Show module names instead of file paths in flamegraph

Display module names instead of full file paths
(/home/user/project/pkg/mod.py → pkg.mod) in flamegraph for
readability.

Author: ivonastojanovic

Lib/profiling/sampling/_flamegraph_assets/flamegraph.js

@@ -64,6 +64,9 @@ function resolveStringIndices(node, table) {
   if (typeof resolved.funcname === 'number') {
     resolved.funcname = resolveString(resolved.funcname, table);
   }
+  if (typeof resolved.module_name === 'number') {
+    resolved.module_name = resolveString(resolved.module_name);
+  }
 
   if (Array.isArray(resolved.source)) {
     resolved.source = resolved.source.map(index =>
@@ -78,6 +81,11 @@ function resolveStringIndices(node, table) {
   return resolved;
 }
 
+// Escape HTML special characters
+function escapeHtml(str) {
+  return str.replace(/&/g, "&amp;").replace(/</g, "&lt;").replace(/>/g, "&gt;");
+}
+
 function selectFlamegraphData() {
   const baseData = isShowingElided ? elidedFlamegraphData : normalData;
 
@@ -228,6 +236,7 @@ function setupLogos() {
 function updateStatusBar(nodeData, rootValue) {
   const funcname = resolveString(nodeData.funcname) || resolveString(nodeData.name) || "--";
   const filename = resolveString(nodeData.filename) || "";
+  const moduleName = resolveString(nodeData.module_name) || "";
   const lineno = nodeData.lineno;
   const timeMs = (nodeData.value / 1000).toFixed(2);
   const percent = rootValue > 0 ? ((nodeData.value / rootValue) * 100).toFixed(1) : "0.0";
@@ -249,8 +258,7 @@ function updateStatusBar(nodeData, rootValue) {
 
   const fileEl = document.getElementById('status-file');
   if (fileEl && filename && filename !== "~") {
-    const basename = filename.split('/').pop();
-    fileEl.textContent = lineno ? `${basename}:${lineno}` : basename;
+    fileEl.textContent = lineno ? `${moduleName}:${lineno}` : moduleName;
   }
 
   const funcEl = document.getElementById('status-func');
@@ -301,6 +309,7 @@ function createPythonTooltip(data) {
 
     const funcname = resolveString(d.data.funcname) || resolveString(d.data.name);
     const filename = resolveString(d.data.filename) || "";
+    const moduleName = escapeHtml(resolveString(d.data.module_name) || "");
     const isSpecialFrame = filename === "~";
 
     // Build source section
@@ -309,7 +318,7 @@ function createPythonTooltip(data) {
       const sourceLines = source
         .map((line) => {
           const isCurrent = line.startsWith("→");
-          const escaped = line.replace(/&/g, "&amp;").replace(/</g, "&lt;").replace(/>/g, "&gt;");
+          const escaped = escapeHtml(line);
           return `<div class="tooltip-source-line${isCurrent ? ' current' : ''}">${escaped}</div>`;
         })
         .join("");
@@ -369,7 +378,7 @@ function createPythonTooltip(data) {
     }
 
     const fileLocationHTML = isSpecialFrame ? "" : `
-      <div class="tooltip-location">${filename}${d.data.lineno ? ":" + d.data.lineno : ""}</div>`;
+      <div class="tooltip-location">${moduleName}${d.data.lineno ? ":" + d.data.lineno : ""}</div>`;
 
     // Differential stats section
     let diffSection = "";
@@ -628,24 +637,24 @@ function updateSearchHighlight(searchTerm, searchInput) {
         const name = resolveString(d.data.name) || "";
         const funcname = resolveString(d.data.funcname) || "";
         const filename = resolveString(d.data.filename) || "";
+        const moduleName = resolveString(d.data.module_name) || "";
         const lineno = d.data.lineno;
         const term = searchTerm.toLowerCase();
 
-        // Check if search term looks like file:line pattern
+        // Check if search term looks like module:line pattern
         const fileLineMatch = term.match(/^(.+):(\d+)$/);
         let matches = false;
 
         if (fileLineMatch) {
-          // Exact file:line matching
           const searchFile = fileLineMatch[1];
           const searchLine = parseInt(fileLineMatch[2], 10);
-          const basename = filename.split('/').pop().toLowerCase();
-          matches = basename.includes(searchFile) && lineno === searchLine;
+          matches = moduleName.toLowerCase().includes(searchFile) && lineno === searchLine;
         } else {
           // Regular substring search
           matches =
             name.toLowerCase().includes(term) ||
             funcname.toLowerCase().includes(term) ||
+            moduleName.toLowerCase().includes(term) ||
             filename.toLowerCase().includes(term);
         }
 
@@ -1047,6 +1056,7 @@ function populateStats(data) {
 
     let filename = resolveString(node.filename);
     let funcname = resolveString(node.funcname);
+    let moduleName = resolveString(node.module_name);
 
     if (!filename || !funcname) {
       const nameStr = resolveString(node.name);
@@ -1061,6 +1071,7 @@ function populateStats(data) {
 
     filename = filename || 'unknown';
     funcname = funcname || 'unknown';
+    moduleName = moduleName || 'unknown';
 
     if (filename !== 'unknown' && funcname !== 'unknown' && node.value > 0) {
       let childrenValue = 0;
@@ -1077,12 +1088,14 @@ function populateStats(data) {
         existing.directPercent = (existing.directSamples / totalSamples) * 100;
         if (directSamples > existing.maxSingleSamples) {
           existing.filename = filename;
+          existing.module_name = moduleName;
           existing.lineno = node.lineno || '?';
           existing.maxSingleSamples = directSamples;
         }
       } else {
         functionMap.set(funcKey, {
           filename: filename,
+          module_name: moduleName,
           lineno: node.lineno || '?',
           funcname: funcname,
           directSamples,
@@ -1117,6 +1130,7 @@ function populateStats(data) {
       const h = hotSpots[i];
       const filename = h.filename || 'unknown';
       const lineno = h.lineno ?? '?';
+      const moduleName = h.module_name || 'unknown';
       const isSpecialFrame = filename === '~' && (lineno === 0 || lineno === '?');
 
       let funcDisplay = h.funcname || 'unknown';
@@ -1127,8 +1141,7 @@ function populateStats(data) {
         if (isSpecialFrame) {
           fileEl.textContent = '--';
         } else {
-          const basename = filename !== 'unknown' ? filename.split('/').pop() : 'unknown';
-          fileEl.textContent = `${basename}:${lineno}`;
+          fileEl.textContent = `${moduleName}:${lineno}`;
         }
       }
       if (percentEl) percentEl.textContent = `${h.directPercent.toFixed(1)}%`;
@@ -1144,8 +1157,9 @@ function populateStats(data) {
     if (card) {
       if (i < hotSpots.length && hotSpots[i]) {
         const h = hotSpots[i];
-        const basename = h.filename !== 'unknown' ? h.filename.split('/').pop() : '';
-        const searchTerm = basename && h.lineno !== '?' ? `${basename}:${h.lineno}` : h.funcname;
+        const moduleName = h.module_name || 'unknown';
+        const hasValidLocation = moduleName !== 'unknown' && h.lineno !== '?';
+        const searchTerm = hasValidLocation ? `${moduleName}:${h.lineno}` : h.funcname;
         card.dataset.searchterm = searchTerm;
         card.onclick = () => searchForHotspot(searchTerm);
         card.style.cursor = 'pointer';
@@ -1281,6 +1295,7 @@ function accumulateInvertedNode(parent, stackFrame, leaf, isDifferential) {
       self: 0,
       children: {},
       filename: stackFrame.filename,
+      module_name: stackFrame.module_name,
       lineno: stackFrame.lineno,
       funcname: stackFrame.funcname,
       source: stackFrame.source,

Lib/profiling/sampling/stack_collector.py

@@ -12,6 +12,7 @@
 from .collector import Collector, extract_lineno
 from .opcode_utils import get_opcode_mapping
 from .string_table import StringTable
+from .module_utils import extract_module_name, get_python_path_info
 
 
 class StackTraceCollector(Collector):
@@ -72,6 +73,7 @@ def __init__(self, *args, **kwargs):
         self._sample_count = 0  # Track actual number of samples (not thread traces)
         self._func_intern = {}
         self._string_table = StringTable()
+        self._module_cache = {}
         self._all_threads = set()
 
         # Thread status statistics (similar to LiveStatsCollector)
@@ -168,19 +170,21 @@ def export(self, filename):
 
     @staticmethod
     @functools.lru_cache(maxsize=None)
-    def _format_function_name(func):
+    def _format_function_name(func, module_name):
         filename, lineno, funcname = func
 
         # Special frames like <GC> and <native> should not show file:line
         if filename == "~" and lineno == 0:
             return funcname
 
-        if len(filename) > 50:
-            parts = filename.split("/")
-            if len(parts) > 2:
-                filename = f".../{'/'.join(parts[-2:])}"
+        return f"{funcname} ({module_name}:{lineno})"
 
-        return f"{funcname} ({filename}:{lineno})"
+    def _get_module_name(self, filename, path_info):
+        module_name = self._module_cache.get(filename)
+        if module_name is None:
+            module_name, _ = extract_module_name(filename, path_info)
+            self._module_cache[filename] = module_name
+        return module_name
 
     def _convert_to_flamegraph_format(self):
         if self._total_samples == 0:
@@ -192,7 +196,7 @@ def _convert_to_flamegraph_format(self):
                 "strings": self._string_table.get_strings()
             }
 
-        def convert_children(children, min_samples):
+        def convert_children(children, min_samples, path_info):
             out = []
             for func, node in children.items():
                 samples = node["samples"]
@@ -202,14 +206,18 @@ def convert_children(children, min_samples):
                 # Intern all string components for maximum efficiency
                 filename_idx = self._string_table.intern(func[0])
                 funcname_idx = self._string_table.intern(func[2])
-                name_idx = self._string_table.intern(self._format_function_name(func))
+                module_name = self._get_module_name(func[0], path_info)
+
+                module_name_idx = self._string_table.intern(module_name)
+                name_idx = self._string_table.intern(self._format_function_name(func, module_name))
 
                 child_entry = {
                     "name": name_idx,
                     "value": samples,
                     "self": node.get("self", 0),
                     "children": [],
                     "filename": filename_idx,
+                    "module_name": module_name_idx,
                     "lineno": func[1],
                     "funcname": funcname_idx,
                     "threads": sorted(list(node.get("threads", set()))),
@@ -228,7 +236,7 @@ def convert_children(children, min_samples):
 
                 # Recurse
                 child_entry["children"] = convert_children(
-                    node["children"], min_samples
+                    node["children"], min_samples, path_info
                 )
                 out.append(child_entry)
 
@@ -239,8 +247,9 @@ def convert_children(children, min_samples):
         # Filter out very small functions (less than 0.1% of total samples)
         total_samples = self._total_samples
         min_samples = max(1, int(total_samples * 0.001))
+        path_info = get_python_path_info()
 
-        root_children = convert_children(self._root["children"], min_samples)
+        root_children = convert_children(self._root["children"], min_samples, path_info)
         if not root_children:
             return {
                 "name": self._string_table.intern("No significant data"),

Lib/test/test_profiling/test_sampling_profiler/test_collectors.py

@@ -435,12 +435,12 @@ def test_flamegraph_collector_basic(self):
         strings = data.get("strings", [])
         name = resolve_name(data, strings)
         self.assertTrue(name.startswith("Program Root: "))
-        self.assertIn("func2 (file.py:20)", name)
+        self.assertIn("func2 (file:20)", name)
         self.assertEqual(data["self"], 0)  # non-leaf: no self time
         children = data.get("children", [])
         self.assertEqual(len(children), 1)
         child = children[0]
-        self.assertIn("func1 (file.py:10)", resolve_name(child, strings))
+        self.assertIn("func1 (file:10)", resolve_name(child, strings))
         self.assertEqual(child["value"], 1)
         self.assertEqual(child["self"], 1)  # leaf: all time is self