Python: CG trace: Don't abuse example dir

python/tools/recorded-call-graph-metrics/.gitignore

@@ -1,8 +1,9 @@
 # Example DB
 cg-trace-example-db/
 
-# Examples traces should be ignored in general
-example-traces/
+# Tests artifacts
+tests/python-traces/
+tests/cg-trace-test-db
 
 # Artifact from building `pip install -e .`
 src/cg_trace.egg-info/

python/tools/recorded-call-graph-metrics/README.md

@@ -4,17 +4,15 @@ also known as _call graph tracing_.
 
 Execute a python program and for each call being made, record the call and callee. This allows us to compare call graph resolution from static analysis with actual data -- that is, can we statically determine the target of each actual call correctly.
 
-This is still in the early stages, and currently only supports a very minimal working example (to show that this approach might work).
-
-The next hurdle is being able to handle multiple calls on the same line, such as
-
-- `foo(); bar()`
-- `foo(bar())`
-- `foo().bar()`
-
 ## How do I give it a spin?
 
-After following setup instructions below, run the `recreate-db.sh` script to create the database `cg-trace-example-db`. Then run the queries inside the `ql/` directory.
+After following setup instructions below, you should be able to reproduce the example trace by running
+
+```
+cg-trace --xml example/simple.xml example/simple.py
+```
+
+You can also run traces for all tests and build a database by running `tests/create-test-db.sh`. Then run the queries inside the `ql/` directory.
 
 
 ## Setup

python/tools/recorded-call-graph-metrics/example/simple.xml

@@ -1,10 +1,10 @@
 <root>
   <info>
     <cg_trace_version>0.0.2</cg_trace_version>
-    <args>--xml example-traces/simple.xml example/simple.py</args>
+    <args>--xml example/simple.xml example/simple.py</args>
     <exit_status>completed</exit_status>
     <elapsed>0.00 seconds</elapsed>
-    <utctimestamp>2020-07-20T12:02:56</utctimestamp>
+    <utctimestamp>2020-07-22T12:14:02</utctimestamp>
   </info>
   <recorded_calls>
     <recorded_call>

python/tools/recorded-call-graph-metrics/tests/create-test-db.sh

@@ -1,31 +1,32 @@
 #!/bin/bash
 
-set -e
-set -x
+set -Eeuo pipefail # see https://vaneyckt.io/posts/safer_bash_scripts_with_set_euxo_pipefail/
 
-if ! pip show cg_trace; then
+SCRIPTDIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" >/dev/null 2>&1 && pwd )"
+
+if ! pip show cg_trace &>/dev/null; then
     echo "You need to follow setup instructions in README"
     exit 1
 fi
 
-DB="cg-trace-example-db"
-SRC="example/"
-XMLDIR="example-traces"
+DB="$SCRIPTDIR/cg-trace-test-db"
+SRC="$SCRIPTDIR/python-src/"
+XMLDIR="$SCRIPTDIR/python-traces/"
 PYTHON_EXTRACTOR=$(codeql resolve extractor --language=python)
 
-
-cg-trace --xml "$XMLDIR"/simple.xml example/simple.py
-cg-trace --xml "$XMLDIR"/builtins.xml example/builtins.py
-cg-trace --xml "$XMLDIR"/multiple-on-one-line.xml example/multiple-on-one-line.py
-cg-trace --xml "$XMLDIR"/class-simple.xml example/class-simple.py
-
-
 rm -rf "$DB"
+rm -rf "$XMLDIR"
+
+mkdir -p "$XMLDIR"
+
+for f in $(ls $SRC); do
+    echo "Tracing $f"
+    cg-trace --xml "$XMLDIR/${f%.py}.xml" "$SRC/$f"
+done
 
 codeql database init --source-root="$SRC" --language=python "$DB"
 codeql database trace-command --working-dir="$SRC" "$DB" "$PYTHON_EXTRACTOR/tools/autobuild.sh"
 codeql database index-files --language xml --include-extension .xml --working-dir="$XMLDIR" "$DB"
 codeql database finalize "$DB"
 
-set +x
 echo "Created database '$DB'"

python/tools/recorded-call-graph-metrics/tests/python-src/simple.py

@@ -0,0 +1,10 @@
+def foo():
+    print('foo')
+
+def bar():
+    print('bar')
+
+foo()
+bar()
+
+foo(); bar()