Renamed BlockToReplication BlockToDesiredReplication.

[arvados.git] / crunch_scripts / run-command

diff --git a/crunch_scripts/run-command b/crunch_scripts/run-command
index fc3134f6518313a9b401b932ce06ed8f326d53e5..1ff63616ef638db376d49cfeecf8655f04dddbb1 100755 (executable)
--- a/crunch_scripts/run-command
+++ b/crunch_scripts/run-command
@@ -57,12 +57,10 @@ else:
     jobp = json.loads(args.script_parameters)
     os.environ['JOB_UUID'] = 'zzzzz-8i9sb-1234567890abcde'
     os.environ['TASK_UUID'] = 'zzzzz-ot0gb-1234567890abcde'
-    os.environ['CRUNCH_SRC'] = '/tmp/crunche-src'
+    os.environ['CRUNCH_SRC'] = '/tmp/crunch-src'
     if 'TASK_KEEPMOUNT' not in os.environ:
         os.environ['TASK_KEEPMOUNT'] = '/keep'
 
-links = []
-
 def sub_tmpdir(v):
     return os.path.join(arvados.current_task().tmpdir, 'tmpdir')
 
@@ -156,6 +154,8 @@ def var_items(p, c, key):
 def expand_item(p, c):
     if isinstance(c, dict):
         if "foreach" in c and "command" in c:
+            # Expand a command template for each item in the specified user
+            # parameter
             var, items = var_items(p, c, "foreach")
             if var is None:
                 raise EvaluationError("Must specify 'var' in foreach")
@@ -166,6 +166,7 @@ def expand_item(p, c):
                 r.append(expand_item(params, c["command"]))
             return r
         elif "list" in c and "index" in c and "command" in c:
+            # extract a single item from a list
             var, items = var_items(p, c, "list")
             if var is None:
                 raise EvaluationError("Must specify 'var' in list")
@@ -175,9 +176,13 @@ def expand_item(p, c):
         elif "regex" in c:
             pattern = re.compile(c["regex"])
             if "filter" in c:
+                # filter list so that it only includes items that match a
+                # regular expression
                 _, items = var_items(p, c, "filter")
                 return [i for i in items if pattern.match(i)]
             elif "group" in c:
+                # generate a list of lists, where items are grouped on common
+                # subexpression match
                 _, items = var_items(p, c, "group")
                 groups = {}
                 for i in items:
@@ -186,6 +191,8 @@ def expand_item(p, c):
                         add_to_group(groups, match)
                 return [groups[k] for k in groups]
             elif "extract" in c:
+                # generate a list of lists, where items are split by
+                # subexpression match
                 _, items = var_items(p, c, "extract")
                 r = []
                 for i in items:
@@ -194,6 +201,7 @@ def expand_item(p, c):
                         r.append(list(match.groups()))
                 return r
         elif "batch" in c and "size" in c:
+            # generate a list of lists, where items are split into a batch size
             _, items = var_items(p, c, "batch")
             sz = int(c["size"])
             r = []
@@ -209,8 +217,8 @@ def expand_item(p, c):
             return expand_item(p, p[m.group(1)])
         else:
             return subst.do_substitution(p, c)
-
-    raise EvaluationError("expand_item() unexpected parameter type %s" % (type(c))
+    else:
+        raise EvaluationError("expand_item() unexpected parameter type %s" % type(c))
 
 # Evaluate in a list context
 # "p" is the parameter scope, "value" will be evaluated
@@ -338,6 +346,8 @@ except Exception as e:
     logger.error(pprint.pformat(taskp))
     sys.exit(1)
 
+# rcode holds the return codes produced by each subprocess
+rcode = {}
 try:
     subprocesses = []
     close_streams = []
@@ -379,7 +389,6 @@ try:
 
     active = 1
     pids = set([s.pid for s in subprocesses])
-    rcode = {}
     while len(pids) > 0:
         (pid, status) = os.wait()
         pids.discard(pid)
@@ -404,33 +413,29 @@ signal.signal(signal.SIGINT, signal.SIG_DFL)
 signal.signal(signal.SIGTERM, signal.SIG_DFL)
 signal.signal(signal.SIGQUIT, signal.SIG_DFL)
 
-for l in links:
-    os.unlink(l)
-
 logger.info("the following output files will be saved to keep:")
 
-subprocess.call(["find", ".", "-type", "f", "-printf", "run-command: %12.12s %h/%f\\n"], stdout=sys.stderr)
+subprocess.call(["find", "-L", ".", "-type", "f", "-printf", "run-command: %12.12s %h/%f\\n"], stdout=sys.stderr, cwd=outdir)
 
 logger.info("start writing output to keep")
 
-if "task.vwd" in taskp:
-    if "task.foreach" in jobp:
-        # This is a subtask, so don't merge with the original collection, that will happen at the end
-        outcollection = vwd.checkin(subst.do_substitution(taskp, taskp["task.vwd"]), outdir, merge=False).manifest_text()
-    else:
-        # Just a single task, so do merge with the original collection
-        outcollection = vwd.checkin(subst.do_substitution(taskp, taskp["task.vwd"]), outdir, merge=True).manifest_text()
-else:
-    outcollection = robust_put.upload(outdir, logger)
+if "task.vwd" in taskp and "task.foreach" in jobp:
+    for root, dirs, files in os.walk(outdir):
+        for f in files:
+            s = os.lstat(os.path.join(root, f))
+            if stat.S_ISLNK(s.st_mode):
+                os.unlink(os.path.join(root, f))
+
+(outcollection, checkin_error) = vwd.checkin(outdir)
 
-# Success if no non-zero return codes
-success = not any([status != 0 for status in rcode.values()])
+# Success if we ran any subprocess, and they all exited 0.
+success = rcode and all(status == 0 for status in rcode.itervalues()) and not checkin_error
 
 api.job_tasks().update(uuid=arvados.current_task()['uuid'],
                                      body={
-                                         'output': outcollection,
+                                         'output': outcollection.manifest_text(),
                                          'success': success,
                                          'progress':1.0
                                      }).execute()
 
-sys.exit(rcode)
+sys.exit(0 if success else 1)