1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292

import io
import os
import re
import tempfile
from pathlib import Path
from typing import List

import fitz  
import img2pdf  
import numpy as np
import torch
import typer
from PIL import Image, ImageDraw, ImageFont
from rich.progress import track
from transformers import AutoModel, AutoTokenizer


def pdf_to_images_high_quality(
    pdf_path: Path, temp_dir: Path, dpi=144, image_format="PNG"
) -> List[Path]:
    image_files = []

    pdf_document = fitz.open(pdf_path)

    zoom = dpi / 72.0
    matrix = fitz.Matrix(zoom, zoom)

    for page_num in range(pdf_document.page_count):
        page = pdf_document[page_num]

        pixmap = page.get_pixmap(matrix=matrix, alpha=False)
        Image.MAX_IMAGE_PIXELS = None

        if image_format.upper() == "PNG":
            img_data = pixmap.tobytes("png")
            img = Image.open(io.BytesIO(img_data))
        else:
            img_data = pixmap.tobytes("png")
            img = Image.open(io.BytesIO(img_data))
            if img.mode in ("RGBA", "LA"):
                background = Image.new("RGB", img.size, (255, 255, 255))
                background.paste(
                    img, mask=img.split()[-1] if img.mode == "RGBA" else None
                )
                img = background  

        img_path = temp_dir / f"{page_num}.png"
        img.save(img_path)
        img.close()
        image_files.append(img_path)

    pdf_document.close()
    return image_files


def pil_to_pdf_img2pdf(pil_images, output_path: Path):
    if not pil_images:
        return

    image_bytes_list = []

    for img in pil_images:
        if img.mode != "RGB":
            img = img.convert("RGB")

        img_buffer = io.BytesIO()
        img.save(img_buffer, format="JPEG", quality=95)
        img_bytes = img_buffer.getvalue()
        image_bytes_list.append(img_bytes)

    try:
        pdf_bytes = img2pdf.convert(image_bytes_list)
        assert pdf_bytes is not None
        with open(output_path, "wb") as f:
            f.write(pdf_bytes)

    except Exception as e:
        print(f"error: {e}")


def re_match(text):
    pattern = r"(<\|ref\|>(.*?)<\|/ref\|><\|det\|>(.*?)<\|/det\|>)"
    matches = re.findall(pattern, text, re.DOTALL)

    mathes_image = []
    mathes_other = []
    for a_match in matches:
        if "<|ref|>image<|/ref|>" in a_match[0]:
            mathes_image.append(a_match[0])
        else:
            mathes_other.append(a_match[0])
    return matches, mathes_image, mathes_other


def extract_coordinates_and_label(ref_text, image_width, image_height):
    try:
        label_type = ref_text[1]
        cor_list = eval(ref_text[2])
    except Exception as e:
        print(e)
        return None

    return (label_type, cor_list)


def draw_bounding_boxes(image, refs, jdx, out_path: Path):
    image_width, image_height = image.size
    img_draw = image.copy()
    draw = ImageDraw.Draw(img_draw)

    overlay = Image.new("RGBA", img_draw.size, (0, 0, 0, 0))
    draw2 = ImageDraw.Draw(overlay)

    
    font = ImageFont.load_default()

    img_idx = 0

    for i, ref in enumerate(refs):
        try:
            result = extract_coordinates_and_label(ref, image_width, image_height)
            if result:
                label_type, points_list = result

                color = (
                    np.random.randint(0, 200),
                    np.random.randint(0, 200),
                    np.random.randint(0, 255),
                )

                color_a = color + (20,)
                for points in points_list:
                    x1, y1, x2, y2 = points

                    x1 = int(x1 / 999 * image_width)
                    y1 = int(y1 / 999 * image_height)

                    x2 = int(x2 / 999 * image_width)
                    y2 = int(y2 / 999 * image_height)

                    if label_type == "image":
                        try:
                            cropped = image.crop((x1, y1, x2, y2))
                            cropped.save(out_path / f"images/{jdx}_{img_idx}.jpg")
                        except Exception as e:
                            print(e)
                            pass
                        img_idx += 1

                    try:
                        if label_type == "title":
                            draw.rectangle([x1, y1, x2, y2], outline=color, width=4)
                            draw2.rectangle(
                                [x1, y1, x2, y2],
                                fill=color_a,
                                outline=(0, 0, 0, 0),
                                width=1,
                            )
                        else:
                            draw.rectangle([x1, y1, x2, y2], outline=color, width=2)
                            draw2.rectangle(
                                [x1, y1, x2, y2],
                                fill=color_a,
                                outline=(0, 0, 0, 0),
                                width=1,
                            )

                        text_x = x1
                        text_y = max(0, y1 - 15)

                        text_bbox = draw.textbbox((0, 0), label_type, font=font)
                        text_width = text_bbox[2] - text_bbox[0]
                        text_height = text_bbox[3] - text_bbox[1]
                        draw.rectangle(
                            [text_x, text_y, text_x + text_width, text_y + text_height],
                            fill=(255, 255, 255, 30),
                        )

                        draw.text((text_x, text_y), label_type, font=font, fill=color)
                    except Exception:
                        pass
        except Exception:
            continue
    img_draw.paste(overlay, (0, 0), overlay)
    return img_draw


def process_image_with_refs(image, ref_texts, jdx, out_path):
    result_image = draw_bounding_boxes(image, ref_texts, jdx, out_path)
    return result_image


app = typer.Typer(help="Convert PDF to Markdown using DeepSeek-OCR")


@app.command()
def convert(
    input_file: Path = typer.Argument(..., help="Input PDF file path"),
    out_path: Path = typer.Option(
        "output", "-o", "--output", help="Output directory for markdown file"
    ),
):
    os.makedirs(out_path / "images", exist_ok=True)
    temp_dir = tempfile.TemporaryDirectory()

    typer.echo(f"📄 Converting {input_file} to images...")
    image_files = pdf_to_images_high_quality(input_file, Path(temp_dir.name))

    MODEL_NAME = "deepseek-ai/DeepSeek-OCR"

    typer.echo("🤖 Loading DeepSeek-OCR model...")
    tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME, trust_remote_code=True)
    model = AutoModel.from_pretrained(
        MODEL_NAME,
        attn_implementation="flash_attention_2",
        trust_remote_code=True,
        use_safetensors=True,
        torch_dtype=torch.bfloat16,  
    )
    model = model.eval().cuda()

    prompt = "<image>\n<|grounding|>Convert the document to markdown."

    mmd_det_path = out_path / (Path(input_file).stem + "_det.md")
    mmd_path = out_path / (Path(input_file).stem + ".md")
    pdf_out_path = out_path / (Path(input_file).stem + "_layouts.pdf")

    contents_det = ""
    contents = ""
    draw_images = []
    jdx = 0

    typer.echo("🔍 Processing pages with OCR...")
    for image_file in track(image_files):
        content = model.infer(
            tokenizer,
            prompt=prompt,
            image_file=image_file,
            output_path=temp_dir.name,
            base_size=1024,
            image_size=640,
            crop_mode=True,
            save_results=False,
            test_compress=True,
            eval_mode=True,
        )

        page_num = "\n<--- Page Split --->"
        contents_det += content + f"\n{page_num}\n"

        matches_ref, matches_images, matches_other = re_match(content)

        with Image.open(image_file) as image_draw:
            result_image = process_image_with_refs(
                image_draw, matches_ref, jdx, out_path
            )

        draw_images.append(result_image)

        for idx, a_match_image in enumerate(matches_images):
            content = content.replace(
                a_match_image, "![](images/" + str(jdx) + "_" + str(idx) + ".jpg)\n"
            )

        for idx, a_match_other in enumerate(matches_other):
            content = (
                content.replace(a_match_other, "")
                .replace("\\coloneqq", ":=")
                .replace("\\eqqcolon", "=:")
                .replace("\n\n\n\n", "\n\n")
                .replace("\n\n\n", "\n\n")
            )

        contents += content + f"\n{page_num}\n"

        jdx += 1

    typer.echo(f"💾 Saving markdown to {mmd_path}...")
    with open(mmd_det_path, "w", encoding="utf-8") as afile:
        afile.write(contents_det)

    with open(mmd_path, "w", encoding="utf-8") as afile:
        afile.write(contents)

    pil_to_pdf_img2pdf(draw_images, pdf_out_path)

    temp_dir.cleanup()
    typer.echo("✅ Conversion completed successfully!")


if __name__ == "__main__":
    app()