leave Python changes for another PR

benchmarks/benchmarks.py

@@ -33,6 +33,74 @@ def load_sequences():
     return sequences
 
 
+class TimeMinHashSuite:
+    def setup(self):
+        self.mh = MinHash(MINHASH_NUM, MINHASH_K, track_abundance=False)
+        self.protein_mh = MinHash(
+            MINHASH_NUM, MINHASH_K, is_protein=True, track_abundance=False
+        )
+        self.sequences = load_sequences()
+
+        self.populated_mh = MinHash(MINHASH_NUM, MINHASH_K, track_abundance=False)
+        for seq in self.sequences:
+            self.populated_mh.add_sequence(seq)
+
+    def time_add_sequence(self):
+        mh = self.mh
+        sequences = self.sequences
+        for seq in sequences:
+            mh.add_sequence(seq)
+
+    def time_add_protein(self):
+        mh = self.protein_mh
+        sequences = self.sequences
+        for seq in sequences:
+            mh.add_protein(seq)
+
+    def time_get_mins(self):
+        mh = self.populated_mh
+        for i in range(GET_MINS_RANGE):
+            mh.get_mins()
+
+    def time_add_hash(self):
+        mh = self.mh
+        for i in range(ADD_HASH_RANGE):
+            mh.add_hash(i)
+
+    def time_add_many(self):
+        mh = self.mh
+        mh.add_many(list(range(ADD_MANY_RANGE)))
+
+    def time_similarity(self):
+        mh = self.mh
+        other_mh = self.populated_mh
+        for i in range(SIMILARITY_TIMES):
+            mh.similarity(other_mh)
+
+    def time_count_common(self):
+        mh = self.mh
+        other_mh = self.populated_mh
+        for i in range(COUNT_COMMON_TIMES):
+            mh.count_common(other_mh)
+
+    def time_merge(self):
+        mh = self.mh
+        other_mh = self.populated_mh
+        for i in range(MERGE_TIMES):
+            mh.merge(other_mh)
+
+    def time_copy(self):
+        mh = self.populated_mh
+        for i in range(COPY_TIMES):
+            mh.__copy__()
+
+    def time_concat(self):
+        mh = self.mh
+        other_mh = self.populated_mh
+        for i in range(CONCAT_TIMES):
+            mh += other_mh
+
+
 class PeakmemMinHashSuite:
     def setup(self):
         self.mh = MinHash(MINHASH_NUM, MINHASH_K, track_abundance=True)
@@ -66,6 +134,33 @@ def peakmem_add_many(self):
 ####################
 
 
+class TimeMinAbundanceSuite(TimeMinHashSuite):
+    def setup(self):
+        TimeMinHashSuite.setup(self)
+        self.mh = MinHash(MINHASH_NUM, MINHASH_K, track_abundance=True)
+
+        self.populated_mh = MinHash(MINHASH_NUM, MINHASH_K, track_abundance=True)
+        for seq in self.sequences:
+            self.populated_mh.add_sequence(seq)
+
+    def time_get_mins_abundance(self):
+        mh = self.populated_mh
+        for i in range(GET_MINS_RANGE):
+            mh.get_mins(with_abundance=True)
+
+    def time_set_abundances(self):
+        mh = self.mh
+        mins = self.populated_mh.get_mins(with_abundance=True)
+        for i in range(SET_ABUNDANCES_RANGE):
+            mh.set_abundances(mins)
+
+    def time_set_abundances_noclear(self):
+        mh = self.mh
+        mins = self.populated_mh.get_mins(with_abundance=True)
+        for i in range(SET_ABUNDANCES_RANGE):
+            mh.set_abundances(mins, clear=False)
+
+
 class PeakmemMinAbundanceSuite(PeakmemMinHashSuite):
     def setup(self):
         PeakmemMinHashSuite.setup(self)
@@ -75,6 +170,35 @@ def setup(self):
 ####################
 
 
+class TimeZipStorageSuite:
+    def setup(self):
+        import zipfile
+
+        self.zipfile = NamedTemporaryFile()
+
+        with zipfile.ZipFile(
+            self.zipfile, mode="w", compression=zipfile.ZIP_STORED
+        ) as storage:
+            for i in range(ZIP_STORAGE_WRITE):
+                # just so we have lots of entries
+                storage.writestr(str(i), b"0")
+            # one big-ish entry
+            storage.writestr("sig1", b"9" * 1_000_000)
+
+    def time_load_from_zipstorage(self):
+        with ZipStorage(self.zipfile.name) as storage:
+            for i in range(ZIP_STORAGE_LOAD):
+                storage.load("sig1")
+
+    def time_load_small_from_zipstorage(self):
+        with ZipStorage(self.zipfile.name) as storage:
+            for i in range(ZIP_STORAGE_LOAD):
+                storage.load("99999")
+
+    def teardown(self):
+        self.zipfile.close()
+
+
 class PeakmemZipStorageSuite:
     def setup(self):
         import zipfile

pyproject.toml

@@ -103,7 +103,6 @@ test = [
   "pytest>=6.2.4,<8.4.0",
   "pytest-cov>=4,<6.0",
   "pytest-xdist>=3.1",
-  "pytest-benchmark>=4.0",
   "pyyaml>=6,<7",
   "diff-cover>=7.3",
   "covdefaults>=2.2.2",

tox.ini

@@ -111,34 +111,6 @@ commands =
     asv machine --yes
     asv continuous latest HEAD {posargs}
 
-[testenv:benchmarks]
-description = run pytest-benchmark for benchmarking
-changedir = {toxinidir}
-commands =
-    pytest \
-      --cov "{envsitepackagesdir}/sourmash" \
-      --cov-config "{toxinidir}/tox.ini" \
-      --cov-report= \
-      --junitxml {toxworkdir}/junit.benchmarks.xml \
-      --benchmark-only \
-      -n 0 \
-      {posargs:tests}
-
-[testenv:codspeed]
-description = run codspeed for benchmarking
-deps =
-    pytest-codspeed
-changedir = {toxinidir}
-commands =
-    pytest \
-      --cov "{envsitepackagesdir}/sourmash" \
-      --cov-config "{toxinidir}/tox.ini" \
-      --cov-report= \
-      --junitxml {toxworkdir}/junit.codspeed.xml \
-      --codspeed \
-      -k benchmarks \
-      {posargs:tests}
-
 [testenv:docs]
 description = invoke sphinx-build to build the HTML docs
 basepython = python3.10
@@ -208,7 +180,7 @@ commands =
     coverage xml -i -o {toxworkdir}/coverage.xml
     coverage html -i -d {toxworkdir}/htmlcov
     diff-cover --compare-branch {env:DIFF_AGAINST:origin/latest} {toxworkdir}/coverage.xml
-depends = py312, py311, py310, pypy3, codspeed
+depends = py312, py311, py310, pypy3
 pass_env = {[testenv]pass_env}
     DIFF_AGAINST
 set_env = COVERAGE_FILE={toxworkdir}/.coverage
@@ -261,7 +233,7 @@ source = src/sourmash/
 python =
     3.10: py310, docs, package_description, coverage
     3.11: py311, coverage
-    3.12: py312, coverage, codspeed
+    3.12: py312, coverage
 
 [flake8]
 max-complexity = 22