Merge pull request #7 from reallistic/cleanup_collector

Cleanup some dead code in the collector, add Pyloot.stop to the README

README.md

@@ -30,6 +30,16 @@ Otherwise, threading.Thread is used.
 """
 loot.start()
 
+"""
+Stop running the collector background thread.
+
+NOTE: This does not do a "final" collection.
+To ensure objects were collected in a short lived execution, call collect_objects().
+
+:param blocking: When true, wait until the thread has died
+"""
+loot.stop()
+
 """
 Return a WSGI compatible application serving the PyLoot remote backend and
 and the website.
@@ -76,7 +86,7 @@ app = DispatcherMiddleware(app, {
 # Embedded code
 from pyloot import PyLoot
 ...
-pyloot = PyLoot(host="127.0.0.1", port=8030)
+pyloot = PyLoot(host="127.0.0.1", port=8000)
 ...
 ```
 ```shell script
@@ -86,7 +96,7 @@ usage: pyloot [-h HOST] [-p PORT] [--help]
 
 optional arguments:
 -h HOST, --host HOST  Host to listen on. (Default: 0.0.0.0)
--p PORT, --port PORT  Port to listen on. (Default: 8030)
+-p PORT, --port PORT  Port to listen on. (Default: 8000)
 --help                show this help message and exit
 ```
 
@@ -111,7 +121,7 @@ app.mount("/_pyloot", WSGIMiddleware(pyloot_wrapper))
 
 # Screenshots
 ### View history of object counts by object group:
-![history screenshot](https://raw.githubusercontent.com/reallistic/pyloot/master/docs/historypage.png)
+![history screenshot](https://raw.githubusercontent.com/reallistic/pyloot/master/docs/history.png)
 
 ### Modify history page size
 ![history screenshot](https://raw.githubusercontent.com/reallistic/pyloot/master/docs/history-pageLimit.png)

pyloot/__init__.py

@@ -55,6 +55,14 @@ def start(self):
         logger.debug("Finished starting pyloot looter thread")
 
     def stop(self, blocking=False):
+        """
+        Stop running the collector background thread.
+
+        NOTE: This does not do a "final" collection.
+        To ensure objects were collected in a short lived execution, call collect_objects().
+
+        :param blocking: When true, wait until the thread has died
+        """
         self._running = False
         if blocking:
             self._thread_ended.wait()

pyloot/collector.py

@@ -39,7 +39,7 @@ def _safe_getattr(obj: object, k: str, default: Any = Exception):
     try:
         from zope.interface.ro import C3
 
-        if isinstance(obj, C3) or issubclass(obj, C3):
+        if isinstance(obj, C3) or issubclass(obj, C3):  # type: ignore
             if k.startswith("ORIG_"):
                 return "__ignored_zope_interface_C3_{}__".format(k)
     except (ImportError, TypeError):
@@ -92,38 +92,21 @@ def _should_include_object(ref: object, ignore_set: Set[int]) -> bool:
     return True
 
 
-def get_child_ids(obj: object, ignore_set: Set[int]) -> List[int]:
+def get_child_ids(obj: object) -> List[int]:
     """
     Return children of the provided object using gc.get_referents
 
     :param obj: The object
-    :param ignore_set: A set of `id()`s which should be ignored.
     :return: List of object ids
     """
     return [id(child) for child in gc.get_referents(obj)]
 
 
-def get_parent_ids(obj: object, ignore_set: Set[int]) -> List[int]:
-    """
-    Return children of the provided object using gc.get_referents
-
-    :param obj: The object
-    :param ignore_set: A set of `id()`s which should be ignored.
-    :return: List of object ids
-    """
-    return [
-        id(parent)
-        for parent in gc.get_referrers(obj)
-        if _should_include_object(parent, ignore_set)
-    ]
-
-
-def get_data(obj: object, ignore_set: Set[int]) -> ObjectDescriptor:
+def get_data(obj: object) -> ObjectDescriptor:
     """
     Return a Object descriptor for the given object
 
     :param obj: The object
-    :param ignore_set: A set of `id()`s which should be ignored.
     :return:
     """
     obj_type = type(obj)
@@ -134,7 +117,7 @@ def get_data(obj: object, ignore_set: Set[int]) -> ObjectDescriptor:
         id=id(obj),
         attrs=_safe_get_attrs(obj),
         parent_ids=[],
-        child_ids=get_child_ids(obj, ignore_set),
+        child_ids=get_child_ids(obj),
         repr=_safe_repr(obj),
     )
 
@@ -162,7 +145,7 @@ def get_object_descriptors(
 
     objs = [obj for obj in objs if _should_include_object(obj, ignore_set)]
     ignore_set.add(id(objs))
-    results = [get_data(obj, ignore_set) for obj in objs]
+    results = [get_data(obj) for obj in objs]
     del objs
     del ignore_set
     child_to_parent: Dict[int, Set[int]] = defaultdict(set)

setup.py

@@ -3,7 +3,7 @@
 
 setup(
     name="pyloot",
-    versioning="distance",
+    versioning="post",
     setup_requires="setupmeta",
     author="Michael Chase",
     description="Multiprocessing compatible memory leak debugger inspired by dozer/dowser",

tests/test_memory.py

@@ -5,7 +5,7 @@
 def test_store_blank_slate():
     backend = InMemoryBackend()
     test_obj = dict(a=1)
-    obj_descr = collector.get_data(test_obj, set())
+    obj_descr = collector.get_data(test_obj)
     backend.store([obj_descr])
 
     descrs = backend.fetch()
@@ -29,7 +29,7 @@ def test_store_blank_slate():
     assert history[0].counts == [1, 1]
 
     test_obj2 = dict(a=1)
-    obj_descr2 = collector.get_data(test_obj2, set())
+    obj_descr2 = collector.get_data(test_obj2)
 
     # new obj should increase data and count
     backend.store([obj_descr, obj_descr2])