Code churn per file

README.md

@@ -34,6 +34,7 @@ Result is a print with aggregated contribution and churn per author for a given
 
 - **-h, --h, --help**    show this help message and exit
 - **-exdir**                   exclude Git repository subdirectory
+-- **--show-file-data**        display line count changes per file
 
 ## Usage Example 1
 
@@ -63,6 +64,73 @@ contribution: 4423
 churn:        -543
 ```
 
+## Usage Example 3
+
+```bash
+python ./gitcodechurn.py after="2018-11-29" before="2021-11-05" author="flacle" dir="/Users/myname/myrepo" --show-file-data
+```
+
+## Output 3
+
+```bash
+author:          flacle
+contribution:    337
+churn:           -19
+-------------------------------------------------------------------------------
+            FILE NAME             |  LINE #  |  ADDED   | REMOVED  
+-------------------------------------------------------------------------------
+         gitcodechurn.py          |    1     |   190    |    0     
+-------------------------------------------------------------------------------
+         gitcodechurn.py          |    2     |    4     |    0     
+-------------------------------------------------------------------------------
+         gitcodechurn.py          |    37    |    2     |    0     
+-------------------------------------------------------------------------------
+         gitcodechurn.py          |    40    |    0     |    1     
+-------------------------------------------------------------------------------
+         gitcodechurn.py          |    42    |    1     |    0     
+-------------------------------------------------------------------------------
+         gitcodechurn.py          |    45    |    0     |    1     
+-------------------------------------------------------------------------------
+         gitcodechurn.py          |    47    |    1     |    0     
+-------------------------------------------------------------------------------
+         gitcodechurn.py          |    50    |    0     |    1     
+-------------------------------------------------------------------------------
+         gitcodechurn.py          |    52    |    1     |    0     
+-------------------------------------------------------------------------------
+         gitcodechurn.py          |    55    |    0     |    1     
+-------------------------------------------------------------------------------
+         gitcodechurn.py          |    57    |    8     |    1     
+-------------------------------------------------------------------------------
+         gitcodechurn.py          |    66    |    2     |    0     
+-------------------------------------------------------------------------------
+         gitcodechurn.py          |    62    |    0     |    1     
+...
+-------------------------------------------------------------------------------
+         gitcodechurn.py          |   200    |    1     |    0     
+-------------------------------------------------------------------------------
+            README.md             |    12    |    2     |    0     
+-------------------------------------------------------------------------------
+            README.md             |    16    |    0     |    1     
+-------------------------------------------------------------------------------
+            README.md             |    18    |    1     |    0     
+-------------------------------------------------------------------------------
+            README.md             |    21    |    11    |    0     
+-------------------------------------------------------------------------------
+            README.md             |    20    |    0     |    1     
+-------------------------------------------------------------------------------
+            README.md             |    33    |    1     |    0     
+-------------------------------------------------------------------------------
+            README.md             |    22    |    0     |    1     
+-------------------------------------------------------------------------------
+            README.md             |    35    |    1     |    0     
+-------------------------------------------------------------------------------
+            README.md             |    24    |    0     |    2     
+-------------------------------------------------------------------------------
+            README.md             |    37    |    3     |    0     
+-------------------------------------------------------------------------------
+            README.md             |    41    |    12    |    0     
+```
+
 Outputs of Usage Example 1 can be used as part of a pipeline that generates bar charts for reports:
 ![contribution vs churn example chart](/chart.png)
 

gitcodechurn.py

@@ -67,6 +67,11 @@ def main():
         default = '',
         help = 'the Git repository subdirectory to be excluded'
     )
+    parser.add_argument(
+        "--show-file-data",
+        action="store_true",
+        help="Display line change information for the analyzed file(s)"
+    )
     args = parser.parse_args()
 
     after  = args.after
@@ -99,6 +104,53 @@ def main():
     # print files in case more granular results are needed
     #print('files: ', files)
 
+    if args.show_file_data:
+        display_file_metrics(files)
+
+
+def display_file_metrics(files):
+    display_file_metrics_header()
+    for file_name, line_change_info in files.items():
+        for line_number, line_diff_stats in line_change_info.items():
+            display_file_metrics_row(file_name, line_number, line_diff_stats)
+
+
+def display_file_metrics_header():
+    print("-" * 79)
+    print(
+        "{file}|{line_number}|{lines_added}|{lines_removed}".format(
+            file=format_column("FILE NAME", 34),
+            line_number=format_column("LINE #", 10),
+            lines_added=format_column("ADDED", 10),
+            lines_removed=format_column("REMOVED", 10),
+        )
+    )
+
+
+def display_file_metrics_row(file_name, line_number, line_diff_stats):
+    added = line_diff_stats.get("lines_added")
+    removed = line_diff_stats.get("lines_removed")
+
+    if added == 0 and removed == 0:
+        return
+    print("-" * 79)
+    print(
+        "{file}|{ln}|{lines_added}|{lines_removed}".format(
+            file=format_column(file_name, 34),
+            ln=format_column(str(line_number), 10),
+            lines_added=format_column(str(added), 10),
+            lines_removed=format_column(str(removed), 10),
+        )
+    )
+
+
+def format_column(text, width):
+    text_length = len(text)
+    total_pad = width - text_length
+    pad_left = total_pad // 2
+    pad_right = total_pad - pad_left
+    return (" " * pad_left) + text + (" " * pad_right)
+
 
 def calculate_statistics(commits, dir, exdir):
     # structured like this: files -> LOC
@@ -142,21 +194,97 @@ def get_loc(commit, dir, files, contribution, churn, exdir):
             new_loc_changes = is_loc_change(result, loc_changes)
             if loc_changes != new_loc_changes:
                 loc_changes = new_loc_changes
-                locc = get_loc_change(loc_changes)
-                for loc in locc:
-                    if loc in files[file]:
-                        files[file][loc] += locc[loc]
-                        churn += abs(locc[loc])
-                    else:
-                        files[file][loc] = locc[loc]
-                        contribution += abs(locc[loc])
+                (removal, addition) = get_loc_change(loc_changes)
+
+                files, contribution, churn = merge_operations(removal, addition, files, contribution, churn, file)
             else:
                 continue
     return [files, contribution, churn]
 
 
-def get_commit_results(command, dir):
-    return get_proc_out(command, dir).splitlines()
+def merge_operations(removal, addition, files, contribution, churn, file):
+    # Ensure all required data is in place
+    ensure_file_exists(files, file)
+
+    file_line_churn_dict = files[file]
+
+    if is_noop(removal, addition):
+        # In the case of a noop, it's not counted in change metrics, but should
+        # be marked as changed to accurately include future churn metrics
+        # An example of this is a diff like:
+        #   "diff --git README.md README.md",
+        #   "index bedbc85..bb033cd 100644",
+        #   "--- README.md",
+        #   "+++ README.md",
+        #   "@@ -8 +8 @@ Code churn has several definitions, the one that to me provides the most value a",
+        #   "-*Reference: https://blog.gitprime.com/why-code-churn-matters/*",
+        #   "+*Reference: https://www.pluralsight.com/blog/teams/why-code-churn-matters*",
+        # In this example, we deleted the line, and then added the line by updating the link
+        # This repo would consider this a "No-Op" as it nets to no change
+        # However, we want to mark line 8 as changed so that all subsequent
+        # changes to line 8 are marked as churn
+        # The thinking behind this is the other updates should have been made
+        # while this change was being made.
+        remove_line_number = removal[0]
+        ensure_line_exists(file_line_churn_dict, remove_line_number)
+        return files, contribution, churn
+
+    for (line_number, lines_removed, lines_added) in compute_changes(removal, addition):
+        # Churn check performed before line modification changes
+        is_churn = is_this_churn(file_line_churn_dict, line_number)
+
+        ensure_line_exists(file_line_churn_dict, line_number)
+        line_count_change_metrics = file_line_churn_dict[line_number]
+
+        line_count_change_metrics["lines_removed"] += lines_removed
+        line_count_change_metrics["lines_added"] += lines_added
+
+        if is_churn:
+            churn += abs(lines_removed) + abs(lines_added)
+        else:
+            contribution += abs(lines_removed) + abs(lines_added)
+
+    return files, contribution, churn
+
+
+def compute_changes(removal, addition):
+    # If both removal and addition affect the same line, net out the change
+    # Returns a list of tuples of type (line_number, lines_removed, lines_added)
+    removed_line_number, lines_removed = removal
+    added_line_number, lines_added = addition
+
+    if removed_line_number == added_line_number:
+        if lines_added >= lines_removed:
+            return [(removed_line_number, 0, (lines_added - lines_removed))]
+        else:
+            return [(removed_line_number, (lines_removed - lines_added), 0)]
+    else:
+        return [
+            (removed_line_number, lines_removed, 0),
+            (added_line_number, 0, lines_added),
+        ]
+
+
+def is_this_churn(file_line_churn_dict, line_number):
+    # The definition of churn is any change to a line
+    # after the first time the line has been changed
+    # This is detected by a line operation logged in the file_line_churn_dict
+    return line_number in file_line_churn_dict
+
+
+def ensure_line_exists(file_line_churn_dict, line_number):
+    if line_number not in file_line_churn_dict:
+        file_line_churn_dict[line_number] = {"lines_removed": 0, "lines_added": 0}
+
+
+def ensure_file_exists(files, file):
+    if file not in files:
+        files[file] = {}
+
+
+def is_noop(removal, addition):
+    # A noop event occurs when a change indicates one delete and one add on the same line
+    return removal == addition
 
 
 # arrives in a format such as -13 +27,5 (no commas mean 1 loc change)
@@ -175,6 +303,8 @@ def get_loc_change(loc_changes):
         left = int(left[1:])
         left_dec = 1
 
+    removal = (left, left_dec)
+
     # additions
     right = loc_changes[loc_changes.find(' ')+1:]
     right_dec = 0
@@ -186,10 +316,10 @@ def get_loc_change(loc_changes):
         right = int(right[1:])
         right_dec = 1
 
-    if left == right:
-        return {left: (right_dec - left_dec)}
-    else:
-        return {left : left_dec, right: right_dec}
+    addition = (right, right_dec)
+
+    return (removal, addition)
+
 
 def is_loc_change(result, loc_changes):
     # search for loc changes (@@ ) and update loc_changes variable