import cairo import pangocairo surface = cairo.PDFSurface("shaped_khmer.pdf", 200, 100) context = cairo.Context(surface) pangocairo_context = pangocairo.CairoContext(context) pangocairo_context.set_antialias(cairo.ANTIALIAS_SUBPIXEL)
import pdfplumber with pdfplumber.open("khmer_document.pdf") as pdf: for page in pdf.pages: text = page.extract_text() print(text) Works for basic extraction but may fail with complex Khmer glyph order. python khmer pdf
layout = pangocairo_context.create_layout() layout.set_text("កម្ពុជា") layout.set_font_description(pango.FontDescription("Khmer OS 12")) import cairo import pangocairo surface = cairo
import fitz # PyMuPDF doc = fitz.open("khmer_document.pdf") for page in doc: text = page.get_text() print(text) pdfplumber extracts text while preserving layout, good for Khmer. encoding="utf-8") as f: yaml.dump(data
pangocairo_context.update_layout(layout) pangocairo_context.show_layout(layout) surface.finish() For scanned Khmer PDFs, convert to images then use Tesseract with Khmer language pack.
with open("data.yaml", "w", encoding="utf-8") as f: yaml.dump(data, f, allow_unicode=True)