LocalFoodAI_lanfr144/scratch/test_pre_length.py

import os
from markdown_pdf import MarkdownPdf, Section
import fitz

docs_dir = r"c:/Users/lanfr144/Documents/DOPRO1/Antigravity/Food/docs"
md_file = os.path.join(docs_dir, "Installation_Guide.md")

with open(md_file, "r", encoding="utf-8") as f:
    md_content = f.read()

# Current CSS
user_css = """
* {
    color: #1a1a1a !important;
}
body {
    font-family: 'Helvetica', 'Arial', sans-serif !important;
    color: #1a1a1a !important;
    background-color: #ffffff !important;
}
h1, h2, h3, h4, h5, h6, h1 *, h2 *, h3 *, h4 *, h5 *, h6 * {
    color: #000000 !important;
    margin-top: 18px !important;
    margin-bottom: 8px !important;
}
pre {
    background-color: #f8f9fa !important;
    border: 1px solid #e9ecef !important;
    padding: 2px !important;
    margin-top: 8px !important;
    margin-bottom: 16px !important;
    border-radius: 3px !important;
    font-family: 'Courier New', 'Courier', monospace !important;
    font-size: 10pt !important;
    white-space: pre-wrap !important;
    word-break: break-all !important;
}
pre code, pre * {
    font-family: 'Courier New', 'Courier', monospace !important;
    font-size: 10pt !important;
    color: #212529 !important;
    background-color: #f8f9fa !important;
    white-space: pre-wrap !important;
    word-break: break-all !important;
}
code {
    font-family: 'Courier New', 'Courier', monospace !important;
    font-size: 10pt !important;
    color: #b02a37 !important;
    background-color: #f8f9fa !important;
    padding: 2px 4px !important;
    border-radius: 3px !important;
    white-space: pre-wrap !important;
    word-break: break-all !important;
}
a, a * {
    color: #0d6efd !important;
}
blockquote, blockquote * {
    color: #555555 !important;
    border-left: 4px solid #ccc !important;
    padding-left: 10px !important;
}
table, tr, td, th, table * {
    color: #1a1a1a !important;
    border-color: #cccccc !important;
}
th {
    background-color: #f2f2f2 !important;
}
ul, li {
    list-style-type: disc !important;
}
"""

pdf = MarkdownPdf()
section = Section(md_content, paper_size="A4", root=docs_dir)
pdf.add_section(section, user_css=user_css)
pdf.save("scratch/test_css_on.pdf")

doc = fitz.open("scratch/test_css_on.pdf")
print("Total pages:", len(doc))
for i in range(len(doc)):
    txt = doc[i].get_text()
    print(f"Page {i+1} text length: {len(txt)}")
    if i == 1:
        print("--- PAGE 2 TEXT CONTENT ---")
        print(txt.replace("\ufb01", "fi"))
doc.close()