Merge pull request #33 from tmaeno/master

added --useNumFilesInSecDSsAsRatio to prun

ChangeLog.txt

@@ -1,5 +1,12 @@
 ** Release Notes
 
+1.5.77
+  * added --useNumFilesInSecDSsAsRatio to prun
+
+1.5.76
+  * For changes in ALRB_USER_PLATFORM
+  * execute_pchain in panda_api.py
+
 1.5.75
   * warning message for missing nickname in IAM
 

pandaclient/MiscUtils.py

@@ -101,9 +101,10 @@ def makeJediJobParam(
         elif outDS:
             dictItem["consolidate"] = ".".join(outDS.split(".")[:2]) + "." + wrappedUuidGen() + "/"
         if nFilesPerJob not in [None, 0]:
-            dictItem["nFilesPerJob"] = nFilesPerJob
-        if useNumFilesAsRatio and nFilesPerJob not in [None, 0]:
-            dictItem["ratio"] = nFilesPerJob
+            if useNumFilesAsRatio:
+                dictItem["ratio"] = nFilesPerJob
+            else:
+                dictItem["nFilesPerJob"] = nFilesPerJob
         if file_list:
             dictItem["files"] = file_list
     if hidden:
@@ -324,7 +325,7 @@ def parse_secondary_datasets_opt(secondaryDSs):
                             if l:
                                 tmpMap[tmpItems[2]]["files"].append(l)
             else:
-                errStr = "Wrong format %s in --secondaryDSs. Must be " "StreamName:nFilesPerJob:DatasetName[:Pattern[:nSkipFiles[:FileNameList]]]" % tmpItem
+                errStr = "Wrong format %s in --secondaryDSs. Must be " "StreamName:nFiles:DatasetName[:Pattern[:nSkipFiles[:FileNameList]]]" % tmpItem
                 return False, errStr
         # set
         secondaryDSs = tmpMap

pandaclient/PandaToolsPkgInfo.py

@@ -1 +1 @@
-release_version = "1.5.76"
+release_version = "1.5.77"

pandaclient/PrunScript.py

@@ -378,7 +378,13 @@ def main(get_taskparams=False, ext_args=None, dry_mode=False):
         action="store",
         dest="secondaryDSs",
         default="",
-        help="List of secondary datasets when the job requires multiple inputs. See PandaRun wiki page for detail",
+        help="List of secondary datasets when the job requires multiple inputs. "
+        "Comma-separated strings in the format of StreamName:nFiles:DatasetName[:Pattern[:nSkipFiles[:FileNameList]]]. "
+        "StreamName is the stream name used in --exec to expand to actual filenames. "
+        "nFiles is the number of files per job by default, while it means the ratio to the number of primary "
+        "files when --useNumFilesInSecDSsAsRatio is set. DatasetName is the dataset name. "
+        "Pattern is used to filter files in the dataset. nSkipFiles is the number of files to skip in the dataset. "
+        "FileNameList is a file listing names of files to be used in the dataset. ",
     )
     group_input.add_argument(
         "--reusableSecondary",
@@ -387,6 +393,14 @@ def main(get_taskparams=False, ext_args=None, dry_mode=False):
         default="",
         help="A comma-separated list of secondary streams which reuse files when all files are used",
     )
+    group_input.add_argument(
+        "--useNumFilesInSecDSsAsRatio",
+        action="store_const",
+        const=True,
+        dest="useNumFilesInSecDSsAsRatio",
+        default=False,
+        help="Set the option when the nFiles field in --secondaryDSs means the ratio to the number of primary files",
+    )
     group_submit.add_argument(
         "--site",
         action="store",
@@ -2509,6 +2523,7 @@ def _onExit(dir, files, del_command):
                 include=tmpMap["pattern"],
                 offset=tmpMap["nSkip"],
                 nFilesPerJob=tmpMap["nFiles"],
+                useNumFilesAsRatio=options.useNumFilesInSecDSsAsRatio,
                 reusableAtt=reusableAtt,
                 outDS=options.outDS,
                 file_list=tmpMap["files"],