798.0,
+                    1968.0
+                ],
+                "score": 0.95,
+                "text": "Fig 4  SSCP analysis of FHIT exon 4. T"
+            },
+            {
+                "category_id": 15,
+                "poly": [
+                    1163.0,
+                    1943.0,
+                    1425.0,
+                    1943.0,
+                    1425.0,
+                    1968.0,
+                    1163.0,
+                    1968.0
+                ],
+                "score": 0.96,
+                "text": "Tumor tissue ; N :Corresponding"
+            },
+            {
+                "category_id": 15,
+                "poly": [
+                    755.0,
+                    1979.0,
+                    1189.0,
+                    1979.0,
+                    1189.0,
+                    2004.0,
+                    755.0,
+                    2004.0
+                ],
+                "score": 0.92,
+                "text": "normal tissue ; M : PBR322/Hae II Marker ; ssDNA"
+            },
+            {
+                "category_id": 15,
+                "poly": [
+                    1207.0,
+                    1979.0,
+                    1422.0,
+                    1979.0,
+                    1422.0,
+                    2004.0,
+                    1207.0,
+                    2004.0
+                ],
+                "score": 0.97,
+                "text": "Single-stranded DNA ; ds-"
+            },
+            {
+                "category_id": 15,
+                "poly": [
+                    755.0,
+                    2015.0,
+                    797.0,
+                    2015.0,
+                    797.0,
+                    2038.0,
+                    755.0,
+                    2038.0
+                ],
+                "score": 1.0,
+                "text": "DNA"
+            },
+            {
+                "category_id": 15,
+                "poly": [
+                    814.0,
+                    2015.0,
+                    996.0,
+                    2015.0,
+                    996.0,
+                    2038.0,
+                    814.0,
+                    2038.0
+                ],
+                "score": 0.98,
+                "text": "Double-stranded DNA"
+            },
+            {
+                "category_id": 15,
+                "poly": [
+                    71.0,
+                    1880.0,
+                    698.0,
+                    1880.0,
+                    698.0,
+                    1902.0,
+                    71.0,
+                    1902.0
+                ],
+                "score": 0.96,
+                "text": "Fig 2Alterations of PCR amplified products of FHIT exon 3,4,5 and"
+            },
+            {
+                "category_id": 15,
+                "poly": [
+                    28.0,
+                    1916.0,
+                    399.0,
+                    1916.0,
+                    399.0,
+                    1937.0,
+                    28.0,
+                    1937.0
+                ],
+                "score": 0.98,
+                "text": "microsatellite marker D3S1300、D3S1312.A"
+            },
+            {
+                "category_id": 15,
+                "poly": [
+                    417.0,
+                    1916.0,
+                    701.0,
+                    1916.0,
+                    701.0,
+                    1937.0,
+                    417.0,
+                    1937.0
+                ],
+                "score": 0.9,
+                "text": "Deletion of exon5(arrows）;B :"
+            },
+            {
+                "category_id": 15,
+                "poly": [
+                    29.0,
+                    1953.0,
+                    700.0,
+                    1953.0,
+                    700.0,
+                    1974.0,
+                    29.0,
+                    1974.0
+                ],
+                "score": 0.95,
+                "text": "Deletion of exon 3 A( arrows）;C : Deletion of microsatellite marker D3S1300,"
+            },
+            {
+                "category_id": 15,
+                "poly": [
+                    28.0,
+                    1989.0,
+                    134.0,
+                    1989.0,
+                    134.0,
+                    2014.0,
+                    28.0,
+                    2014.0
+                ],
+                "score": 1.0,
+                "text": "D3S1312.T"
+            },
+            {
+                "category_id": 15,
+                "poly": [
+                    149.0,
+                    1989.0,
+                    696.0,
+                    1989.0,
+                    696.0,
+                    2014.0,
+                    149.0,
+                    2014.0
+                ],
+                "score": 0.96,
+                "text": "Tumor ; N : Corresponding normal tissue ; L : Corresponding lymph"
+            },
+            {
+                "category_id": 15,
+                "poly": [
+                    30.0,
+                    2027.0,
+                    634.0,
+                    2027.0,
+                    634.0,
+                    2047.0,
+                    30.0,
+                    2047.0
+                ],
+                "score": 0.94,
+                "text": "node tissue;M :DL2000 DNA marker;L1:Lewis ;A :A549;S SPAC-1"
+            },
+            {
+                "category_id": 15,
+                "poly": [
+                    801.0,
+                    1259.0,
+                    1427.0,
+                    1259.0,
+                    1427.0,
+                    1280.0,
+                    801.0,
+                    1280.0
+                ],
+                "score": 0.94,
+                "text": "Fig 3SSCP analysis of FHIT exon 3.The arrow indicateda deletion of"
+            },
+            {
+                "category_id": 15,
+                "poly": [
+                    757.0,
+                    1294.0,
+                    1424.0,
+                    1294.0,
+                    1424.0,
+                    1318.0,
+                    757.0,
+                    1318.0
+                ],
+                "score": 0.96,
+                "text": "exon 3 of 41T. T : Tumor tissue ; N : Corresponding normal tissue ; M PBR322/"
+            },
+            {
+                "category_id": 15,
+                "poly": [
+                    755.0,
+                    1329.0,
+                    1218.0,
+                    1329.0,
+                    1218.0,
+                    1355.0,
+                    755.0,
+                    1355.0
+                ],
+                "score": 0.95,
+                "text": "Hae Il Marker / ssDNA : Single-stranded DNA ; dsDNA"
+            },
+            {
+                "category_id": 15,
+                "poly": [
+                    1236.0,
+                    1329.0,
+                    1418.0,
+                    1329.0,
+                    1418.0,
+                    1355.0,
+                    1236.0,
+                    1355.0
+                ],
+                "score": 1.0,
+                "text": "Double-strandedDNA"
+            },
+            {
+                "category_id": 15,
+                "poly": [
+                    910.0,
+                    1217.0,
+                    1269.0,
+                    1217.0,
+                    1269.0,
+                    1241.0,
+                    910.0,
+                    1241.0
+                ],
+                "score": 1.0,
+                "text": "图3FHIT基因外显子3的SSCP分析"
+            },
+            {
+                "category_id": 15,
+                "poly": [
+                    909.0,
+                    1904.0,
+                    1269.0,
+                    1904.0,
+                    1269.0,
+                    1927.0,
+                    909.0,
+                    1927.0
+                ],
+                "score": 1.0,
+                "text": "图4FHIT基因外显子4的SSCP分析"
+            },
+            {
+                "category_id": 15,
+                "poly": [
+                    374.0,
+                    563.0,
+                    1077.0,
+                    563.0,
+                    1077.0,
+                    583.0,
+                    374.0,
+                    583.0
+                ],
+                "score": 0.99,
+                "text": "图1FHIT基因外显子3、4、5、8和微卫星灶的PCR扩增产物琼脂糖电泳图"
+            },
+            {
+                "category_id": 15,
+                "poly": [
+                    1351.0,
+                    81.0,
+                    1376.0,
+                    81.0,
+                    1376.0,
+                    102.0,
+                    1351.0,
+                    102.0
+                ],
+                "score": 1.0,
+                "text": "13"
+            },
+            {
+                "category_id": 15,
+                "poly": [
+                    207.0,
+                    600.0,
+                    1245.0,
+                    600.0,
+                    1245.0,
+                    624.0,
+                    207.0,
+                    624.0
+                ],
+                "score": 0.96,
+                "text": "Fig 1 Agarose electrophoresis of PCR products of exor( A)3 ,4 ,5 ,8 and three microsatellite markers( B）of FHIT gene"
+            },
+            {
+                "category_id": 15,
+                "poly": [
+                    309.0,
+                    634.0,
+                    1142.0,
+                    634.0,
+                    1142.0,
+                    662.0,
+                    309.0,
+                    662.0
+                ],
+                "score": 0.97,
+                "text": "M1 :DL2000 DNA marker ; M2 PBR322/Hae Il marker ; T :Tumor ; N :Corresponding normal tissue"
+            },
+            {
+                "category_id": 15,
+                "poly": [
+                    73.0,
+                    1840.0,
+                    651.0,
+                    1840.0,
+                    651.0,
+                    1864.0,
+                    73.0,
+                    1864.0
+                ],
+                "score": 1.0,
+                "text": "图2FHIT基因外显子和微卫星灶PCR扩增产物缺失电泳图"
+            },
+            {
+                "category_id": 15,
+                "poly": [
+                    207.0,
+                    600.0,
+                    1245.0,
+                    600.0,
+                    1245.0,
+                    625.0,
+                    207.0,
+                    625.0
+                ],
+                "score": 0.96,
+                "text": "Fig 1  Agarose electrophoresis of PCR products of exor A)3 ,4 ,5 ,8 and three microsatellite markers( B)of FHIT gene"
+            },
+            {
+                "category_id": 15,
+                "poly": [
+                    309.0,
+                    635.0,
+                    1142.0,
+                    635.0,
+                    1142.0,
+                    661.0,
+                    309.0,
+                    661.0
+                ],
+                "score": 0.97,
+                "text": "M1 :DL2000 DNA marker ; M2 PBR322/Hae Il marker ; T Tumor ; N :Corresponding normal tissue"
+            }
+        ],
+        "page_info": {
+            "page_no": 0,
+            "height": 2080,
+            "width": 1472
+        }
+    }
+]