swarm_reviews.py #1

#!/usr/bin/env python3
# -*- coding: utf-8 -*-

# ==============================================================================
# Copyright and license info is available in the LICENSE file included with
# the Server Deployment Package (SDP), and also available online:
# https://swarm.workshop.perforce.com/projects/perforce-software-sdp/view/main/LICENSE
# ------------------------------------------------------------------------------

"""
NAME:
    SwarmReviews.py

DESCRIPTION:
    This script prints Swarm reviews as a csv file.

    Change id, Review id, Review state, Submitter ID, Submitter Name, Date submitted,
    Date Review Created, Date Review Last Updated,
    No of Up Votes, No of Down Votes, Test status, Num Comments, Num Open Tasks, Num Closed Tasks,
    Job ID, Jira Ticket ID (via Job), Jira Ticket Type, Jira Ticket State,
    Review Type, Scope of Review, Supporting Docs, Deadline for Review, Actual Review Effort,
    Jira Ticket Summary (via Job), Review Description

    You can run like this:

        p4 login -ap
        [note the output ticket and use with -t flag in next command]
        for /f "usebackq" %p in (`p4 dirs //depot/*`) do python swarm_reviews.py -t 1234567ABDBD %p/... >> rep.csv
        
    Alternatively may be easier:
        p4 login -ap
        [Note ticket as above]
        p4 dirs //streams_depot/* > dirs.txt
        [Optionally edit the list of paths in dirs.txt]
        python swarm_reviews.py -t 1234567ABDBD -i dirs.txt -e /... > rep.csv

"""

# Python 2.7/3.3 compatibility.
from __future__ import print_function

import sys
import requests
from requests.adapters import HTTPAdapter
from requests.packages.urllib3.util.retry import Retry
import os
import textwrap
import argparse
import logging
import P4
import time
import json
import re
from collections import defaultdict, OrderedDict
from six.moves import range
import six

script_name = os.path.basename(os.path.splitext(__file__)[0])

# Avoid utf encoding issues when writing to stdout
sys.stdout = os.fdopen(sys.stdout.buffer.fileno(), 'w', encoding='utf8')

LOGDIR = os.getenv('LOGS', '/p4/1/logs')

DEFAULT_LOG_FILE = "log-%s.log" % script_name
if os.path.exists(LOGDIR):
    DEFAULT_LOG_FILE = os.path.join(LOGDIR, "%s.log" % script_name)
DEFAULT_VERBOSITY = 'DEBUG'
LOGGER_NAME = 'P4Triggers'

def formatdate(epoch):
    if isinstance(epoch, six.string_types):
        epoch = int(epoch)
    t = time.localtime(epoch)
    return time.strftime("%Y-%m-%d %H:%M:%S", t)

def csv_escape(field):
    result = str(field).replace("\n", " | ")
    if "," in result:
        return '"%s"' % result
    return result

class Review(object):
    "Wrapper around Swarm JSON"

    def __init__(self, json):
        if 'review' in json:
            self.json = json['review']
        else:
            self.json = json
        self.comments = []

    def addComments(self, comments):
        self.comments = comments

    def numComments(self):
        return len([c for c in self.comments if c['taskState'] == 'comment'])

    def numOpenTasks(self):
        return len([c for c in self.comments if c['taskState'] == 'open'])

    def numClosedTasks(self):
        return len([c for c in self.comments if c['taskState'] == 'addressed'])

    def getval(self, k):
        try:
            if k in self.json:
                return str(self.json[k])
        except:
            pass
        return ""
        
    def id(self):
        return self.getval('id')

    def state(self):
        return self.getval('state')

    def author(self):
        return self.getval('author')

    def updated(self):
        return self.getval('updated')

    def numVotes(self, value):
        count = 0
        try:
            if 'participants' in self.json:
                for p in six.iteritems(self.json['participants']):
                    if p and p[1] and 'vote' in p[1]:
                        if 'value' in p[1]['vote'] and p[1]['vote']['value'] == value:
                            count += 1
        except:
            pass
        return count

    def numVotesUp(self):
        return self.numVotes(1)

    def numVotesDown(self):
        return self.numVotes(-1)

    def testStatus(self):
        return self.getval('testStatus')

    def description(self):
        return self.getval('description')

    def getField(self, f):
        if 'description' not in self.json:
            return ""
        refield = re.compile("^%s:\s*(.*)" % re.escape(f), re.MULTILINE)
        m = refield.search(self.json['description'])
        if m:
            return m.group(1)

    def updateValuesFromTemplateDescription(self, rec):
        """Parse following:
            Review Type: [Screening | WalkThrough | Inspection]
            Scope of Review: [Delta Review | Full Review]
            Supporting Documents:  [list of documents with versions]
            Deadline for the Review:  [date]
            Review Meeting (required for walkthrough or inspection):
              Planned:  [date, participants]
              Performed:  [date, participants]
            Actual Review Effort: [hours]
            reviewType,reviewScope,supportingDocs,reviewDeadline,reviewEffort,
        """
        rec['reviewType'] = self.getField('Review Type')
        rec['reviewScope'] = self.getField('Scope of Review')
        rec['supportingDocs'] = self.getField('Supporting Documents')
        rec['reviewDeadline'] = self.getField('Deadline for the Review')
        rec['reviewEffort'] = self.getField('Actual Review Effort')

class SwarmComments(object):
    """Parses all swarm comment objects - directly from p4 keys output to avoid overhead of
    individual API calls
    """

    def __init__(self, comments):
        "Comments is output from p4.run_keys('-e', 'swarm-comment-*')"
        self.review_comments = defaultdict(list)
        for c in comments:
            j = json.loads(c['value'])
            if 'topic' in j and j['topic'].startswith('review'):
                id = j['topic'].split('/')[1]
                self.review_comments[id].append(j)

    def getReviewComments(self, id):
        if id in self.review_comments:
            return self.review_comments[id]
        return []

class SwarmReviews(object):
    """See module doc string for details"""

    def __init__(self, *args, **kwargs):
        self.parse_args(__doc__, args)

    def parse_args(self, doc, args):
        """Common parsing and setting up of args"""
        desc = textwrap.dedent(doc)
        parser = argparse.ArgumentParser(
            formatter_class=argparse.RawDescriptionHelpFormatter,
            description=desc,
            epilog="Copyright (c) 2008-2018 Perforce Software, Inc."
        )
        self.add_parse_args(parser)     # Should be implemented by subclass
        self.options = parser.parse_args(args=args)
        self.init_logger()
        self.logger.debug("Command Line Options: %s\n" % self.options)

    def add_parse_args(self, parser, default_log_file=None, default_verbosity=None):
        """Default trigger arguments - common to all triggers
        :param default_verbosity:
        :param default_log_file:
        :param parser:
        """
        if not default_log_file:
            default_log_file = DEFAULT_LOG_FILE
        if not default_verbosity:
            default_verbosity = DEFAULT_VERBOSITY
        parser.add_argument('-p', '--port', default=None,
                            help="Perforce server port - set using %%serverport%%. Default: $P4PORT")
        parser.add_argument('-u', '--user', default=None, help="Perforce user. Default: $P4USER")
        parser.add_argument('-L', '--log', default=default_log_file, help="Default: " + default_log_file)
        parser.add_argument('-i', '--input', help="Name of a file containing a list of paths, one per line.")
        parser.add_argument('-e', '--ext', help="Extension text to add to all files in input list. Default: /...", default=None)
        parser.add_argument('-t', '--ticket', required=True, default=None,
                            help="Perforce ticket for use with Swarm. Output of 'p4 login -ap' for specified or current user")
        parser.add_argument('path', nargs='*', help="Perforce depot path(s) - required if no input file specified")
        parser.add_argument('-v', '--verbosity',
                            nargs='?',
                            const="INFO",
                            default=default_verbosity,
                            choices=('DEBUG', 'WARNING', 'INFO', 'ERROR', 'FATAL'),
                            help="Output verbosity level. Default is: " + default_verbosity)

    def init_logger(self, logger_name=None):
        if not logger_name:
            logger_name = LOGGER_NAME
        self.logger = logging.getLogger(logger_name)
        self.logger.setLevel(self.options.verbosity)
        logformat = '%(levelname)s %(asctime)s %(filename)s %(lineno)d: %(message)s'
        logging.basicConfig(format=logformat, filename=self.options.log, level=self.options.verbosity)
        formatter = logging.Formatter('%(message)s')
        ch = logging.StreamHandler(sys.stderr)
        ch.setLevel(logging.INFO)
        ch.setFormatter(formatter)
        self.logger.addHandler(ch)

    def get_swarm_base_url(self, p4):
        p = p4.run_property("-l", "-n", "P4.Swarm.URL")
        url = p[0]['value']
        if url[-1] == '/':
            url = url[:-1]
        r = self.session.get("%s/api/version" % url)
        j = r.json()
        if 'apiVersions' in j:
            ver = j['apiVersions'][-1]
            # Unfortunately Swarm chose to only support some of the API calls in version 10 - yuck!
            if ver == 10:
                ver = 9
            return "%s/api/v%d" % (url, ver)
        return "%s/api/v4" % url

    def get_swarm_reviews(self, p4, changes):
        auth = (p4.user, self.options.ticket)
        base_url = self.get_swarm_base_url(p4)
        reviews = []
        chunk_size = 40
        chunks = [changes[i:i + chunk_size] for i in range(0, len(changes), chunk_size)]
        self.logger.info("Chunks to process: %d" % len(chunks))
        for chunk in chunks:
            self.logger.info("Processing Swarm chunk")
            chg_search = ["change[]=%s" % c['change'] for c in chunk]
            url = '%s/reviews?%s' % (base_url, "&".join(chg_search))
            self.logger.debug("Get: %s" % (url))
            try:
                resp = self.session.get(url, auth=auth)
                if resp.status_code == 200:
                    json = resp.json()
                    self.logger.debug("Result: %s" % (json))
                    if 'reviews' in json:
                        for r in json['reviews']:
                            reviews.append(r)
                else:
                    self.logger.warn("%d: %s" % (resp.status_code, resp.reason))
            except Exception as e:
                self.logger.exception(e)
        return reviews

    def run(self):
        """Runs script"""
        try:
            self.logger.debug("%s: starting" % script_name)

            p4 = P4.P4()
            if self.options.port:
                p4.port = self.options.port
            if self.options.user:
                p4.user = self.options.user
            p4.connect()

            self.session = requests.Session()
            retry = Retry(connect=3, backoff_factor=0.5)
            adapter = HTTPAdapter(max_retries=retry)
            self.session.mount('http://', adapter)
            self.session.mount('https://', adapter)

            users = {}
            for u in p4.run_users():
                users[u['User']] = u

            swarm_comments = SwarmComments(p4.run_keys("-e", "swarm-comment-*"))

            fields = "path,commit,authorID,authorName,dateSubmitted,reviewID,reviewState,dateUpdated,numVotesUp,numVotesDown," \
                  "testStatus,numComments,numOpenTasks,numClosedTasks,job,JiraID,JiraStatus,JiraType," \
                  "reviewType,reviewScope,supportingDocs,reviewDeadline,reviewEffort,JiraSummary,reviewDescription".split(",")
            print(",".join(fields))
            paths = []
            if len(self.options.path) > 0:
                paths.extend(self.options.path)
            if self.options.input:
                with open(self.options.input, "r") as f:
                    paths.extend([p.rstrip() for p in f.readlines()])
            for path in paths:
                if self.options.ext:
                    path += self.options.ext
                self.logger.info("Processing %s" % path)
                changes = p4.run_changes(path)
                if not changes:
                    self.logger.warning("No changes found for: %s" % path)
                    rec = OrderedDict()
                    for f in fields:  # Have fields in correct order
                        rec[f] = ""
                    rec['path'] = path
                    rec['commit'] = "No changes"
                    msg = ",".join([csv_escape(m) for m in six.itervalues(rec)])
                    print(msg)
                    continue

                reviews = self.get_swarm_reviews(p4, changes)
                reviews_by_chg = {}
                for r in reviews:
                    for c in r['commits']:
                        reviews_by_chg[c] = r

                fixes = defaultdict(list)
                for f in p4.run_fixes(path):
                    fixes[f['Change']].append(f)
                jobs = {}
                for j in p4.run_jobs(path):
                    jobs[j['Job']] = j

                for c in changes:
                    rec = OrderedDict()
                    for f in fields:    # Have fields in correct order
                        rec[f] = ""
                    rec['path'] = path
                    rec['commit'] = c['change']
                    rec['authorID'] = c['user']
                    try:
                        rec['authorName'] = users[c['user']]['FullName']
                    except KeyError:
                        rec['authorName'] = c['user']
                    rec['dateSubmitted'] = formatdate(c['time'])
                    if int(c['change']) in reviews_by_chg:
                        try:
                            r = Review(reviews_by_chg[int(c['change'])])
                            r.addComments(swarm_comments.getReviewComments(r.id()))
                            r.updateValuesFromTemplateDescription(rec)
                            rec['reviewDescription'] = r.description()
                            rec['reviewID'] = r.id()
                            rec['reviewState'] = r.state()
                            rec['dateUpdated'] = formatdate(r.updated())
                            rec['numVotesUp'] = str(r.numVotesUp())
                            rec['numVotesDown'] = str(r.numVotesDown())
                            rec['testStatus'] = r.testStatus()
                            rec['numComments'] = r.numComments()
                            rec['numOpenTasks'] = r.numOpenTasks()
                            rec['numClosedTasks'] = r.numClosedTasks()
                        except Exception as e:  # Ignore most errors
                            self.logger.exception(e)
                    if c['change'] in fixes:
                        j = fixes[c['change']][0]['Job']
                        if j in jobs:
                            rec['job'] = j
                            if 'JiraIssue' in jobs[j]:
                                rec['JiraID'] = jobs[j]['JiraIssue']
                            if 'JiraStatus' in jobs[j]:
                                rec['JiraStatus'] = jobs[j]['JiraStatus']
                            if 'JiraType' in jobs[j]:
                                rec['JiraType'] = jobs[j]['JiraType']
                            rec['JiraSummary'] = jobs[j]['Description'].rstrip()
                    try:
                        msg = ",".join([csv_escape(m) for m in six.itervalues(rec)])
                        print(msg)
                    except Exception as e:
                        self.logger.exception(e)
                        msg = ",".join([csv_escape(m).decode('cp1252').encode('utf-8') for m in six.itervalues(rec)])
                        print(msg)
        except Exception as e:
            self.logger.exception(e)
            print(str(e))

if __name__ == '__main__':
    """ Main Program"""
    obj = SwarmReviews(*sys.argv[1:])
    sys.exit(obj.run())