Fix _parse_file and add new test

Signed-off-by: konstantin <[email protected]>

Author: zurk

lookout/style/format/features.py

@@ -148,6 +148,26 @@ def to_comment(self, correct_y: int) -> Comment:
         comment.text = comment.text.replace("<", "`").replace(">", "`")
         return comment
 
+    def to_code(self, correct_y: int) -> str:
+        """
+        Writes the code with regard to the correct node class.
+        :param correct_y: the index of the correct node class.
+        :return: code string
+        """
+        if correct_y == CLASS_INDEX[CLS_NOOP]:
+            return ""
+        if correct_y == CLASS_INDEX[CLS_NOOP]:
+            comment.text = "format: %s at column %d should be removed" % (
+                CLASSES[self.y], self.start.col)
+        elif self.y == CLASS_INDEX[CLS_NOOP]:
+            comment.text = "format: %s should be inserted at column %d" % (
+                CLASSES[correct_y], self.start.col)
+        else:
+            comment.text = "format: replace %s with %s at column %d" % (
+                CLASSES[self.y], CLASSES[correct_y], self.start.col)
+        comment.text = comment.text.replace("<", "`").replace(">", "`")
+        return comment
+
 
 CLS_SPACE = "<space>"
 CLS_SPACE_INC = "<+space>"
@@ -159,6 +179,18 @@ def to_comment(self, correct_y: int) -> Comment:
 CLS_SINGLE_QUOTE = "'"
 CLS_DOUBLE_QUOTE = '"'
 CLS_NOOP = "<noop>"
+CLASSES_REPR = {
+    CLS_SPACE: " ",
+    CLS_TAB: "\t",
+    CLS_NEWLINE: "\n",
+    CLS_SPACE_INC: "<nan>",
+    CLS_SPACE_DEC: "<nan>",
+    CLS_TAB_INC: "<nan>",
+    CLS_TAB_DEC: "<nan>",
+    CLS_SINGLE_QUOTE: "'",
+    CLS_DOUBLE_QUOTE: '"',
+    CLS_NOOP: "",
+}
 CLASSES = (CLS_SPACE, CLS_TAB, CLS_NEWLINE, CLS_SPACE_INC, CLS_SPACE_DEC,
            CLS_TAB_INC, CLS_TAB_DEC, CLS_SINGLE_QUOTE, CLS_DOUBLE_QUOTE, CLS_NOOP)
 CLASS_INDEX = {cls: i for i, cls in enumerate(CLASSES)}
@@ -639,6 +671,8 @@ def _parse_file(self, contents: str, root: bblfsh.Node, path: str) -> \
         parser = self.tokens.PARSER
         searchsorted = numpy.searchsorted
         for node in node_tokens:
+            if node.start_position.offset < pos:
+                continue
             if node.start_position.offset > pos:
                 sumlen = 0
                 diff = contents[pos:node.start_position.offset]

lookout/style/format/tests/for_parse_test.js.xz

@@ -25,6 +25,15 @@ def setUpClass(cls):
             cls.uast = bblfsh.Node.FromString(fin.read())
         cls.extractor = FeatureExtractor("javascript", parents_depth=2, siblings_window=5)
 
+    def test_parse_file2(self):
+        test_js_code_filepath = Path(__file__).parent / "for_parse_test.js.xz"
+        with lzma.open(str(test_js_code_filepath), mode="rt") as f:
+            code = f.read()
+        uast = bblfsh.BblfshClient("0.0.0.0:9432").parse(
+            filename="", language="javascript", contents=code.encode()).uast
+        nodes, parents = self.extractor._parse_file(code, uast, test_js_code_filepath)
+        self.assertEqual("".join(n.value for n in nodes), code)
+
     def test_parse_file(self):
         nodes, parents = self.extractor._parse_file(self.contents, self.uast, "test_file")
         text = []