Only iterate static or dynamic symbols (#9)

As a response to lief-project/LIEF#962 only
iterate static or dynamic symbols.

Added a simple Makefile to demonstrate shadowing variables.

.gitignore

@@ -8,3 +8,5 @@ sqlelf.egg-info/
 .vscode/
 dist/
 result
+examples/shadowed-symbols/exe
+examples/**/*.so

examples/shadowed-symbols/Makefile

@@ -0,0 +1,25 @@
+all: exe
+
+# This library depends libx2.so soname and calls h() from it
+y/liby.so: x/libx2.so
+	@mkdir -p $(dir $@)
+	echo 'extern int foo(); int g() { return foo(); }' | $(CC) -o $@ -shared -x c - -Lx -l:libx2.so '-Wl,--no-as-needed,--enable-new-dtags,-rpath,$$ORIGIN/../x'
+
+# This library has both file and soname libx.so
+x/libx.so:
+	@mkdir -p $(dir $@)
+	echo 'int foo(){return 12;}' | $(CC) -o $@ -shared -x c -
+
+# This library has both file and soname libx.so
+x/libx2.so:
+	@mkdir -p $(dir $@)
+	echo 'int foo(){return 1000;}' | $(CC) -o $@ -shared -x c -
+
+# This links to b/liby.so and c/libx.so, and gets libx.so and liby.so in DT_NEEDED, no paths.
+exe: y/liby.so x/libx.so
+	echo 'extern int g(); extern int foo(); int main(){ printf("\%d\n", g() + foo()); }' | \
+	$(CC) -o $@ -include stdio.h -x c - -Ly -Lx -l:liby.so '-Wl,--no-as-needed,--enable-new-dtags,-rpath,$$ORIGIN/y' \
+		  -l:libx.so '-Wl,--no-as-needed,--enable-new-dtags,-rpath,$$ORIGIN/x'
+
+clean:
+	rm -rf -- x y exe

sqlelf/cli.py

@@ -60,7 +60,7 @@ def start(args=sys.argv[1:], stdin=sys.stdin):
 
     # If none of the inputs are valid files, simply return
     if len(filenames) == 0:
-        return
+        sys.exit("No valid ELF files were provided")
 
     binaries: list[Binary] = [Binary(filename) for filename in filenames]
 

sqlelf/elf/symbol.py

@@ -5,6 +5,7 @@
 
 import apsw
 import apsw.ext
+import lief
 
 from sqlelf.elf.binary import Binary
 from sqlelf.elf.section import section_name as elf_section_name
@@ -16,7 +17,7 @@ def generator() -> Iterator[dict[str, Any]]:
             # super important that these accessors are pulled out of the tight loop
             # as they can be costly
             binary_path = binary.path
-            for symbol in binary.symbols:
+            for symbol in symbols(binary):
                 # The section index can be special numbers like 65521 or 65522
                 # that refer to special sections so they can't be indexed
                 section_name: str | None = next(
@@ -56,6 +57,25 @@ def generator() -> Iterator[dict[str, Any]]:
     return generator
 
 
+def symbols(binary: Binary) -> Iterator[lief.ELF.Symbol]:
+    """Use heuristic to either get static symbols or dynamic symbol table
+
+    The static symbol table is a superset of the dynamic symbol table.
+    However it is often stripped from binaries as it's not needed beyond
+    debugging.
+
+    This method uses the simplest heuristic of checking for it's existence
+    to return the static symbol table.
+
+    A bad actor is free to strip arbitrarily from the static symbol table
+    and it would affect this method.
+    """
+    static_symbols = binary.static_symbols
+    if len(static_symbols) > 0:
+        return static_symbols
+    return binary.dynamic_symbols
+
+
 def register(connection: apsw.Connection, binaries: list[Binary]):
     generator = elf_symbols(binaries)
     # setup columns and access by providing an example of the first entry returned

tests/test_cli.py

@@ -15,7 +15,8 @@ def test_cli_single_file_arguments():
     cli.start(["/bin/ls"], stdin)
 
 def test_cli_single_non_existent_file_arguments():
-    cli.start(["does_not_exist"])
+    with pytest.raises(SystemExit) as err:
+        cli.start(["does_not_exist"])
 
 def test_cli_prompt_single_file_arguments():
     stdin = StringIO(".exit 56\n")