Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

feat(replays): Scrub PII from texts mutation nodes #1796

Merged
merged 5 commits into from
Jan 31, 2023
Merged

feat(replays): Scrub PII from texts mutation nodes #1796

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
c16f470
Scrub PII from texts mutation nodes
cmanallen Jan 30, 2023
653d991
Update changelog
cmanallen Jan 30, 2023
24e92b0
Add event parsing and scrubbing coverage
cmanallen Jan 30, 2023
a50c04c
Assert texts objects are scrubbed for PII
cmanallen Jan 30, 2023
e03b7d9
ref: Streamline import
jan-auer Jan 31, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions CHANGELOG.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,6 +12,7 @@

- Fix type errors in replay recording parsing. ([#1765](https://github.com/getsentry/relay/pull/1765))
- Remove error and session sample rate fields from replay-event parser. ([#1791](https://github.com/getsentry/relay/pull/1791))
- Scrub replay recording PII from mutation "texts" vector. ([#1796](https://github.com/getsentry/relay/pull/1796))

## 23.1.0

Expand Down
71 changes: 41 additions & 30 deletions relay-replays/src/recording/mod.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -138,11 +138,11 @@ impl RecordingProcessor<'_> {
) -> Result<(), ProcessingAction> {
match variant {
IncrementalSourceDataVariant::Mutation(mutation) => {
for text in &mut mutation.texts {
self.strip_pii(&mut text.value)?
}
for addition in &mut mutation.adds {
match self.recurse_snapshot_node(&mut addition.node) {
Ok(_) => {}
Err(e) => return Err(e),
}
self.recurse_snapshot_node(&mut addition.node)?
}
}
IncrementalSourceDataVariant::Input(input) => self.strip_pii(&mut input.text)?,
Expand Down Expand Up @@ -442,14 +442,20 @@ struct InputIncrementalSourceData {
#[derive(Debug, Serialize, Deserialize)]
#[serde(rename_all = "camelCase")]
struct MutationIncrementalSourceData {
texts: Vec<Value>,
texts: Vec<Text>,
attributes: Vec<Value>,
removes: Vec<Value>,
adds: Vec<MutationAdditionIncrementalSourceData>,
#[serde(skip_serializing_if = "Option::is_none")]
is_attach_iframe: Option<Value>,
}

#[derive(Debug, Serialize, Deserialize)]
struct Text {
id: i32,
value: String,
}

#[derive(Debug)]
struct DefaultIncrementalSourceData {
pub source: u8,
Expand All @@ -466,11 +472,11 @@ struct MutationAdditionIncrementalSourceData {

#[cfg(test)]
mod tests {
use crate::recording;
use crate::recording::Event;
use assert_json_diff::assert_json_eq;
use serde_json::{Error, Value};

use crate::recording::{self, Event};

fn loads(bytes: &[u8]) -> Result<Vec<Event>, Error> {
serde_json::from_slice(bytes)
}
Expand Down Expand Up @@ -629,20 +635,9 @@ mod tests {

recording::strip_pii(&mut events).unwrap();

let aa = events.pop().unwrap();
if let recording::Event::T3(bb) = aa {
if let recording::IncrementalSourceDataVariant::Mutation(mut cc) = bb.data {
let dd = cc.adds.pop().unwrap();
if let recording::NodeVariant::T2(mut ee) = dd.node.variant {
let ff = ee.child_nodes.pop().unwrap();
if let recording::NodeVariant::T3(gg) = ff.variant {
assert_eq!(gg.text_content, "[ip]");
return;
}
}
}
}
unreachable!();
let parsed = serde_json::to_string(&events).unwrap();
assert!(parsed.contains("\"value\":\"[ip]\"")); // Assert texts were mutated.
assert!(parsed.contains("\"textContent\":\"[ip]\"")) // Assert text node was mutated.
}

#[test]
Expand Down Expand Up @@ -683,24 +678,40 @@ mod tests {
assert_json_eq!(input_parsed, input_raw);
}

// Event coverage
// Event Parsing and Scrubbing.

#[test]
fn test_rrweb_event_3_parsing() {
fn test_scrub_pii_full_snapshot_event() {
let payload = include_bytes!("../../tests/fixtures/rrweb-event-2.json");
let mut events: Vec<recording::Event> = serde_json::from_slice(payload).unwrap();
recording::strip_pii(&mut events).unwrap();

let scrubbed_result = serde_json::to_string(&events).unwrap();
assert!(scrubbed_result.contains("\"attributes\":{\"src\":\"#\"}"));
assert!(scrubbed_result.contains("\"textContent\":\"my ssn is ***********\""));
}

#[test]
fn test_scrub_pii_incremental_snapshot_event() {
let payload = include_bytes!("../../tests/fixtures/rrweb-event-3.json");
let mut events: Vec<recording::Event> = serde_json::from_slice(payload).unwrap();
recording::strip_pii(&mut events).unwrap();

let input_parsed: recording::Event = serde_json::from_slice(payload).unwrap();
let input_raw: Value = serde_json::from_slice(payload).unwrap();
assert_json_eq!(input_parsed, input_raw)
let scrubbed_result = serde_json::to_string(&events).unwrap();
assert!(scrubbed_result.contains("\"textContent\":\"[creditcard]\""));
assert!(scrubbed_result.contains("\"value\":\"***********\""));
}

#[test]
fn test_rrweb_event_5_parsing() {
fn test_scrub_pii_custom_event() {
let payload = include_bytes!("../../tests/fixtures/rrweb-event-5.json");
let mut events: Vec<recording::Event> = serde_json::from_slice(payload).unwrap();
recording::strip_pii(&mut events).unwrap();

let input_parsed: Vec<recording::Event> = serde_json::from_slice(payload).unwrap();
let input_raw: Value = serde_json::from_slice(payload).unwrap();
assert_json_eq!(input_parsed, input_raw);
let scrubbed_result = serde_json::to_string(&events).unwrap();
assert!(scrubbed_result.contains("\"description\":\"[creditcard]\""));
assert!(scrubbed_result.contains("\"description\":\"https://sentry.io?ip-address=[ip]\""));
assert!(scrubbed_result.contains("\"message\":\"[email]\""));
}
}

Expand Down
82 changes: 82 additions & 0 deletions relay-replays/tests/fixtures/rrweb-event-2.json
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,82 @@
[
{
"type": 2,
"data": {
"node": {
"type": 0,
"childNodes": [
{
"type": 1,
"name": "html",
"publicId": "",
"systemId": "",
"id": 2
},
{
"type": 2,
"tagName": "html",
"attributes": {
"lang": "en"
},
"childNodes": [
{
"type": 2,
"tagName": "head",
"attributes": {},
"childNodes": [],
"id": 4
},
{
"type": 2,
"tagName": "body",
"attributes": {},
"childNodes": [
{
"type": 2,
"tagName": "a",
"attributes": {
"href": "https://github.com/getsentry/sourcemaps.io",
"class": "github-corner",
"aria-label": "View source on Github"
},
"childNodes": [
{
"type": 2,
"tagName": "div",
"attributes": {},
"childNodes": [
{
"type": 3,
"textContent": "my ssn is 111-11-1111",
"id": 244
}
],
"id": 18
},
{
"type": 2,
"tagName": "img",
"attributes": {"src": "https://www.sentry.io"},
"childNodes": [],
"id": 19
}
],
"id": 17
}
],
"id": 16
}
],
"id": 3
}
],
"id": 1
},
"initialOffset": {
"left": 0,
"top": 0
}
},
"timestamp": 1658770770472
}
]
71 changes: 38 additions & 33 deletions relay-replays/tests/fixtures/rrweb-event-3.json
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,33 +1,38 @@
{
"type": 3,
"data": {
"source": 1,
"positions": [
{
"x": 2412,
"y": 131,
"id": 24,
"timeOffset": -402
},
{
"x": 2393,
"y": 133,
"id": 24,
"timeOffset": -232
},
{
"x": 2390,
"y": 125,
"id": 24,
"timeOffset": -64
},
{
"x": 2404,
"y": 62,
"id": 24,
"timeOffset": -10
}
]
},
"timestamp": 1658770772443
}
[
{
"type": 3,
"data": {
"source": 0,
"texts": [
{
"id": 44,
"value": "111-11-1111"
}
],
"attributes": [],
"removes": [],
"isAttachIframe": true,
"adds": [
{
"parentId": 74,
"previousId": null,
"nextId": null,
"node": {
"type": 2,
"tagName": "h1",
"attributes": {},
"childNodes": [
{
"type": 3,
"textContent": "4111-1111-1111-1111",
"id": 284
}
],
"id": 283
}
}
]
},
"timestamp": 1665063926945
}
]
6 changes: 3 additions & 3 deletions relay-replays/tests/fixtures/rrweb-event-5.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,7 +6,7 @@
"tag": "performanceSpan",
"payload": {
"op": "memory",
"description": "memory",
"description": "4111-1111-1111-1111",
"startTimestamp": 1658770772.902,
"endTimestamp": 1658770772.902,
"data": {
Expand All @@ -26,7 +26,7 @@
"tag": "performanceSpan",
"payload": {
"op": "navigation.navigate",
"description": "https://sentry.io",
"description": "https://sentry.io?ip-address=127.0.0.1",
"startTimestamp": 1665063926.125,
"endTimestamp": 1665063926.833,
"data": {
Expand All @@ -45,7 +45,7 @@
"timestamp": 1674135065.772,
"type": "default",
"category": "ui.input",
"message": "something happened",
"message": "[email protected]",
"data": {
"nodeId": 14235
}
Expand Down
7 changes: 6 additions & 1 deletion relay-replays/tests/fixtures/rrweb-pii-ip-address.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,7 +3,12 @@
"type": 3,
"data": {
"source": 0,
"texts": [],
"texts": [
{
"id": 1,
"value": "127.0.0.1"
}
],
"attributes": [],
"removes": [],
"adds": [
Expand Down