Add tests to test all the logging capabilities

Currently , there are no tests to test textattack attack logging into
files such as csv, text and json.

Commit adds tests to test logging in the 3 above mentioned formats.

.gitignore

@@ -47,3 +47,4 @@ checkpoints/
 
 .vscode
 *.csv
+!tests/sample_outputs/csv_attack_log.csv

tests/sample_outputs/csv_attack_log.csv

@@ -0,0 +1,3 @@
+"original_text","perturbed_text","original_score","perturbed_score","original_output","perturbed_output","ground_truth_output","num_queries","result_type"
+"lovingly photographed in the manner of a golden book sprung to [[life]] , stuart little 2 [[manages]] [[sweetness]] largely without stickiness .","lovingly photographed in the manner of a golden book sprung to [[ife]] , stuart little 2 [[manager]] [[seetness]] largely without stickiness .",0.09334743022918701,0.6904040575027466,1,0,1,28,"Successful"
+"[[consistently]] [[clever]] and [[suspenseful]] .","[[conisstently]] [[celver]] and [[Huspenseful]] .",0.009427368640899658,0.8219608664512634,1,0,1,16,"Successful"

tests/sample_outputs/json_attack_summary.json

@@ -0,0 +1,13 @@
+{
+    "Attack Results": {
+        "Number of successful attacks:": 2,
+        "Number of failed attacks:": 0,
+        "Number of skipped attacks:": 0,
+        "Original accuracy:": 100.0,
+        "Accuracy under attack:": 0.0,
+        "Attack success rate:": 100.0,
+        "Average perturbed word %:": 45.0,
+        "Average num. words per input:": 12.0,
+        "Avg num queries:": 22.0
+    }
+}

tests/sample_outputs/txt_attack_log.txt

@@ -0,0 +1,21 @@
+--------------------------------------------- Result 1 ---------------------------------------------
+[[Positive (91%)]] --> [[Negative (69%)]]
+
+lovingly photographed in the manner of a golden book sprung to [[life]] , stuart little 2 [[manages]] [[sweetness]] largely without stickiness .
+
+lovingly photographed in the manner of a golden book sprung to [[ife]] , stuart little 2 [[manager]] [[seetness]] largely without stickiness .
+--------------------------------------------- Result 2 ---------------------------------------------
+[[Positive (99%)]] --> [[Negative (82%)]]
+
+[[consistently]] [[clever]] and [[suspenseful]] .
+
+[[conisstently]] [[celver]] and [[Huspenseful]] .
+Number of successful attacks: 2
+Number of failed attacks: 0
+Number of skipped attacks: 0
+Original accuracy: 100.0%
+Accuracy under attack: 0.0%
+Attack success rate: 100.0%
+Average perturbed word %: 45.0%
+Average num. words per input: 12.0
+Avg num queries: 22.0

tests/test_command_line/test_loggers.py

@@ -0,0 +1,84 @@
+import json
+import os
+
+from helpers import run_command_and_get_result
+import pytest
+
+DEBUG = False
+
+"""
+Attack command-line tests in the format (name, args, sample_output_file)
+"""
+
+"""
+ list_test_params data structure requires
+ 1) test name
+ 2) logger filetype - json/text/csv. # Future Work : Tests for Wandb and Visdom
+ 3) logger file name
+ 4) sample log file
+"""
+
+list_test_params = [
+    (
+        "json_summary_logger",
+        "json",
+        "textattack attack --recipe deepwordbug --model lstm-mr --num-examples 2 --log-summary-to-json attack_summary.json",
+        "attack_summary.json",
+        "tests/sample_outputs/json_attack_summary.json",
+    ),
+    (
+        "csv_logger",
+        "csv",
+        "textattack attack --recipe deepwordbug --model lstm-mr --num-examples 2 --log-to-csv attack_log.csv",
+        "attack_log.csv",
+        "tests/sample_outputs/csv_attack_log.csv",
+    ),
+    (
+        "txt_logger",
+        "txt",
+        "textattack attack --recipe deepwordbug --model lstm-mr --num-examples 2 --log-to-txt attack_log.txt",
+        "attack_log.txt",
+        "tests/sample_outputs/txt_attack_log.txt",
+    ),
+]
+
+
+@pytest.mark.parametrize(
+    "name, filetype, command, test_log_file, sample_log_file", list_test_params
+)
+def test_logger(name, filetype, command, test_log_file, sample_log_file):
+    # Run command and validate outputs.
+    result = run_command_and_get_result(command)
+
+    assert result.stdout is not None
+    assert result.stderr is not None
+    assert result.returncode == 0
+    assert os.path.exists(test_log_file), f"{test_log_file} did not get generated"
+
+    if filetype == "json":
+        with open(sample_log_file) as f:
+            desired_dictionary = json.load(f)
+
+        with open(test_log_file) as f:
+            test_dictionary = json.load(f)
+
+        assert (
+            desired_dictionary == test_dictionary
+        ), f"{filetype} file {test_log_file} differs from {sample_log_file}"
+
+    elif filetype == "csv" or filetype == "txt":
+        assert (
+            os.system(f"diff {test_log_file} {sample_log_file}") == 0
+        ), f"{filetype} file {test_log_file} differs from {sample_log_file}"
+
+    # cleanup
+    os.remove(test_log_file)
+
+
+#     result = run_command_and_get_result(command)
+#     stdout = result.stdout.decode().strip()
+#     print("stdout =>", stdout)
+#     stderr = result.stderr.decode().strip()
+#     print("stderr =>", stderr)
+
+#     assert stdout == desired_text