|
0981a08…
|
noreply
|
1 |
"""Microsoft OneNote source connector using the m365 CLI (cli-microsoft365). |
|
0981a08…
|
noreply
|
2 |
|
|
0981a08…
|
noreply
|
3 |
Fetches pages from OneNote notebooks via the `m365` CLI tool. |
|
0981a08…
|
noreply
|
4 |
Outputs plain text suitable for KG ingestion. |
|
0981a08…
|
noreply
|
5 |
|
|
0981a08…
|
noreply
|
6 |
Requires: npm install -g @pnp/cli-microsoft365 |
|
0981a08…
|
noreply
|
7 |
Auth: m365 login (interactive) |
|
0981a08…
|
noreply
|
8 |
Docs: https://pnp.github.io/cli-microsoft365/ |
|
0981a08…
|
noreply
|
9 |
""" |
|
0981a08…
|
noreply
|
10 |
|
|
0981a08…
|
noreply
|
11 |
import json |
|
0981a08…
|
noreply
|
12 |
import logging |
|
0981a08…
|
noreply
|
13 |
import re |
|
0981a08…
|
noreply
|
14 |
import shutil |
|
0981a08…
|
noreply
|
15 |
import subprocess |
|
0981a08…
|
noreply
|
16 |
from pathlib import Path |
|
0981a08…
|
noreply
|
17 |
from typing import Any, List, Optional |
|
0981a08…
|
noreply
|
18 |
|
|
0981a08…
|
noreply
|
19 |
from video_processor.sources.base import BaseSource, SourceFile |
|
0981a08…
|
noreply
|
20 |
|
|
0981a08…
|
noreply
|
21 |
logger = logging.getLogger(__name__) |
|
0981a08…
|
noreply
|
22 |
|
|
0981a08…
|
noreply
|
23 |
|
|
0981a08…
|
noreply
|
24 |
def _run_m365(args: List[str], timeout: int = 30) -> Any: |
|
0981a08…
|
noreply
|
25 |
"""Run an m365 CLI command and return parsed JSON output.""" |
|
0981a08…
|
noreply
|
26 |
cmd = ["m365"] + args + ["--output", "json"] |
|
0981a08…
|
noreply
|
27 |
proc = subprocess.run(cmd, capture_output=True, text=True, timeout=timeout) |
|
0981a08…
|
noreply
|
28 |
if proc.returncode != 0: |
|
0981a08…
|
noreply
|
29 |
raise RuntimeError(f"m365 {' '.join(args)} failed: {proc.stderr.strip()}") |
|
0981a08…
|
noreply
|
30 |
try: |
|
0981a08…
|
noreply
|
31 |
return json.loads(proc.stdout) |
|
0981a08…
|
noreply
|
32 |
except json.JSONDecodeError: |
|
0981a08…
|
noreply
|
33 |
return proc.stdout.strip() |
|
0981a08…
|
noreply
|
34 |
|
|
0981a08…
|
noreply
|
35 |
|
|
0981a08…
|
noreply
|
36 |
def _html_to_text(html: str) -> str: |
|
0981a08…
|
noreply
|
37 |
"""Strip HTML tags and decode entities to produce plain text. |
|
0981a08…
|
noreply
|
38 |
|
|
0981a08…
|
noreply
|
39 |
Uses only stdlib ``re`` — no external dependencies. |
|
0981a08…
|
noreply
|
40 |
""" |
|
0981a08…
|
noreply
|
41 |
# Remove script/style blocks entirely |
|
0981a08…
|
noreply
|
42 |
text = re.sub(r"<(script|style)[^>]*>.*?</\1>", "", html, flags=re.DOTALL | re.IGNORECASE) |
|
0981a08…
|
noreply
|
43 |
# Replace <br>, <p>, <div>, <li>, <tr> with newlines for readability |
|
0981a08…
|
noreply
|
44 |
text = re.sub(r"<br\s*/?>", "\n", text, flags=re.IGNORECASE) |
|
0981a08…
|
noreply
|
45 |
text = re.sub(r"</(p|div|li|tr|h[1-6])>", "\n", text, flags=re.IGNORECASE) |
|
0981a08…
|
noreply
|
46 |
# Strip remaining tags |
|
0981a08…
|
noreply
|
47 |
text = re.sub(r"<[^>]+>", "", text) |
|
0981a08…
|
noreply
|
48 |
# Decode common HTML entities |
|
0981a08…
|
noreply
|
49 |
entity_map = { |
|
0981a08…
|
noreply
|
50 |
"&": "&", |
|
0981a08…
|
noreply
|
51 |
"<": "<", |
|
0981a08…
|
noreply
|
52 |
">": ">", |
|
0981a08…
|
noreply
|
53 |
""": '"', |
|
0981a08…
|
noreply
|
54 |
"'": "'", |
|
0981a08…
|
noreply
|
55 |
"'": "'", |
|
0981a08…
|
noreply
|
56 |
" ": " ", |
|
0981a08…
|
noreply
|
57 |
} |
|
0981a08…
|
noreply
|
58 |
for entity, char in entity_map.items(): |
|
0981a08…
|
noreply
|
59 |
text = text.replace(entity, char) |
|
0981a08…
|
noreply
|
60 |
# Decode numeric entities ({ and ) |
|
0981a08…
|
noreply
|
61 |
text = re.sub(r"&#x([0-9a-fA-F]+);", lambda m: chr(int(m.group(1), 16)), text) |
|
0981a08…
|
noreply
|
62 |
text = re.sub(r"&#(\d+);", lambda m: chr(int(m.group(1))), text) |
|
0981a08…
|
noreply
|
63 |
# Collapse excessive blank lines |
|
0981a08…
|
noreply
|
64 |
text = re.sub(r"\n{3,}", "\n\n", text) |
|
0981a08…
|
noreply
|
65 |
return text.strip() |
|
0981a08…
|
noreply
|
66 |
|
|
0981a08…
|
noreply
|
67 |
|
|
0981a08…
|
noreply
|
68 |
class OneNoteSource(BaseSource): |
|
0981a08…
|
noreply
|
69 |
""" |
|
0981a08…
|
noreply
|
70 |
Fetch pages from OneNote notebooks via the m365 CLI. |
|
0981a08…
|
noreply
|
71 |
|
|
0981a08…
|
noreply
|
72 |
Usage: |
|
0981a08…
|
noreply
|
73 |
source = OneNoteSource() # all notebooks |
|
0981a08…
|
noreply
|
74 |
source = OneNoteSource(notebook_name="Work Notes") # specific notebook |
|
0981a08…
|
noreply
|
75 |
source = OneNoteSource(notebook_name="Work", section_name="Meetings") |
|
0981a08…
|
noreply
|
76 |
files = source.list_videos() |
|
0981a08…
|
noreply
|
77 |
source.download_all(files, Path("./notes")) |
|
0981a08…
|
noreply
|
78 |
""" |
|
0981a08…
|
noreply
|
79 |
|
|
0981a08…
|
noreply
|
80 |
def __init__( |
|
0981a08…
|
noreply
|
81 |
self, |
|
0981a08…
|
noreply
|
82 |
notebook_name: Optional[str] = None, |
|
0981a08…
|
noreply
|
83 |
section_name: Optional[str] = None, |
|
0981a08…
|
noreply
|
84 |
): |
|
0981a08…
|
noreply
|
85 |
self.notebook_name = notebook_name |
|
0981a08…
|
noreply
|
86 |
self.section_name = section_name |
|
0981a08…
|
noreply
|
87 |
|
|
0981a08…
|
noreply
|
88 |
def authenticate(self) -> bool: |
|
0981a08…
|
noreply
|
89 |
"""Check if m365 CLI is installed and logged in.""" |
|
0981a08…
|
noreply
|
90 |
if not shutil.which("m365"): |
|
0981a08…
|
noreply
|
91 |
logger.error("m365 CLI not found. Install with: npm install -g @pnp/cli-microsoft365") |
|
0981a08…
|
noreply
|
92 |
return False |
|
0981a08…
|
noreply
|
93 |
try: |
|
0981a08…
|
noreply
|
94 |
result = _run_m365(["status"], timeout=10) |
|
0981a08…
|
noreply
|
95 |
if isinstance(result, dict) and result.get("connectedAs"): |
|
0981a08…
|
noreply
|
96 |
return True |
|
0981a08…
|
noreply
|
97 |
if isinstance(result, str) and "Logged in" in result: |
|
0981a08…
|
noreply
|
98 |
return True |
|
0981a08…
|
noreply
|
99 |
logger.error("m365 not logged in. Run: m365 login") |
|
0981a08…
|
noreply
|
100 |
return False |
|
0981a08…
|
noreply
|
101 |
except (RuntimeError, subprocess.TimeoutExpired): |
|
0981a08…
|
noreply
|
102 |
logger.error("m365 not logged in. Run: m365 login") |
|
0981a08…
|
noreply
|
103 |
return False |
|
0981a08…
|
noreply
|
104 |
|
|
0981a08…
|
noreply
|
105 |
def list_videos( |
|
0981a08…
|
noreply
|
106 |
self, |
|
0981a08…
|
noreply
|
107 |
folder_id: Optional[str] = None, |
|
0981a08…
|
noreply
|
108 |
folder_path: Optional[str] = None, |
|
0981a08…
|
noreply
|
109 |
patterns: Optional[List[str]] = None, |
|
0981a08…
|
noreply
|
110 |
) -> List[SourceFile]: |
|
0981a08…
|
noreply
|
111 |
"""List OneNote pages across notebooks/sections. Returns SourceFile per page.""" |
|
0981a08…
|
noreply
|
112 |
files: List[SourceFile] = [] |
|
0981a08…
|
noreply
|
113 |
|
|
0981a08…
|
noreply
|
114 |
# Step 1: List notebooks |
|
0981a08…
|
noreply
|
115 |
try: |
|
0981a08…
|
noreply
|
116 |
notebooks = _run_m365(["onenote", "notebook", "list"], timeout=60) |
|
0981a08…
|
noreply
|
117 |
except RuntimeError as e: |
|
0981a08…
|
noreply
|
118 |
logger.error(f"Failed to list OneNote notebooks: {e}") |
|
0981a08…
|
noreply
|
119 |
return [] |
|
0981a08…
|
noreply
|
120 |
|
|
0981a08…
|
noreply
|
121 |
if not isinstance(notebooks, list): |
|
0981a08…
|
noreply
|
122 |
notebooks = [] |
|
0981a08…
|
noreply
|
123 |
|
|
0981a08…
|
noreply
|
124 |
# Filter notebooks by name if specified |
|
0981a08…
|
noreply
|
125 |
if self.notebook_name: |
|
0981a08…
|
noreply
|
126 |
notebooks = [ |
|
0981a08…
|
noreply
|
127 |
nb |
|
0981a08…
|
noreply
|
128 |
for nb in notebooks |
|
0981a08…
|
noreply
|
129 |
if self.notebook_name.lower() in nb.get("displayName", "").lower() |
|
0981a08…
|
noreply
|
130 |
] |
|
0981a08…
|
noreply
|
131 |
|
|
0981a08…
|
noreply
|
132 |
for notebook in notebooks: |
|
0981a08…
|
noreply
|
133 |
notebook_id = notebook.get("id", "") |
|
0981a08…
|
noreply
|
134 |
notebook_name = notebook.get("displayName", "Untitled Notebook") |
|
0981a08…
|
noreply
|
135 |
|
|
0981a08…
|
noreply
|
136 |
# Step 2: List sections in this notebook |
|
0981a08…
|
noreply
|
137 |
try: |
|
0981a08…
|
noreply
|
138 |
sections = _run_m365( |
|
0981a08…
|
noreply
|
139 |
["onenote", "section", "list", "--notebookId", notebook_id], |
|
0981a08…
|
noreply
|
140 |
timeout=60, |
|
0981a08…
|
noreply
|
141 |
) |
|
0981a08…
|
noreply
|
142 |
except RuntimeError as e: |
|
0981a08…
|
noreply
|
143 |
logger.warning(f"Failed to list sections for notebook '{notebook_name}': {e}") |
|
0981a08…
|
noreply
|
144 |
continue |
|
0981a08…
|
noreply
|
145 |
|
|
0981a08…
|
noreply
|
146 |
if not isinstance(sections, list): |
|
0981a08…
|
noreply
|
147 |
sections = [] |
|
0981a08…
|
noreply
|
148 |
|
|
0981a08…
|
noreply
|
149 |
# Filter sections by name if specified |
|
0981a08…
|
noreply
|
150 |
if self.section_name: |
|
0981a08…
|
noreply
|
151 |
sections = [ |
|
0981a08…
|
noreply
|
152 |
s |
|
0981a08…
|
noreply
|
153 |
for s in sections |
|
0981a08…
|
noreply
|
154 |
if self.section_name.lower() in s.get("displayName", "").lower() |
|
0981a08…
|
noreply
|
155 |
] |
|
0981a08…
|
noreply
|
156 |
|
|
0981a08…
|
noreply
|
157 |
for section in sections: |
|
0981a08…
|
noreply
|
158 |
section_id = section.get("id", "") |
|
0981a08…
|
noreply
|
159 |
section_name = section.get("displayName", "Untitled Section") |
|
0981a08…
|
noreply
|
160 |
|
|
0981a08…
|
noreply
|
161 |
# Step 3: List pages in this section |
|
0981a08…
|
noreply
|
162 |
try: |
|
0981a08…
|
noreply
|
163 |
pages = _run_m365( |
|
0981a08…
|
noreply
|
164 |
["onenote", "page", "list", "--sectionId", section_id], |
|
0981a08…
|
noreply
|
165 |
timeout=60, |
|
0981a08…
|
noreply
|
166 |
) |
|
0981a08…
|
noreply
|
167 |
except RuntimeError as e: |
|
0981a08…
|
noreply
|
168 |
logger.warning(f"Failed to list pages in section '{section_name}': {e}") |
|
0981a08…
|
noreply
|
169 |
continue |
|
0981a08…
|
noreply
|
170 |
|
|
0981a08…
|
noreply
|
171 |
if not isinstance(pages, list): |
|
0981a08…
|
noreply
|
172 |
pages = [] |
|
0981a08…
|
noreply
|
173 |
|
|
0981a08…
|
noreply
|
174 |
for page in pages: |
|
0981a08…
|
noreply
|
175 |
page_id = page.get("id", "") |
|
0981a08…
|
noreply
|
176 |
title = page.get("title", "Untitled Page").strip() or "Untitled Page" |
|
0981a08…
|
noreply
|
177 |
modified = page.get("lastModifiedDateTime") |
|
0981a08…
|
noreply
|
178 |
# Build a path for organizational context |
|
0981a08…
|
noreply
|
179 |
page_path = f"{notebook_name}/{section_name}/{title}" |
|
0981a08…
|
noreply
|
180 |
|
|
0981a08…
|
noreply
|
181 |
files.append( |
|
0981a08…
|
noreply
|
182 |
SourceFile( |
|
0981a08…
|
noreply
|
183 |
name=title, |
|
0981a08…
|
noreply
|
184 |
id=str(page_id), |
|
0981a08…
|
noreply
|
185 |
size_bytes=None, |
|
0981a08…
|
noreply
|
186 |
mime_type="text/html", |
|
0981a08…
|
noreply
|
187 |
modified_at=modified, |
|
0981a08…
|
noreply
|
188 |
path=page_path, |
|
0981a08…
|
noreply
|
189 |
) |
|
0981a08…
|
noreply
|
190 |
) |
|
0981a08…
|
noreply
|
191 |
|
|
0981a08…
|
noreply
|
192 |
logger.info(f"Found {len(files)} page(s) in OneNote") |
|
0981a08…
|
noreply
|
193 |
return files |
|
0981a08…
|
noreply
|
194 |
|
|
0981a08…
|
noreply
|
195 |
def download(self, file: SourceFile, destination: Path) -> Path: |
|
0981a08…
|
noreply
|
196 |
"""Download a OneNote page's content as a text file.""" |
|
0981a08…
|
noreply
|
197 |
destination = Path(destination) |
|
0981a08…
|
noreply
|
198 |
destination.parent.mkdir(parents=True, exist_ok=True) |
|
0981a08…
|
noreply
|
199 |
|
|
0981a08…
|
noreply
|
200 |
try: |
|
0981a08…
|
noreply
|
201 |
result = _run_m365( |
|
0981a08…
|
noreply
|
202 |
["onenote", "page", "get", "--id", file.id], |
|
0981a08…
|
noreply
|
203 |
timeout=60, |
|
0981a08…
|
noreply
|
204 |
) |
|
0981a08…
|
noreply
|
205 |
except RuntimeError as e: |
|
0981a08…
|
noreply
|
206 |
raise RuntimeError(f"Failed to fetch OneNote page {file.id}: {e}") from e |
|
0981a08…
|
noreply
|
207 |
|
|
0981a08…
|
noreply
|
208 |
# Extract HTML content from the result |
|
0981a08…
|
noreply
|
209 |
if isinstance(result, dict): |
|
0981a08…
|
noreply
|
210 |
html = result.get("content", result.get("body", {}).get("content", "")) |
|
0981a08…
|
noreply
|
211 |
if not html: |
|
0981a08…
|
noreply
|
212 |
# Fallback: serialize the whole response |
|
0981a08…
|
noreply
|
213 |
html = json.dumps(result, indent=2) |
|
0981a08…
|
noreply
|
214 |
elif isinstance(result, str): |
|
0981a08…
|
noreply
|
215 |
html = result |
|
0981a08…
|
noreply
|
216 |
else: |
|
0981a08…
|
noreply
|
217 |
html = str(result) |
|
0981a08…
|
noreply
|
218 |
|
|
0981a08…
|
noreply
|
219 |
text = _html_to_text(html) |
|
0981a08…
|
noreply
|
220 |
destination.write_text(text, encoding="utf-8") |
|
0981a08…
|
noreply
|
221 |
logger.info(f"Saved page '{file.name}' to {destination}") |
|
0981a08…
|
noreply
|
222 |
return destination |