Support year selection, output repo language.

Author: baijiangliang

main.py

@@ -1,7 +1,6 @@
 # coding: utf8
 """Generate annual report for programmers."""
 import os
-import time
 import traceback
 from datetime import datetime
 from typing import Dict, Any
@@ -18,11 +17,18 @@ def get_user_info() -> Dict[str, Any]:
     """ Get information from user's inputs. """
     print('Sometimes it is the people no one imagines anything of, who do the things no one can '
           'imagine.\n' + ' ' * 70 + '-- Alan Turing')
-    print('欢迎使用程序员年度总结生成器')
+    print('欢迎使用程序员年度报告生成器')
     while True:
         name = input('请输入你的名字，按回车继续\n').strip()
         if name:
             break
+    now = datetime.now()
+    recent_year = now.year
+    if now.month == 1:
+        recent_year -= 1
+    year = input('请输入你想要生成报告的年份，不输入则使用 {0}\n'.format(recent_year)).strip()
+    if not year:
+        year = recent_year
     print('请输入你近一年使用过的 git 邮箱，不输入则使用 ${0} 的结果'.format(const.GIT_EMAIL_CMD))
     emails = []
     while True:
@@ -59,34 +65,18 @@ def get_user_info() -> Dict[str, Any]:
         encrypt = True
     info = {
         'name': name,
+        'year': int(year),
         'emails': list(set(emails)),
         'git_inputs': list(set(git_inputs)),
         'encrypt': encrypt,
     }
     return info
 
 
-def get_time_info() -> Dict[str, int]:
-    """ Get recent year's info. """
-    now = datetime.now()
-    recent_year = now.year
-    if now.month == 1:
-        recent_year -= 1
-    begin_ts = time.mktime(datetime(year=recent_year, month=1, day=1).timetuple())
-    end_ts = time.mktime(datetime(year=recent_year + 1, month=1, day=1).timetuple())
-    year_ends = {
-        'year': recent_year,
-        'begin': int(begin_ts),
-        'end': int(end_ts),
-    }
-    return year_ends
-
-
 def main():
     ctx = util.DotDict()
     ctx.run_dir = RUN_DIR
     ctx.update(get_user_info())
-    ctx.update(get_time_info())
     ctx.update(check_linguist(ctx))
     print('\nContext:')
     for key, val in ctx.items():

report.py

@@ -79,34 +79,36 @@ def write_stat(self):
         self.report.writerow(title)
         self.report.writerow('')
 
-        summary = self.repos.get_commit_summary()
         self.report.writerow(['Summary'])
+        summary = self.repos.get_commit_summary()
         names = ['projects', 'commits', 'merges', 'changes']
         for name in names[:3]:
             self.report.writerow([name, summary[name]])
         self.report.writerow([names[-1], summary['insert'] + summary['delete']])
         self.report.writerow('')
 
-        lang_stat = self.repos.get_language_stat()
-        self.report.writerow(['Coding stat by language'])
-        headers = ['language', 'commits', 'insertions', 'deletions', 'changes']
+        self.report.writerow(['Coding stat by repo'])
+        headers = ['name', 'language', 'commits', 'merges', 'insertions', 'deletions', 'changes']
         self.report.writerow(headers)
-        for lang, stat in lang_stat.items():
+        for repo in self.repos.repos:
+            repo_stat = repo.get_commit_summary()
             row = [
-                lang, stat['commits'], stat['insert'], stat['delete'],
-                stat['insert'] + stat['delete'],
+                repo.name, repo.language, repo_stat['commits'], repo_stat['merges'],
+                repo_stat['insert'], repo_stat['delete'], repo_stat['insert'] + repo_stat['delete'],
             ]
             self.report.writerow(row)
         self.report.writerow('')
 
-        self.report.writerow(['Coding stat by repo'])
-        headers = ['name', 'commits', 'merges', 'insertions', 'deletions', 'changes']
+        self.report.writerow(['Coding stat by language'])
+        lang_stat = self.repos.get_language_stat()
+        headers = ['language', 'commits', 'insertions', 'deletions', 'changes']
         self.report.writerow(headers)
-        for repo in self.repos.repos:
-            repo_stat = repo.get_commit_summary()
+        sorted_lang = sorted(lang_stat.keys(), key=lambda x: lang_stat[x]['weight'], reverse=True)
+        for lang in sorted_lang:
+            stat = lang_stat[lang]
             row = [
-                repo.name, repo_stat['commits'], repo_stat['merges'], repo_stat['insert'],
-                repo_stat['delete'], repo_stat['insert'] + repo_stat['delete'],
+                lang, stat['commits'], stat['insert'], stat['delete'],
+                stat['insert'] + stat['delete'],
             ]
             self.report.writerow(row)
         self.report.writerow('')
@@ -115,10 +117,13 @@ def write_stat(self):
         self.report.writerow(['Merge stat by name'])
         headers = ['name', 'merge', 'merged_by', 'merges']
         self.report.writerow(headers)
-        for name, stat in merge_stat.items():
+        sorted_names = sorted(merge_stat,
+                              key=lambda x: merge_stat[x]['merge'] + merge_stat[x]['merged_by'],
+                              reverse=True)
+        for name in sorted_names:
+            stat = merge_stat[name]
             row = [
-                name, stat.get('merge', 0), stat.get('merged_by', 0),
-                stat.get('merge', 0) + stat.get('merged_by', 0),
+                name, stat['merge'], stat['merged_by'], stat['merge'] + stat['merged_by']
             ]
             self.report.writerow(row)
         self.report.writerow('')
@@ -225,16 +230,16 @@ def draw_language_stat(self):
         weights = [lang_stat[key]['weight'] for key in labels]
         percents = util.get_percents(weights)
         weighted = [(labels[i], percents[i]) for i in range(len(labels))]
-        weighted.sort(key=lambda x: x[1], reverse=True)
         res = []
-        other_index = 0
-        for i, item in enumerate(weighted):
+        other_percent = 0
+        for item in weighted:
             if item[1] > 2:
                 res.append(item)
             else:
-                other_index = i
-        other_percent = sum(list(item[1] for item in weighted[other_index:])) + 0.01
-        res.append(('other', other_percent))
+                other_percent += item[1]
+        if other_percent > 0:
+            other_percent += 0.001
+            res.append(('other', other_percent))
         labels, weights = [item[0] for item in res], [item[1] for item in res]
         plt.pie(weights, labels=labels, autopct='%1.1f%%', startangle=90)
         plt.title('Programming languages by weight')

repository.py

@@ -72,13 +72,15 @@ def __init__(self, git_url_or_path: str, ctx: util.DotDict):
         self.name = util.encrypt_string(repo_name, ctx.encrypt)
         self.git_url = util.run(const.GIT_REMOTE_URL_CMD, check=False)
         self.ctx = ctx
+        self.language = ''
         self.linguist_enabled = False
         self.linguist_res = {}
-        self.analyze_by_linguist()
         self.commit_list = []
         self.commit_dict = {}
         self.user_commits = []
+        self.analyze_by_linguist()
         self.parse_git_commits()
+        self.get_repo_language()
         print('{0} loaded successfully!'.format(repo_name))
 
     def parse_git_commits(self):
@@ -91,7 +93,8 @@ def parse_git_commits(self):
             if line.strip() == 'master':
                 branch = 'master'
                 break
-        git_log_cmd = git_log_tmpl.format(branch=branch, begin=self.ctx.begin, end=self.ctx.end,
+        begin, end = util.get_year_ends(self.ctx.year)
+        git_log_cmd = git_log_tmpl.format(branch=branch, begin=begin, end=end,
                                           fmt=const.GIT_LOG_FORMAT)
         git_log = util.run(git_log_cmd)
         commit_logs = git_log.split(const.GIT_COMMIT_SEPARATOR)
@@ -116,8 +119,7 @@ def parse_git_log(self, commit_log: str) -> Any:
                         author=lines[2], email=lines[3], timestamp=int(lines[4]))
         commit.subject = lines[5]
         commit.num_stat = [line.strip() for line in lines[6:] if line.strip()]
-        if commit.email in self.ctx.emails:
-            self.parse_commit_stat(commit)
+        self.parse_commit_stat(commit)
         return commit
 
     def analyze_by_linguist(self):
@@ -136,6 +138,12 @@ def analyze_by_linguist(self):
                 self.linguist_res[file] = lang
         self.linguist_enabled = True
 
+    def get_repo_language(self):
+        stat = self.get_language_stat(only_user=False)
+        if not stat:
+            return
+        self.language = max(stat.keys(), key=lambda x: stat[x]['weight'])
+
     def get_commit_summary(self) -> util.DotDict:
         summary = {
             'commits': 0,
@@ -151,10 +159,12 @@ def get_commit_summary(self) -> util.DotDict:
             summary['delete'] += commit.code_del
         return util.DotDict(summary)
 
-    def get_language_stat(self) -> Dict[str, Any]:
+    def get_language_stat(self, only_user=True) -> Dict[str, Any]:
         """ Get each used language's commit stat. """
         res = {}
-        for commit in self.user_commits:
+        for commit in self.commit_list:
+            if only_user and commit.email not in self.ctx.emails:
+                continue
             for lang, stat in commit.lang_stat.items():
                 if lang not in res:
                     res[lang] = {
@@ -382,19 +392,24 @@ def get_merge_stat(self) -> Dict[str, Dict[str, Any]]:
                 # user merges others' commit
                 elif commit.email in self.ctx.emails:
                     if merged.email not in merges:
-                        merges[merged.email] = {}
+                        merges[merged.email] = {
+                            'merge': 1,
+                            'merged_by': 0,
+                        }
                         authors[merged.email] = {merged.author}
                     else:
-                        merges[merged.email]['merge'] = merges[merged.email].get('merge', 0) + 1
+                        merges[merged.email]['merge'] += 1
                         authors[merged.email].add(merged.author)
                 # user's commit merged by others
                 elif merged.email in self.ctx.emails:
                     if commit.email not in merges:
-                        merges[commit.email] = {}
+                        merges[commit.email] = {
+                            'merge': 0,
+                            'merged_by': 1
+                        }
                         authors[commit.email] = {commit.author}
                     else:
-                        merges[commit.email]['merged_by'] = merges[commit.email].get('merged_by',
-                                                                                     0) + 1
+                        merges[commit.email]['merged_by'] += 1
                         authors[commit.email].add(commit.author)
         result = {}
         for email, stat in merges.items():

tests.py

@@ -16,8 +16,6 @@ def setUp(self):
             'git_inputs': [run_dir],
             'encrypt': True,
             'year': 2018,
-            'begin': 1514736000,
-            'end': 1546272000,
             'linguist_enabled': False
         })
 

util.py

@@ -1,8 +1,9 @@
 # coding: utf8
 import os
 import subprocess
+import time
 from  datetime import datetime
-from typing import List, Any
+from typing import List, Any, Tuple
 
 import const
 
@@ -46,6 +47,12 @@ def timestamp_to_fixed_day(timestamp: int) -> datetime:
     return dt.replace(year=2018, month=1, day=1)
 
 
+def get_year_ends(year: int) -> Tuple[int, int]:
+    begin_ts = time.mktime(datetime(year=year, month=1, day=1).timetuple())
+    end_ts = time.mktime(datetime(year=year + 1, month=1, day=1).timetuple())
+    return int(begin_ts), int(end_ts)
+
+
 def is_ascii(s: str) -> bool:
     return all(ord(c) < 128 for c in s)