Spaces:
Runtime error
Runtime error
import base64 | |
import json | |
import os | |
import time | |
import zipfile | |
from pathlib import Path | |
import re | |
import uuid | |
import pymupdf | |
os.system('pip uninstall -y magic-pdf') | |
os.system('pip install git+https://github.com/opendatalab/MinerU.git@dev') | |
os.system('wget https://github.com/opendatalab/MinerU/raw/dev/scripts/download_models_hf.py -O download_models_hf.py') | |
os.system('python download_models_hf.py') | |
with open('/home/user/magic-pdf.json', 'r') as file: | |
data = json.load(file) | |
data['device-mode'] = "cuda" | |
if os.getenv('apikey'): | |
data['llm-aided-config']['title_aided']['api_key'] = os.getenv('apikey') | |
data['llm-aided-config']['title_aided']['enable'] = True | |
with open('/home/user/magic-pdf.json', 'w') as file: | |
json.dump(data, file, indent=4) | |
os.system('cp -r paddleocr /home/user/.paddleocr') | |
from gradio_pdf import PDF | |
import gradio as gr | |
from loguru import logger | |
from magic_pdf.data.data_reader_writer import FileBasedDataReader | |
from magic_pdf.libs.hash_utils import compute_sha256 | |
from magic_pdf.tools.common import do_parse, prepare_env | |
def create_css(): | |
return """ | |
/* μ 체 μ€νμΌ */ | |
.gradio-container { | |
background: linear-gradient(135deg, #EFF6FF 0%, #F5F3FF 100%); | |
max-width: 1200px !important; | |
margin: 0 auto !important; | |
padding: 2rem !important; | |
} | |
/* μ λͺ© μ€νμΌ */ | |
.title-area { | |
text-align: center; | |
margin-bottom: 2rem; | |
padding: 1rem; | |
background: white; | |
border-radius: 1rem; | |
box-shadow: 0 4px 6px -1px rgba(0, 0, 0, 0.1); | |
} | |
.title-area h1 { | |
background: linear-gradient(90deg, #2563EB 0%, #7C3AED 100%); | |
-webkit-background-clip: text; | |
-webkit-text-fill-color: transparent; | |
font-size: 2.5rem; | |
font-weight: bold; | |
margin-bottom: 0.5rem; | |
} | |
.title-area p { | |
color: #6B7280; | |
font-size: 1.1rem; | |
} | |
/* μ»΄ν¬λνΈ μ€νμΌλ§ */ | |
.gr-box, .gr-panel { | |
border: 2px solid #E0E7FF !important; | |
border-radius: 12px !important; | |
box-shadow: 0 4px 6px -1px rgba(0, 0, 0, 0.1) !important; | |
background: white !important; | |
} | |
/* νμΌ μ λ‘λ μμ */ | |
.file-upload { | |
border: 2px dashed #93C5FD !important; | |
border-radius: 8px !important; | |
padding: 2rem !important; | |
background: #F0F9FF !important; | |
transition: all 0.3s ease; | |
} | |
.file-upload:hover { | |
background: #E0F2FE !important; | |
border-color: #60A5FA !important; | |
} | |
/* λ²νΌ μ€νμΌλ§ */ | |
.gr-button.primary-button { | |
background: linear-gradient(90deg, #2563EB 0%, #7C3AED 100%) !important; | |
color: white !important; | |
border: none !important; | |
border-radius: 8px !important; | |
padding: 0.75rem 1.5rem !important; | |
font-weight: bold !important; | |
transition: opacity 0.2s !important; | |
} | |
.gr-button.primary-button:hover { | |
opacity: 0.9 !important; | |
} | |
.gr-button.secondary-button { | |
background: white !important; | |
color: #4B5563 !important; | |
border: 1px solid #D1D5DB !important; | |
border-radius: 8px !important; | |
padding: 0.75rem 1.5rem !important; | |
} | |
.gr-button.secondary-button:hover { | |
background: #F9FAFB !important; | |
} | |
/* μ¬λΌμ΄λ μ€νμΌλ§ */ | |
.gr-slider { | |
background: #E0E7FF !important; | |
} | |
.gr-slider .gr-slider-handle { | |
background: #4F46E5 !important; | |
} | |
/* 체ν¬λ°μ€ μ€νμΌλ§ */ | |
.gr-checkbox { | |
border-color: #6366F1 !important; | |
} | |
.gr-checkbox:checked { | |
background-color: #4F46E5 !important; | |
} | |
/* ν μ€νμΌλ§ */ | |
.gr-tabs { | |
border-bottom: 2px solid #E0E7FF !important; | |
} | |
.gr-tab-button { | |
color: #6B7280 !important; | |
padding: 0.75rem 1rem !important; | |
font-weight: 500 !important; | |
} | |
.gr-tab-button.selected { | |
color: #4F46E5 !important; | |
border-bottom: 2px solid #4F46E5 !important; | |
} | |
/* λ§ν¬λ€μ΄ μΆλ ₯ μμ */ | |
.markdown-output { | |
background: white !important; | |
border-radius: 8px !important; | |
padding: 1rem !important; | |
box-shadow: inset 0 2px 4px rgba(0, 0, 0, 0.05) !important; | |
} | |
""" | |
def read_fn(path): | |
disk_rw = FileBasedDataReader(os.path.dirname(path)) | |
return disk_rw.read(os.path.basename(path)) | |
def parse_pdf(doc_path, output_dir, end_page_id, is_ocr, layout_mode, formula_enable, table_enable, language): | |
os.makedirs(output_dir, exist_ok=True) | |
try: | |
file_name = f"{str(Path(doc_path).stem)}_{time.time()}" | |
pdf_data = read_fn(doc_path) | |
if is_ocr: | |
parse_method = "ocr" | |
else: | |
parse_method = "auto" | |
local_image_dir, local_md_dir = prepare_env(output_dir, file_name, parse_method) | |
do_parse( | |
output_dir, | |
file_name, | |
pdf_data, | |
[], | |
parse_method, | |
False, | |
end_page_id=end_page_id, | |
layout_model=layout_mode, | |
formula_enable=formula_enable, | |
table_enable=table_enable, | |
lang=language, | |
f_dump_orig_pdf=False, | |
) | |
return local_md_dir, file_name | |
except Exception as e: | |
logger.exception(e) | |
def compress_directory_to_zip(directory_path, output_zip_path): | |
try: | |
with zipfile.ZipFile(output_zip_path, 'w', zipfile.ZIP_DEFLATED) as zipf: | |
for root, dirs, files in os.walk(directory_path): | |
for file in files: | |
file_path = os.path.join(root, file) | |
arcname = os.path.relpath(file_path, directory_path) | |
zipf.write(file_path, arcname) | |
return 0 | |
except Exception as e: | |
logger.exception(e) | |
return -1 | |
def image_to_base64(image_path): | |
with open(image_path, "rb") as image_file: | |
return base64.b64encode(image_file.read()).decode('utf-8') | |
def replace_image_with_base64(markdown_text, image_dir_path): | |
pattern = r'\!\[(?:[^\]]*)\]\(([^)]+)\)' | |
def replace(match): | |
relative_path = match.group(1) | |
full_path = os.path.join(image_dir_path, relative_path) | |
base64_image = image_to_base64(full_path) | |
return f"" | |
return re.sub(pattern, replace, markdown_text) | |
def to_markdown(file_path, end_pages, is_ocr, layout_mode, formula_enable, table_enable, language): | |
file_path = to_pdf(file_path) | |
if end_pages > 20: | |
end_pages = 20 | |
local_md_dir, file_name = parse_pdf(file_path, './output', end_pages - 1, is_ocr, | |
layout_mode, formula_enable, table_enable, language) | |
archive_zip_path = os.path.join("./output", compute_sha256(local_md_dir) + ".zip") | |
zip_archive_success = compress_directory_to_zip(local_md_dir, archive_zip_path) | |
if zip_archive_success == 0: | |
logger.info("μμΆ μ±κ³΅") | |
else: | |
logger.error("μμΆ μ€ν¨") | |
md_path = os.path.join(local_md_dir, file_name + ".md") | |
with open(md_path, 'r', encoding='utf-8') as f: | |
txt_content = f.read() | |
md_content = replace_image_with_base64(txt_content, local_md_dir) | |
new_pdf_path = os.path.join(local_md_dir, file_name + "_layout.pdf") | |
return md_content, txt_content, archive_zip_path, new_pdf_path | |
def to_pdf(file_path): | |
with pymupdf.open(file_path) as f: | |
if f.is_pdf: | |
return file_path | |
else: | |
pdf_bytes = f.convert_to_pdf() | |
unique_filename = f"{uuid.uuid4()}.pdf" | |
tmp_file_path = os.path.join(os.path.dirname(file_path), unique_filename) | |
with open(tmp_file_path, 'wb') as tmp_pdf_file: | |
tmp_pdf_file.write(pdf_bytes) | |
return tmp_file_path | |
latex_delimiters = [{"left": "$$", "right": "$$", "display": True}, | |
{"left": '$', "right": '$', "display": False}] | |
def init_model(): | |
from magic_pdf.model.doc_analyze_by_custom_model import ModelSingleton | |
try: | |
model_manager = ModelSingleton() | |
txt_model = model_manager.get_model(False, False) | |
logger.info(f"txt_model init final") | |
ocr_model = model_manager.get_model(True, False) | |
logger.info(f"ocr_model init final") | |
return 0 | |
except Exception as e: | |
logger.exception(e) | |
return -1 | |
model_init = init_model() | |
logger.info(f"model_init: {model_init}") | |
latin_lang = [ | |
'af', 'az', 'bs', 'cs', 'cy', 'da', 'de', 'es', 'et', 'fr', 'ga', 'hr', | |
'hu', 'id', 'is', 'it', 'ku', 'la', 'lt', 'lv', 'mi', 'ms', 'mt', 'nl', | |
'no', 'oc', 'pi', 'pl', 'pt', 'ro', 'rs_latin', 'sk', 'sl', 'sq', 'sv', | |
'sw', 'tl', 'tr', 'uz', 'vi', 'french', 'german' | |
] | |
arabic_lang = ['ar', 'fa', 'ug', 'ur'] | |
cyrillic_lang = [ | |
'ru', 'rs_cyrillic', 'be', 'bg', 'uk', 'mn', 'abq', 'ady', 'kbd', 'ava', | |
'dar', 'inh', 'che', 'lbe', 'lez', 'tab' | |
] | |
devanagari_lang = [ | |
'hi', 'mr', 'ne', 'bh', 'mai', 'ang', 'bho', 'mah', 'sck', 'new', 'gom', | |
'sa', 'bgc' | |
] | |
other_lang = ['ch', 'en', 'korean', 'japan', 'chinese_cht', 'ta', 'te', 'ka'] | |
all_lang = ['', 'auto'] | |
all_lang.extend([*other_lang, *latin_lang, *arabic_lang, *cyrillic_lang, *devanagari_lang]) | |
if __name__ == "__main__": | |
with gr.Blocks(title="OCR FLEX", css=create_css()) as demo: | |
# νμ΄ν μμ | |
with gr.Row(elem_classes="title-area"): | |
gr.HTML(""" | |
<h1>OCR FLEX</h1> | |
<p>PDFμ μ΄λ―Έμ§μμ ν μ€νΈλ₯Ό λΉ λ₯΄κ³ μ ννκ² μΆμΆνμΈμ</p> | |
""") | |
with gr.Row(): | |
# μΌμͺ½ ν¨λ | |
with gr.Column(variant='panel', scale=5): | |
file = gr.File( | |
label="PDF λλ μ΄λ―Έμ§ νμΌμ μ λ‘λνμΈμ", | |
file_types=[".pdf", ".png", ".jpeg", ".jpg"], | |
elem_classes="file-upload" | |
) | |
max_pages = gr.Slider( | |
1, 20, 10, | |
step=1, | |
label='μ΅λ λ³ν νμ΄μ§ μ', | |
elem_classes="custom-slider" | |
) | |
with gr.Row(): | |
layout_mode = gr.Dropdown( | |
["layoutlmv3", "doclayout_yolo"], | |
label="λ μ΄μμ λͺ¨λΈ", | |
value="doclayout_yolo", | |
elem_classes="custom-dropdown" | |
) | |
language = gr.Dropdown( | |
all_lang, | |
label="μΈμ΄", | |
value='auto', | |
elem_classes="custom-dropdown" | |
) | |
with gr.Row(): | |
formula_enable = gr.Checkbox( | |
label="μμ μΈμ νμ±ν", | |
value=True, | |
elem_classes="custom-checkbox" | |
) | |
is_ocr = gr.Checkbox( | |
label="OCR κ°μ νμ±ν", | |
value=False, | |
elem_classes="custom-checkbox" | |
) | |
table_enable = gr.Checkbox( | |
label="ν μΈμ νμ±ν(ν μ€νΈ)", | |
value=True, | |
elem_classes="custom-checkbox" | |
) | |
with gr.Row(): | |
change_bu = gr.Button( | |
"λ³ν", | |
elem_classes="primary-button" | |
) | |
clear_bu = gr.ClearButton( | |
value="μ΄κΈ°ν", | |
elem_classes="secondary-button" | |
) | |
pdf_show = PDF( | |
label='PDF 미리보기', | |
interactive=False, | |
visible=True, | |
height=800, | |
elem_classes="pdf-preview" | |
) | |
with gr.Accordion("μμ :", open=False): | |
example_root = os.path.join(os.path.dirname(__file__), "examples") | |
gr.Examples( | |
examples=[os.path.join(example_root, _) for _ in os.listdir(example_root) if | |
_.endswith("pdf")], | |
inputs=file | |
) | |
# μ€λ₯Έμͺ½ ν¨λ | |
with gr.Column(variant='panel', scale=5): | |
output_file = gr.File( | |
label="λ³ν κ²°κ³Ό", | |
interactive=False, | |
elem_classes="output-file" | |
) | |
with gr.Tabs() as tabs: | |
with gr.Tab("λ§ν¬λ€μ΄ λ λλ§"): | |
md = gr.Markdown( | |
label="λ§ν¬λ€μ΄ λ λλ§", | |
height=1100, | |
show_copy_button=True, | |
latex_delimiters=latex_delimiters, | |
line_breaks=True, | |
elem_classes="markdown-output" | |
) | |
with gr.Tab("λ§ν¬λ€μ΄ ν μ€νΈ"): | |
md_text = gr.TextArea( | |
lines=45, | |
show_copy_button=True, | |
elem_classes="markdown-text" | |
) | |
# μ΄λ²€νΈ νΈλ€λ¬ | |
file.change( | |
fn=to_pdf, | |
inputs=file, | |
outputs=pdf_show | |
) | |
change_bu.click( | |
fn=to_markdown, | |
inputs=[ | |
file, | |
max_pages, | |
is_ocr, | |
layout_mode, | |
formula_enable, | |
table_enable, | |
language | |
], | |
outputs=[ | |
md, | |
md_text, | |
output_file, | |
pdf_show | |
], | |
api_name=False | |
) | |
clear_bu.add([file, md, pdf_show, md_text, output_file, is_ocr]) | |
# μ± μ€ν | |
demo.launch(ssr_mode=True) |