initial steps in ecpanded data collection

debian_queries.py

@@ -0,0 +1,68 @@
+import os
+import requests
+import wget
+from perceval.backends.core.git import Git
+from git import Repo
+import shutil
+
+
+def debian_vcs_query(debian_vcs_link):
+    upstream_repo_link = ""
+    project_name = debian_vcs_link.split("/")[-1]
+    output_directory = "/data/users/mgaughan/tmp1/" + project_name
+    print(output_directory)
+    #url = debian_vcs_link + '/-/blob/master/debian/upstream/metadata'
+    try:
+        Repo.clone_from(debian_vcs_link, output_directory)
+    except:
+        print("error cloning")
+        return upstream_repo_link
+    try:
+        upstream_metadata = open(output_directory + "/debian/upstream/metadata", "r").read().split("\n")
+        upstream_repo_array = upstream_metadata[3].split(":")[-2:]
+        upstream_repo_link = ":".join(upstream_repo_array)
+        print(upstream_repo_link)
+    except:
+        print("no file!")
+        upstream_repo_link = ""
+    #filename = wget.download(url, out=output_directory)
+    #print(type(filename))
+    shutil.rmtree(output_directory, ignore_errors=True)
+    return upstream_repo_link
+
+
+def debian_query(package_name):
+    headers = {'content-type': 'application/json', 'Accept-Charset': 'UTF-8'}
+    first_api_url = "https://sources.debian.org/api/src/" + package_name
+    try:
+        first_response = requests.get(url = first_api_url, headers=headers)
+        first_response_dict = first_response.json()
+    except:
+        print('error with the first debian request')
+        first_response_dict = {}
+        return ("")
+    if first_response_dict == {'error': 404}:
+        print('not found in debian system')
+        return ("")
+    #print(first_response_dict)
+    most_recent_package_version = first_response_dict['versions'][0]['version']
+    #print(first_response_dict['versions'][0]['version'])
+    second_api_url = "https://sources.debian.org/api/info/package/" + package_name + "/" + most_recent_package_version + "/"
+    try:
+        second_response = requests.get(url = second_api_url, headers=headers)
+        second_response_dict = second_response.json()
+    except:
+        print('error with the second debian request')
+        second_response_dict = {}
+        return ("")
+    if 'vcs_browser' not in second_response_dict['pkg_infos'].keys():
+        print('no vcs link')
+        return ("")
+    print(second_response_dict)
+    print(second_response_dict['pkg_infos']['vcs_browser'])
+    return second_response_dict['pkg_infos']['vcs_browser']
+
+
+if __name__ == "__main__":
+    #main()
+    debian_vcs_query("https://salsa.debian.org/debian/0xffff")

expanded_data_collection.py

@@ -2,10 +2,12 @@ import csv
 import requests
 import os
 import datetime as dt
+import wget
 
 import perceval_tasks as pt
 import github_api_req as gha
 import gh_gsql_req as ghs
+import debian_queries as dqs
 
 
 key = os.environ.get('KKEXKEY')
@@ -21,17 +23,23 @@ def main():
         successful_count = 0
         for row in spamreader:
             index += 1
-            if index > 10:
+            if index > 20:
                 break
             project_dict = {}
             project_dict["project_name"] = row[0]
             project_dict["underproduction_mean"] = row[16]
             project_dict["underproduction_low"] = row[17]
             project_dict["underproduction_high"] = row[18]
-            project_dict["vcs_link"] = debian_query(project_dict["project_name"])
-            if project_dict["vcs_link"] == "":
+            project_dict["debian_vcs_link"] = dqs.debian_query(project_dict["project_name"])
+            if project_dict["debian_vcs_link"] == "":
                 continue
-            perceval_data = pt.main(project_dict["vcs_link"], early_cutoff)
+            if "github" in project_dict["debian_vcs_link"]:
+                project_dict["upstream_vcs_link"] = clean_gh_vcs_link(project_dict["debian_vcs_link"])
+            else:
+                project_dict["upstream_vcs_link"] = dqs.debian_vcs_query(project_dict["debian_vcs_link"])
+            if project_dict["upstream_vcs_link"] == "":
+                continue
+            perceval_data = pt.main(project_dict["upstream_vcs_link"], early_cutoff)
             if perceval_data == {}:
                 continue
             project_dict['age_of_project'], project_dict['contributors'], project_dict['collaborators'] = perceval_data['age_of_project'], perceval_data['contributors'], perceval_data['collaborators']
@@ -46,37 +54,12 @@ def main():
         dict_writer.writerows(array_of_projects)
             
 
-def debian_query(package_name):
-    headers = {'content-type': 'application/json', 'Accept-Charset': 'UTF-8'}
-    first_api_url = "https://sources.debian.org/api/src/" + package_name
-    try:
-        first_response = requests.get(url = first_api_url, headers=headers)
-        first_response_dict = first_response.json()
-    except:
-        print('error with the first debian request')
-        first_response_dict = {}
-        return ("")
-    if first_response_dict == {'error': 404}:
-        print('not found in debian system')
-        return ("")
-    #print(first_response_dict)
-    most_recent_package_version = first_response_dict['versions'][0]['version']
-    #print(first_response_dict['versions'][0]['version'])
-    second_api_url = "https://sources.debian.org/api/info/package/" + package_name + "/" + most_recent_package_version + "/"
-    try:
-        second_response = requests.get(url = second_api_url, headers=headers)
-        second_response_dict = second_response.json()
-    except:
-        print('error with the second debian request')
-        second_response_dict = {}
-        return ("")
-    if 'vcs_browser' not in second_response_dict['pkg_infos'].keys():
-        print('no vcs link')
-        return ("")
-    print(second_response_dict)
-    print(second_response_dict['pkg_infos']['vcs_browser'])
-    return second_response_dict['pkg_infos']['vcs_browser']
+def clean_gh_vcs_link(debian_vcs_link):
+    url_array = debian_vcs_link.split("/")
+    new_url = "/".join(url_array[:5])
+    return new_url
 
 if __name__ == "__main__":
     main()
-    #debian_query("zurl")
+    #clean_gh_vcs_link("https://github.com/kilobyte/3270font/tree/debian/")
+    #debian_vcs_query("https://salsa.debian.org/debian/0xffff/")

perceval_tasks.py

@@ -13,6 +13,7 @@ def main(vcs_path, begin_date):
         perceval_info['age_of_project'] = get_repo_age(perceval_info['list_of_commits'])
         perceval_info['contributors'], perceval_info['collaborators'] = get_all_actors(perceval_info['list_of_commits'])
         del perceval_info['list_of_commits']
+        print(perceval_info)
         return perceval_info
     else:
         print('error, no commits found?')
@@ -22,6 +23,7 @@ def main(vcs_path, begin_date):
 
 # this is the primary function for getting the list of commits from perceval
 def get_perceval_log(vcs_path, begin_date):
+    vcs_path = vcs_path.strip()
     print(vcs_path)
     try:
         repo_dir = '/data/users/mgaughan/tmp/' + str(vcs_path.split('/')[-1])
@@ -63,3 +65,5 @@ def get_all_actors(all_commits):
             authors.remove(committer)
     return len(authors), len(committers)
 
+if __name__ == "__main__":
+    main(" https://github.com/pali/0xFFFF".strip(),dt.datetime(2008,2, 8))