Merge pull request #133 from githubnext/copilot/fix-perf-benchmark-autoloop

Fix autoloop creating multiple PRs by preserving exact branch names

Author: mrjf

.github/workflows/autoloop.lock.yml

@@ -45,6 +45,7 @@ safe-outputs:
     title-prefix: "[Autoloop] "
     labels: [automation, autoloop]
     protected-files: fallback-to-issue
+    preserve-branch-name: true
     max: 1
   push-to-pull-request-branch:
     target: "*"
@@ -434,10 +435,26 @@ steps:
       # Look up existing PR for the selected program's canonical branch
       existing_pr = None
       head_branch = None
+
+      def verify_pr_is_open(pr_number):
+          """Check if a PR is still open via the GitHub API. Returns True if open."""
+          try:
+              verify_url = f"https://api.github.com/repos/{repo}/pulls/{pr_number}"
+              verify_req = urllib.request.Request(verify_url, headers={
+                  "Authorization": f"token {github_token}",
+                  "Accept": "application/vnd.github.v3+json",
+              })
+              with urllib.request.urlopen(verify_req, timeout=30) as verify_resp:
+                  pr_data = json.loads(verify_resp.read().decode())
+              return pr_data.get("state") == "open"
+          except Exception:
+              return True  # If we can't verify, assume it's open (best effort)
+
       if selected:
           head_branch = f"autoloop/{selected}"
           owner = repo.split("/")[0] if "/" in repo else ""
           if owner:
+              # Strategy 1: exact branch match (works when branch has no framework suffix)
               try:
                   pr_api_url = (
                       f"https://api.github.com/repos/{repo}/pulls"
@@ -451,22 +468,54 @@ steps:
                       open_prs = json.loads(pr_resp.read().decode())
                   if open_prs:
                       existing_pr = open_prs[0]["number"]
-                      print(f"  Found existing PR #{existing_pr} for branch {head_branch}")
-                  else:
-                      print(f"  No existing PR found for branch {head_branch}")
+                      print(f"  Found existing PR #{existing_pr} for exact branch {head_branch}")
               except Exception as e:
-                  print(f"  Warning: could not check for existing PRs: {e}")
+                  print(f"  Warning: could not check for existing PRs by exact branch: {e}")
+
+              # Strategy 2: search by title and branch prefix (catches framework-generated
+              # hash suffixes like autoloop/name-a1b2c3d4e5f6g7h8 created by create-pull-request)
+              if existing_pr is None:
+                  try:
+                      title_marker = f"[Autoloop: {selected}]"
+                      branch_prefix = head_branch  # e.g. autoloop/perf-comparison
+                      list_url = (
+                          f"https://api.github.com/repos/{repo}/pulls"
+                          f"?state=open&per_page=100&sort=created&direction=desc"
+                      )
+                      list_req = urllib.request.Request(list_url, headers={
+                          "Authorization": f"token {github_token}",
+                          "Accept": "application/vnd.github.v3+json",
+                      })
+                      with urllib.request.urlopen(list_req, timeout=30) as list_resp:
+                          all_open_prs = json.loads(list_resp.read().decode())
+                      # Match branch names: exact canonical name or canonical + framework hash suffix
+                      branch_pattern = re.compile(r'^' + re.escape(branch_prefix) + r'(-[0-9a-f]{16})?$')
+                      for pr in all_open_prs:
+                          pr_title = pr.get("title", "")
+                          pr_head_ref = pr.get("head", {}).get("ref", "")
+                          if title_marker in pr_title or branch_pattern.match(pr_head_ref):
+                              existing_pr = pr["number"]
+                              print(f"  Found existing PR #{existing_pr} by title/branch-prefix (branch: {pr_head_ref})")
+                              break
+                      if existing_pr is None:
+                          print(f"  No existing PR found for program {selected}")
+                  except Exception as e:
+                      print(f"  Warning: could not search for existing PRs by title/prefix: {e}")
           else:
               print(f"  Warning: could not parse owner from GITHUB_REPOSITORY='{repo}'")
 
-          # Also check the state file for a recorded PR number as fallback
+          # Strategy 3: check the state file for a recorded PR number as fallback
           if existing_pr is None:
               state = read_program_state(selected)
               pr_field = state.get("pr") or ""
               pr_match = re.match(r'^#?(\d+)$', pr_field.strip())
               if pr_match:
-                  existing_pr = int(pr_match.group(1))
-                  print(f"  Found PR #{existing_pr} from state file for {selected}")
+                  pr_num = int(pr_match.group(1))
+                  if verify_pr_is_open(pr_num):
+                      existing_pr = pr_num
+                      print(f"  Found open PR #{existing_pr} from state file for {selected}")
+                  else:
+                      print(f"  PR #{pr_num} from state file is no longer open — ignoring")
 
       result = {
           "selected": selected,

.github/workflows/autoloop.md

@@ -7,7 +7,14 @@
   },
   "files": {
     "ignoreUnknown": false,
-    "ignore": ["dist/**", "node_modules/**", "*.d.ts", "playground/**/*.js", "playground/serve.ts"]
+    "ignore": [
+      "dist/**",
+      "node_modules/**",
+      "*.d.ts",
+      "playground/**/*.js",
+      "playground/serve.ts",
+      "benchmarks/**"
+    ]
   },
   "formatter": {
     "enabled": true,

biome.json

@@ -227,7 +227,9 @@ export function setAttr(obj: object, key: string, value: unknown): void {
  */
 export function deleteAttr(obj: object, key: string): void {
   const existing = registry.get(obj);
-  if (existing === undefined) return;
+  if (existing === undefined) {
+    return;
+  }
   const { [key]: _removed, ...rest } = existing;
   if (Object.keys(rest).length === 0) {
     registry.delete(obj);

src/core/attrs.ts

@@ -102,10 +102,14 @@ export class DataFrame {
    * @param columns - Ordered map of column name → Series (all same length and index).
    * @param index   - Row index (must match each Series' length).
    */
-  constructor(columns: ReadonlyMap<string, Series<Scalar>>, index: Index<Label>) {
+  constructor(
+    columns: ReadonlyMap<string, Series<Scalar>>,
+    index: Index<Label>,
+    columnNames?: readonly string[],
+  ) {
     this._columns = columns;
     this.index = index;
-    this.columns = new Index<string>([...columns.keys()]);
+    this.columns = new Index<string>(columnNames ?? [...columns.keys()]);
   }
 
   /**
@@ -208,7 +212,7 @@ export class DataFrame {
 
   /** `[nRows, nCols]` — mirrors `pandas.DataFrame.shape`. */
   get shape(): [number, number] {
-    return [this.index.size, this._columns.size];
+    return [this.index.size, this.columns.size];
   }
 
   /** Always `2`. */
@@ -218,12 +222,12 @@ export class DataFrame {
 
   /** Total number of cells (`nRows * nCols`). */
   get size(): number {
-    return this.index.size * this._columns.size;
+    return this.index.size * this.columns.size;
   }
 
   /** `true` when the DataFrame has no rows or no columns. */
   get empty(): boolean {
-    return this.index.size === 0 || this._columns.size === 0;
+    return this.index.size === 0 || this.columns.size === 0;
   }
 
   // ─── column access ────────────────────────────────────────────────────────

src/core/frame.ts

@@ -82,23 +82,28 @@ export function insertColumn(
   }
 
   // Rebuild the column map, inserting the new column at position `loc`.
+  const colNames: string[] = [];
   const colMap = new Map<string, Series<Scalar>>();
   let idx = 0;
 
   for (const colName of df.columns.values) {
     if (idx === loc) {
-      colMap.set(column, series);
+      colNames.push(column);
     }
+    colNames.push(colName);
     colMap.set(colName, df.col(colName));
     idx++;
   }
 
   // Handle insertion at the end (loc === nCols).
   if (loc === nCols) {
-    colMap.set(column, series);
+    colNames.push(column);
   }
 
-  return new DataFrame(colMap, df.index);
+  // Always add the new column data to the map (last-wins for duplicate names).
+  colMap.set(column, series);
+
+  return new DataFrame(colMap, df.index, colNames);
 }
 
 // ─── popColumn ────────────────────────────────────────────────────────────────

src/core/insert_pop.ts

@@ -100,7 +100,7 @@ function collectSuffixes(
     // Sort numerically when both look like integers, otherwise lexicographically.
     const na = Number(a);
     const nb = Number(b);
-    if (!Number.isNaN(na) && !Number.isNaN(nb)) {
+    if (!(Number.isNaN(na) || Number.isNaN(nb))) {
       return na - nb;
     }
     return a < b ? -1 : a > b ? 1 : 0;

src/reshape/wide_to_long.ts

@@ -108,7 +108,9 @@ export function catFromCodes(
   const { ordered = false, name = null } = opts;
   const cats = uniqueKeys(categories);
   const values: Scalar[] = codes.map((code) => {
-    if (code === -1) return null;
+    if (code === -1) {
+      return null;
+    }
     if (code < -1 || code >= cats.length) {
       throw new RangeError(`catFromCodes: code ${code} is out of range [0, ${cats.length - 1}]`);
     }
@@ -210,9 +212,13 @@ export function catDiffCategories(a: CatSeriesLike, b: CatSeriesLike): CatSeries
 export function catEqualCategories(a: CatSeriesLike, b: CatSeriesLike): boolean {
   const aSet = new Set((a.cat.categories.values as Scalar[]).map(String));
   const bSet = new Set((b.cat.categories.values as Scalar[]).map(String));
-  if (aSet.size !== bSet.size) return false;
+  if (aSet.size !== bSet.size) {
+    return false;
+  }
   for (const c of aSet) {
-    if (!bSet.has(c)) return false;
+    if (!bSet.has(c)) {
+      return false;
+    }
   }
   return true;
 }
@@ -243,12 +249,16 @@ export function catSortByFreq(
   const { ascending = false } = opts;
   const cats = series.cat.categories.values as Scalar[];
   const freq = new Map<string, number>();
-  for (const c of cats) freq.set(String(c), 0);
+  for (const c of cats) {
+    freq.set(String(c), 0);
+  }
   for (const v of series.values) {
     if (!isMissing(v)) {
       const k = String(v);
       const prev = freq.get(k);
-      if (prev !== undefined) freq.set(k, prev + 1);
+      if (prev !== undefined) {
+        freq.set(k, prev + 1);
+      }
     }
   }
   const sorted = [...cats].sort((a, b) => {
@@ -306,7 +316,9 @@ export function catToOrdinal(series: CatSeriesLike, order: readonly Scalar[]): C
 export function catFreqTable(series: CatSeriesLike): Record<string, number> {
   const cats = series.cat.categories.values as Scalar[];
   const freq: Record<string, number> = {};
-  for (const c of cats) freq[String(c)] = 0;
+  for (const c of cats) {
+    freq[String(c)] = 0;
+  }
   for (const v of series.values) {
     if (!isMissing(v)) {
       const k = String(v);
@@ -358,7 +370,9 @@ export function catCrossTab(
   const counts = new Map<string, Map<string, number>>();
   for (const r of rowCats) {
     const row = new Map<string, number>();
-    for (const c of colCats) row.set(String(c), 0);
+    for (const c of colCats) {
+      row.set(String(c), 0);
+    }
     counts.set(String(r), row);
   }
 
@@ -368,18 +382,26 @@ export function catCrossTab(
   for (let i = 0; i < n; i++) {
     const av = aVals[i];
     const bv = bVals[i];
-    if (isMissing(av) || isMissing(bv)) continue;
+    if (isMissing(av) || isMissing(bv)) {
+      continue;
+    }
     const row = counts.get(String(av));
-    if (row === undefined) continue;
+    if (row === undefined) {
+      continue;
+    }
     const prev = row.get(String(bv));
-    if (prev !== undefined) row.set(String(bv), prev + 1);
+    if (prev !== undefined) {
+      row.set(String(bv), prev + 1);
+    }
   }
 
   // Compute total for normalization
   let total = 0;
   if (normalize) {
     for (const row of counts.values()) {
-      for (const v of row.values()) total += v;
+      for (const v of row.values()) {
+        total += v;
+      }
     }
   }
 
@@ -399,7 +421,11 @@ export function catCrossTab(
     const rowTotals: Scalar[] = rowCats.map((r) => {
       let sum = 0;
       const row = counts.get(String(r));
-      if (row) for (const v of row.values()) sum += v;
+      if (row) {
+        for (const v of row.values()) {
+          sum += v;
+        }
+      }
       return normalize && total > 0 ? sum / total : sum;
     });
     data[marginsName] = rowTotals;
@@ -430,7 +456,9 @@ export function catCrossTab(
     // Ensure all column arrays have the same length
     for (const col of allCols) {
       const arr = data[col];
-      if (arr === undefined) data[col] = rowLabels.map(() => 0);
+      if (arr === undefined) {
+        data[col] = rowLabels.map(() => 0);
+      }
     }
   }
 

src/stats/categorical_ops.ts

@@ -113,7 +113,7 @@ function applyWindow(
   minN: number,
   agg: (nums: number[], n: number) => Scalar,
 ): SeriesLike {
-  const { values, index, name } = series;
+  const { values, name } = series;
   const n = values.length;
   const minPeriods = opts.minPeriods ?? window;
   const effectiveMin = Math.max(minN, minPeriods);

src/stats/window_extended.ts

@@ -82,8 +82,12 @@ function validNums(slice: readonly Scalar[]): number[] {
 /** Convert a raw window slice to `null`-substituted numeric array. */
 function rawWindow(slice: readonly Scalar[]): (number | null)[] {
   return slice.map((v): number | null => {
-    if (isMissing(v)) return null;
-    if (typeof v === "number") return v;
+    if (isMissing(v)) {
+      return null;
+    }
+    if (typeof v === "number") {
+      return v;
+    }
     return null;
   });
 }