docs: update parsing of benchmarks/experiments for table

move existing table to a legacy catalog of benchmarks/experiments

docs/Makefile

@@ -22,7 +22,10 @@ legacysystemconfigs:
 tags:
 	./generate-benchmark-list.py $(WORKSPACE_PATH)
 
-.PHONY: help legacysystemconfigs systemconfigs tags Makefile
+legacytags:
+	./generate-legacy-benchmark-list.py $(WORKSPACE_PATH)
+
+.PHONY: help legacysystemconfigs systemconfigs legacytags tags Makefile
 
 # Catch-all target: route all unknown targets to Sphinx using the new
 # "make mode" option.  $(O) is meant as a shortcut for $(SPHINXOPTS).

docs/conf.py

@@ -24,6 +24,13 @@
     ]
 )
 
+subprocess.call(
+    [
+        "make",
+        "legacytags",
+    ]
+)
+
 subprocess.call(
     [
         "make",

docs/generate-benchmark-list.py

@@ -18,7 +18,7 @@ def construct_tag_groups(tag_groups, tag_dicts, dictionary):
 
 
 def benchpark_benchmarks(benchmarks):
-    experiments_dir = "../legacy/experiments"
+    experiments_dir = "../experiments"
     for x in os.listdir(experiments_dir):
         benchmarks.append(f"{x}")
     return benchmarks

docs/generate-legacy-benchmark-list.py

@@ -0,0 +1,94 @@
+#!/usr/bin/env python3
+
+import pandas as pd
+import yaml
+import os
+import sys
+import subprocess
+
+
+def construct_tag_groups(tag_groups, tag_dicts, dictionary):
+    # everything is a dict
+    for k, v in dictionary.items():
+        if isinstance(v, list):
+            tag_groups.append(k)
+            tag_dicts[k] = v
+        else:
+            print("ERROR in construct_tag_groups")
+
+
+def benchpark_benchmarks(benchmarks):
+    experiments_dir = "../legacy/experiments"
+    for x in os.listdir(experiments_dir):
+        benchmarks.append(f"{x}")
+    return benchmarks
+
+
+def main(workspace):
+    benchmarks = list()
+    benchpark_benchmarks(benchmarks)
+
+    f = "../taxonomy.yaml"
+    with open(f, "r") as stream:
+        try:
+            data = yaml.safe_load(stream)
+        except yaml.YAMLError as exc:
+            print(exc)
+
+    tag_groups = []
+    tag_dicts = {}
+    for k, v in data.items():
+        if k == "benchpark-tags":
+            construct_tag_groups(tag_groups, tag_dicts, v)
+        else:
+            print("ERROR in top level construct_tag_groups")
+
+    main = dict()
+
+    tags_taggroups = {}
+    for bmark in benchmarks:
+        tags_taggroups[bmark] = {}
+        for k, v in tag_dicts.items():
+            tags_taggroups[bmark][k] = []
+
+    for bmark in benchmarks:
+        # call benchpark tags -a bmark workspace
+        cmd = ["../bin/benchpark", "tags", "-a", bmark, workspace]
+        try:
+            byte_data = subprocess.run(cmd, capture_output=True, check=True)
+        except subprocess.CalledProcessError as e:
+            print(f"Failed cmd: {cmd}\nOutput: {e.stdout}\nError: {e.stderr}")
+            raise
+        tags = str(byte_data.stdout, "utf-8")
+        tags = (
+            tags.replace("[", "")
+            .replace("]", "")
+            .replace("'", "")
+            .replace(" ", "")
+            .replace("\n", "")
+            .split(",")
+        )
+        for t in tags:
+            for k, v in tag_dicts.items():
+                if t in v:
+                    print("appending", t, "at key", k)
+                    tags_taggroups[bmark][k].append(t)
+        main[bmark] = tags_taggroups[bmark]
+
+    df = pd.DataFrame(main)
+    df.to_csv("legacy-benchmark-list.csv")
+
+    #################
+    # Tables
+    # columns: benchmarks (i.e., amg2023)
+    # rows: tag groups (i.e., application domain).  Each cell should hopefully have a tag - and some might have multiple
+
+
+if __name__ == "__main__":
+    try:
+        workspace = sys.argv[1]
+    except IndexError:
+        print("Usage: " + os.path.basename(__file__) + " <workspace>")
+        sys.exit(1)
+
+    main(workspace)

docs/index.rst

@@ -60,3 +60,4 @@
 
    legacy-getting-started
    legacy-system-list
+   legacy-benchmark-list

docs/legacy-benchmark-list.rst

@@ -0,0 +1,13 @@
+.. Copyright 2023 Lawrence Livermore National Security, LLC and other
+   Benchpark Project Developers. See the top-level COPYRIGHT file for details.
+
+   SPDX-License-Identifier: Apache-2.0
+
+==========================
+Benchmarks and Experiments
+==========================
+
+.. csv-table:: Current Benchpark tags by Benchmark and tag groups.
+   :file: legacy-benchmark-list.csv
+   :header-rows: 1
+   :align: left