חילוץ טקסט מקבצי OneNote באמצעות Python

pip install aspose-note

from aspose.note import Document, RichText

doc = Document("MyNotes.one")
for rt in doc.GetChildNodes(RichText):
    if rt.Text:
        print(rt.Text)

from aspose.note import Document, RichText

doc = Document("MyNotes.one")
lines = [rt.Text for rt in doc.GetChildNodes(RichText) if rt.Text]

with open("extracted.txt", "w", encoding="utf-8") as f:
    f.write("\n".join(lines))

print(f"Saved {len(lines)} text blocks to extracted.txt")

from aspose.note import Document, Page, RichText

doc = Document("MyNotes.one")
for page in doc.GetChildNodes(Page):
    title = (
        page.Title.TitleText.Text
        if page.Title and page.Title.TitleText
        else "(untitled)"
    )
    page_texts = [rt.Text for rt in page.GetChildNodes(RichText) if rt.Text]
    print(f"\n=== {title} ===")
    for text in page_texts:
        print(text)

from aspose.note import Document, RichText

doc = Document("MyNotes.one")
for rt in doc.GetChildNodes(RichText):
    for run in rt.Runs:
        if run.Style.IsHyperlink and run.Style.HyperlinkAddress:
            print(f"{run.Text!r}  ->  {run.Style.HyperlinkAddress}")

from aspose.note import Document, RichText

doc = Document("MyNotes.one")
for rt in doc.GetChildNodes(RichText):
    for run in rt.Runs:
        s = run.Style
        if any([s.Bold, s.Italic, s.Underline]):
            flags = ", ".join(f for f, v in [
                ("bold", s.Bold), ("italic", s.Italic), ("underline", s.Underline)
            ] if v)
            print(f"[{flags}] {run.Text.strip()!r}")

import io, urllib.request
from aspose.note import Document, RichText

##Example: load from bytes already in memory
one_bytes = open("MyNotes.one", "rb").read()
doc = Document(io.BytesIO(one_bytes))
texts = [rt.Text for rt in doc.GetChildNodes(RichText) if rt.Text]
print(f"Extracted {len(texts)} text block(s)")

import sys
if hasattr(sys.stdout, "reconfigure"):
    sys.stdout.reconfigure(encoding="utf-8", errors="replace")