Make telemetry opt-in for v4.4 (#132)

Author: sidmohan0

README.md

@@ -25,6 +25,10 @@ pip install datafog[nlp-advanced]
 pip install datafog[all]
 ```
 
+Python 3.13 support is certified for the core SDK and CLI. Optional extras such
+as `nlp`, `nlp-advanced`, `ocr`, `distributed`, and `all` are available but not
+yet certified on Python 3.13.
+
 ## Quick Start
 
 ```python
@@ -132,9 +136,15 @@ datafog hash-text "john@example.com"
 
 ## Telemetry
 
-DataFog includes anonymous telemetry by default.
+DataFog telemetry is disabled by default.
+
+To opt in:
+
+```bash
+export DATAFOG_TELEMETRY=1
+```
 
-To opt out:
+To force telemetry off:
 
 ```bash
 export DATAFOG_NO_TELEMETRY=1

datafog/telemetry.py

@@ -1,10 +1,13 @@
 """
-Anonymous, opt-out usage telemetry for DataFog.
+Anonymous, opt-in usage telemetry for DataFog.
 
 Collects anonymous usage data to help the DataFog team understand which engines,
 functions, and features are actually used. No text content is ever sent.
 
-Opt out by setting either environment variable:
+Telemetry is disabled by default. Opt in by setting:
+    DATAFOG_TELEMETRY=1
+
+Force telemetry off by setting either environment variable:
     DATAFOG_NO_TELEMETRY=1
     DO_NOT_TRACK=1
 """
@@ -29,13 +32,18 @@
 _scope = threading.local()
 
 
+def _env_truthy(name: str) -> bool:
+    """Return True when an environment variable explicitly opts in/out."""
+    return os.environ.get(name, "").strip().lower() in {"1", "true", "yes", "on"}
+
+
 def _is_telemetry_enabled() -> bool:
-    """Check if telemetry is enabled (opt-out via env vars)."""
-    if os.environ.get("DATAFOG_NO_TELEMETRY", "").strip() == "1":
+    """Check if telemetry is enabled (opt-in, with opt-out overrides)."""
+    if _env_truthy("DATAFOG_NO_TELEMETRY"):
         return False
-    if os.environ.get("DO_NOT_TRACK", "").strip() == "1":
+    if _env_truthy("DO_NOT_TRACK"):
         return False
-    return True
+    return _env_truthy("DATAFOG_TELEMETRY")
 
 
 def _get_anonymous_id() -> str:

docs/v5-compatibility-matrix.rst

@@ -127,9 +127,9 @@ Compatibility Matrix
      - Deferred for v5.1+ overhaul.
      - Keep compatibility where practical; no runtime package installs.
    * - Telemetry
-     - Default-on opt-out telemetry.
+     - Opt-in telemetry.
      - Trust-critical behavior change.
-     - Make opt-in or no-network-by-default.
+     - Keep no-network-by-default.
    * - ``*_lean`` and ``*_original`` modules
      - Parallel historical implementations.
      - Remove or make private after migration path.

scripts/generate_changelog.py

@@ -7,6 +7,16 @@
 from datetime import datetime
 
 
+def get_current_version():
+    """Read the current package version from datafog/__about__.py."""
+    try:
+        with open("datafog/__about__.py") as f:
+            match = re.search(r'^__version__ = "([^"]+)"', f.read(), re.M)
+            return match.group(1) if match else None
+    except OSError:
+        return None
+
+
 def get_latest_tag():
     """Get the latest git tag."""
     try:
@@ -65,6 +75,7 @@ def generate_changelog(beta=False, alpha=False):
     """Generate changelog content."""
     latest_tag = get_latest_tag()
     commits = get_commits_since_tag(latest_tag)
+    current_version = get_current_version()
 
     if not commits:
         return "No changes since last release."
@@ -85,6 +96,32 @@ def generate_changelog(beta=False, alpha=False):
         changelog = "# What's New\n\n"
         changelog += f"*Released: {datetime.now().strftime('%Y-%m-%d')}*\n\n"
 
+    if not alpha and not beta and current_version == "4.4.0":
+        changelog += "## Python 3.13 Support Scope\n\n"
+        changelog += (
+            "Python 3.13 support is certified for the core SDK and CLI: "
+            "`pip install datafog` and `pip install datafog[cli]`.\n\n"
+        )
+        changelog += (
+            "Optional extras including `nlp`, `nlp-advanced`, `ocr`, "
+            "`distributed`, and `all` are available but not yet certified on "
+            "Python 3.13. They will be validated separately based on user "
+            "demand.\n\n"
+        )
+        changelog += "## v5 Migration Bridge\n\n"
+        changelog += (
+            "This release adds the v5-preview top-level APIs `datafog.scan`, "
+            "`datafog.redact`, and `datafog.protect` while keeping the legacy "
+            "`datafog.detect` and `datafog.process` APIs working with targeted "
+            "migration warnings.\n\n"
+        )
+        changelog += "## Privacy Defaults\n\n"
+        changelog += (
+            "Telemetry is now opt-in. DataFog does not send telemetry unless "
+            "`DATAFOG_TELEMETRY=1` is explicitly set. `DATAFOG_NO_TELEMETRY=1` "
+            "and `DO_NOT_TRACK=1` continue to force telemetry off.\n\n"
+        )
+
     if categories["features"]:
         changelog += "## 🚀 New Features\n"
         for commit in categories["features"]: