BUG: Fix to_json lines with escaped characters

Updates existing to_json methodology by adding is_escaping variable,
which ensures escaped chars are handled correctly.

Bug description: A simple check of whether the prior char is a backslash
is insufficient because the backslash may itself be escaped.

A test is also included (previously included in #14693).

xref #14693
xref #15096

Author: rouzazari

pandas/io/tests/json/test_pandas.py

@@ -966,9 +966,11 @@ def test_to_jsonl(self):
         expected = '{"a":1,"b":2}\n{"a":1,"b":2}'
         self.assertEqual(result, expected)
 
-        df = DataFrame([["foo}", "bar"], ['foo"', "bar"]], columns=['a', 'b'])
+        df = DataFrame([["foo}", "bar"], ['foo"', "bar"], ["foo\\", "bar"]],
+                       columns=['a', 'b'])
         result = df.to_json(orient="records", lines=True)
-        expected = '{"a":"foo}","b":"bar"}\n{"a":"foo\\"","b":"bar"}'
+        expected = ('{"a":"foo}","b":"bar"}\n{"a":"foo\\"","b":"bar"}\n'
+                    '{"a":"foo\\\\","b":"bar"}')
         self.assertEqual(result, expected)
         assert_frame_equal(pd.read_json(result, lines=True), df)
 

pandas/lib.pyx

@@ -1099,6 +1099,7 @@ def convert_json_to_lines(object arr):
     """
     cdef:
         Py_ssize_t i = 0, num_open_brackets_seen = 0, in_quotes = 0, length
+        Py_ssize_t is_escaping = 0
         ndarray[uint8_t] narr
         unsigned char v, comma, left_bracket, right_brack, newline
 
@@ -1113,8 +1114,10 @@ def convert_json_to_lines(object arr):
     length = narr.shape[0]
     for i in range(length):
         v = narr[i]
-        if v == quote and i > 0 and narr[i - 1] != backslash:
+        if v == quote and i > 0 and not is_escaping::
             in_quotes = ~in_quotes
+        if v == backslash or is_escaping:
+            is_escaping = ~is_escaping
         if v == comma: # commas that should be \n
             if num_open_brackets_seen == 0 and not in_quotes:
                 narr[i] = newline