chem-pdf2ppt 2.0.0

package/references/visual_style_en.md

@@ -0,0 +1,172 @@
+# Chemistry Academic PPT Visual Standards
+
+This document defines the visual design standards for chemistry academic PPTs, targeting **professional, publication-quality** presentations.
+
+## Color Schemes
+
+### Preset Themes
+
+#### 1. Academic Classic (academic) — Default, General Chemistry
+
+| Role | Hex | Usage |
+|------|-----|-------|
+| Primary | `#003366` | Titles, section slide backgrounds |
+| Background | `#FFFFFF` | Content slide backgrounds |
+| Body text | `#333333` | Main text |
+| Accent | `#B41E1E` | Decorative lines, key data highlights |
+| Light bg | `#F0F4F8` | Summary slide backgrounds, table zebra stripes |
+| Muted | `#8C8C8C` | Page numbers, figure captions, source labels |
+
+#### 2. Molecular Tech (molecular) — Computational Chemistry / Materials
+
+| Role | Hex | Usage |
+|------|-----|-------|
+| Primary | `#1A5276` | Titles, section slide backgrounds |
+| Background | `#F8F9FA` | Content slide backgrounds |
+| Body text | `#2C3E50` | Main text |
+| Accent | `#E74C3C` | Decorative lines, key data highlights |
+| Light bg | `#EBF0F5` | Summary slide backgrounds, table zebra stripes |
+
+#### 3. Green Chemistry (green) — Catalysis / Energy / Environment
+
+| Role | Hex | Usage |
+|------|-----|-------|
+| Primary | `#1E5631` | Titles, section slide backgrounds |
+| Background | `#F7F9F4` | Content slide backgrounds |
+| Body text | `#333333` | Main text |
+| Accent | `#D4A017` | Decorative lines, key data highlights |
+| Light bg | `#EEF3E9` | Summary slide backgrounds, table zebra stripes |
+
+#### 4. Nature Style (nature) — CNS Journal Presentations
+
+| Role | Hex | Usage |
+|------|-----|-------|
+| Primary | `#222222` | Titles, section slide backgrounds |
+| Background | `#FFFFFF` | Content slide backgrounds |
+| Body text | `#444444` | Main text |
+| Accent | `#0066CC` | Decorative lines, key data highlights |
+| Light bg | `#F8F8F8` | Summary slide backgrounds, table zebra stripes |
+
+### Color Principles
+
+- One theme per deck — do not mix themes
+- Accent color only for decorative lines, key numbers, emphasis marks — ≤ 5% of slide area
+- Dark section slides must use white or near-white text for sufficient contrast
+- No fluorescent colors, rainbow gradients, or decorative shadows
+
+## Typography
+
+### Recommended Fonts
+
+| Usage | Chinese | English / Numbers | Size |
+|------|---------|-------------------|------|
+| Cover title | Microsoft YaHei Bold / Source Han Sans Bold | Arial Bold / Helvetica Bold | 36–42pt |
+| Section divider | Microsoft YaHei Bold / Source Han Sans Bold | Arial Bold | 36–40pt |
+| Slide title | Microsoft YaHei Bold / Source Han Sans Bold | Arial Bold | 28–34pt |
+| Body bullets | Microsoft YaHei Regular / Source Han Sans Regular | Arial Regular | 16–20pt |
+| Figure captions/sources | Microsoft YaHei Regular / Source Han Sans Regular | Arial Regular | 9–11pt |
+| Table content | Microsoft YaHei Regular / Source Han Sans Regular | Arial Regular | 12–14pt |
+| Data highlights | Microsoft YaHei Bold / Source Han Sans Bold | Arial Bold | 24–28pt |
+
+### Typography Principles
+
+- Title at least 10pt larger than body — clear hierarchy
+- Subscripts in chemical formulas (H₂O, SO₄²⁻) — accept plain text if styling unavailable
+- Variables in italics (*E*a, *k*cat) — not mandatory
+- No more than 3 font sizes per slide
+
+## Layout Standards
+
+### Canvas
+
+- 16:9 widescreen (13.333" × 7.5")
+- Horizontal padding ≥ 0.5" (0.7" recommended)
+- Vertical padding ≥ 0.3"
+
+### Title Position
+
+- Content slide titles fixed at top-left (0.7", 0.3"), left-aligned
+- Thin decorative line below title (1.5pt, accent color) at 0.1–0.15"
+- Section divider titles may be centered or left-aligned (recommended: left-aligned + vertical accent bar)
+
+### Figure Layouts (4 modes)
+
+1. **figure_right** (default): best for landscape figures with detailed captions
+   - Left text: 0.7"–6.5" (5.8" wide)
+   - Right figure: 7.2"–12.7" (5.5" wide)
+
+2. **figure_top**: best for wide figures with short descriptions
+   - Figure area: 0.7"–12.6" (3.0–3.5" tall)
+   - Text below in remaining space
+
+3. **figure_left**: best for portrait figures or figure-emphasis slides
+   - Left figure: 5.5" wide
+   - Right text: 6.2" wide
+
+4. **figure_full**: best for complex mechanism diagrams, multi-panel figures
+   - Figure: 0.5"–12.8" (full width)
+   - Brief caption at bottom
+
+**Selection principle**: let the figure dictate the layout. Wide figures → `figure_top`, tall figures → `figure_left/right`, complex figures → `figure_full`.
+
+## Chemistry-Specific Visual Elements
+
+### Reaction Schemes
+
+- Use → or chemistry arrow to connect reactants and products
+- Reaction conditions above or below arrow (small font)
+- Example: `A + B ──→ C (yield: 85%)`
+- Use monospace or Arial font
+
+### Data Highlights
+
+- Key numbers (FE%, TOF, yield, selectivity) may be enlarged and boldfaced
+- Use accent color for breakthrough data
+- Do not highlight more than 3 numbers simultaneously
+
+### Energy / Free Energy Diagrams
+
+- Preserve original figure; annotate key barrier values on the slide
+- If redrawing: X-axis = reaction coordinate, Y-axis = energy (eV or kcal/mol)
+- Label the rate-determining step barrier
+
+### Characterization Data Interpretation Template
+
+```
+XRD:    → Phase confirmed as [...], crystallite size [...] nm (Scherrer)
+TEM:    → Morphology: [...], particle size distribution [...] nm
+HRTEM:  → Lattice fringe [...] nm, corresponding to [...] plane
+XPS:    → [...] element in [...] valence state, binding energy [...] eV
+BET:    → Surface area [...] m²/g, pore size [...] nm
+```
+
+### Performance Data Table Template
+
+```
+| Catalyst   | FE(C₂₊)% | j (mA/cm²) | Stability/h | Electrolyte | Ref      |
+|------------|----------|------------|-------------|-------------|----------|
+| Ru₁/Cu     | 82%      | 300        | 100         | 1M KOH      | This work |
+| Cu NPs     | 45%      | 150        | 20          | 1M KOH      | [1]       |
+| Benchmark  | 60%      | 200        | 50          | 1M KHCO₃    | [2]       |
+```
+
+## Readability Checklist
+
+- [ ] Body text ≥ 16pt
+- [ ] Figure text legible at presentation scale
+- [ ] Sufficient color contrast (dark text on light background, or vice versa)
+- [ ] No text overflow or overlapping
+- [ ] Figures not cropped or distorted
+- [ ] Page numbers consecutive
+- [ ] One core message per slide
+
+## Don'ts
+
+- No decorative images, clip art, or emoji
+- No gradient backgrounds
+- No dark backgrounds on content slides
+- No more than 6 bullet points per slide
+- No copy-pasting paper paragraphs verbatim
+- No figures too small to read
+- No cramming multiple dense figures onto one slide
+- No accent bar below titles on regular content slides (section slides only)

package/requirements.txt

@@ -0,0 +1,20 @@
+# PDF2PPT — Chemistry Academic PPT Generator
+# 化学学术论文 → PowerPoint 转换器
+
+# PDF reading and text extraction
+pymupdf>=1.23.0
+pdfplumber>=0.10.0
+
+# PPTX creation
+python-pptx>=0.6.23
+
+# Image processing
+Pillow>=10.0.0
+
+# PDF page to image conversion (optional, requires Poppler)
+pdf2image>=1.17.0
+
+# Note: poppler-utils is a system dependency for pdf2image
+#   macOS:  brew install poppler
+#   Linux:  sudo apt-get install poppler-utils
+#   Windows: https://github.com/oschwartz10612/poppler-windows/releases/

package/scripts/analyze_paper.py

@@ -0,0 +1,334 @@
+# -*- coding: utf-8 -*-
+"""
+PDF 论文分析脚本 — 化学论文类型识别、章节结构、图表位置、关键信息提取
+Chemistry Paper Analyzer with JSON output and robust encoding handling
+"""
+import fitz
+import sys
+import os
+import re
+import json
+from collections import defaultdict
+
+
+def _safe_print(msg):
+    """Windows-safe print."""
+    try:
+        print(msg)
+    except UnicodeEncodeError:
+        print(msg.encode('ascii', errors='replace').decode('ascii'))
+
+
+# ============================================================
+# 化学关键词库 / Chemistry Keyword Database
+# ============================================================
+
+CHEM_KEYWORDS = {
+    "催化": ["cataly", "TOF", "turnover", "selectivity", "conversion",
+             "catalyst", "active site", "Sabatier", "d-band", "overpotential"],
+    "材料": ["MOF", "COF", "perovskite", "zeolite", "framework",
+             "polymer", "nanosheet", "nanoparticle", "quantum dot"],
+    "有机合成": ["synthesis", "yield", "substrate scope", "coupling",
+                 "cross-coupling", "organocatal", "asymmetric", "functional group"],
+    "计算化学": [
+        "DFT", "density functional", "VASP", "Gaussian", "CP2K",
+        "Quantum ESPRESSO", "molecular dynamics", "AIMD", "ab initio",
+        "MD simulation", "free energy", "transition state",
+        "k-point", "pseudopotential", "PAW", "basis set",
+        "B3LYP", "PBE", "RPBE", "SCAN", "HSE", "GGA", "PBE0",
+        "Hartree-Fock", "coupled cluster", "CCSD", "MP2",
+        "Monte Carlo", "metadynamics", "enhanced sampling",
+        "reaction coordinate", "diabatic", "adiabatic",
+        "reorganization energy", "Marcus theory"
+    ],
+    "电化学": ["electrochem", "ORR", "OER", "HER", "CO2RR", "NRR",
+               "Li-ion", "battery", "supercapacitor", "electrolyte",
+               "faradaic efficiency", "Tafel", "RHE", "SHE"],
+    "光谱/表征": ["XRD", "XPS", "TEM", "SEM", "STEM", "HAADF",
+                  "NMR", "IR", "Raman", "EXAFS", "XANES", "BET",
+                  "EPR", "UV-vis", "AFM", "FTIR", "spectroscopy"],
+    "环境/大气": ["atmospheric", "aerosol", "SOA", "PM2.5", "oxidation",
+                  "OH radical", "ozone", "photochem", "tropospheric"],
+    "能源": ["solar cell", "perovskite solar", "water splitting",
+             "hydrogen evolution", "photocatal", "fuel cell", "battery"],
+    "辐射化学": ["radiolysis", "pulse radiolysis", "hydrated electron",
+                 "solvated electron", "transient absorption"],
+}
+
+# 表征关键词（仅在方法/实验段落出现时才算实验信号）
+CHARACTERIZATION_TERMS = [
+    "XRD", "XPS", "TEM", "SEM", "STEM", "HAADF", "NMR", "IR",
+    "Raman", "EXAFS", "XANES", "BET", "EPR", "UV-vis", "AFM", "FTIR"
+]
+
+# 计算关键词（强信号）
+COMPUTATIONAL_STRONG = [
+    "DFT", "density functional theory", "VASP", "Gaussian", "CP2K",
+    "Quantum ESPRESSO", "ab initio molecular dynamics", "AIMD",
+    "PBE0", "B3LYP", "PBE", "Hartree-Fock", "CCSD", "MP2",
+    "transition state", "reaction coordinate", "diabatic",
+    "pseudopotential", "basis set", "k-point", "cutoff",
+    "molecular dynamics simulation", "trajector"
+]
+
+
+def analyze_pdf(pdf_path, verbose=True, output_json=None):
+    """分析 PDF 论文的完整结构和内容
+
+    Returns:
+        dict: 论文全部分析结果
+    """
+    if not os.path.exists(pdf_path):
+        raise FileNotFoundError(f"PDF not found: {pdf_path}")
+
+    errors = []
+    doc = fitz.open(pdf_path)
+    result = {
+        "source": os.path.basename(pdf_path),
+        "title": "",
+        "paper_type": "experimental",
+        "paper_type_confidence": "low",
+        "subfields": [],
+        "sections": [],
+        "figures_detected": defaultdict(list),
+        "tables_detected": defaultdict(list),
+        "chemical_methods": [],
+        "total_pages": len(doc),
+        "has_supporting_info": False,
+        "errors": errors,
+    }
+
+    full_text = ""
+    first_page_text = ""
+    abstract_text = ""
+
+    # 提取所有文本
+    for page_num in range(len(doc)):
+        try:
+            page = doc[page_num]
+            text = page.get_text()
+            full_text += text
+            if page_num == 0:
+                first_page_text = text
+            if page_num <= 1:
+                abstract_text += text
+        except Exception as e:
+            errors.append(f"Page {page_num+1} text extraction failed: {e}")
+
+    full_text_lower = full_text.lower()
+
+    # ---- 标题提取 ----
+    lines = [l.strip() for l in first_page_text.split('\n') if l.strip()]
+    if lines:
+        result["title"] = lines[0]
+
+    # ---- 章节检测 ----
+    section_patterns = [
+        (r'^\s*(?:Abstract|摘要)\s*$', "Abstract"),
+        (r'^\s*(?:Introduction|引言|绪论)\s*$', "Introduction"),
+        (r'^\s*(?:Method|Experimental|Computational|计算方法?|实验方法?|Theory)\s*', "Methods"),
+        (r'^\s*(?:Result|Discussion|结果|讨论)\s*', "Results"),
+        (r'^\s*(?:Conclusion|结论|Summary|总结)\s*$', "Conclusions"),
+        (r'^\s*(?:Supporting Information|SI|附录|补充|References?)\s*', "SI/References"),
+    ]
+
+    for page_num in range(len(doc)):
+        try:
+            page = doc[page_num]
+            text = page.get_text()
+            for line in text.split('\n'):
+                line_stripped = line.strip()
+                if not line_stripped or len(line_stripped) > 80:
+                    continue
+                for pat, label in section_patterns:
+                    if re.match(pat, line_stripped, re.I):
+                        result["sections"].append({
+                            "page": page_num + 1,
+                            "heading": line_stripped,
+                            "label": label
+                        })
+                        break
+        except Exception:
+            pass
+
+    # ---- 图表检测 (version-compatible clustering) ----
+    def _cluster_compat(page, x_tol=3, y_tol=3):
+        """兼容 PyMuPDF 新旧版本的图形聚类"""
+        if hasattr(page, 'cluster_drawings'):
+            try:
+                return page.cluster_drawings(x_tolerance=x_tol, y_tolerance=y_tol)
+            except Exception:
+                pass
+        try:
+            drawings = page.get_drawings()
+        except Exception:
+            return []
+        if not drawings:
+            return []
+        rects = []
+        for d in drawings:
+            r = d.get('rect')
+            if r and r.x1 - r.x0 > 0.5 and r.y1 - r.y0 > 0.5:
+                rects.append(fitz.Rect(r.x0 - x_tol, r.y0 - y_tol, r.x1 + x_tol, r.y1 + y_tol))
+        if not rects:
+            return []
+        rects.sort(key=lambda r: (r.y0, r.x0))
+        clusters = [rects[0]]
+        for r in rects[1:]:
+            merged = False
+            for i, c in enumerate(clusters):
+                if r.intersects(c):
+                    clusters[i] = c | r
+                    merged = True
+                    break
+            if not merged:
+                clusters.append(r)
+        return [fitz.Rect(c.x0 + x_tol, c.y0 + y_tol, c.x1 - x_tol, c.y1 - y_tol) & page.rect for c in clusters]
+
+    for page_num in range(len(doc)):
+        try:
+            page = doc[page_num]
+            rects = _cluster_compat(page)
+            for rect in rects:
+                if rect.width > 100 and rect.height > 80:
+                    nearby_rect = fitz.Rect(
+                        rect.x0 - 10, rect.y1,
+                        rect.x1 + 10, min(rect.y1 + 120, page.rect.y1)
+                    )
+                    nearby_text = page.get_text(clip=nearby_rect)
+                    caption = ""
+                    if nearby_text:
+                        for line in nearby_text.split('\n'):
+                            m = re.match(r'(Fig|Figure|Table|Scheme)\s*[\.\s]', line, re.I)
+                            if m:
+                                caption = line.strip()[:150]
+                                break
+
+                    if caption:
+                        if re.match(r'(Fig|Figure|Scheme)', caption, re.I):
+                            result["figures_detected"][page_num + 1].append({
+                                "rect": [rect.x0, rect.y0, rect.x1, rect.y1],
+                                "caption": caption,
+                                "size": f"{rect.width:.0f}x{rect.height:.0f}"
+                            })
+                        elif re.match(r'Table', caption, re.I):
+                            result["tables_detected"][page_num + 1].append({
+                                "rect": [rect.x0, rect.y0, rect.x1, rect.y1],
+                                "caption": caption,
+                                "size": f"{rect.width:.0f}x{rect.height:.0f}"
+                            })
+        except Exception:
+            pass
+
+    # ---- 论文类型判定 (改进逻辑) ----
+    # 在正文前1/3部分（通常不含大量参考文献）检测信号
+    text_first_third = full_text[:len(full_text) // 3].lower()
+
+    computational_score = 0
+    for kw in COMPUTATIONAL_STRONG:
+        if kw.lower() in full_text_lower:
+            # 强信号在方法部分（前1/3）权重更高
+            if kw.lower() in text_first_third:
+                computational_score += 3
+            else:
+                computational_score += 1
+
+    # 表征关键词检测 — 只在 methods/results 区域有意义
+    char_in_text = sum(1 for t in CHARACTERIZATION_TERMS if t.lower() in full_text_lower)
+    char_in_first_third = sum(1 for t in CHARACTERIZATION_TERMS if t.lower() in text_first_third)
+
+    # 判定逻辑
+    has_strong_char = char_in_first_third >= 2
+    has_strong_comp = computational_score >= 5
+
+    if has_strong_comp and not has_strong_char:
+        result["paper_type"] = "computational"
+        result["paper_type_confidence"] = "high"
+    elif has_strong_char and has_strong_comp:
+        result["paper_type"] = "hybrid"
+        result["paper_type_confidence"] = "medium"
+    elif has_strong_comp and char_in_text >= 2:
+        result["paper_type"] = "hybrid"
+        result["paper_type_confidence"] = "low"
+        errors.append("Computational signal strong but characterization terms found in text (may be from references). Classified as hybrid with low confidence.")
+    elif has_strong_char:
+        result["paper_type"] = "experimental"
+        result["paper_type_confidence"] = "high"
+    else:
+        result["paper_type"] = "experimental"
+        result["paper_type_confidence"] = "low"
+        errors.append("Insufficient signals for confident paper type classification.")
+
+    # ---- 化学子领域识别 ----
+    for subfield, keywords in CHEM_KEYWORDS.items():
+        match_count = sum(1 for kw in keywords if kw.lower() in full_text_lower)
+        if match_count >= 2:
+            result["subfields"].append(subfield)
+
+    # ---- 特征方法提取 ----
+    method_patterns = [
+        (r'(CP2K|VASP|Gaussian\s*\d+|Q-Chem\s*\d+|GROMACS|LAMMPS|Quantum\s+ESPRESSO)', 2),
+        (r'(PBE0|B3LYP|PBE|RPBE|SCAN|HSE06|GGA|LDA|meta-GGA)', 1),
+        (r'(impregnation|sol-gel|hydrothermal|solvothermal|co-precipitation|calcination|pyrolysis)', 1),
+        (r'(XRD|XPS|TEM|SEM|STEM|HAADF-STEM|NMR|IR|Raman|EXAFS|XANES|BET|EPR)', 1),
+    ]
+    for pattern, weight in method_patterns:
+        for m in re.finditer(pattern, full_text, re.I):
+            method = m.group(0).strip()
+            if method and method not in result["chemical_methods"]:
+                result["chemical_methods"].append(method)
+
+    # ---- 支持信息检测 ----
+    if 'supporting information' in full_text_lower or 'supplementary' in full_text_lower:
+        result["has_supporting_info"] = True
+
+    doc.close()
+
+    # ---- 输出 ----
+    if verbose:
+        _safe_print(f"PDF Analysis: {result['source']}")
+        _safe_print(f"  Pages: {result['total_pages']}")
+        _safe_print(f"  Paper Type: {result['paper_type']} (confidence: {result['paper_type_confidence']})")
+        _safe_print(f"  Subfields: {', '.join(result['subfields']) or '未识别'}")
+        _safe_print(f"  Sections: {len(result['sections'])} detected")
+        _safe_print(f"  Figures: {sum(len(v) for v in result['figures_detected'].values())} detected")
+        _safe_print(f"  Methods: {', '.join(result['chemical_methods'][:8])}")
+        if errors:
+            _safe_print(f"  [!] {len(errors)} issue(s):")
+            for e in errors:
+                _safe_print(f"      - {e}")
+
+    if output_json:
+        # 将 defaultdict 转为普通 dict 以便 JSON 序列化
+        result_out = result.copy()
+        result_out["figures_detected"] = {
+            k: v for k, v in result["figures_detected"].items()
+        }
+        result_out["tables_detected"] = {
+            k: v for k, v in result["tables_detected"].items()
+        }
+        with open(output_json, 'w', encoding='utf-8') as f:
+            json.dump(result_out, f, indent=2, ensure_ascii=False)
+        _safe_print(f"  JSON report: {output_json}")
+
+    return result
+
+
+def main():
+    if len(sys.argv) < 2:
+        print("Usage: python analyze_paper.py <pdf_file> [--json output.json]")
+        print("Example: python analyze_paper.py paper.pdf --json report.json")
+        return
+
+    pdf_path = sys.argv[1]
+    output_json = None
+
+    for i, arg in enumerate(sys.argv):
+        if arg == "--json" and i + 1 < len(sys.argv):
+            output_json = sys.argv[i + 1]
+
+    analyze_pdf(pdf_path, verbose=True, output_json=output_json)
+
+
+if __name__ == "__main__":
+    main()

package/scripts/convert_to_images.py

@@ -0,0 +1,67 @@
+"""
+PDF转图片脚本 - 将PDF每一页转换为高清图片
+PDF to Images Converter - Converts PDF pages to high-resolution images
+"""
+from pdf2image import convert_from_path
+import os
+import sys
+
+def pdf_to_images(pdf_path, output_dir="pdf_images", dpi=200, fmt="PNG"):
+    """
+    将PDF转换为图片
+    
+    Args:
+        pdf_path: PDF文件路径
+        output_dir: 输出目录
+        dpi: 图片分辨率（越高越清晰，但文件越大）
+        fmt: 输出格式（PNG, JPEG）
+    """
+    if not os.path.exists(pdf_path):
+        print(f"Error: File not found: {pdf_path}")
+        return
+    
+    # 创建输出目录
+    if not os.path.exists(output_dir):
+        os.makedirs(output_dir)
+    
+    print(f"Converting PDF: {pdf_path}")
+    print(f"Output directory: {output_dir}")
+    print(f"DPI: {dpi}, Format: {fmt}")
+    print("-" * 40)
+    
+    try:
+        images = convert_from_path(pdf_path, dpi=dpi)
+        print(f"Successfully converted {len(images)} pages")
+        
+        for i, image in enumerate(images):
+            page_num = i + 1
+            image_path = os.path.join(output_dir, f"page_{page_num}.{fmt.lower()}")
+            image.save(image_path, fmt)
+            print(f"  Saved: page_{page_num}.{fmt.lower()}")
+        
+        print("-" * 40)
+        print(f"Conversion complete! {len(images)} pages saved to: {output_dir}")
+        return len(images)
+        
+    except Exception as e:
+        print(f"Error: {e}")
+        return 0
+
+def main():
+    if len(sys.argv) < 2:
+        print("Usage: python convert_to_images.py <pdf_file> [output_dir] [dpi]")
+        print("Examples:")
+        print("  python convert_to_images.py paper.pdf")
+        print("  python convert_to_images.py paper.pdf images 300")
+        print("  python convert_to_images.py paper.pdf pages 150 jpg")
+        return
+    
+    pdf_path = sys.argv[1]
+    output_dir = sys.argv[2] if len(sys.argv) > 2 else "pdf_images"
+    dpi = int(sys.argv[3]) if len(sys.argv) > 3 else 200
+    fmt = sys.argv[4].upper() if len(sys.argv) > 4 else "PNG"
+    
+    pdf_to_images(pdf_path, output_dir, dpi, fmt)
+
+if __name__ == "__main__":
+    main()