From d791fa7adb14991c972b6166f39155ff314b7d1e Mon Sep 17 00:00:00 2001
From: Peter Amstutz <peter.amstutz@curii.com>
Date: Tue, 2 Aug 2022 10:46:54 -0400
Subject: [PATCH] 19280: More metrics

Arvados-DCO-1.1-Signed-off-by: Peter Amstutz <peter.amstutz@curii.com>
---
 sdk/cwl/arvados_cwl/runner.py | 78 +++++++++++++++++++----------------
 1 file changed, 43 insertions(+), 35 deletions(-)

diff --git a/sdk/cwl/arvados_cwl/runner.py b/sdk/cwl/arvados_cwl/runner.py
index b00a30ad5f..2582c0a3a3 100644
--- a/sdk/cwl/arvados_cwl/runner.py
+++ b/sdk/cwl/arvados_cwl/runner.py
@@ -311,17 +311,19 @@ def upload_dependencies(arvrunner, name, document_loader,
 
     metadata = scanobj
 
-    sc_result = scandeps(uri, scanobj,
-                         loadref_fields,
-                         set(("$include", "location")),
-                         loadref, urljoin=document_loader.fetcher.urljoin,
-                         nestdirs=False)
-
-    optional_deps = scandeps(uri, scanobj,
-                                  loadref_fields,
-                                  set(("$schemas",)),
-                                  loadref, urljoin=document_loader.fetcher.urljoin,
-                                  nestdirs=False)
+    with Perf(metrics, "scandeps include, location"):
+        sc_result = scandeps(uri, scanobj,
+                             loadref_fields,
+                             set(("$include", "location")),
+                             loadref, urljoin=document_loader.fetcher.urljoin,
+                             nestdirs=False)
+
+    with Perf(metrics, "scandeps $schemas"):
+        optional_deps = scandeps(uri, scanobj,
+                                      loadref_fields,
+                                      set(("$schemas",)),
+                                      loadref, urljoin=document_loader.fetcher.urljoin,
+                                      nestdirs=False)
 
     sc_result.extend(optional_deps)
 
@@ -351,30 +353,34 @@ def upload_dependencies(arvrunner, name, document_loader,
             sc.append(obj)
         collect_uuids(obj)
 
-    visit_class(workflowobj, ("File", "Directory"), collect_uuids)
-    visit_class(sc_result, ("File", "Directory"), collect_uploads)
+    with Perf(metrics, "collect uuids"):
+        visit_class(workflowobj, ("File", "Directory"), collect_uuids)
+
+    with Perf(metrics, "collect uploads"):
+        visit_class(sc_result, ("File", "Directory"), collect_uploads)
 
     # Resolve any collection uuids we found to portable data hashes
     # and assign them to uuid_map
     uuid_map = {}
     fetch_uuids = list(uuids.keys())
-    while fetch_uuids:
-        # For a large number of fetch_uuids, API server may limit
-        # response size, so keep fetching from API server has nothing
-        # more to give us.
-        lookups = arvrunner.api.collections().list(
-            filters=[["uuid", "in", fetch_uuids]],
-            count="none",
-            select=["uuid", "portable_data_hash"]).execute(
-                num_retries=arvrunner.num_retries)
+    with Perf(metrics, "fetch_uuids"):
+        while fetch_uuids:
+            # For a large number of fetch_uuids, API server may limit
+            # response size, so keep fetching from API server has nothing
+            # more to give us.
+            lookups = arvrunner.api.collections().list(
+                filters=[["uuid", "in", fetch_uuids]],
+                count="none",
+                select=["uuid", "portable_data_hash"]).execute(
+                    num_retries=arvrunner.num_retries)
 
-        if not lookups["items"]:
-            break
+            if not lookups["items"]:
+                break
 
-        for l in lookups["items"]:
-            uuid_map[l["uuid"]] = l["portable_data_hash"]
+            for l in lookups["items"]:
+                uuid_map[l["uuid"]] = l["portable_data_hash"]
 
-        fetch_uuids = [u for u in fetch_uuids if u not in uuid_map]
+            fetch_uuids = [u for u in fetch_uuids if u not in uuid_map]
 
     normalizeFilesDirs(sc)
 
@@ -420,12 +426,13 @@ def upload_dependencies(arvrunner, name, document_loader,
         else:
             del discovered[d]
 
-    mapper = ArvPathMapper(arvrunner, sc, "",
-                           "keep:%s",
-                           "keep:%s/%s",
-                           name=name,
-                           single_collection=True,
-                           optional_deps=optional_deps)
+    with Perf(metrics, "mapper"):
+        mapper = ArvPathMapper(arvrunner, sc, "",
+                               "keep:%s",
+                               "keep:%s/%s",
+                               name=name,
+                               single_collection=True,
+                               optional_deps=optional_deps)
 
     keeprefs = set()
     def addkeepref(k):
@@ -469,8 +476,9 @@ def upload_dependencies(arvrunner, name, document_loader,
         p["location"] = "keep:%s%s" % (uuid_map[uuid], gp.groups()[1] if gp.groups()[1] else "")
         p[collectionUUID] = uuid
 
-    visit_class(workflowobj, ("File", "Directory"), setloc)
-    visit_class(discovered, ("File", "Directory"), setloc)
+    with Perf(metrics, "setloc"):
+        visit_class(workflowobj, ("File", "Directory"), setloc)
+        visit_class(discovered, ("File", "Directory"), setloc)
 
     if discovered_secondaryfiles is not None:
         for d in discovered:
-- 
2.30.2