Spaces:
Build error
Build error
Update app.py
Browse files
app.py
CHANGED
@@ -6,11 +6,9 @@ import os
|
|
6 |
import shutil
|
7 |
import fitz # PyMuPDF
|
8 |
from PIL import Image
|
|
|
9 |
import io
|
10 |
-
|
11 |
-
# Imports for new formats
|
12 |
-
from docx import Document
|
13 |
-
import openpyxl
|
14 |
|
15 |
from reportlab.platypus import SimpleDocTemplate, Paragraph, Spacer, Table, TableStyle, PageBreak, BaseDocTemplate, Frame, PageTemplate, Image as ReportLabImage
|
16 |
from reportlab.lib.styles import getSampleStyleSheet, ParagraphStyle
|
@@ -27,8 +25,10 @@ LAYOUTS = {
|
|
27 |
"A4 Landscape": {"size": landscape(A4)},
|
28 |
"Letter Portrait": {"size": letter},
|
29 |
"Letter Landscape": {"size": landscape(letter)},
|
|
|
|
|
30 |
}
|
31 |
-
OUTPUT_DIR = CWD / "
|
32 |
PREVIEW_DIR = CWD / "previews"
|
33 |
FONT_DIR = CWD
|
34 |
|
@@ -36,263 +36,289 @@ FONT_DIR = CWD
|
|
36 |
OUTPUT_DIR.mkdir(exist_ok=True)
|
37 |
PREVIEW_DIR.mkdir(exist_ok=True)
|
38 |
|
|
|
39 |
# --- Font & Emoji Handling ---
|
40 |
-
EMOJI_FONT_PATH = None
|
41 |
-
EMOJI_IMAGE_CACHE = {}
|
42 |
|
43 |
-
def
|
44 |
-
"""Finds and registers all .ttf files from the
|
45 |
-
|
46 |
text_font_names = []
|
|
|
47 |
|
48 |
-
|
49 |
-
|
50 |
-
|
51 |
-
|
52 |
-
|
53 |
-
raise FileNotFoundError("CRITICAL: 'NotoColorEmoji-Regular.ttf' not found. This file is required for emoji support.")
|
54 |
-
|
55 |
-
for font_path in FONT_DIR.glob("*.ttf"):
|
56 |
try:
|
57 |
font_name = font_path.stem
|
58 |
pdfmetrics.registerFont(TTFont(font_name, str(font_path)))
|
59 |
-
pdfmetrics.
|
60 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
61 |
text_font_names.append(font_name)
|
62 |
except Exception as e:
|
63 |
print(f"Could not register font {font_path.name}: {e}")
|
64 |
|
65 |
if not text_font_names:
|
|
|
66 |
text_font_names.append('Helvetica')
|
67 |
-
return sorted(text_font_names)
|
68 |
|
69 |
-
|
70 |
-
"
|
71 |
-
|
72 |
-
|
73 |
-
|
74 |
-
try:
|
75 |
-
rect = fitz.Rect(0, 0, size_pt * 1.5, size_pt * 1.5)
|
76 |
-
doc = fitz.open()
|
77 |
-
page = doc.new_page(width=rect.width, height=rect.height)
|
78 |
-
page.insert_font(fontname="emoji", fontfile=EMOJI_FONT_PATH)
|
79 |
-
page.insert_text(fitz.Point(0, size_pt * 1.1), emoji_char, fontname="emoji", fontsize=size_pt)
|
80 |
-
pix = page.get_pixmap(alpha=True, dpi=300)
|
81 |
-
doc.close()
|
82 |
-
img_buffer = io.BytesIO(pix.tobytes("png"))
|
83 |
-
img_buffer.seek(0)
|
84 |
-
EMOJI_IMAGE_CACHE[(emoji_char, size_pt)] = img_buffer
|
85 |
-
return img_buffer
|
86 |
-
except Exception as e:
|
87 |
-
print(f"ERROR: Could not render emoji '{emoji_char}': {e}")
|
88 |
-
return None
|
89 |
|
90 |
-
|
91 |
-
|
92 |
-
|
93 |
-
|
94 |
-
|
95 |
-
|
96 |
-
|
97 |
|
|
|
98 |
emoji_pattern = re.compile(f"([{re.escape(''.join(map(chr, range(0x1f600, 0x1f650))))}"
|
99 |
f"{re.escape(''.join(map(chr, range(0x1f300, 0x1f5ff))))}"
|
100 |
f"{re.escape(''.join(map(chr, range(0x1f900, 0x1f9ff))))}"
|
101 |
-
f"{re.escape(''.join(map(chr, range(0x2600, 0x26ff))))}
|
102 |
-
|
103 |
-
|
104 |
-
|
105 |
-
|
106 |
-
|
107 |
-
|
108 |
-
|
109 |
-
|
110 |
-
|
111 |
-
|
112 |
-
|
113 |
-
flowables.append(img)
|
114 |
-
else:
|
115 |
-
formatted_part = re.sub(r'\*\*(.*?)\*\*', r'<b>\1</b>', part)
|
116 |
-
flowables.append(Paragraph(formatted_part, style))
|
117 |
-
return Table([flowables], colWidths=[None] * len(flowables), style=[('VALIGN', (0, 0), (-1, -1), 'MIDDLE')]) if flowables else Spacer(0,0)
|
118 |
-
|
119 |
-
story, first_heading = [], True
|
120 |
-
for line in markdown_text.split('\n'):
|
121 |
-
stripped = line.strip()
|
122 |
-
if stripped.startswith("# "):
|
123 |
-
if not first_heading: story.append(PageBreak())
|
124 |
-
story.append(text_to_flowables(stripped[2:], style_h1)); first_heading = False
|
125 |
-
elif stripped.startswith("## "):
|
126 |
-
story.append(text_to_flowables(stripped[3:], style_h2))
|
127 |
-
elif stripped.startswith(("- ", "* ")):
|
128 |
-
story.append(text_to_flowables(stripped[2:], ParagraphStyle(parent=style_body, leftIndent=20, bulletIndent=10)))
|
129 |
-
elif stripped:
|
130 |
-
story.append(text_to_flowables(stripped, style_body))
|
131 |
else:
|
132 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
133 |
return story
|
134 |
|
135 |
-
def create_pdf(md_content, font_name, pagesize, num_columns):
|
136 |
-
"""Generates a PDF file from markdown content."""
|
137 |
-
md_buffer = io.BytesIO()
|
138 |
-
story = markdown_to_story(md_content, font_name)
|
139 |
-
if num_columns > 1:
|
140 |
-
doc = BaseDocTemplate(md_buffer, pagesize=pagesize, leftMargin=0.5*inch, rightMargin=0.5*inch, topMargin=0.5*inch, bottomMargin=0.5*inch)
|
141 |
-
frame_width = (doc.width / num_columns) - (num_columns - 1) * 0.1*inch
|
142 |
-
frames = [Frame(doc.leftMargin + i * (frame_width + 0.2*inch), doc.bottomMargin, frame_width, doc.height) for i in range(num_columns)]
|
143 |
-
doc.addPageTemplates([PageTemplate(id='MultiCol', frames=frames)])
|
144 |
-
else:
|
145 |
-
doc = SimpleDocTemplate(md_buffer, pagesize=pagesize)
|
146 |
-
doc.build(story)
|
147 |
-
return md_buffer
|
148 |
-
|
149 |
-
def create_docx(md_content):
|
150 |
-
"""Generates a DOCX file from markdown content."""
|
151 |
-
document = Document()
|
152 |
-
for line in md_content.split('\n'):
|
153 |
-
line = line.strip()
|
154 |
-
if line.startswith('# '): document.add_heading(line[2:], level=1)
|
155 |
-
elif line.startswith('## '): document.add_heading(line[3:], level=2)
|
156 |
-
elif line.startswith(('- ','* ')): document.add_paragraph(line[2:], style='List Bullet')
|
157 |
-
elif line:
|
158 |
-
p = document.add_paragraph()
|
159 |
-
parts = re.split(r'(\*\*.*?\*\*)', line)
|
160 |
-
for part in parts:
|
161 |
-
if part.startswith('**') and part.endswith('**'):
|
162 |
-
p.add_run(part[2:-2]).bold = True
|
163 |
-
else:
|
164 |
-
p.add_run(part)
|
165 |
-
return document
|
166 |
-
|
167 |
-
def create_xlsx(md_content):
|
168 |
-
"""Generates an XLSX file, splitting content by H1 headers into columns."""
|
169 |
-
workbook = openpyxl.Workbook()
|
170 |
-
sheet = workbook.active
|
171 |
-
sections = re.split(r'\n# ', '\n' + md_content)
|
172 |
-
if not sections[0].strip(): sections.pop(0)
|
173 |
-
|
174 |
-
for c_idx, section in enumerate(sections, 1):
|
175 |
-
lines = section.split('\n')
|
176 |
-
sheet.cell(row=1, column=c_idx, value=lines[0].strip()) # Header
|
177 |
-
for r_idx, line_content in enumerate(lines[1:], 2):
|
178 |
-
sheet.cell(row=r_idx, column=c_idx, value=line_content.strip())
|
179 |
-
return workbook
|
180 |
|
181 |
def create_pdf_preview(pdf_path: Path):
|
182 |
preview_path = PREVIEW_DIR / f"{pdf_path.stem}.png"
|
183 |
try:
|
184 |
-
doc = fitz.open(pdf_path)
|
185 |
-
|
186 |
-
pix = page.get_pixmap(dpi=150)
|
187 |
-
pix.save(str(preview_path))
|
188 |
-
doc.close()
|
189 |
return str(preview_path)
|
190 |
except Exception as e:
|
191 |
-
print(f"
|
192 |
-
return None
|
193 |
|
194 |
# --- Main API Function ---
|
195 |
-
def
|
196 |
-
if not files: raise gr.Error("Please upload at least one Markdown
|
197 |
-
if not
|
|
|
198 |
|
199 |
shutil.rmtree(OUTPUT_DIR, ignore_errors=True); shutil.rmtree(PREVIEW_DIR, ignore_errors=True)
|
200 |
OUTPUT_DIR.mkdir(); PREVIEW_DIR.mkdir()
|
201 |
-
EMOJI_IMAGE_CACHE.clear()
|
202 |
|
203 |
-
|
204 |
-
md_contents = []
|
205 |
for f in files:
|
206 |
-
|
207 |
-
|
208 |
-
|
209 |
-
|
210 |
-
md_content = "\n\n# ".join(md_contents)
|
211 |
-
if not md_content.startswith("#"): md_content = "# " + md_content
|
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 |
-
filename = f"Outline_{time_str}.xlsx"
|
238 |
-
output_path = OUTPUT_DIR / filename
|
239 |
-
xlsx_book.save(output_path)
|
240 |
-
generated_files.append(output_path)
|
241 |
-
except Exception as e:
|
242 |
-
print(f"Failed to generate {format_choice}: {e}")
|
243 |
-
gr.Warning(f"Failed to generate {format_choice}. See console for details.")
|
244 |
|
245 |
-
|
246 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
247 |
|
248 |
-
return
|
249 |
|
250 |
# --- Gradio UI Definition ---
|
251 |
-
|
252 |
-
|
253 |
-
|
254 |
-
|
255 |
-
|
256 |
-
|
257 |
-
|
258 |
-
|
259 |
-
|
260 |
-
|
261 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
262 |
|
263 |
with gr.Row():
|
264 |
with gr.Column(scale=1):
|
265 |
gr.Markdown("### βοΈ Generation Settings")
|
266 |
-
uploaded_files = gr.File(label="Upload Markdown Files", file_count="multiple", file_types=[".md"])
|
267 |
-
output_formats = gr.CheckboxGroup(choices=["PDF", "DOCX", "XLSX"], label="Select Output Formats", value=["PDF"])
|
268 |
|
269 |
-
with gr.
|
270 |
-
|
271 |
-
|
272 |
-
num_columns_slider = gr.Slider(label="Text Columns (for PDF)", minimum=1, maximum=4, step=1, value=1)
|
273 |
|
274 |
-
|
|
|
|
|
|
|
275 |
|
276 |
with gr.Column(scale=2):
|
277 |
-
gr.Markdown("### πΌοΈ
|
278 |
-
gallery_output = gr.Gallery(label="PDF Previews", show_label=False, elem_id="gallery", columns=3, height="auto", object_fit="contain")
|
279 |
-
log_output = gr.Markdown(label="Generation Log", value="
|
280 |
-
downloadable_files_output = gr.Files(label="Download Generated
|
281 |
|
282 |
-
|
283 |
-
gr.Warning("The application is in a degraded state. Required font files are missing. Please check the console log and add the necessary files.")
|
284 |
-
else:
|
285 |
-
generate_btn.click(
|
286 |
-
fn=generate_outputs_api,
|
287 |
-
inputs=[uploaded_files, output_formats, selected_layouts, selected_fonts, num_columns_slider],
|
288 |
-
outputs=[gallery_output, log_output, downloadable_files_output]
|
289 |
-
)
|
290 |
|
291 |
if __name__ == "__main__":
|
292 |
-
|
293 |
-
demo.launch()
|
294 |
-
else:
|
295 |
-
print("\n" + "="*60)
|
296 |
-
print("Application launch aborted due to missing font files.")
|
297 |
-
print("Please ensure 'NotoColorEmoji-Regular.ttf' is in the project directory.")
|
298 |
-
print("="*60)
|
|
|
6 |
import shutil
|
7 |
import fitz # PyMuPDF
|
8 |
from PIL import Image
|
9 |
+
from collections import defaultdict
|
10 |
import io
|
11 |
+
from pypdf import PdfWriter
|
|
|
|
|
|
|
12 |
|
13 |
from reportlab.platypus import SimpleDocTemplate, Paragraph, Spacer, Table, TableStyle, PageBreak, BaseDocTemplate, Frame, PageTemplate, Image as ReportLabImage
|
14 |
from reportlab.lib.styles import getSampleStyleSheet, ParagraphStyle
|
|
|
25 |
"A4 Landscape": {"size": landscape(A4)},
|
26 |
"Letter Portrait": {"size": letter},
|
27 |
"Letter Landscape": {"size": landscape(letter)},
|
28 |
+
"Legal Portrait": {"size": legal},
|
29 |
+
"Legal Landscape": {"size": landscape(legal)},
|
30 |
}
|
31 |
+
OUTPUT_DIR = CWD / "generated_pdfs"
|
32 |
PREVIEW_DIR = CWD / "previews"
|
33 |
FONT_DIR = CWD
|
34 |
|
|
|
36 |
OUTPUT_DIR.mkdir(exist_ok=True)
|
37 |
PREVIEW_DIR.mkdir(exist_ok=True)
|
38 |
|
39 |
+
|
40 |
# --- Font & Emoji Handling ---
|
|
|
|
|
41 |
|
42 |
+
def register_local_fonts():
|
43 |
+
"""Finds and registers all .ttf files from the application's base directory."""
|
44 |
+
print("--- Font Registration Process Starting ---")
|
45 |
text_font_names = []
|
46 |
+
emoji_font_name = None
|
47 |
|
48 |
+
print(f"Scanning for fonts in: {FONT_DIR.absolute()}")
|
49 |
+
font_files = list(FONT_DIR.glob("*.ttf"))
|
50 |
+
print(f"Found {len(font_files)} .ttf files: {[f.name for f in font_files]}")
|
51 |
+
|
52 |
+
for font_path in font_files:
|
|
|
|
|
|
|
53 |
try:
|
54 |
font_name = font_path.stem
|
55 |
pdfmetrics.registerFont(TTFont(font_name, str(font_path)))
|
56 |
+
pdfmetrics.registerFont(TTFont(f"{font_name}-Bold", str(font_path)))
|
57 |
+
pdfmetrics.registerFont(TTFont(f"{font_name}-Italic", str(font_path)))
|
58 |
+
pdfmetrics.registerFont(TTFont(f"{font_name}-BoldItalic", str(font_path)))
|
59 |
+
pdfmetrics.registerFontFamily(font_name, normal=font_name, bold=f"{font_name}-Bold", italic=f"{font_name}-Italic", boldItalic=f"{font_name}-BoldItalic")
|
60 |
+
|
61 |
+
if "notocoloremoji-regular" in font_name.lower():
|
62 |
+
emoji_font_name = font_name
|
63 |
+
elif "notoemoji" not in font_name.lower(): # Exclude other symbol fonts from text selection
|
64 |
text_font_names.append(font_name)
|
65 |
except Exception as e:
|
66 |
print(f"Could not register font {font_path.name}: {e}")
|
67 |
|
68 |
if not text_font_names:
|
69 |
+
print("WARNING: No text fonts found. Adding 'Helvetica' as a default.")
|
70 |
text_font_names.append('Helvetica')
|
|
|
71 |
|
72 |
+
print(f"Successfully registered user-selectable fonts: {text_font_names}")
|
73 |
+
print(f"Emoji font set to: {emoji_font_name}")
|
74 |
+
print("--- Font Registration Process Finished ---")
|
75 |
+
return sorted(text_font_names), emoji_font_name
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
76 |
|
77 |
+
def apply_emoji_font(text: str, emoji_font_name: str) -> str:
|
78 |
+
"""
|
79 |
+
Intelligently wraps emoji characters in a <font> tag, preserving existing HTML-like tags.
|
80 |
+
This prevents invalid nested tags like <b><font>...</font></b> which ReportLab handles poorly.
|
81 |
+
"""
|
82 |
+
if not emoji_font_name:
|
83 |
+
return text
|
84 |
|
85 |
+
# Regex to find emojis
|
86 |
emoji_pattern = re.compile(f"([{re.escape(''.join(map(chr, range(0x1f600, 0x1f650))))}"
|
87 |
f"{re.escape(''.join(map(chr, range(0x1f300, 0x1f5ff))))}"
|
88 |
f"{re.escape(''.join(map(chr, range(0x1f900, 0x1f9ff))))}"
|
89 |
+
f"{re.escape(''.join(map(chr, range(0x2600, 0x26ff))))}"
|
90 |
+
f"{re.escape(''.join(map(chr, range(0x2700, 0x27bf))))}]+)")
|
91 |
+
|
92 |
+
# Regex to split the string by existing tags (<b>, <i>)
|
93 |
+
tag_pattern = re.compile(r"(<[^>]+>)")
|
94 |
+
parts = tag_pattern.split(text)
|
95 |
+
|
96 |
+
result = []
|
97 |
+
for part in parts:
|
98 |
+
if tag_pattern.match(part):
|
99 |
+
# It's a tag, append it as is
|
100 |
+
result.append(part)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
101 |
else:
|
102 |
+
# It's text, apply emoji font to any emojis within this segment
|
103 |
+
result.append(emoji_pattern.sub(fr'<font name="{emoji_font_name}">\1</font>', part))
|
104 |
+
|
105 |
+
return "".join(result)
|
106 |
+
|
107 |
+
|
108 |
+
# --- PDF Generation & Handling ---
|
109 |
+
|
110 |
+
def markdown_to_story(markdown_text: str, font_name: str, emoji_font: str):
|
111 |
+
"""
|
112 |
+
Converts markdown to a ReportLab story, with enhanced styling and page breaks.
|
113 |
+
This version correctly separates structural parsing from content formatting.
|
114 |
+
"""
|
115 |
+
styles = getSampleStyleSheet()
|
116 |
+
# Define styles for various markdown elements
|
117 |
+
style_normal = ParagraphStyle('BodyText', fontName=font_name, spaceAfter=6, leading=14, fontSize=10)
|
118 |
+
style_h1 = ParagraphStyle('h1', parent=styles['h1'], fontName=font_name, spaceBefore=12, fontSize=24, leading=28, textColor=colors.darkblue)
|
119 |
+
style_h2 = ParagraphStyle('h2', parent=styles['h2'], fontName=font_name, fontSize=18, leading=22, spaceBefore=10)
|
120 |
+
style_h3 = ParagraphStyle('h3', parent=styles['h3'], fontName=font_name, fontSize=14, leading=18, spaceBefore=8)
|
121 |
+
style_code = ParagraphStyle('Code', fontName='Courier', backColor=colors.whitesmoke, textColor=colors.darkred, borderWidth=1, borderColor=colors.lightgrey, padding=8, leading=12, fontSize=9)
|
122 |
+
style_table_header = ParagraphStyle('TableHeader', parent=style_normal, fontName=font_name + "-Bold" if font_name != 'Helvetica' else 'Helvetica-Bold')
|
123 |
+
|
124 |
+
story = []
|
125 |
+
lines = markdown_text.split('\n')
|
126 |
+
|
127 |
+
in_code_block, in_table = False, False
|
128 |
+
code_block_text, table_data = "", []
|
129 |
+
first_heading = True
|
130 |
+
|
131 |
+
for line in lines:
|
132 |
+
stripped_line = line.strip()
|
133 |
+
|
134 |
+
if stripped_line.startswith("```"):
|
135 |
+
if in_code_block:
|
136 |
+
story.append(Paragraph(code_block_text.replace('\n', '<br/>'), style_code)); story.append(Spacer(1, 0.1 * inch))
|
137 |
+
in_code_block = False; code_block_text = ""
|
138 |
+
else: in_code_block = True
|
139 |
+
continue
|
140 |
+
if in_code_block:
|
141 |
+
code_block_text += line.replace('&', '&').replace('<', '<').replace('>', '>') + '\n'
|
142 |
+
continue
|
143 |
+
|
144 |
+
if stripped_line.startswith('|'):
|
145 |
+
if not in_table: in_table = True
|
146 |
+
if all(c in '-|: ' for c in stripped_line): continue
|
147 |
+
cells = [cell.strip() for cell in stripped_line.strip('|').split('|')]
|
148 |
+
table_data.append(cells)
|
149 |
+
continue
|
150 |
+
if in_table:
|
151 |
+
in_table = False
|
152 |
+
if table_data:
|
153 |
+
header_content = [apply_emoji_font(re.sub(r'\*\*(.*?)\*\*', r'<b>\1</b>', cell), emoji_font) for cell in table_data[0]]
|
154 |
+
header = [Paragraph(cell, style_table_header) for cell in header_content]
|
155 |
+
|
156 |
+
formatted_rows = []
|
157 |
+
for row in table_data[1:]:
|
158 |
+
formatted_cells = [apply_emoji_font(re.sub(r'\*\*(.*?)\*\*', r'<b>\1</b>', cell), emoji_font) for cell in row]
|
159 |
+
formatted_rows.append([Paragraph(cell, style_normal) for cell in formatted_cells])
|
160 |
+
|
161 |
+
table = Table([header] + formatted_rows, hAlign='LEFT', repeatRows=1)
|
162 |
+
table.setStyle(TableStyle([('BACKGROUND', (0, 0), (-1, 0), colors.lightgrey), ('GRID', (0, 0), (-1, -1), 1, colors.darkgrey), ('VALIGN', (0,0), (-1,-1), 'MIDDLE')]))
|
163 |
+
story.append(table); story.append(Spacer(1, 0.2 * inch))
|
164 |
+
table_data = []
|
165 |
+
|
166 |
+
if not stripped_line:
|
167 |
+
story.append(Spacer(1, 0.1 * inch))
|
168 |
+
continue
|
169 |
+
|
170 |
+
# Default content is the whole stripped line
|
171 |
+
content = stripped_line
|
172 |
+
style = style_normal
|
173 |
+
extra_args = {}
|
174 |
+
|
175 |
+
# Detect structural elements and extract the raw content
|
176 |
+
if stripped_line.startswith("# "):
|
177 |
+
if not first_heading: story.append(PageBreak())
|
178 |
+
content = stripped_line.lstrip('# '); style = style_h1; first_heading = False
|
179 |
+
elif stripped_line.startswith("## "):
|
180 |
+
content = stripped_line.lstrip('## '); style = style_h2
|
181 |
+
elif stripped_line.startswith("### "):
|
182 |
+
content = stripped_line.lstrip('### '); style = style_h3
|
183 |
+
elif stripped_line.startswith(("- ", "* ")):
|
184 |
+
content = stripped_line[2:]; extra_args['bulletText'] = 'β’'
|
185 |
+
|
186 |
+
# Now, format the extracted content
|
187 |
+
# Apply markdown formatting for bold/italic
|
188 |
+
formatted_content = re.sub(r'_(.*?)_', r'<i>\1</i>', re.sub(r'\*\*(.*?)\*\*', r'<b>\1</b>', content))
|
189 |
+
# Then apply emoji font to the already formatted line
|
190 |
+
final_content = apply_emoji_font(formatted_content, emoji_font)
|
191 |
+
|
192 |
+
story.append(Paragraph(final_content, style, **extra_args))
|
193 |
return story
|
194 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
195 |
|
196 |
def create_pdf_preview(pdf_path: Path):
|
197 |
preview_path = PREVIEW_DIR / f"{pdf_path.stem}.png"
|
198 |
try:
|
199 |
+
doc = fitz.open(pdf_path); page = doc.load_page(0); pix = page.get_pixmap()
|
200 |
+
pix.save(str(preview_path)); doc.close()
|
|
|
|
|
|
|
201 |
return str(preview_path)
|
202 |
except Exception as e:
|
203 |
+
print(f"Could not create preview for {pdf_path.name}: {e}"); return None
|
|
|
204 |
|
205 |
# --- Main API Function ---
|
206 |
+
def generate_pdfs_api(files, layouts, fonts, num_columns, page_w_mult, page_h_mult, progress=gr.Progress(track_tqdm=True)):
|
207 |
+
if not files: raise gr.Error("Please upload at least one Markdown or Image file.")
|
208 |
+
if not layouts: raise gr.Error("Please select at least one page layout.")
|
209 |
+
if not fonts: raise gr.Error("Please select at least one font.")
|
210 |
|
211 |
shutil.rmtree(OUTPUT_DIR, ignore_errors=True); shutil.rmtree(PREVIEW_DIR, ignore_errors=True)
|
212 |
OUTPUT_DIR.mkdir(); PREVIEW_DIR.mkdir()
|
|
|
213 |
|
214 |
+
grouped_files = defaultdict(lambda: {'md': [], 'img': []})
|
|
|
215 |
for f in files:
|
216 |
+
file_path = Path(f.name)
|
217 |
+
stem = file_path.stem.split('_')[0] if '_' in file_path.stem else file_path.stem
|
218 |
+
if file_path.suffix.lower() == '.md': grouped_files[stem]['md'].append(file_path)
|
219 |
+
elif file_path.suffix.lower() in ['.png', '.jpg', '.jpeg']: grouped_files[stem]['img'].append(file_path)
|
|
|
|
|
220 |
|
221 |
+
log_updates, generated_pdf_paths = "", []
|
222 |
+
|
223 |
+
for stem, assets in progress.tqdm(grouped_files.items(), desc="Processing File Groups"):
|
224 |
+
for layout_name in layouts:
|
225 |
+
for font_name in fonts:
|
226 |
+
merger = PdfWriter()
|
227 |
+
|
228 |
+
if assets['md']:
|
229 |
+
md_content = "\n".join([p.read_text(encoding='utf-8') for p in assets['md']])
|
230 |
+
md_buffer = io.BytesIO()
|
231 |
+
story = markdown_to_story(md_content, font_name, EMOJI_FONT_NAME)
|
232 |
+
|
233 |
+
base_w, base_h = LAYOUTS[layout_name]["size"]
|
234 |
+
pagesize = (base_w * page_w_mult, base_h * page_h_mult)
|
235 |
+
|
236 |
+
if num_columns > 1:
|
237 |
+
doc = BaseDocTemplate(md_buffer, pagesize=pagesize, leftMargin=0.5*inch, rightMargin=0.5*inch, topMargin=0.5*inch, bottomMargin=0.5*inch)
|
238 |
+
frame_width = (doc.width / num_columns) - (num_columns - 1) * 0.1*inch
|
239 |
+
frames = [Frame(doc.leftMargin + i * (frame_width + 0.2*inch), doc.bottomMargin, frame_width, doc.height) for i in range(num_columns)]
|
240 |
+
doc.addPageTemplates([PageTemplate(id='MultiCol', frames=frames)])
|
241 |
+
else:
|
242 |
+
doc = SimpleDocTemplate(md_buffer, pagesize=pagesize)
|
243 |
+
doc.build(story)
|
244 |
+
merger.append(fileobj=md_buffer)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
245 |
|
246 |
+
for img_path in assets['img']:
|
247 |
+
with Image.open(img_path) as img: img_width, img_height = img.size
|
248 |
+
img_buffer = io.BytesIO()
|
249 |
+
doc = SimpleDocTemplate(img_buffer, pagesize=(img_width, img_height), leftMargin=0, rightMargin=0, topMargin=0, bottomMargin=0)
|
250 |
+
doc.build([ReportLabImage(img_path, width=img_width, height=img_height)])
|
251 |
+
merger.append(fileobj=img_buffer)
|
252 |
+
|
253 |
+
if len(merger.pages) > 0:
|
254 |
+
time_str = datetime.datetime.now().strftime('%m-%d-%a_%I%M%p').upper()
|
255 |
+
filename = f"{stem}_{time_str}_{layout_name.replace(' ','-')}_{page_w_mult}x{page_h_mult}_{font_name}_Cols{num_columns}.pdf"
|
256 |
+
output_path = OUTPUT_DIR / filename
|
257 |
+
with open(output_path, "wb") as f: merger.write(f)
|
258 |
+
generated_pdf_paths.append(output_path)
|
259 |
+
log_updates += f"Generated: {filename}\n"
|
260 |
+
|
261 |
+
gallery_previews = [create_pdf_preview(p) for p in generated_pdf_paths]
|
262 |
+
final_gallery = [g for g in gallery_previews if g is not None]
|
263 |
|
264 |
+
return final_gallery, log_updates, [str(p) for p in generated_pdf_paths]
|
265 |
|
266 |
# --- Gradio UI Definition ---
|
267 |
+
AVAILABLE_FONTS, EMOJI_FONT_NAME = register_local_fonts()
|
268 |
+
SAMPLE_MARKDOWN = """# Deities Guide: Mythology and Moral Lessons
|
269 |
+
|
270 |
+
1. π **Introduction**
|
271 |
+
- **Purpose**: Explore deities, spirits, saints, and beings with their epic stories and morals!
|
272 |
+
- **Usage**: A guide for learning and storytelling across traditions. οΈ
|
273 |
+
- **Themes**: Justice βοΈ, faith π, hubris ποΈ, redemption β¨, cosmic order π.
|
274 |
+
|
275 |
+
2. π οΈ **Core Concepts of Divinity**
|
276 |
+
- **Powers**: Creation π, omniscience ποΈβπ¨οΈ, shapeshifting π¦ across entities.
|
277 |
+
- **Life Cycle**: Mortality β³, immortality βΎοΈ, transitions like saints and avatars π.
|
278 |
+
- **Communication**: Omens ποΈ, visions ποΈ, miracles β¨ from gods and spirits.
|
279 |
+
|
280 |
+
# βοΈ Arthurian Legends
|
281 |
+
- **Merlin, Morgan le Fay, Arthur**: Mentor π§, rival π§ββοΈ, son π.
|
282 |
+
- **Relation**: Family tests loyalty π€.
|
283 |
+
- **Lesson**: Honor ποΈ vs. betrayal π‘οΈ.
|
284 |
+
|
285 |
+
# ποΈ Greek Mythology
|
286 |
+
- **Zeus, Hera, Athena**: Father β‘, mother π, daughter π¦.
|
287 |
+
- **Relation**: Family rules with tension π©οΈ.
|
288 |
+
- **Lesson**: Hubris ΰ€
ΰ€Ήΰ€ΰ€ΰ€Ύΰ€° meets wisdom π§ .
|
289 |
+
|
290 |
+
# ποΈ Hindu Trimurti
|
291 |
+
- **Brahma, Vishnu, Shiva**: Creator Brahma, preserver Vishnu, destroyer Shiva.
|
292 |
+
- **Relation**: Divine trio cycles existence π.
|
293 |
+
- **Lesson**: Balance βοΈ sustains life π.
|
294 |
+
"""
|
295 |
+
with open(CWD / "sample.md", "w", encoding="utf-8") as f: f.write(SAMPLE_MARKDOWN)
|
296 |
+
|
297 |
+
with gr.Blocks(theme=gr.themes.Soft(), title="Advanced PDF Generator") as demo:
|
298 |
+
gr.Markdown("# π Advanced PDF Layout Engine")
|
299 |
+
gr.Markdown("Upload Markdown/Image files. The app finds local `.ttf` fonts. Group assets with a common name (e.g., `Doc_part1.md`, `Doc_img1.png`) to combine them. `# Headers` create automatic page breaks.")
|
300 |
|
301 |
with gr.Row():
|
302 |
with gr.Column(scale=1):
|
303 |
gr.Markdown("### βοΈ Generation Settings")
|
304 |
+
uploaded_files = gr.File(label="Upload Markdown & Image Files", file_count="multiple", file_types=[".md", ".png", ".jpg", ".jpeg"])
|
|
|
305 |
|
306 |
+
with gr.Row():
|
307 |
+
page_w_mult_slider = gr.Slider(label="Page Width Multiplier", minimum=1, maximum=5, step=1, value=1)
|
308 |
+
page_h_mult_slider = gr.Slider(label="Page Height Multiplier", minimum=1, maximum=2, step=1, value=1)
|
|
|
309 |
|
310 |
+
num_columns_slider = gr.Slider(label="Number of Text Columns", minimum=1, maximum=4, step=1, value=1)
|
311 |
+
selected_layouts = gr.CheckboxGroup(choices=list(LAYOUTS.keys()), label="Select Base Page Layout", value=["A4 Portrait"])
|
312 |
+
selected_fonts = gr.CheckboxGroup(choices=AVAILABLE_FONTS, label="Select Text Font", value=[AVAILABLE_FONTS[0]] if AVAILABLE_FONTS else [])
|
313 |
+
generate_btn = gr.Button("π Generate PDFs", variant="primary")
|
314 |
|
315 |
with gr.Column(scale=2):
|
316 |
+
gr.Markdown("### πΌοΈ PDF Preview Gallery")
|
317 |
+
gallery_output = gr.Gallery(label="Generated PDF Previews", show_label=False, elem_id="gallery", columns=3, height="auto", object_fit="contain")
|
318 |
+
log_output = gr.Markdown(label="Generation Log", value="Logs will appear here...")
|
319 |
+
downloadable_files_output = gr.Files(label="Download Generated PDFs")
|
320 |
|
321 |
+
generate_btn.click(fn=generate_pdfs_api, inputs=[uploaded_files, selected_layouts, selected_fonts, num_columns_slider, page_w_mult_slider, page_h_mult_slider], outputs=[gallery_output, log_output, downloadable_files_output])
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
322 |
|
323 |
if __name__ == "__main__":
|
324 |
+
demo.launch()
|
|
|
|
|
|
|
|
|
|
|
|