Mẫu dữ liệu quản lý vấn đề

import xml.etree.ElementTree as ET
import pandas as pd
from datetime import datetime
import io

# CONFIGURATION
# Path to the exported Jira XML file
INPUT_FILE = 'jira_export.xml'
OUTPUT_FILE = 'process_mining_event_log.csv'
SOURCE_SYSTEM = 'Jira Service Management'

# JQL USED FOR EXPORT (Reference Only)
# project = "Problem Management" AND created >= "2023-01-01" ORDER BY created DESC

def parse_jira_xml(xml_file):
    tree = ET.parse(xml_file)
    channel = tree.getroot().find('channel')
    
    events = []
    
    # Namespaces are usually not required for standard Jira RSS XML, 
    # but we handle standard tags directly.
    
    for item in channel.findall('item'):
        # 1. Extract Header Attributes
        problem_key = item.find('key').text
        summary = item.find('summary').text if item.find('summary') is not None else ''
        created_str = item.find('created').text
        resolution_date_node = item.find('resolutiondate')
        due_date_node = item.find('duedate')
        
        # Parse base timestamps
        # Jira XML format example: Wed, 4 Oct 2023 14:30:00 +0000
        fmt = '%a, %d %b %Y %H:%M:%S %z'
        created_dt = datetime.strptime(created_str, fmt)
        
        # --- ACTIVITY 1: Problem Record Created ---
        events.append({
            'ProblemKey': problem_key,
            'ActivityName': 'Problem Record Created',
            'EventTimestamp': created_dt,
            'UserKey': item.find('reporter').get('username') if item.find('reporter') is not None else None,
            'SourceSystem': SOURCE_SYSTEM,
            'ProblemSummary': summary,
            'Priority': item.find('priority').text if item.find('priority') is not None else None
        })

        # --- ACTIVITY: Incident Linked / Change Request Linked ---
        # Check <issuelinks> tag
        issue_links = item.find('issuelinks')
        if issue_links is not None:
            for linktype in issue_links.findall('issuelinktype'):
                link_name = linktype.find('name').text # e.g., 'Relates', 'Blocks', 'Caused'
                for outward in linktype.findall('outwardlinks'):
                    for issue_link in outward.findall('issuelink'):
                        linked_key = issue_link.find('issuekey').text
                        # Heuristic: Identify link type based on key pattern or link description
                        activity = None
                        if 'INC' in linked_key: # Customize for your Incident Prefix
                            activity = 'Incident Linked to Problem'
                        elif 'CHG' in linked_key: # Customize for your Change Prefix
                            activity = 'Change Request Linked'
                        
                        if activity:
                             events.append({
                                'ProblemKey': problem_key,
                                'ActivityName': activity,
                                'EventTimestamp': created_dt, # Link timestamp often not in XML, defaulting to created or requires API. fallback.
                                'UserKey': 'System',
                                'SourceSystem': SOURCE_SYSTEM,
                                'ProblemSummary': f'Linked to {linked_key}',
                                'Priority': None
                            })

        # --- PROCESS HISTORY (Status, Priority, Assignment, etc.) ---
        # Note: History is not always in standard RSS. 
        # If utilizing the 'SearchRequestView' XML, history is included inside <customfields> or specific expansion.
        # However, typically standard RSS does NOT include full history.
        # Assuming this XML export includes the Activity Stream or we rely on Custom Field dates for specific milestones.
        
        # LOGIC FOR INFERRING ACTIVITIES FROM CURRENT STATE & DATES (If History missing)
        # Or parsing <subtasks> or <comments> as proxies.
        
        # --- ACTIVITY: SLA Breached ---
        if resolution_date_node is not None and due_date_node is not None:
            res_date = datetime.strptime(resolution_date_node.text, fmt)
            due_date = datetime.strptime(due_date_node.text, fmt)
            
            if res_date > due_date:
                events.append({
                    'ProblemKey': problem_key,
                    'ActivityName': 'SLA Breached',
                    'EventTimestamp': res_date,
                    'UserKey': 'System',
                    'SourceSystem': SOURCE_SYSTEM,
                    'ProblemSummary': summary,
                    'Priority': None
                })

        # --- ACTIVITY: Problem Record Closed / Resolved ---
        if resolution_date_node is not None:
             events.append({
                'ProblemKey': problem_key,
                'ActivityName': 'Problem Record Closed',
                'EventTimestamp': datetime.strptime(resolution_date_node.text, fmt),
                'UserKey': item.find('assignee').get('username') if item.find('assignee') is not None else None,
                'SourceSystem': SOURCE_SYSTEM,
                'ProblemSummary': summary,
                'Priority': None
            })

        # --- MOCKING HISTORY PARSING IF XML HAS HISTORY PLUGIN EXTENSION ---
        # (Standard Jira RSS export is weak on history. 
        # This block assumes the existence of a custom history export or parsing comment logs for keywords).
        
        comments = item.find('comments')
        if comments:
            for comment in comments.findall('comment'):
                body = comment.text.lower()
                c_date = datetime.strptime(comment.get('created'), fmt)
                author = comment.get('author')
                
                act_name = None
                if 'root cause' in body:
                    act_name = 'Root Cause Identified'
                elif 'workaround' in body:
                    act_name = 'Workaround Updated'
                elif 'investigation' in body:
                    act_name = 'Investigation Started'
                
                if act_name:
                    events.append({
                        'ProblemKey': problem_key,
                        'ActivityName': act_name,
                        'EventTimestamp': c_date,
                        'UserKey': author,
                        'SourceSystem': SOURCE_SYSTEM,
                        'ProblemSummary': summary,
                        'Priority': None
                    })

    return pd.DataFrame(events)

# EXECUTION
try:
    df = parse_jira_xml(INPUT_FILE)
    
    # Add global extraction timestamp
    df['LastDataUpdate'] = datetime.now()
    
    # Formatting
    if not df.empty:
        df['EventTimestamp'] = pd.to_datetime(df['EventTimestamp']).dt.strftime('%Y-%m-%d %H:%M:%S')
        
        # Output
        df.to_csv(OUTPUT_FILE, index=False)
        print(f"Success: Extracted {len(df)} events to {OUTPUT_FILE}")
    else:
        print("No events found. Check XML format or JQL.")
        
except Exception as e:
    print(f"Error processing file: {str(e)}")

import requests
import pandas as pd
import json
from datetime import datetime
import base64

class=class="hl-string">"hl-comment"># ==========================================
class=class="hl-string">"hl-comment"># CONFIGURATION
class=class="hl-string">"hl-comment"># ==========================================
CONFIG = {
    class="hl-string">"JIRA_URL": class="hl-string">"https://your-domain.atlassian.net",
    class="hl-string">"USER_EMAIL": class="hl-string">"your-email@example.com",
    class="hl-string">"API_TOKEN": class="hl-string">"your-api-token-or-pat",
    class="hl-string">"PROJECT_KEY": class="hl-string">"PM",  class=class="hl-string">"hl-comment"># Problem Management Project Key
    class="hl-string">"LOOKBACK_DAYS": class="hl-string">"180"
}

class=class="hl-string">"hl-comment"># Mappings: Key = Jira Field Name/ID, Value = Activity Name context
class=class="hl-string">"hl-comment"># You must replace class="hl-string">'customfield_XXXXX' with your actual field IDs found via API
FIELD_MAPPING = {
    class="hl-string">"priority": class="hl-string">"Problem Priority Changed",
    class="hl-string">"assignee": class="hl-string">"Assigned to Support Group", class=class="hl-string">"hl-comment"># Fallback if support group field unused
    class="hl-string">"customfield_10010": class="hl-string">"Assigned to Support Group", class=class="hl-string">"hl-comment"># Replace with Support Group Field ID
    class="hl-string">"customfield_10011": class="hl-string">"Workaround Updated",      class=class="hl-string">"hl-comment"># Replace with Workaround Field ID
    class="hl-string">"customfield_10012": class="hl-string">"Root Cause Identified",   class=class="hl-string">"hl-comment"># Replace with Root Cause Field ID
    class="hl-string">"Link": class="hl-string">"Link Event",                           class=class="hl-string">"hl-comment"># Internal name for issue links
    class="hl-string">"issuelinks": class="hl-string">"Link Event"
}

class=class="hl-string">"hl-comment"># Map Jira Workflow Statuses to Standard Activities
class=class="hl-string">"hl-comment"># Update the keys (left side) to match your actual Jira Status names
STATUS_MAPPING = {
    class="hl-string">"Under Investigation": class="hl-string">"Investigation Started",
    class="hl-string">"In Progress": class="hl-string">"Investigation Started",
    class="hl-string">"Root Cause Identified": class="hl-string">"Root Cause Identified",
    class="hl-string">"Implementing": class="hl-string">"Permanent Fix Applied",
    class="hl-string">"Fixed": class="hl-string">"Permanent Fix Applied",
    class="hl-string">"Resolved": class="hl-string">"Resolution Verified",
    class="hl-string">"Verified": class="hl-string">"Resolution Verified",
    class="hl-string">"In Review": class="hl-string">"Post Implementation Review",
    class="hl-string">"Closed": class="hl-string">"Problem Record Closed",
    class="hl-string">"Breached": class="hl-string">"SLA Breached" class=class="hl-string">"hl-comment"># Only if you have a specific status for this
}

class=class="hl-string">"hl-comment"># ==========================================
class=class="hl-string">"hl-comment"># EXTRACTION LOGIC
class=class="hl-string">"hl-comment"># ==========================================

def get_auth_header():
    class=class="hl-string">"hl-comment"># For Cloud (Basic Auth)
    creds = fclass="hl-string">"{CONFIG['USER_EMAIL']}:{CONFIG['API_TOKEN']}"
    encoded = base64.b64encode(creds.encode()).decode()
    return {class="hl-string">"Authorization": fclass="hl-string">"Basic {encoded}", class="hl-string">"Content-Type": class="hl-string">"application/json"}
    class=class="hl-string">"hl-comment"># For Data Center (PAT), uncomment below:
    class=class="hl-string">"hl-comment"># return {class="hl-string">"Authorization": fclass="hl-string">"Bearer {CONFIG['API_TOKEN']}", class="hl-string">"Content-Type": class="hl-string">"application/json"}

def fetch_issues():
    url = fclass="hl-string">"{CONFIG['JIRA_URL']}/rest/api/2/search"
    jql = fclass="hl-string">"project = {CONFIG['PROJECT_KEY']} AND issuetype = Problem AND created >= -{CONFIG['LOOKBACK_DAYS']}d"
    
    all_issues = []
    start_at = 0
    max_results = 50
    
    while True:
        params = {
            class="hl-string">"jql": jql,
            class="hl-string">"startAt": start_at,
            class="hl-string">"maxResults": max_results,
            class="hl-string">"expand": class="hl-string">"changelog"
        }
        
        try:
            response = requests.get(url, headers=get_auth_header(), params=params)
            response.raise_for_status()
            data = response.json()
        except Exception as e:
            print(fclass="hl-string">"API Error: {e}")
            break
            
        issues = data.get(class="hl-string">"issues", [])
        if not issues:
            break
            
        all_issues.extend(issues)
        start_at += max_results
        print(fclass="hl-string">"Fetched {len(all_issues)} issues so far...")
        
    return all_issues

def parse_activity(issue):
    events = []
    
    class=class="hl-string">"hl-comment"># Basic Issue Data
    key = issue[class="hl-string">"key"]
    fields = issue[class="hl-string">"fields"]
    summary = fields.get(class="hl-string">"summary", class="hl-string">"")
    class=class="hl-string">"hl-comment"># Current values (for attributes)
    current_priority = fields.get(class="hl-string">"priority", {}).get(class="hl-string">"name", class="hl-string">"")
    current_assignee = fields.get(class="hl-string">"assignee", {}).get(class="hl-string">"displayName") if fields.get(class="hl-string">"assignee") else class="hl-string">"Unassigned"
    
    class=class="hl-string">"hl-comment"># 1. Problem Record Created
    created_time = fields[class="hl-string">"created"]
    events.append({
        class="hl-string">"ProblemKey": key,
        class="hl-string">"ActivityName": class="hl-string">"Problem Record Created",
        class="hl-string">"EventTimestamp": created_time,
        class="hl-string">"UserKey": fields.get(class="hl-string">"creator", {}).get(class="hl-string">"displayName", class="hl-string">"System"),
        class="hl-string">"SupportGroup": current_assignee,
        class="hl-string">"Priority": current_priority,
        class="hl-string">"ProblemSummary": summary,
        class="hl-string">"SourceSystem": class="hl-string">"Jira Service Management"
    })

    class=class="hl-string">"hl-comment"># Process Changelog
    changelog = issue.get(class="hl-string">"changelog", {}).get(class="hl-string">"histories", [])
    
    for history in changelog:
        timestamp = history[class="hl-string">"created"]
        user = history.get(class="hl-string">"author", {}).get(class="hl-string">"displayName", class="hl-string">"Unknown")
        
        for item in history[class="hl-string">"items"]:
            field = item.get(class="hl-string">"field")
            field_id = item.get(class="hl-string">"fieldId", class="hl-string">"")
            from_str = item.get(class="hl-string">"fromString", class="hl-string">"")
            to_str = item.get(class="hl-string">"toString", class="hl-string">"")
            
            class=class="hl-string">"hl-comment"># 2, 5, 7, 9, 10, 11, 13, 14. Status Changes
            if field == class="hl-string">"status":
                activity = STATUS_MAPPING.get(to_str)
                
                class=class="hl-string">"hl-comment"># Handle Reopened logic explicitly
                if to_str in [class="hl-string">"Open", class="hl-string">"Reopened"] and from_str in [class="hl-string">"Resolved", class="hl-string">"Closed"]:
                    activity = class="hl-string">"Problem Reopened"
                
                if activity:
                    events.append({
                        class="hl-string">"ProblemKey": key,
                        class="hl-string">"ActivityName": activity,
                        class="hl-string">"EventTimestamp": timestamp,
                        class="hl-string">"UserKey": user,
                        class="hl-string">"SupportGroup": current_assignee,
                        class="hl-string">"Priority": current_priority,
                        class="hl-string">"ProblemSummary": summary,
                        class="hl-string">"SourceSystem": class="hl-string">"Jira Service Management"
                    })

            class=class="hl-string">"hl-comment"># 2. Priority Changed
            elif field == class="hl-string">"priority":
                events.append({
                    class="hl-string">"ProblemKey": key,
                    class="hl-string">"ActivityName": class="hl-string">"Problem Priority Changed",
                    class="hl-string">"EventTimestamp": timestamp,
                    class="hl-string">"UserKey": user,
                    class="hl-string">"SupportGroup": current_assignee,
                    class="hl-string">"Priority": to_str,
                    class="hl-string">"ProblemSummary": summary,
                    class="hl-string">"SourceSystem": class="hl-string">"Jira Service Management"
                })

            class=class="hl-string">"hl-comment"># 3. Assigned to Support Group (via custom field or Assignee)
            elif field_id == class="hl-string">"customfield_10010" or field == class="hl-string">"assignee":
                events.append({
                    class="hl-string">"ProblemKey": key,
                    class="hl-string">"ActivityName": class="hl-string">"Assigned to Support Group",
                    class="hl-string">"EventTimestamp": timestamp,
                    class="hl-string">"UserKey": user,
                    class="hl-string">"SupportGroup": to_str,
                    class="hl-string">"Priority": current_priority,
                    class="hl-string">"ProblemSummary": summary,
                    class="hl-string">"SourceSystem": class="hl-string">"Jira Service Management"
                })

            class=class="hl-string">"hl-comment"># 6. Workaround Updated
            elif field_id == class="hl-string">"customfield_10011":
                if to_str: class=class="hl-string">"hl-comment"># Only if value is added/changed to something
                    events.append({
                        class="hl-string">"ProblemKey": key,
                        class="hl-string">"ActivityName": class="hl-string">"Workaround Updated",
                        class="hl-string">"EventTimestamp": timestamp,
                        class="hl-string">"UserKey": user,
                        class="hl-string">"SupportGroup": current_assignee,
                        class="hl-string">"Priority": current_priority,
                        class="hl-string">"ProblemSummary": summary,
                        class="hl-string">"SourceSystem": class="hl-string">"Jira Service Management"
                    })

            class=class="hl-string">"hl-comment"># 7. Root Cause Identified (via Field Population)
            elif field_id == class="hl-string">"customfield_10012":
                if to_str:
                    events.append({
                        class="hl-string">"ProblemKey": key,
                        class="hl-string">"ActivityName": class="hl-string">"Root Cause Identified",
                        class="hl-string">"EventTimestamp": timestamp,
                        class="hl-string">"UserKey": user,
                        class="hl-string">"SupportGroup": current_assignee,
                        class="hl-string">"Priority": current_priority,
                        class="hl-string">"ProblemSummary": summary,
                        class="hl-string">"SourceSystem": class="hl-string">"Jira Service Management"
                    })

            class=class="hl-string">"hl-comment"># 4 & 8. Links (Incident Linked / Change Request Linked)
            elif field == class="hl-string">"Link":
                class=class="hl-string">"hl-comment"># This requires heuristic checking of the linked key or link type
                class=class="hl-string">"hl-comment"># Assuming standard keys: INC-xxxx for Incident, CHG-xxxx or CR-xxxx for Change
                if class="hl-string">"INC-" in to_str or class="hl-string">"Incident" in to_str:
                     events.append({
                        class="hl-string">"ProblemKey": key,
                        class="hl-string">"ActivityName": class="hl-string">"Incident Linked",
                        class="hl-string">"EventTimestamp": timestamp,
                        class="hl-string">"UserKey": user,
                        class="hl-string">"SupportGroup": current_assignee,
                        class="hl-string">"Priority": current_priority,
                        class="hl-string">"ProblemSummary": summary,
                        class="hl-string">"SourceSystem": class="hl-string">"Jira Service Management"
                    })
                elif class="hl-string">"CHG-" in to_str or class="hl-string">"CR-" in to_str or class="hl-string">"Change" in to_str:
                     events.append({
                        class="hl-string">"ProblemKey": key,
                        class="hl-string">"ActivityName": class="hl-string">"Change Request Linked",
                        class="hl-string">"EventTimestamp": timestamp,
                        class="hl-string">"UserKey": user,
                        class="hl-string">"SupportGroup": current_assignee,
                        class="hl-string">"Priority": current_priority,
                        class="hl-string">"ProblemSummary": summary,
                        class="hl-string">"SourceSystem": class="hl-string">"Jira Service Management"
                    })
            
            class=class="hl-string">"hl-comment"># 12. SLA Breached (Detection Logic)
            class=class="hl-string">"hl-comment"># Detection depends heavily on plugins. Common method: check for SLA status field change
            elif class="hl-string">"SLA" in field or class="hl-string">"Time to Resolution" in field:
                if class="hl-string">"breach" in to_str.lower() or class="hl-string">"fail" in to_str.lower():
                    events.append({
                        class="hl-string">"ProblemKey": key,
                        class="hl-string">"ActivityName": class="hl-string">"SLA Breached",
                        class="hl-string">"EventTimestamp": timestamp,
                        class="hl-string">"UserKey": user,
                        class="hl-string">"SupportGroup": current_assignee,
                        class="hl-string">"Priority": current_priority,
                        class="hl-string">"ProblemSummary": summary,
                        class="hl-string">"SourceSystem": class="hl-string">"Jira Service Management"
                    })

    return events

class=class="hl-string">"hl-comment"># ==========================================
class=class="hl-string">"hl-comment"># MAIN EXECUTION
class=class="hl-string">"hl-comment"># ==========================================
if __name__ == class="hl-string">"__main__":
    print(class="hl-string">"Starting extraction...")
    issues = fetch_issues()
    
    all_events = []
    for issue in issues:
        all_events.extend(parse_activity(issue))
        
    df = pd.DataFrame(all_events)
    
    class=class="hl-string">"hl-comment"># Add LastDataUpdate
    df[class="hl-string">"LastDataUpdate"] = datetime.now().isoformat()
    
    class=class="hl-string">"hl-comment"># Export
    filename = class="hl-string">"jira_problem_event_log.csv"
    df.to_csv(filename, index=False)
    print(fclass="hl-string">"Extraction complete. Saved {len(df)} rows to {filename}")