Merge pull request #14 from mehdigolzadeh/mapping_onlypredict

Added new features.
Version updated to 1.0.0

README.md

@@ -59,7 +59,7 @@ _The default start-date is 6 months before the current date.
 
 `--verbose` **To have verbose output result**
 > Example: $ bodegha repo_owner/repo_name --verbose --key <token>
- 
+
 _The default value is false, if you don't pass this parameter the output will only be the accounts and their type_
 
 `--min-comments MIN_COMMENTS` 		**Minimum number of pull request and issue comments that are required to analyze an account**
@@ -87,6 +87,11 @@ _This group of parameters is the type of output, e.g., if you pass --json you wi
 
 > Example: $ bodegha repo_owner/repo_name --exclude mehdigolzadeh alexandredecan tommens --key <token>
 
+## As of version 1.0.0
+`--only-predicted` **Only list accounts that the prediction is available**
+> Example: $ bodegha repo_owner/repo_name --only-predicted
+ 
+
 ## Examples of BoDeGHa output (for illustration purposes only)
 ```
 $ bodegha request/request --key <my token> --start-date 01-01-2017  --verbose

bodegha.py

@@ -358,7 +358,7 @@ def myrunner(function, ret, *args, **kwargs):
     return ret[0]
 
 
-def progress(repository, accounts, exclude, date, verbose, min_comments, max_comments, apikey, output_type):
+def progress(repository, accounts, exclude, date, verbose, min_comments, max_comments, apikey, output_type, only_predicted):
     download_progress = tqdm(
         total=25, desc='Downloading comments', smoothing=.1,
         bar_format='{desc}: {percentage:3.0f}%|{bar}', leave=False)
@@ -383,10 +383,10 @@ def progress(repository, accounts, exclude, date, verbose, min_comments, max_com
         .sort_values('created_at', ascending=False)
         .groupby('author').head(100)
     )
-    if exclude != []:
+    if len(exclude) > 0:
         df = df[~df["author"].isin(exclude)]
 
-    if accounts != []:
+    if len(accounts) > 0:
         df = df[lambda x: x['author'].isin(accounts)]
 
     if(len(df) < 1):
@@ -429,11 +429,45 @@ def progress(repository, accounts, exclude, date, verbose, min_comments, max_com
     prediction_progress.set_description(tasks[1])
     result = run_function_in_thread(
         prediction_progress, predict, 25, args=(model, df_clusters))
+
+    result = result.sort_values(['prediction', 'account']).assign(patterns= lambda x: x['patterns'].astype('Int64'))
+
+    if only_predicted == True:
+        result = result.append(  
+            (
+                comments[lambda x: ~x['author'].isin(result['account'])][['author','body']]
+                .groupby('author', as_index=False)
+                .count()
+                .assign(
+                    emptycomments=np.nan,
+                    patterns=np.nan,
+                    dispersion=np.nan,
+                    prediction="Unknown",
+                )
+                .rename(columns={'author':'account','body':'comments','emptycomments':'empty comments'})
+            ),ignore_index=True,sort=True)
+
+        for identity in (set(accounts) - set(result['account'])):
+            result = result.append({
+                'account': identity,
+                'comments':np.nan,
+                'empty comments':np.nan,
+                'patterns':np.nan,
+                'dispersion':np.nan,
+                'prediction':"Not found",
+            },ignore_index=True,sort=True)
+
     if verbose is False:
-        result = result[['account', 'prediction']]
-    result = result.set_index('account').sort_values(['prediction', 'account'])
+        result = result.set_index('account')[['prediction']]
+    else:
+        result = (
+            result
+            .set_index('account')
+            [['comments', 'empty comments', 'patterns', 'dispersion','prediction']]
+        )
 
     prediction_progress.close()
+
 
     if output_type == 'json':
         return (result.reset_index().to_json(orient='records'))
@@ -450,11 +484,12 @@ def arg_parser():
     parser.add_argument(
         '--accounts', metavar='ACCOUNT', required=False, default=list(), type=str, nargs='*',
         help='User login of one or more accounts. Example: \
---accounts mehdigolzadeh alexandredecan tommens')
+--accounts mehdijuliani melgibson tomgucci')
     parser.add_argument(
         '--exclude', metavar='ACCOUNT', required=False, default=list(), type=str, nargs='*',
         help='List of accounts to be excluded in the analysis. Example: \
---exclude mehdigolzadeh alexandredecan tommens')
+--exclude mehdijuliani melgibson tomgucci')
+
     parser.add_argument(
         '--start-date', type=str, required=False,
         default=None, help='Starting date of comments to be considered')
@@ -470,7 +505,10 @@ def arg_parser():
     parser.add_argument(
         '--key', metavar='APIKEY', required=True, type=str, default='',
         help='GitHub APIv4 key to download comments from GitHub GraphQL API')
-
+    parser.add_argument(
+        '--only-predicted', action="store_false", required=False, default=True,
+        help='Only list accounts that the prediction is available.')
+
     group2 = parser.add_mutually_exclusive_group()
     group2.add_argument('--text', action='store_true', help='Print results as text.')
     group2.add_argument('--csv', action='store_true', help='Print results as csv.')
@@ -521,7 +559,8 @@ def cli():
                     min_comments,
                     max_comments,
                     apikey,
-                    output_type
+                    output_type,
+                    args.only_predicted,
                 ))
     except BodeghaError as e:
         sys.exit(e)

setup.py

@@ -4,7 +4,7 @@
 
 
 __package__ = 'bodegha'
-__version__ = '0.2.3'
+__version__ = '1.0.0'
 __licence__ = 'LGPL3'
 __maintainer__ = 'Mehdi Golzadeh'
 __email__ = '[email protected]'