Enable function chaining for filter and reduce. Remove obsolete.

darshan-util/pydarshan/Makefile

 .PHONY: clean clean-test clean-pyc clean-build docs help
-.DEFAULT_GOAL := help
 
-help:
-	@python -c "$$PRINT_HELP_PYSCRIPT" < $(MAKEFILE_LIST)
 
-clean: clean-build clean-pyc clean-test ## remove all build, test, coverage and Python artifacts
+clean: clean-build clean-pyc clean-test  # remove all build, test, coverage and Python artifacts
 
-clean-build: ## remove build artifacts
+clean-build:  # remove build artifacts
 	rm -rf build/
 	rm -rf dist/
 	rm -rf .eggs/
@@ -14,34 +11,36 @@ clean-build: ## remove build artifacts
 	find . -name '*.egg-info' -exec rm -fr {} +
 	find . -name '*.egg' -exec rm -f {} +
 
-clean-pyc: ## remove Python file artifacts
+clean-pyc:  # remove Python file artifacts
 	find . -name '*.pyc' -exec rm -f {} +
 	find . -name '*.pyo' -exec rm -f {} +
 	find . -name '*~' -exec rm -f {} +
 	find . -name '__pycache__' -exec rm -fr {} +
 
-clean-test: ## remove test and coverage artifacts
+clean-test:  # remove test and coverage artifacts
 	rm -rf .tox/
 	rm -f .coverage
 	rm -rf htmlcov/
 	rm -rf .pytest_cache
 
-lint: ## check style with flake8
+
+lint:  # check style with flake8
 	flake8 darshan tests
 
-test: ## run tests quickly with the default Python
+test:  # run tests quickly with the default Python
 	py.test --verbose
 
-test-all: ## run tests on every Python version with tox
+test-all:  # run tests on every Python version with tox
 	tox
 
-coverage: ## check code coverage quickly with the default Python
+coverage:  # check code coverage quickly with the default Python
 	coverage run --source darshan -m pytest
 	coverage report -m
 	coverage html
-	$(BROWSER) htmlcov/index.html
+	xdg-open htmlcov/index.html
+
 
-docs: ## generate Sphinx HTML documentation, including API docs
+docs:  # generate Sphinx HTML documentation, including API docs
 	rm -f docs/darshan.rst
 	rm -f docs/darshan.backend.rst
 	rm -f docs/darshan.plots.rst
@@ -53,16 +52,19 @@ docs: ## generate Sphinx HTML documentation, including API docs
 docs-show: docs
 	xdg-open docs/_build/html/index.html
 
-servedocs: docs ## compile the docs watching for changes
+servedocs: docs  # compile the docs watching for changes
 	watchmedo shell-command -p '*.rst' -c '$(MAKE) -C docs html' -R -D .
 
-release: dist ## package and upload a release
-	twine upload dist/*
 
-dist: clean ## builds source and wheel package
+release: dist # package and upload a release
+	twine upload --repository testpypi dist/*
+	#twine upload dist/*
+
+dist: clean  # builds source and wheel package
 	python setup.py sdist
 	python setup.py bdist_wheel
 	ls -l dist
 
-install: clean ## install the package to the active Python's site-packages
+
+install: clean  # install the package to the active Python's site-packages
 	python setup.py install

darshan-util/pydarshan/darshan/__init__.py

@@ -5,12 +5,18 @@
 __version__ = '0.1.0'
 
 
+
+options = {
+
+}
+
+
 #from darshan.backend.cffi_backend import *
 from darshan.report import DarshanReport
 
 
 
-def enable_experimental(verbose=True):
+def enable_experimental(verbose=False):
     """
     Enable experimental features such as aggregation methods for reports.
 

darshan-util/pydarshan/darshan/experimental/aggregators/agg_ioops.py

@@ -103,7 +103,7 @@ def agg_ioops(self, mode='append'):
 
 
     # cleanup and prepare for json serialization?
-    tmp = json.dumps(ctx, cls=NumpyEncoder)
+    tmp = json.dumps(ctx, cls=DarshanReportJSONEncoder)
     ctx = json.loads(tmp)
 
     

darshan-util/pydarshan/darshan/experimental/aggregators/create_time_summary.py

@@ -43,12 +43,8 @@ def create_time_summary(self, mode="append"):
 
 
 
-
     # overwrite existing summary entry
     if mode == "append":
         self.summary['time_summary'] = ctx
-    
 
     return ctx
-
-

darshan-util/pydarshan/darshan/experimental/aggregators/filter.py

 from darshan.report import *
 
 import sys
+import copy
+import re
 
 
-def filter(self, mods=None, name_records=None, data_format='numpy', mode='append'):
+def filter(self, mods=None, name_records=None, pattern=None, regex=None):
     """
     Return filtered list of records.
 
@@ -15,11 +17,12 @@ def filter(self, mods=None, name_records=None, data_format='numpy', mode='append
         None
     """
 
-
+    
+    r = copy.deepcopy(self)
 
 
     # convienience
-    recs = self.records
+    recs = r.records
     ctx = {}
 
 
@@ -36,38 +39,51 @@ def filter(self, mods=None, name_records=None, data_format='numpy', mode='append
         name_records = None
 
     
-    # change inputs to whitelists
+    # whitelist all mods
     if mods == None:
-        mods = self.records.keys()
+        mods = r.records.keys()
 
 
-    if name_records == None:
-        name_records = list(self.name_records.keys())
+    if pattern != None:
+        pattern = pattern.replace("*", "(.*?)")
+    elif regex:
+        pattern = regex
 
+
+    # whitelist name_records
+    if name_records == None and pattern == None and regex == None:
+        # allow all name records if no critirium provided
+        name_records = list(r.name_records.keys())
     else:
         resolve_table = {}
-        for key, value in self.name_records.items():
+        ids = []
+
+        for key, value in r.name_records.items():
             resolve_table[key] = key
             resolve_table[value] = key
 
-        ids = []
-        for nrec in name_records:
-            if nrec in resolve_table:
-                ids.append(resolve_table[nrec])
-
-        # TODO: decide if overwriting kargs is really a good idea.. currently considering it a sanitation step
-        name_records = ids
+            # whitelist names that match pattern
+            if pattern != None or regex != None:
 
+                if re.match(pattern, value):
+                    print("YES", pattern, value)
+                    ids.append(key)
+                else:
+                    print("NO", pattern, value)
 
+        # convert filenames/name_records mix into list of ids only
+        if name_records != None:
+            for nrec in name_records:
+                if nrec in resolve_table:
+                    ids.append(resolve_table[nrec])
 
-    print(mods)
-    print(name_records)
+        # TODO: decide if overwriting kargs is really a good idea.. currently considering it a sanitation step
+        name_records = ids
 
 
     if name_records != None:
-
         # aggragate
-        for mod, recs in self.records.items():
+        for mod, recs in r.records.items():
 
             if mod not in mods:
                 continue
@@ -82,13 +98,7 @@ def filter(self, mods=None, name_records=None, data_format='numpy', mode='append
                     ctx[mod].append(rec)
 
 
+    r.records = ctx
 
 
-    if mode == 'append':
-        name = 'filter'
-        if name not in self.summary:
-            self.summary[name] = {}
-        self.data[name] = ctx
-    
-    return ctx
-
+    return r

darshan-util/pydarshan/darshan/experimental/aggregators/merge.py

+from darshan.report import *
+
+import sys
+import copy
+
+def merge(self, other, reduce_first=False):
+    """
+    Merge two darshan reports and return a new combined report.
+
+
+    Args:
+        mods:           Name(s) of modules to preserve (reduced)
+        name_records:   Id(s)/Name(s) of name_records to preserve (reduced)
+
+
+    Return:
+        None
+    """
+
+
+    # new report
+    nr = DarshanReport()
+
+    # keep provenance?
+    if self.provenance_enabled or other.provenance_enabled:
+        # Currently, assume logs remain in memomry to create prov. tree on demand
+        # Alternative: maintain a tree with simpler refs? (modified reports would not work then)
+
+        #nr.provenance_reports[self.filename] = copy.copy(self)
+        #nr.provenance_reports[other.filename] = copy.copy(other)
+
+        nr.provenance_reports[self.filename] = None
+        nr.provenance_reports[other.filename] = None
+
+        nr.provenance_log.append(("add", self, other, datetime.datetime.now()))
+
+
+    # update metadata helper
+    def update_metadata(report, force=False):
+        if force:
+            nr.start_time = report.start_time
+            nr.end_time = report.end_time
+            return
+
+        if report.start_time < nr.start_time:
+            nr.start_time = report.start_time
+
+        if report.end_time > nr.end_time:
+            nr.end_time = report.end_time
+
+
+    update_metadata(self, force=True)
+    update_metadata(other)
+
+
+    # copy over records (references, under assumption single records are not altered)
+    for report in [self, other]:
+        for key, records in report.data['records'].items():
+            #print(report, key)
+            if key not in nr.records:
+                nr.records[key] = copy.copy(records)
+            else:
+                nr.records[key] += copy.copy(records)
+
+        for key, mod in report.modules.items():
+            if key not in nr.modules:
+                nr.modules[key] = copy.copy(mod)
+                # TODO: invalidate len/counters
+
+        for key, counter in report.counters.items():
+            if key not in nr.counters:
+                nr.counters[key] = copy.copy(counter)
+                # TODO: invalidate len/counters
+
+
+        for key, nrec in report.name_records.items():
+            if key not in nr.counters:
+                nr.name_records[key] = copy.copy(nrec)
+                # TODO: verify colliding name_records?
+
+    return nr

darshan-util/pydarshan/darshan/experimental/aggregators/mod_agg_iohist.py

@@ -48,7 +48,7 @@ def mod_agg_iohist(self, mod, mode='append'):
         name = "%s_%s_%s" % (typ, name[-2], name[-1])
         return name
 
-    tmp = json.dumps(ctx[mod], cls=NumpyEncoder)
+    tmp = json.dumps(ctx[mod], cls=DarshanReportJSONEncoder)
     tmp = json.loads(tmp)
     cn = backend.counter_names(mod)
     c = dict(zip(cn, tmp))

darshan-util/pydarshan/darshan/experimental/aggregators/print_module_records.py

+from darshan.report import *
+
+def print_module_records(self, mode='append'):
+    """
+    Compile the I/O operations summary for the current report.
+
+    Args:
+        mode (str): Whether to 'append' (default) or to 'return' aggregation. 
+
+    Return:
+        None or dict: Depending on mode
+    """
+
+    
+    descriptions = {
+        'POSIX': """
+# *******************************************************
+# POSIX module data
+# *******************************************************
+
+# description of POSIX counters:
+#   POSIX_*: posix operation counts.
+#   READS,WRITES,OPENS,SEEKS,STATS,MMAPS,SYNCS,FILENOS,DUPS are types of operations.
+#   POSIX_RENAME_SOURCES/TARGETS: total count file was source or target of a rename operation
+#   POSIX_RENAMED_FROM: Darshan record ID of the first rename source, if file was a rename target
+#   POSIX_MODE: mode that file was opened in.
+#   POSIX_BYTES_*: total bytes read and written.
+#   POSIX_MAX_BYTE_*: highest offset byte read and written.
+#   POSIX_CONSEC_*: number of exactly adjacent reads and writes.
+#   POSIX_SEQ_*: number of reads and writes from increasing offsets.
+#   POSIX_RW_SWITCHES: number of times access alternated between read and write.
+#   POSIX_*_ALIGNMENT: memory and file alignment.
+#   POSIX_*_NOT_ALIGNED: number of reads and writes that were not aligned.
+#   POSIX_MAX_*_TIME_SIZE: size of the slowest read and write operations.
+#   POSIX_SIZE_*_*: histogram of read and write access sizes.
+#   POSIX_STRIDE*_STRIDE: the four most common strides detected.
+#   POSIX_STRIDE*_COUNT: count of the four most common strides.
+#   POSIX_ACCESS*_ACCESS: the four most common access sizes.
+#   POSIX_ACCESS*_COUNT: count of the four most common access sizes.
+#   POSIX_*_RANK: rank of the processes that were the fastest and slowest at I/O (for shared files).
+#   POSIX_*_RANK_BYTES: bytes transferred by the fastest and slowest ranks (for shared files).
+#   POSIX_F_*_START_TIMESTAMP: timestamp of first open/read/write/close.
+#   POSIX_F_*_END_TIMESTAMP: timestamp of last open/read/write/close.
+#   POSIX_F_READ/WRITE/META_TIME: cumulative time spent in read, write, or metadata operations.
+#   POSIX_F_MAX_*_TIME: duration of the slowest read and write operations.
+#   POSIX_F_*_RANK_TIME: fastest and slowest I/O time for a single rank (for shared files).
+#   POSIX_F_VARIANCE_RANK_*: variance of total I/O time and bytes moved for all ranks (for shared files).
+
+# WARNING: POSIX module log format version <=3 has the following limitations:
+# - No support for the following counters to properly instrument dup, fileno, and rename operations:
+# 	- POSIX_FILENOS
+# 	- POSIX_DUPS
+# 	- POSIX_RENAME_SOURCES
+# 	- POSIX_RENAME_TARGETS
+# 	- POSIX_RENAMED_FROM
+
+#<module>	<rank>	<record id>	<counter>	<value>	<file name>	<mount pt>	<fs type>
+        """,
+
+        "MPI-IO": """
+# *******************************************************
+# MPI-IO module data
+# *******************************************************
+
+# description of MPIIO counters:
+#   MPIIO_INDEP_*: MPI independent operation counts.
+#   MPIIO_COLL_*: MPI collective operation counts.
+#   MPIIO_SPLIT_*: MPI split collective operation counts.
+#   MPIIO_NB_*: MPI non blocking operation counts.
+#   READS,WRITES,and OPENS are types of operations.
+#   MPIIO_SYNCS: MPI file sync operation counts.
+#   MPIIO_HINTS: number of times MPI hints were used.
+#   MPIIO_VIEWS: number of times MPI file views were used.
+#   MPIIO_MODE: MPI-IO access mode that file was opened with.
+#   MPIIO_BYTES_*: total bytes read and written at MPI-IO layer.
+#   MPIIO_RW_SWITCHES: number of times access alternated between read and write.
+#   MPIIO_MAX_*_TIME_SIZE: size of the slowest read and write operations.
+#   MPIIO_SIZE_*_AGG_*: histogram of MPI datatype total sizes for read and write operations.
+#   MPIIO_ACCESS*_ACCESS: the four most common total access sizes.
+#   MPIIO_ACCESS*_COUNT: count of the four most common total access sizes.
+#   MPIIO_*_RANK: rank of the processes that were the fastest and slowest at I/O (for shared files).
+#   MPIIO_*_RANK_BYTES: total bytes transferred at MPI-IO layer by the fastest and slowest ranks (for shared files).
+#   MPIIO_F_*_START_TIMESTAMP: timestamp of first MPI-IO open/read/write/close.
+#   MPIIO_F_*_END_TIMESTAMP: timestamp of last MPI-IO open/read/write/close.
+#   MPIIO_F_READ/WRITE/META_TIME: cumulative time spent in MPI-IO read, write, or metadata operations.
+#   MPIIO_F_MAX_*_TIME: duration of the slowest MPI-IO read and write operations.
+#   MPIIO_F_*_RANK_TIME: fastest and slowest I/O time for a single rank (for shared files).
+#   MPIIO_F_VARIANCE_RANK_*: variance of total I/O time and bytes moved for all ranks (for shared files).
+
+# WARNING: MPIIO module log format version <=2 does not support the following counters:
+# - MPIIO_F_CLOSE_START_TIMESTAMP
+# - MPIIO_F_OPEN_END_TIMESTAMP
+
+#<module>	<rank>	<record id>	<counter>	<value>	<file name>	<mount pt>	<fs type>
+        """,
+
+        "LUSTRE": """
+# *******************************************************
+# LUSTRE module data
+# *******************************************************
+
+# description of LUSTRE counters:
+#   LUSTRE_OSTS: number of OSTs across the entire file system.
+#   LUSTRE_MDTS: number of MDTs across the entire file system.
+#   LUSTRE_STRIPE_OFFSET: OST ID offset specified when the file was created.
+#   LUSTRE_STRIPE_SIZE: stripe size for file in bytes.
+#   LUSTRE_STRIPE_WIDTH: number of OSTs over which the file is striped.
+#   LUSTRE_OST_ID_*: indices of OSTs over which the file is striped.
+
+#<module>	<rank>	<record id>	<counter>	<value>	<file name>	<mount pt>	<fs type>
+        """,
+
+        "STDIO": """
+# *******************************************************
+# STDIO module data
+# *******************************************************
+
+# description of STDIO counters:
+#   STDIO_{OPENS|FDOPENS|WRITES|READS|SEEKS|FLUSHES} are types of operations.
+#   STDIO_BYTES_*: total bytes read and written.
+#   STDIO_MAX_BYTE_*: highest offset byte read and written.
+#   STDIO_*_RANK: rank of the processes that were the fastest and slowest at I/O (for shared files).
+#   STDIO_*_RANK_BYTES: bytes transferred by the fastest and slowest ranks (for shared files).
+#   STDIO_F_*_START_TIMESTAMP: timestamp of the first call to that type of function.
+#   STDIO_F_*_END_TIMESTAMP: timestamp of the completion of the last call to that type of function.
+#   STDIO_F_*_TIME: cumulative time spent in different types of functions.
+#   STDIO_F_*_RANK_TIME: fastest and slowest I/O time for a single rank (for shared files).
+#   STDIO_F_VARIANCE_RANK_*: variance of total I/O time and bytes moved for all ranks (for shared files).
+
+# WARNING: STDIO module log format version 1 has the following limitations:
+# - No support for properly instrumenting fdopen operations (STDIO_FDOPENS)
+
+#<module>	<rank>	<record id>	<counter>	<value>	<file name>	<mount pt>	<fs type>
+        """,
+     
+
+    }
+
+
+
+    pass

darshan-util/pydarshan/darshan/experimental/aggregators/records_as_dict.py

+from darshan.report import *
+
+def records_as_dict(self, mode='append', recs=None):
+    """
+    Convert all counters to dictionaries with human-readable names.
+
+    Args:
+        mode (str): Whether to 'append' (default) or to 'return' aggregation. 
+
+    Return:
+        None or dict: Depending on mode
+    """
+
+    # convienience
+    if recs is None:
+        recs = self.records
+
+
+    ctx = {}
+
+    # aggragate
+    for mod in recs:
+
+        if mod in ['DXT_POSIX']:
+            continue
+
+
+        # check records for module are present
+        if mod not in ctx:
+            ctx[mod] = []
+
+        for rec in recs[mod]:
+            c = None
+            fc = None
+
+            if 'counters' in rec:
+                c = dict(zip(self.counters[mod]['counters'], rec['counters'].tolist()))
+            
+            if 'fcounters' in rec:
+                fc = dict(zip(self.counters[mod]['fcounters'], rec['fcounters'].tolist()))
+
+            if rec['id'] in self.name_records:
+                nrec = self.name_records[rec['id']]
+            else:
+                nrec = None
+
+            ctx[mod].append({'id': rec['id'], 'rank': rec['rank'], 'counters': c, 'fcounters': fc, 'name_record': nrec})
+
+
+
+    return ctx

darshan-util/pydarshan/darshan/experimental/aggregators/reduce.py

@@ -17,8 +17,11 @@ def reduce(self, operation="sum", mods=None, name_records=None, mode='append', d
     """
 
 
+    r = copy.deepcopy(self)
+
+
     # convienience
-    recs = self.records
+    recs = r.records
     ctx = {}
 
 
@@ -43,15 +46,15 @@ def reduce(self, operation="sum", mods=None, name_records=None, mode='append', d
     
     # change inputs to whitelists
     if mods == None:
-        mods = self.records.keys()
+        mods = r.records.keys()
 
 
     if name_records == None:
-        name_records = list(self.name_records.keys())
+        name_records = list(r.name_records.keys())
 
     else:
         resolve_table = {}
-        for key, value in self.name_records.items():
+        for key, value in r.name_records.items():
             resolve_table[key] = key
             resolve_table[value] = key
 
@@ -72,7 +75,7 @@ def reduce(self, operation="sum", mods=None, name_records=None, mode='append', d
     if name_records != None:
 
         # aggragate
-        for mod, recs in self.records.items():
+        for mod, recs in r.records.items():
 
             if mod not in mods:
                 continue
@@ -121,12 +124,7 @@ def reduce(self, operation="sum", mods=None, name_records=None, mode='append', d
             result[mod].append(rec)            
 
 
+    r.records = result
 
 
-    if mode == 'append':
-        name = 'reduction'
-        if name not in self.summary:
-            self.summary[name] = {}
-        self.data[name] = ctx
-
-    return result
+    return r

darshan-util/pydarshan/darshan/experimental/plots/cffi_matplotlib.py

-# -*- coding: utf-8 -*-
-
-
-import matplotlib
-import matplotlib.pyplot as plt
-import numpy as np
-
-import darshan.backend.cffi_backend as backend
-
-
-
-def plot_access_histogram(log, filter=None, data=None):
-    """
-    Plots a histogram of access sizes for specified module.