chore: keep real names out of code and tests
The daily-summary cleaner hard-coded two contacts' names in its opening phrases; a pattern now strips "Today, <Name> and I discussed" for any contact (with a test). Video path tests use a neutral folder name. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
1 parent
b7ad2dae16
commit
ccead2d7c0
2 files changed
+44
-16
No files matched your search
@@ -1,8 +1,9 @@
|
||||
use anyhow::Result;
|
||||
use chrono::{NaiveDate, Utc};
|
||||
use opentelemetry::trace::Status;
|
||||
use regex::Regex;
|
||||
use std::collections::HashMap;
|
||||
use std::sync::{Arc, Mutex};
|
||||
use std::sync::{Arc, LazyLock, Mutex};
|
||||
use tokio::time::sleep;
|
||||
use tracing_opentelemetry::OpenTelemetrySpanExt;
|
||||
|
||||
@@ -112,14 +113,20 @@ pub fn strip_summary_boilerplate(summary: &str) -> String {
|
||||
}
|
||||
}
|
||||
|
||||
// "Today, <contact> and I discussed …" for any contact name (one or two
|
||||
// capitalised words), so no names are baked into the code.
|
||||
static NAMED_OPENING: LazyLock<Regex> = LazyLock::new(|| {
|
||||
Regex::new(
|
||||
r"^(?:[Tt]oday,?\s+)?\p{Lu}[\p{L}'’-]*(?:\s+\p{Lu}[\p{L}'’-]*)?\s+and\s+I\s+discussed",
|
||||
)
|
||||
.expect("valid regex")
|
||||
});
|
||||
if let Some(m) = NAMED_OPENING.find(&text) {
|
||||
text = text[m.end()..].trim_start().to_string();
|
||||
}
|
||||
|
||||
// Remove common opening phrases that add no semantic value
|
||||
let opening_phrases = [
|
||||
"Today, Melissa and I discussed",
|
||||
"Today, Amanda and I discussed",
|
||||
"Today Melissa and I discussed",
|
||||
"Today Amanda and I discussed",
|
||||
"Melissa and I discussed",
|
||||
"Amanda and I discussed",
|
||||
"Today, I discussed",
|
||||
"Today I discussed",
|
||||
"The conversation covered",
|
||||
@@ -376,3 +383,24 @@ async fn generate_and_store_daily_summary(
|
||||
result?;
|
||||
Ok(())
|
||||
}
|
||||
|
||||
#[cfg(test)]
|
||||
mod tests {
|
||||
use super::strip_summary_boilerplate;
|
||||
|
||||
#[test]
|
||||
fn strips_a_named_opening_for_any_contact() {
|
||||
for raw in [
|
||||
"Today, Sam and I discussed the trip.",
|
||||
"Today Sam Lee and I discussed the trip.",
|
||||
"Jordan and I discussed the trip.",
|
||||
] {
|
||||
assert_eq!(strip_summary_boilerplate(raw), "the trip.", "{raw}");
|
||||
}
|
||||
assert_eq!(
|
||||
strip_summary_boilerplate("The weather and I discussed nothing"),
|
||||
"The weather and I discussed nothing",
|
||||
"a lowercase second word isn't a name"
|
||||
);
|
||||
}
|
||||
}
|
||||
Reference in new issue
Block a user