Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
|
@@ -27,8 +27,8 @@ LAYOUTS = {
|
|
| 27 |
OUTPUT_DIR = Path("generated_pdfs")
|
| 28 |
OUTPUT_DIR.mkdir(exist_ok=True)
|
| 29 |
|
| 30 |
-
# Path for the required emoji font file
|
| 31 |
-
EMOJI_FONT_PATH = Path("
|
| 32 |
|
| 33 |
# Regex to find and wrap emojis for ReportLab
|
| 34 |
EMOJI_PATTERN = re.compile(
|
|
@@ -52,7 +52,6 @@ EMOJI_PATTERN = re.compile(
|
|
| 52 |
|
| 53 |
class PDFGenerator:
|
| 54 |
"""
|
| 55 |
-
An object-oriented approach to generating PDFs.
|
| 56 |
Handles font registration, markdown parsing, and PDF creation.
|
| 57 |
"""
|
| 58 |
def __init__(self, font_path: Path):
|
|
@@ -69,9 +68,8 @@ class PDFGenerator:
|
|
| 69 |
Registers the TTF font file with ReportLab if the file exists.
|
| 70 |
"""
|
| 71 |
if font_path.exists():
|
| 72 |
-
pdfmetrics.registerFont(TTFont(self.emoji_font_name, font_path))
|
| 73 |
else:
|
| 74 |
-
# Provide a helpful error in the web app if the font is missing
|
| 75 |
st.error(f"Emoji font not found at '{font_path}'. Emojis will not be rendered. Please download it.")
|
| 76 |
self.emoji_font_name = "Helvetica" # Fallback to a standard font
|
| 77 |
|
|
@@ -80,11 +78,8 @@ class PDFGenerator:
|
|
| 80 |
π To make emojis appear so grand, wrap them with a font command.
|
| 81 |
Finds all emojis and wraps them in ReportLab <font> tags.
|
| 82 |
"""
|
| 83 |
-
# If the emoji font failed to register, don't try to use it.
|
| 84 |
if self.emoji_font_name != "NotoEmoji":
|
| 85 |
return text
|
| 86 |
-
|
| 87 |
-
# The lambda function takes each matched emoji (m) and wraps it.
|
| 88 |
return EMOJI_PATTERN.sub(lambda m: f'<font name="{self.emoji_font_name}">{m.group(0)}</font>', text)
|
| 89 |
|
| 90 |
def _markdown_to_story(self, markdown_text: str) -> list:
|
|
@@ -93,13 +88,10 @@ class PDFGenerator:
|
|
| 93 |
Converts a markdown string π into a list of ReportLab Flowables (a 'story').
|
| 94 |
"""
|
| 95 |
styles = getSampleStyleSheet()
|
| 96 |
-
|
| 97 |
-
# Define custom styles for different markdown elements
|
| 98 |
style_normal = styles['BodyText']
|
| 99 |
style_h1 = styles['h1']
|
| 100 |
style_h2 = styles['h2']
|
| 101 |
style_h3 = styles['h3']
|
| 102 |
-
# Use a monospaced font for code blocks
|
| 103 |
style_code = ParagraphStyle('Code', parent=styles['Normal'], fontName='Courier', textColor=colors.darkred)
|
| 104 |
|
| 105 |
story = []
|
|
@@ -108,7 +100,6 @@ class PDFGenerator:
|
|
| 108 |
code_block_text = ""
|
| 109 |
|
| 110 |
for line in lines:
|
| 111 |
-
# Handle code blocks (```)
|
| 112 |
if line.strip().startswith("```"):
|
| 113 |
if in_code_block:
|
| 114 |
story.append(Paragraph(code_block_text, style_code))
|
|
@@ -119,33 +110,35 @@ class PDFGenerator:
|
|
| 119 |
continue
|
| 120 |
|
| 121 |
if in_code_block:
|
| 122 |
-
# Escape HTML-sensitive characters and preserve line breaks within code
|
| 123 |
escaped_line = line.replace('&', '&').replace('<', '<').replace('>', '>')
|
| 124 |
code_block_text += escaped_line + '<br/>'
|
| 125 |
continue
|
| 126 |
-
|
| 127 |
-
# This is where we process each line for emojis BEFORE creating a Paragraph
|
| 128 |
-
processed_line = self._wrap_emojis_for_reportlab(line)
|
| 129 |
|
| 130 |
-
#
|
| 131 |
-
if
|
| 132 |
-
|
| 133 |
-
|
| 134 |
-
|
| 135 |
-
|
| 136 |
-
story.append(Paragraph(
|
| 137 |
-
elif
|
| 138 |
-
|
| 139 |
-
|
| 140 |
-
|
| 141 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 142 |
story.append(Spacer(1, 0.2 * inch))
|
| 143 |
else:
|
| 144 |
-
# Handle bold
|
| 145 |
-
formatted_line = re.sub(r'\*\*(.*?)\*\*', r'<b>\1</b>',
|
| 146 |
formatted_line = re.sub(r'_(.*?)_', r'<i>\1</i>', formatted_line)
|
| 147 |
-
|
| 148 |
-
|
|
|
|
| 149 |
return story
|
| 150 |
|
| 151 |
def create_pdf(self, md_asset: Path, layout_name: str, layout_properties: dict):
|
|
@@ -155,26 +148,18 @@ class PDFGenerator:
|
|
| 155 |
"""
|
| 156 |
try:
|
| 157 |
md_content = md_asset.read_text(encoding="utf-8")
|
| 158 |
-
|
| 159 |
date_str = datetime.datetime.now().strftime("%Y-%m-%d")
|
| 160 |
output_filename = f"{md_asset.stem}_{layout_name.replace(' ', '-')}_{date_str}.pdf"
|
| 161 |
output_path = OUTPUT_DIR / output_filename
|
| 162 |
|
| 163 |
-
# The SimpleDocTemplate handles the page creation and content flow
|
| 164 |
doc = SimpleDocTemplate(
|
| 165 |
str(output_path),
|
| 166 |
pagesize=layout_properties.get("size", A4),
|
| 167 |
-
rightMargin=inch,
|
| 168 |
-
|
| 169 |
-
topMargin=inch,
|
| 170 |
-
bottomMargin=inch
|
| 171 |
)
|
| 172 |
-
|
| 173 |
story = self._markdown_to_story(md_content)
|
| 174 |
-
|
| 175 |
-
# The .build() method takes the story and renders the PDF
|
| 176 |
doc.build(story)
|
| 177 |
-
|
| 178 |
except Exception as e:
|
| 179 |
st.error(f"Failed to process {md_asset.name} with ReportLab: {e}")
|
| 180 |
|
|
@@ -184,7 +169,6 @@ class PDFGenerator:
|
|
| 184 |
def get_file_download_link(file_path: Path) -> str:
|
| 185 |
"""
|
| 186 |
π To grab your file and not delay, a special link is paved today.
|
| 187 |
-
Generates a base64-encoded download link for a file.
|
| 188 |
"""
|
| 189 |
with open(file_path, "rb") as f:
|
| 190 |
data = base64.b64encode(f.read()).decode()
|
|
@@ -193,10 +177,8 @@ def get_file_download_link(file_path: Path) -> str:
|
|
| 193 |
def display_file_explorer():
|
| 194 |
"""
|
| 195 |
π To see your files, both old and new, this handy explorer gives a view.
|
| 196 |
-
Renders a simple file explorer in the Streamlit app for MD and PDF files.
|
| 197 |
"""
|
| 198 |
st.header("π File Explorer")
|
| 199 |
-
|
| 200 |
st.subheader("Source Markdown Files (.md)")
|
| 201 |
md_files = list(Path(".").glob("*.md"))
|
| 202 |
if not md_files:
|
|
@@ -210,7 +192,6 @@ def display_file_explorer():
|
|
| 210 |
st.markdown(get_file_download_link(md_file), unsafe_allow_html=True)
|
| 211 |
|
| 212 |
st.subheader("Generated PDF Files")
|
| 213 |
-
# Sort PDFs by modification time to show the newest first
|
| 214 |
pdf_files = sorted(list(OUTPUT_DIR.glob("*.pdf")), key=lambda p: p.stat().st_mtime, reverse=True)
|
| 215 |
if not pdf_files:
|
| 216 |
st.info("No PDFs generated yet. Click the button above to start.")
|
|
@@ -222,7 +203,6 @@ def display_file_explorer():
|
|
| 222 |
with col2:
|
| 223 |
st.markdown(get_file_download_link(pdf_file), unsafe_allow_html=True)
|
| 224 |
|
| 225 |
-
|
| 226 |
# --- Main App Execution ---
|
| 227 |
|
| 228 |
def main():
|
|
@@ -233,41 +213,33 @@ def main():
|
|
| 233 |
st.title("π Markdown to PDF Generator")
|
| 234 |
st.markdown("This tool converts all `.md` files in this directory to PDF. It now supports emojis! π")
|
| 235 |
|
| 236 |
-
|
| 237 |
-
if not list(Path(".").glob("*.md")):
|
| 238 |
with open("sample.md", "w", encoding="utf-8") as f:
|
| 239 |
f.write("# Sample Document π\n\nThis is a sample markdown file. **ReportLab** is creating the PDF. Emojis like π and π‘ should now appear correctly.\n\n### Features\n- Item 1\n- Item 2\n\n```\ndef hello_world():\n print(\"Hello, PDF! π\")\n```\n")
|
| 240 |
st.rerun()
|
| 241 |
|
| 242 |
-
# Instantiate our generator. It will handle font setup on its own.
|
| 243 |
pdf_generator = PDFGenerator(EMOJI_FONT_PATH)
|
| 244 |
|
| 245 |
if st.button("π Generate PDFs from all Markdown Files", type="primary"):
|
| 246 |
markdown_files = list(Path(".").glob("*.md"))
|
| 247 |
-
|
| 248 |
if not markdown_files:
|
| 249 |
st.warning("No `.md` files found. Please add a markdown file to the directory.")
|
| 250 |
else:
|
| 251 |
total_pdfs = len(markdown_files) * len(LAYOUTS)
|
| 252 |
progress_bar = st.progress(0, text="Starting PDF generation...")
|
| 253 |
pdf_count = 0
|
| 254 |
-
|
| 255 |
with st.spinner("Generating PDFs... Please wait."):
|
| 256 |
for md_file in markdown_files:
|
| 257 |
st.info(f"Processing: **{md_file.name}**")
|
| 258 |
for name, properties in LAYOUTS.items():
|
| 259 |
-
# Use the instance method to create the PDF
|
| 260 |
pdf_generator.create_pdf(md_file, name, properties)
|
| 261 |
pdf_count += 1
|
| 262 |
progress_bar.progress(pdf_count / total_pdfs, f"Generated {pdf_count}/{total_pdfs} PDFs...")
|
| 263 |
-
|
| 264 |
st.success("β
PDF generation complete!")
|
| 265 |
st.balloons()
|
| 266 |
-
# Rerun to refresh the file explorer immediately
|
| 267 |
st.rerun()
|
| 268 |
|
| 269 |
display_file_explorer()
|
| 270 |
|
| 271 |
-
|
| 272 |
if __name__ == "__main__":
|
| 273 |
main()
|
|
|
|
| 27 |
OUTPUT_DIR = Path("generated_pdfs")
|
| 28 |
OUTPUT_DIR.mkdir(exist_ok=True)
|
| 29 |
|
| 30 |
+
# β οΈ UPDATED: Path for the required NON-COLOR emoji font file.
|
| 31 |
+
EMOJI_FONT_PATH = Path("NotoEmoji-Regular.ttf")
|
| 32 |
|
| 33 |
# Regex to find and wrap emojis for ReportLab
|
| 34 |
EMOJI_PATTERN = re.compile(
|
|
|
|
| 52 |
|
| 53 |
class PDFGenerator:
|
| 54 |
"""
|
|
|
|
| 55 |
Handles font registration, markdown parsing, and PDF creation.
|
| 56 |
"""
|
| 57 |
def __init__(self, font_path: Path):
|
|
|
|
| 68 |
Registers the TTF font file with ReportLab if the file exists.
|
| 69 |
"""
|
| 70 |
if font_path.exists():
|
| 71 |
+
pdfmetrics.registerFont(TTFont(self.emoji_font_name, str(font_path)))
|
| 72 |
else:
|
|
|
|
| 73 |
st.error(f"Emoji font not found at '{font_path}'. Emojis will not be rendered. Please download it.")
|
| 74 |
self.emoji_font_name = "Helvetica" # Fallback to a standard font
|
| 75 |
|
|
|
|
| 78 |
π To make emojis appear so grand, wrap them with a font command.
|
| 79 |
Finds all emojis and wraps them in ReportLab <font> tags.
|
| 80 |
"""
|
|
|
|
| 81 |
if self.emoji_font_name != "NotoEmoji":
|
| 82 |
return text
|
|
|
|
|
|
|
| 83 |
return EMOJI_PATTERN.sub(lambda m: f'<font name="{self.emoji_font_name}">{m.group(0)}</font>', text)
|
| 84 |
|
| 85 |
def _markdown_to_story(self, markdown_text: str) -> list:
|
|
|
|
| 88 |
Converts a markdown string π into a list of ReportLab Flowables (a 'story').
|
| 89 |
"""
|
| 90 |
styles = getSampleStyleSheet()
|
|
|
|
|
|
|
| 91 |
style_normal = styles['BodyText']
|
| 92 |
style_h1 = styles['h1']
|
| 93 |
style_h2 = styles['h2']
|
| 94 |
style_h3 = styles['h3']
|
|
|
|
| 95 |
style_code = ParagraphStyle('Code', parent=styles['Normal'], fontName='Courier', textColor=colors.darkred)
|
| 96 |
|
| 97 |
story = []
|
|
|
|
| 100 |
code_block_text = ""
|
| 101 |
|
| 102 |
for line in lines:
|
|
|
|
| 103 |
if line.strip().startswith("```"):
|
| 104 |
if in_code_block:
|
| 105 |
story.append(Paragraph(code_block_text, style_code))
|
|
|
|
| 110 |
continue
|
| 111 |
|
| 112 |
if in_code_block:
|
|
|
|
| 113 |
escaped_line = line.replace('&', '&').replace('<', '<').replace('>', '>')
|
| 114 |
code_block_text += escaped_line + '<br/>'
|
| 115 |
continue
|
|
|
|
|
|
|
|
|
|
| 116 |
|
| 117 |
+
# Process the line for markdown syntax first
|
| 118 |
+
if line.startswith("# "):
|
| 119 |
+
final_text = self._wrap_emojis_for_reportlab(line[2:])
|
| 120 |
+
story.append(Paragraph(final_text, style_h1))
|
| 121 |
+
elif line.startswith("## "):
|
| 122 |
+
final_text = self._wrap_emojis_for_reportlab(line[3:])
|
| 123 |
+
story.append(Paragraph(final_text, style_h2))
|
| 124 |
+
elif line.startswith("### "):
|
| 125 |
+
final_text = self._wrap_emojis_for_reportlab(line[4:])
|
| 126 |
+
story.append(Paragraph(final_text, style_h3))
|
| 127 |
+
elif line.strip().startswith(("* ", "- ")):
|
| 128 |
+
final_text = self._wrap_emojis_for_reportlab(line.strip()[2:])
|
| 129 |
+
story.append(Paragraph(f"β’ {final_text}", style_normal))
|
| 130 |
+
elif re.match(r'^\d+\.\s', line.strip()):
|
| 131 |
+
final_text = self._wrap_emojis_for_reportlab(line.strip())
|
| 132 |
+
story.append(Paragraph(final_text, style_normal))
|
| 133 |
+
elif line.strip() == "":
|
| 134 |
story.append(Spacer(1, 0.2 * inch))
|
| 135 |
else:
|
| 136 |
+
# Handle bold/italics, then wrap emojis in the final string
|
| 137 |
+
formatted_line = re.sub(r'\*\*(.*?)\*\*', r'<b>\1</b>', line)
|
| 138 |
formatted_line = re.sub(r'_(.*?)_', r'<i>\1</i>', formatted_line)
|
| 139 |
+
final_text = self._wrap_emojis_for_reportlab(formatted_line)
|
| 140 |
+
story.append(Paragraph(final_text, style_normal))
|
| 141 |
+
|
| 142 |
return story
|
| 143 |
|
| 144 |
def create_pdf(self, md_asset: Path, layout_name: str, layout_properties: dict):
|
|
|
|
| 148 |
"""
|
| 149 |
try:
|
| 150 |
md_content = md_asset.read_text(encoding="utf-8")
|
|
|
|
| 151 |
date_str = datetime.datetime.now().strftime("%Y-%m-%d")
|
| 152 |
output_filename = f"{md_asset.stem}_{layout_name.replace(' ', '-')}_{date_str}.pdf"
|
| 153 |
output_path = OUTPUT_DIR / output_filename
|
| 154 |
|
|
|
|
| 155 |
doc = SimpleDocTemplate(
|
| 156 |
str(output_path),
|
| 157 |
pagesize=layout_properties.get("size", A4),
|
| 158 |
+
rightMargin=inch, leftMargin=inch,
|
| 159 |
+
topMargin=inch, bottomMargin=inch
|
|
|
|
|
|
|
| 160 |
)
|
|
|
|
| 161 |
story = self._markdown_to_story(md_content)
|
|
|
|
|
|
|
| 162 |
doc.build(story)
|
|
|
|
| 163 |
except Exception as e:
|
| 164 |
st.error(f"Failed to process {md_asset.name} with ReportLab: {e}")
|
| 165 |
|
|
|
|
| 169 |
def get_file_download_link(file_path: Path) -> str:
|
| 170 |
"""
|
| 171 |
π To grab your file and not delay, a special link is paved today.
|
|
|
|
| 172 |
"""
|
| 173 |
with open(file_path, "rb") as f:
|
| 174 |
data = base64.b64encode(f.read()).decode()
|
|
|
|
| 177 |
def display_file_explorer():
|
| 178 |
"""
|
| 179 |
π To see your files, both old and new, this handy explorer gives a view.
|
|
|
|
| 180 |
"""
|
| 181 |
st.header("π File Explorer")
|
|
|
|
| 182 |
st.subheader("Source Markdown Files (.md)")
|
| 183 |
md_files = list(Path(".").glob("*.md"))
|
| 184 |
if not md_files:
|
|
|
|
| 192 |
st.markdown(get_file_download_link(md_file), unsafe_allow_html=True)
|
| 193 |
|
| 194 |
st.subheader("Generated PDF Files")
|
|
|
|
| 195 |
pdf_files = sorted(list(OUTPUT_DIR.glob("*.pdf")), key=lambda p: p.stat().st_mtime, reverse=True)
|
| 196 |
if not pdf_files:
|
| 197 |
st.info("No PDFs generated yet. Click the button above to start.")
|
|
|
|
| 203 |
with col2:
|
| 204 |
st.markdown(get_file_download_link(pdf_file), unsafe_allow_html=True)
|
| 205 |
|
|
|
|
| 206 |
# --- Main App Execution ---
|
| 207 |
|
| 208 |
def main():
|
|
|
|
| 213 |
st.title("π Markdown to PDF Generator")
|
| 214 |
st.markdown("This tool converts all `.md` files in this directory to PDF. It now supports emojis! π")
|
| 215 |
|
| 216 |
+
if not any(Path(".").glob("*.md")):
|
|
|
|
| 217 |
with open("sample.md", "w", encoding="utf-8") as f:
|
| 218 |
f.write("# Sample Document π\n\nThis is a sample markdown file. **ReportLab** is creating the PDF. Emojis like π and π‘ should now appear correctly.\n\n### Features\n- Item 1\n- Item 2\n\n```\ndef hello_world():\n print(\"Hello, PDF! π\")\n```\n")
|
| 219 |
st.rerun()
|
| 220 |
|
|
|
|
| 221 |
pdf_generator = PDFGenerator(EMOJI_FONT_PATH)
|
| 222 |
|
| 223 |
if st.button("π Generate PDFs from all Markdown Files", type="primary"):
|
| 224 |
markdown_files = list(Path(".").glob("*.md"))
|
|
|
|
| 225 |
if not markdown_files:
|
| 226 |
st.warning("No `.md` files found. Please add a markdown file to the directory.")
|
| 227 |
else:
|
| 228 |
total_pdfs = len(markdown_files) * len(LAYOUTS)
|
| 229 |
progress_bar = st.progress(0, text="Starting PDF generation...")
|
| 230 |
pdf_count = 0
|
|
|
|
| 231 |
with st.spinner("Generating PDFs... Please wait."):
|
| 232 |
for md_file in markdown_files:
|
| 233 |
st.info(f"Processing: **{md_file.name}**")
|
| 234 |
for name, properties in LAYOUTS.items():
|
|
|
|
| 235 |
pdf_generator.create_pdf(md_file, name, properties)
|
| 236 |
pdf_count += 1
|
| 237 |
progress_bar.progress(pdf_count / total_pdfs, f"Generated {pdf_count}/{total_pdfs} PDFs...")
|
|
|
|
| 238 |
st.success("β
PDF generation complete!")
|
| 239 |
st.balloons()
|
|
|
|
| 240 |
st.rerun()
|
| 241 |
|
| 242 |
display_file_explorer()
|
| 243 |
|
|
|
|
| 244 |
if __name__ == "__main__":
|
| 245 |
main()
|