Fix relative paths in wasm backend source maps. Fixes #9837 (#9882)

Author: kripken

tests/test_core.py

@@ -7181,6 +7181,12 @@ def encode_utf8(data):
       else:
         return data
 
+    def source_map_file_loc(name):
+      if shared.Settings.WASM_BACKEND:
+        return name
+      # in fastcomp, we have the absolute path, which is not good
+      return os.path.abspath(name)
+
     data = json.load(open(map_filename))
     if str is bytes:
       # Python 2 compatibility
@@ -7190,7 +7196,7 @@ def encode_utf8(data):
       # the output file.
       self.assertPathsIdentical(map_referent, data['file'])
     assert len(data['sources']) == 1, data['sources']
-    self.assertPathsIdentical(os.path.abspath('src.cpp'), data['sources'][0])
+    self.assertPathsIdentical(source_map_file_loc('src.cpp'), data['sources'][0])
     if hasattr(data, 'sourcesContent'):
       # the sourcesContent attribute is optional, but if it is present it
       # needs to containt valid source text.
@@ -7203,7 +7209,7 @@ def encode_utf8(data):
       mappings = encode_utf8(mappings)
     seen_lines = set()
     for m in mappings:
-      self.assertPathsIdentical(os.path.abspath('src.cpp'), m['source'])
+      self.assertPathsIdentical(source_map_file_loc('src.cpp'), m['source'])
       seen_lines.add(m['originalLine'])
     # ensure that all the 'meaningful' lines in the original code get mapped
     # when optimizing, the binaryen optimizer may remove some of them (by inlining, etc.)
@@ -8243,7 +8249,7 @@ def test_ubsan_full_static_cast(self, args):
     'g4': ('-g4', [
       "src.cpp:3:12: runtime error: reference binding to null pointer of type 'int'",
       'in main ',
-      '/src.cpp:3:8'
+      'src.cpp:3:8'
     ]),
   })
   @no_fastcomp('ubsan not supported on fastcomp')

tests/test_other.py

@@ -8894,7 +8894,8 @@ def test_wasm_sourcemap(self):
                     '--dwarfdump-output',
                     path_from_root('tests', 'other', 'wasm_sourcemap', 'foo.wasm.dump'),
                     '-o', 'a.out.wasm.map',
-                    path_from_root('tests', 'other', 'wasm_sourcemap', 'foo.wasm')]
+                    path_from_root('tests', 'other', 'wasm_sourcemap', 'foo.wasm'),
+                    '--basepath=' + os.getcwd()]
     run_process(wasm_map_cmd)
     output = open('a.out.wasm.map').read()
     # has "sources" entry with file (includes also `--prefix =wasm-src:///` replacement)
@@ -8909,12 +8910,35 @@ def test_wasm_sourcemap_dead(self):
                     '--dwarfdump-output',
                     path_from_root('tests', 'other', 'wasm_sourcemap_dead', 't.wasm.dump'),
                     '-o', 'a.out.wasm.map',
-                    path_from_root('tests', 'other', 'wasm_sourcemap_dead', 't.wasm')]
+                    path_from_root('tests', 'other', 'wasm_sourcemap_dead', 't.wasm'),
+                    '--basepath=' + os.getcwd()]
     run_process(wasm_map_cmd, stdout=PIPE, stderr=PIPE)
     output = open('a.out.wasm.map').read()
     # has only two entries
     self.assertRegexpMatches(output, r'"mappings":\s*"[A-Za-z0-9+/]+,[A-Za-z0-9+/]+"')
 
+  @no_fastcomp()
+  def test_wasm_sourcemap_relative_paths(self):
+    def test(infile, source_map_added_dir=''):
+      expected_source_map_path = os.path.join(source_map_added_dir, 'a.cpp')
+      print(infile, expected_source_map_path)
+      shutil.copyfile(path_from_root('tests', 'hello_123.c'), infile)
+      infiles = [
+        infile,
+        os.path.abspath(infile),
+        './' + infile
+      ]
+      for curr in infiles:
+        print('  ', curr)
+        run_process([PYTHON, EMCC, curr, '-g4'])
+        with open('a.out.wasm.map', 'r') as f:
+          self.assertIn('"%s"' % expected_source_map_path, str(f.read()))
+
+    test('a.cpp')
+
+    os.mkdir('inner')
+    test(os.path.join('inner', 'a.cpp'), 'inner')
+
   def test_wasm_producers_section(self):
     # no producers section by default
     run_process([PYTHON, EMCC, path_from_root('tests', 'hello_world.c')])
@@ -9690,17 +9714,17 @@ def test_lsan_leaks(self, ext):
   @parameterized({
     'c': ['c', [
       r'in malloc.*a\.out\.wasm\+0x',
-      r'(?im)in f (/|[a-z]:).*/test_lsan_leaks\.c:6:21$',
-      r'(?im)in main (/|[a-z]:).*/test_lsan_leaks\.c:10:16$',
-      r'(?im)in main (/|[a-z]:).*/test_lsan_leaks\.c:12:3$',
-      r'(?im)in main (/|[a-z]:).*/test_lsan_leaks\.c:13:3$',
+      r'(?im)in f (|[/a-z\.]:).*/test_lsan_leaks\.c:6:21$',
+      r'(?im)in main (|[/a-z\.]:).*/test_lsan_leaks\.c:10:16$',
+      r'(?im)in main (|[/a-z\.]:).*/test_lsan_leaks\.c:12:3$',
+      r'(?im)in main (|[/a-z\.]:).*/test_lsan_leaks\.c:13:3$',
     ]],
     'cpp': ['cpp', [
       r'in operator new\[\]\(unsigned long\).*a\.out\.wasm\+0x',
-      r'(?im)in f\(\) (/|[a-z]:).*/test_lsan_leaks\.cpp:4:21$',
-      r'(?im)in main (/|[a-z]:).*/test_lsan_leaks\.cpp:8:16$',
-      r'(?im)in main (/|[a-z]:).*/test_lsan_leaks\.cpp:10:3$',
-      r'(?im)in main (/|[a-z]:).*/test_lsan_leaks\.cpp:11:3$',
+      r'(?im)in f\(\) (|[/a-z\.]:).*/test_lsan_leaks\.cpp:4:21$',
+      r'(?im)in main (|[/a-z\.]:).*/test_lsan_leaks\.cpp:8:16$',
+      r'(?im)in main (|[/a-z\.]:).*/test_lsan_leaks\.cpp:10:3$',
+      r'(?im)in main (|[/a-z\.]:).*/test_lsan_leaks\.cpp:11:3$',
     ]],
   })
   @no_fastcomp('lsan not supported on fastcomp')

tools/shared.py

@@ -2871,10 +2871,14 @@ def path_to_system_js_libraries(library_name):
 
   @staticmethod
   def emit_wasm_source_map(wasm_file, map_file):
+    # source file paths must be relative to the location of the map (which is
+    # emitted alongside the wasm)
+    base_path = os.path.dirname(os.path.abspath(Settings.WASM_BINARY_FILE))
     sourcemap_cmd = [PYTHON, path_from_root('tools', 'wasm-sourcemap.py'),
                      wasm_file,
                      '--dwarfdump=' + LLVM_DWARFDUMP,
-                     '-o',  map_file]
+                     '-o',  map_file,
+                     '--basepath=' + base_path]
     check_call(sourcemap_cmd)
 
   @staticmethod

tools/wasm-sourcemap.py

@@ -11,7 +11,7 @@
 """
 
 import argparse
-from collections import OrderedDict, namedtuple
+from collections import OrderedDict
 import json
 import logging
 from math import floor, log
@@ -39,6 +39,7 @@ def parse_args():
   parser.add_argument('-u', '--source-map-url', nargs='?', help='specifies sourceMappingURL section contest')
   parser.add_argument('--dwarfdump', help="path to llvm-dwarfdump executable")
   parser.add_argument('--dwarfdump-output', nargs='?', help=argparse.SUPPRESS)
+  parser.add_argument('--basepath', help='base path for source files, which will be relative to this')
   return parser.parse_args()
 
 
@@ -73,7 +74,13 @@ def resolve(self, name):
 # SourceMapPrefixes contains resolver for file names that are:
 #  - "sources" is for names that output to source maps JSON
 #  - "load" is for paths that used to load source text
-SourceMapPrefixes = namedtuple('SourceMapPrefixes', 'sources, load')
+class SourceMapPrefixes:
+  def __init__(self, sources, load):
+    self.sources = sources
+    self.load = load
+
+  def provided(self):
+    return bool(self.sources.prefixes or self.load.prefixes)
 
 
 def encode_vlq(n):
@@ -243,11 +250,10 @@ def read_dwarf_entries(wasm, options):
   return sorted(entries, key=lambda entry: entry['address'])
 
 
-def build_sourcemap(entries, code_section_offset, prefixes, collect_sources):
+def build_sourcemap(entries, code_section_offset, prefixes, collect_sources, base_path):
   sources = []
   sources_content = [] if collect_sources else None
   mappings = []
-
   sources_map = {}
   last_address = 0
   last_source_id = 0
@@ -264,7 +270,13 @@ def build_sourcemap(entries, code_section_offset, prefixes, collect_sources):
       column = 1
     address = entry['address'] + code_section_offset
     file_name = entry['file']
-    source_name = prefixes.sources.resolve(file_name)
+    # if prefixes were provided, we use that; otherwise, we emit a relative
+    # path
+    if prefixes.provided():
+      source_name = prefixes.sources.resolve(file_name)
+    else:
+      file_name = os.path.relpath(os.path.abspath(file_name), base_path)
+      source_name = file_name
     if source_name not in sources_map:
       source_id = len(sources)
       sources_map[source_name] = source_id
@@ -311,7 +323,7 @@ def main():
   prefixes = SourceMapPrefixes(sources=Prefixes(options.prefix), load=Prefixes(options.load_prefix))
 
   logger.debug('Saving to %s' % options.output)
-  map = build_sourcemap(entries, code_section_offset, prefixes, options.sources)
+  map = build_sourcemap(entries, code_section_offset, prefixes, options.sources, options.basepath)
   with open(options.output, 'w') as outfile:
     json.dump(map, outfile, separators=(',', ':'))