misc data collection updates

src/helper_scripts/frameworks_for_collection.csv

@@ -0,0 +1,24 @@
+url
+https://github.com/MW-Peachy/Peachy
+https://github.com/addwiki/mediawiki-api-base
+https://github.com/addwiki/mediawiki-api
+https://github.com/hamstar/Wikimate
+https://github.com/alexz-enwp/wikitools
+https://github.com/Riamse/ceterach
+https://github.com/eldur/jwbf
+https://github.com/CXuesong/WikiClientLibrary
+https://github.com/mer-c/wiki-java
+https://github.com/greencardamom/BotWikiAwk
+https://github.com/mwclient/mwclient
+https://github.com/earwig/mwparserfromhell
+https://github.com/barrust/mediawiki
+https://github.com/ankostis/MatWiki
+https://github.com/WPCleaner/wpcleaner
+https://github.com/fastily/jwiki
+https://github.com/wikimedia-gadgets/mock-mediawiki
+https://github.com/kanasimi/wikiapi
+https://github.com/FTB-Gamepedia/MediaWiki-Butt-Ruby
+https://github.com/wikimedia/mediawiki-ruby-api
+https://github.com/kenpratt/wikipedia-client
+https://github.com/jpatokal/mediawiki-gateway
+https://github.com/AccelerationNet/cl-mediawiki

src/helper_scripts/repo_collection.py

@@ -8,32 +8,28 @@ file_location = "/data/users/mgaughan/mw-repo-lifecycles/commit_data/"
 repo_location = "/data/users/mgaughan/mw-repo-lifecycles/repo_artifacts/"
 
 cst = datetime.timezone(datetime.timedelta(hours=-6))
-'''
+
 repos = {
-        "core" : {
+        "parsoid" : {
-            "url": "https://gerrit.wikimedia.org/r/mediawiki/core",
+            "url": "https://gerrit.wikimedia.org/r/mediawiki/services/parsoid",
-            "from_date": datetime.datetime(2012, 1, 1, 00, 00, 00, tzinfo=cst),
+            "from_date": datetime.datetime(2010, 1, 1, 00, 00, 00, tzinfo=cst),
-            "to_date": datetime.datetime(2014, 12, 31, 00, 00, 00, tzinfo=cst)
+            "to_date": datetime.datetime(2024, 12, 31, 00, 00, 00, tzinfo=cst)
             },
-        "visualeditor": {
+}
-            "url": "https://gerrit.wikimedia.org/r/VisualEditor/VisualEditor",
-            "from_date": datetime.datetime(2012, 1, 1, 00, 00, 00, tzinfo=cst),
-            "to_date": datetime.datetime(2013, 5, 1, 00, 00, 00, tzinfo=cst)
-            }
-        }
 '''
-repos = {
+        "parsoid" : {
-        "visualeditor": {
             "url": "https://gerrit.wikimedia.org/r/pywikibot/core",
             "from_date": datetime.datetime(2010, 1, 1, 00, 00, 00, tzinfo=cst),
             "to_date": datetime.datetime(2024, 12, 31, 00, 00, 00, tzinfo=cst)
             }
-}
+        }
+'''
+
 for repo in repos.keys():
     print(repos[repo])
     print(repos[repo]['url'])
     print(repos[repo]["from_date"])
     print(repos[repo]["to_date"])
     repo_info = repos[repo]
-    repo_lifecycle(repo_info['url'], repo_location + repo + "/tmp", repo_info["from_date"], repo_info["to_date"], to_save=True, csv_loc_prefix=file_location)
+    repo_lifecycle(repo_info['url'], repo_location + repo + "/tmp", repo_info["from_date"], repo_info["to_date"], to_save=False, csv_loc_prefix=file_location)
 

src/lib/gerrit_get.py

@@ -0,0 +1,58 @@
+#from phabricator import Phabricator
+import os, sys
+import json
+import numpy as np
+import pandas as pd
+import requests
+import re
+import datetime
+import time
+from urllib.parse import quote_plus
+
+from requests.auth import HTTPDigestAuth
+from pygerrit2 import GerritRestAPI, HTTPBasicAuth
+
+# GET /changes/?q=status:abandoned&q=before:{date}&q={TERM}
+# GET https://gerrit.wikimedia.org/r/changes/?q=status:abandoned+visualeditor
+
+def query_changes(
+    query_terms, 
+    limit = 100, 
+    api_url_base = 'https://gerrit.wikimedia.org/',
+    sleep = 10
+):
+    
+    time.sleep(sleep)
+    to_query = 1
+    after = None
+
+    data = []
+
+    while to_query == 1:
+        time.sleep(sleep)
+        joined_terms = "+".join(query_terms)
+        params = {
+            'q'  : joined_terms,
+        }
+        
+        api_url = f"{api_url_base}r/changes/?q={joined_terms}"
+        #add no-limit and API key somewhere
+
+        #auth = HTTPBasicAuth("ggonnemm", "1V6txZh5X+N3JpDMm5zqZM2M7ewA5D09g4ABOZAl5Q")
+        response = requests.get(api_url, headers={'Content-Type': 'application/json'})
+        
+        result = json.load(response.text[5:])
+        #print(response.text)
+        ## the data
+        data_tmp = result
+        data += data_tmp
+        print(data[-1])
+        ## check if there are more results to query
+        break
+    return data
+
+
+if __name__ == "__main__":
+    query_strings = ['before:2016-12-31', 'visualeditor']
+    results = query_changes(query_strings)
+    print(results)

src/lib/gerrit_query.sh

@@ -0,0 +1,8 @@
+ #!/bin/bash
+ 
+API_URL_BASE="https://gerrit.wikimedia.org/r/changes"
+QUERY_STRING="before:2013-03-29+visualeditor"
+
+API_URL="${API_URL_BASE}/?q=${QUERY_STRING}"
+
+curl -X GET "$API_URL" -H "Accept: application/json" -o response.json

src/lib/phab_get.py

@@ -36,12 +36,17 @@ def query_task_tag(
 
     data = []
 
+    # for bot frameworks
+    # listed on the help page as of 2-12-2024
+    # utilizing git as their VCS
+
     while to_query == 1:
         time.sleep(sleep)
         params = {
             'api.token'  : api_token,
             'constraints[query]':[tag_term], ## term that task is searched for with
-            #'constraints[projects]':["VisualEditor"], ## term that task is tagged with
+            # seemed to be artificially limiting the data that was returned, unrealistically low count values
+            #'constraints[projects]':[tag_term], ## term that task is tagged with
             'constraints[createdStart]':ts1, ## timestamp task creation (min)
             'constraints[createdEnd]':ts2, ## timestamp task creation (max)
             'limit':limit,
@@ -127,10 +132,11 @@ if __name__ == "__main__":
     token = "api-wurg254ciq5uvfxlr4rszn5ynpy4"
     api_base = 'https://phabricator.wikimedia.org/api/'
 
-    p_ts1 = int(datetime.datetime.timestamp(datetime.datetime(2024, 6, 10, 0, 0, 0)))
+    p_ts1 = int(datetime.datetime.timestamp(datetime.datetime(2010, 1, 1, 0, 0, 0)))
-    p_ts2 = int(datetime.datetime.timestamp(datetime.datetime(2024, 10, 10, 0, 0, 0)))
+    p_ts2 = int(datetime.datetime.timestamp(datetime.datetime(2024, 12, 31, 0, 0, 0)))
+    for tag in tags:
 
-    p_data = query_task_tag("Parsoid", ts1=p_ts1, ts2=p_ts2)
+        p_data = query_task_tag(tag, ts1=p_ts1, ts2=p_ts2)
 
         for entry in p_data:
             task_id = entry['phid']
@@ -139,7 +145,7 @@ if __name__ == "__main__":
             comments = {}
             for item in transactions:
                 comments[item['id']] = item['comments']
-        entry['task_comments'] = comments
+            entry['task_commeSigbjorn Finnents'] = comments
-    DATA_PREFIX = "/data/users/mgaughan/mw-repo-lifecycles/phab_data/"
+        DATA_PREFIX = "/data/users/mgaughan/mw-repo-lifecycles/phab_data/" + tag
-    with open(DATA_PREFIX + "parsoid/" + "2024_6_10_to_2024_10_10.json", "w") as outfile1:
+        with open(DATA_PREFIX + "/" + "2010_1_1_to_2024_12_31.json", "w") as outfile1:
             json.dump(p_data, outfile1)

src/lib/spark-warehouse/activity_isolation.py

@@ -24,7 +24,7 @@ def toArray(str):
 
 if __name__ == "__main__":
 
-    mediawiki_history_path = "/data_ext/users/nws8519/mw-repo-lifecycles/wiki_activity_data/single_activity_files/"
+    mediawiki_history_path = "/data_ext/users/nws8519/mw-repo-lifecycles/wiki_activity_data/yearly_activity_files/"
 
     # Note: string unescaping and array conversion is done later
     mediawiki_history_schema = StructType([
@@ -200,4 +200,4 @@ if __name__ == "__main__":
         sort(F.desc("day")). \
         show(10, False)
 
-    activity_count_df.write.format("csv").save("012825_nonbot_single.csv")
+    activity_count_df.write.format("csv").save("020925_nonbot_yearly.csv")