updated scripts

get_spec_file.py

@@ -26,16 +26,23 @@ def get_file(vcs_link, commit_hash, is_readme):
     repo = Repo.clone_from(vcs_link, full_temp_path)
     commit = repo.commit(commit_hash)
     #getting the name of the file from the root directory
+    target_filename = ""
     for filename in os.listdir(full_temp_path):
         if is_readme:
-            target_filename = "README.md"
+            #target_filename = "README.md"
             if "README" in filename:
                 target_filename = filename
         else:
-            target_filename = "CONTRIBUTING.md"
+            #target_filename = "CONTRIBUTING.md"
             if "CONTRIBUTING" in filename:
                 target_filename = filename
+    if target_filename == "":
+        return "NoFile"
+    try:
         targetfile = commit.tree / target_filename
+    except KeyError:
+        # why would a file not be in the commit tree? but would be in the directory?
+        return "KeyError"
     if is_readme:
         last_path = "readme"
     else:
@@ -45,6 +52,7 @@ def get_file(vcs_link, commit_hash, is_readme):
             file.write(f.read().decode('utf-8'))
         file.close()
     shutil.rmtree(full_temp_path, ignore_errors=True)
+    return "NoError"
 
 def for_all_files():
     #toggle this based on readme or contributing files
@@ -52,6 +60,7 @@ def for_all_files():
     csv_path = "kk_031624_pr_did.csv"
     index = -1
     with open(csv_path, 'r') as file:
+        with open('031824_spec_errors.csv', "w") as writing_file:
             #csv_reader = csv.DictReader(file)
             lines = [line for line in file]
             for row in tqdm(csv.reader(lines), total=len(lines)):
@@ -61,7 +70,9 @@ def for_all_files():
                 if row[0] == "":
                     continue
                 #print(row[0])
-            get_file(row[0], row[2], readme_is)
+                return_value = get_file(row[0], row[2], readme_is)
+                if return_value != "NoError":
+                    writing_file.write(row[0], row[2], readme_is, return_value)
                 #get_file('https://github.com/tqdm/tqdm', 'fbe7952cce11e8073378b063bdae7ab277a96eb8', True)
 
 if __name__ == "__main__":

pr_data_get.py

@@ -34,7 +34,7 @@ def file_get_pr(upstream_vcs_link, me_read):
     else:
         full_temp_path = temp_dir + upstream_vcs_link.split('/')[- 1] + ".git"
     print(upstream_vcs_link)
-    if upstream_vcs_link == "https://gitlab.com/ubports/core":
+    if upstream_vcs_link == "https://gitlab.com/ubports/core" or upstream_vcs_link == "https://gitlab.freedesktop.org/xorg/lib":
         shutil.rmtree(full_temp_path, ignore_errors=True)
         return {}
     repo = Git(uri=upstream_vcs_link, gitpath=full_temp_path)
@@ -145,11 +145,12 @@ def pr_count(start, end, commits, author_roster, commit_roster):
             return [by_week, by_week_merge, new_authors, new_committers, author_roster, commit_roster]
 
 def for_files():
+    #csv_path = "final_data/deb_contribfile_roster.csv"
     csv_path = "final_data/deb_readme_roster.csv"
     count = 0 
     with open(csv_path, 'r') as file:
         csv_reader = csv.DictReader(file)
-        with open('kk_test_031624_pr_did.csv', "w") as writing_file:
+        with open('kk_031624_pr_did.csv', "w") as writing_file:
             # this would also have to get switched fro the cont dataset
             keys = ['upstream_vcs_link', "first_readme", "readme_commit_hash", "before_allcom_read", "before_mrg_read", "after_allcom_read", "after_mrg_read", 'before_auth_new', 'after_commit_new', 'after_auth_new', 'before_commit_new']
             dict_writer = csv.DictWriter(writing_file, keys)
@@ -158,7 +159,10 @@ def for_files():
                 count += 1
                 print(row['upstream_vcs_link'])
                 # this would have to get switched to false for the cont dataset
+                try:
                     dict_row = file_get_pr(row['upstream_vcs_link'].strip(), True)
+                except:
+                    dict_row = {}
                 dict_writer.writerow(dict_row)