Add data graph stress tests

tests/stress/helpers.py

@@ -0,0 +1,53 @@
+import random
+
+import set_path
+from i_data import Data_type
+from i_structures.chemical_formulae import common_chem_elements
+from i_calculations.topas import get_pattern_name
+
+
+common_chem_elements = common_chem_elements.split()
+POSSIBLE_CONTENT = list(range(42))
+
+
+def gen_data_item(data_type):
+    meta = {}
+    content = random.choice(POSSIBLE_CONTENT)
+
+    if data_type == Data_type.calculation:
+        meta["name"] = gen_chem_formula()
+
+    elif data_type == Data_type.structure:
+        meta["name"] = gen_chem_formula()
+
+    elif data_type == Data_type.property:
+        meta["name"] = gen_chem_formula()  # FIXME
+
+    elif data_type == Data_type.pattern:
+        meta["name"] = get_pattern_name()
+
+    else:
+        raise RuntimeError
+
+    return (meta, content, data_type)
+
+
+def gen_chem_formula():
+    els = set()
+    pseudo_formula = ""
+
+    for _ in range(random.randint(1, 7)):
+        els.add(random.choice(common_chem_elements))
+
+    els = list(els)
+
+    for n in range(len(els)):
+        coeff = random.randint(1, 10)
+        coeff = "" if coeff == 1 else str(coeff)
+        pseudo_formula += els[n] + coeff
+
+    return pseudo_formula
+
+
+if __name__ == "__main__":
+    print(gen_chem_formula())

tests/stress/set_path.py

@@ -0,0 +1,8 @@
+import sys, os.path
+
+INCL_PATH = os.path.realpath(
+    os.path.normpath(os.path.join(os.path.dirname(os.path.abspath(__file__)), "../../"))
+)
+
+if not INCL_PATH in sys.path:
+    sys.path.insert(0, INCL_PATH)

tests/stress/test_prepare.py

@@ -0,0 +1,62 @@
+#!/usr/bin/env python
+
+import sys
+import random
+import logging
+
+from helpers import gen_data_item
+
+import set_path
+from i_data import Data_type
+from utils import get_data_storage
+
+
+FAKE_NODES = 5000
+FAKE_LINKS = 5000
+
+ALLOWED_TRANSITIONS = (
+    (Data_type.structure, Data_type.property),
+    (Data_type.structure, Data_type.pattern),
+    (Data_type.pattern, Data_type.structure),
+    (Data_type.pattern, Data_type.property),
+    (Data_type.property, Data_type.structure),
+    (Data_type.property, Data_type.pattern),
+)
+
+db = get_data_storage()
+used_uuids = {}
+
+for _ in range(FAKE_NODES):
+    selected_dtype = random.choice(
+        (
+            Data_type.calculation,
+            Data_type.structure,
+            Data_type.property,
+            Data_type.pattern,
+        )
+    )
+    used_uuids.setdefault(selected_dtype, []).append(
+        db.put_item(*gen_data_item(selected_dtype))
+    )
+
+count_links, num_iters = 0, 0
+
+while True:
+    source_dtype, target_dtype = random.choice(ALLOWED_TRANSITIONS)
+    source, target = random.choice(used_uuids[source_dtype]), random.choice(
+        used_uuids[target_dtype]
+    )
+
+    if db.put_link(source, target):
+        count_links += 1
+    # else:
+    #    logging.warning('Duplicate link between %s and %s occured on %s-th step' % (source, target, count_links))
+
+    if count_links == FAKE_LINKS:
+        break
+
+    num_iters += 1
+    if num_iters > FAKE_LINKS * 3:
+        break
+
+db.close()

tests/stress/test_stress.py

@@ -0,0 +1,39 @@
+#!/usr/bin/env python
+
+import sys
+import random
+import logging
+
+from helpers import POSSIBLE_CONTENT
+
+import set_path
+from utils import get_data_storage
+
+
+db = get_data_storage()
+
+accumulated = []
+
+for _ in range(100):
+    logging.warning("=" * 100)
+
+    found = db.search_item(random.choice(POSSIBLE_CONTENT))
+    assert found
+
+    item = db.get_item(found["uuid"], with_links=True)
+
+    accumulated += [found["uuid"]]
+    accumulated += item["children"]
+    accumulated += item["parents"]
+    logging.warning("Collected %s items" % len(accumulated))
+
+    items = db.get_items(accumulated, with_links=True)
+
+    children, parents = [], []
+    for one in items:
+        children += one["children"]
+        parents += one["parents"]
+
+    logging.warning("Got %s children and %s parents" % (len(children), len(parents)))
+
+db.close()