Merge pull request diffblue#342 from diffblue/add_java_app_stats_script

Added script for computation of statistics of a Java application.

Author: marek-trtik

scripts/jstats.py

@@ -0,0 +1,77 @@
+import argparse
+import os
+import json
+
+
+def _parse_cmd_line():
+    parser = argparse.ArgumentParser(
+        description="The script computes a statistics of java files under the passed directory. "
+                    "There is also added count of 'web.xml' files to the statistics, in case "
+                    "the directory hold Java web application(s).")
+    parser.add_argument("-V", "--version", action="store_true",
+                        help="Prints a version string.")
+    parser.add_argument("root_dir", type=str,
+                        help="A directory under which Java files (and web.xml files) will be considered for the statistics.")
+    parser.add_argument("-o", "--output", type=str, default="jstats_result.json",
+                         help="The file into which the statistics will be written to. The file does not have to exist. "
+                              "The data are saved in JSON format. If the option is not specified, then the results will "
+                              "be written into file 'jstats_result.json' under the current working directory.")
+    return parser.parse_args()
+
+
+def count_lines_in_file(file_pathname):
+    with open(file_pathname, "r") as ifile:
+        content = [line.strip() for line in ifile.readlines() if len(line.strip()) > 0]
+    return len(content)
+
+
+def compute_statistics_of_java_files_under_directory(search_root_dir):
+    java_files = [os.path.join(root, filename)
+                  for root, dir_name, file_names in os.walk(search_root_dir)
+                  for filename in file_names
+                  if os.path.splitext(filename)[1].lower() == ".java"]
+    file_lines = map(count_lines_in_file, java_files)
+    if len(java_files) == 0:
+        num_lines_total = 0
+        num_lines_per_file_min = 0
+        num_lines_per_file_max = 0
+        average_num_lines = 0
+    else:
+        num_lines_total = sum(file_lines)
+        num_lines_per_file_min = min(file_lines)
+        num_lines_per_file_max = max(file_lines)
+        average_num_lines = num_lines_total / len(java_files)
+    statistics = dict()
+    statistics["count_of_java_files"] = len(java_files)
+    statistics["minimal_number_of_lines_in_java_files"] = num_lines_per_file_min
+    statistics["maximal_number_of_lines_in_java_files"] = num_lines_per_file_max
+    statistics["average_number_of_lines_in_java_files"] = average_num_lines
+    statistics["total_number_of_lines_in_java_files"] = num_lines_total
+    return statistics
+
+
+def _main():
+    cmdline = _parse_cmd_line()
+
+    if cmdline.version:
+        print("v.1.0")
+        return 0
+
+    cmdline.root_dir = os.path.abspath(cmdline.root_dir)
+    if not os.path.isdir(cmdline.root_dir):
+        print("ERROR: cannot find/access the passed directory '" + cmdline.root_dir + "'.")
+        return 1
+    if os.path.isdir(cmdline.output):
+        print("ERROR: cannot find/access the output file '" + cmdline.output + "'.")
+        return 1
+
+    stats = compute_statistics_of_java_files_under_directory(cmdline.root_dir)
+    stats["num_web.xml_files"] = len([0 for _, _, file_names in os.walk(cmdline.root_dir) for filename in file_names if filename == "web.xml"])
+    with open(cmdline.output, "w") as ofile:
+        ofile.write(json.dumps(stats, sort_keys=True, indent=4))
+    return 0
+
+
+if __name__ == "__main__":
+    exit(_main())
+