Final completeness by time window: ranges keep their posting span, release_final() over blocks of done ranges

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

edge/src/indexer.rs

@@ -78,6 +78,9 @@ pub struct Batch {
     pub from: i64,
     pub next: i64,
     pub headers: u64,
+    /// postedLow / postedHigh: the batch's posting times without the earliest and latest 1% (postedSpan).
+    pub posted_low: Option<i64>,
+    pub posted_high: Option<i64>,
     pub releases: Vec<ReleaseChunk>,
     pub samples: Vec<(ReleaseChunk, &'static str, String)>,
     pub candidates: Vec<Candidate>,
@@ -277,6 +280,16 @@ fn first_track(text: &str) -> Option<usize> {
     None
 }
 
+/// postedSpan: posting times without the earliest and latest 1% (floor(n / 100) at each end).
+pub fn posted_span(mut dates: Vec<i64>) -> (Option<i64>, Option<i64>) {
+    if dates.is_empty() {
+        return (None, None);
+    }
+    dates.sort_unstable();
+    let k = dates.len() / 100;
+    (Some(dates[k]), Some(dates[dates.len() - 1 - k]))
+}
+
 /// nameBesideQuotes (lib/headers/subject.ts): a file name elsewhere when the quoted text has no extension.
 pub fn name_beside_quotes(quoted: &str, before_counter: &str) -> Option<String> {
     if has_extension(quoted) {
@@ -633,6 +646,8 @@ pub struct RangeIndexer {
     count_order: Vec<(&'static str, String)>,
     completed: FxHashMap<String, Vec<Vec<Seg>>>,
     completed_order: Vec<String>,
+    /// Posting times of this batch's segments (posted_span at take).
+    dates: Vec<i64>,
 }
 
 impl RangeIndexer {
@@ -653,6 +668,7 @@ impl RangeIndexer {
             count_order: Vec::new(),
             completed: FxHashMap::default(),
             completed_order: Vec::new(),
+            dates: Vec::new(),
         }
     }
 
@@ -683,6 +699,7 @@ impl RangeIndexer {
             });
         }
         self.batch.headers += headers;
+        self.dates.extend(segments.iter().filter_map(|s| s.posted_at));
         for owner in crossposts {
             self.count("crosspost", owner, 1, 0);
         }
@@ -725,6 +742,7 @@ impl RangeIndexer {
         self.scanner.files.clear();
         let next = self.batch.next;
         let mut out = std::mem::replace(&mut self.batch, Batch { from: next, next, ..Default::default() });
+        (out.posted_low, out.posted_high) = posted_span(std::mem::take(&mut self.dates));
         for id in self.count_order.drain(..) {
             let (articles, files) = self.counts.remove(&id).unwrap();
             out.counts.push((id.0, id.1, articles, files));
@@ -1015,3 +1033,18 @@ mod beside_quotes_tests {
     }
 }
 }
+
+#[cfg(test)]
+mod posted_span_tests {
+    use super::posted_span;
+
+    #[test]
+    fn without_the_earliest_and_latest_percent() {
+        assert_eq!(posted_span(vec![]), (None, None));
+        let mut dates: Vec<i64> = (0..200).map(|i| 1_000_000 + i).collect();
+        dates.push(5);
+        dates.push(9_999_999_999);
+        // 202 dates: 2 trimmed at each end, so the two wild ones go (postedSpan).
+        assert_eq!(posted_span(dates), (Some(1_000_001), Some(1_000_198)));
+    }
+}

edge/src/main.rs

@@ -96,7 +96,11 @@ fn dump_batch(batch: &Batch, out: &mut impl Write) {
         out.write_all(line.as_bytes()).unwrap();
         line.clear();
     };
-    line.push_str(&format!("{{\"t\":\"b\",\"from\":{},\"next\":{},\"headers\":{}}}", batch.from, batch.next, batch.headers));
+    let opt = |t: Option<i64>| t.map_or("null".to_string(), |t| t.to_string());
+    line.push_str(&format!(
+        "{{\"t\":\"b\",\"from\":{},\"next\":{},\"headers\":{},\"postedLow\":{},\"postedHigh\":{}}}",
+        batch.from, batch.next, batch.headers, opt(batch.posted_low), opt(batch.posted_high)
+    ));
     emit(&mut line);
     let release = |r: &indexer::ReleaseChunk, kind: &str, line: &mut String| {
         line.push_str(&format!("{{\"t\":\"{kind}\",\"upload\":{},\"key\":", r.upload));

edge/src/wire.rs

@@ -74,10 +74,13 @@ struct WireFragment<'a> {
 }
 
 #[derive(Serialize)]
+#[serde(rename_all = "camelCase")]
 struct WireBatch<'a> {
     from: i64,
     next: i64,
     headers: u64,
+    posted_low: Option<i64>,
+    posted_high: Option<i64>,
     releases: Vec<WireRelease<'a>>,
     samples: Vec<WireRelease<'a>>,
     candidates: Vec<WireCandidate<'a>>,
@@ -134,6 +137,8 @@ pub fn to_json(batch: &Batch) -> Vec<u8> {
         from: batch.from,
         next: batch.next,
         headers: batch.headers,
+        posted_low: batch.posted_low,
+        posted_high: batch.posted_high,
         releases: batch.releases.iter().map(|r| release(r, None)).collect(),
         samples: batch.samples.iter().map(|(r, rule, key)| release(r, Some((rule, key.as_str())))).collect(),
         candidates: batch.candidates.iter().map(candidate).collect(),