Switch branches/tags
Nothing to show
Find file
Fetching contributors…
Cannot retrieve contributors at this time
510 lines (396 sloc) 14.9 KB
#!/usr/bin/env python
# Copyright 2009 Google Inc. All Rights Reserved.
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# See the License for the specific language governing permissions and
# limitations under the License.
"""pytracker is a Python wrapper around the Tracker API."""
__author__ = ' (Doug Coker)'
import calendar
import cookielib
import re
import time
import urllib
import urllib2
import xml.dom
from xml.dom import minidom
import xml.parsers.expat
import xml.sax.saxutils
# Some fields specify UTC, some GMT?
_TRACKER_DATETIME_RE = re.compile(r'^\d{4}/\d{2}/\d{2} .*(GMT|UTC)$')
def TrackerDatetimeToYMD(pdt):
assert _TRACKER_DATETIME_RE.match(pdt)
pdt = pdt.split()[0]
pdt = pdt.replace('/', '-')
return pdt
class Tracker(object):
"""Tracker API."""
def __init__(self, project_id, auth,
If you are debugging API calls, you may want to use a non-HTTPS API URL:
project_id: the Tracker ID (integer).
auth: a TrackerAuth instance.
base_api_url: the base URL of the HTTP API (with trailing /).
self.project_id = project_id
self.base_api_url = base_api_url
cookies = cookielib.CookieJar()
self.opener = urllib2.build_opener(urllib2.HTTPCookieProcessor(cookies))
self.token = auth.EstablishAuthToken(self.opener)
def _Api(self, request, method, body=None):
url = self.base_api_url + 'projects/%d/%s' % (self.project_id,
headers = {}
if self.token:
headers['X-TrackerToken'] = self.token
if not body and method == 'GET':
# Do a GET
req = urllib2.Request(url, None, headers)
headers['Content-Type'] = 'application/xml'
req = urllib2.Request(url, body, headers)
req.get_method = lambda: method
res =
except urllib2.HTTPError, e:
message = "HTTP Status Code: %s\nMessage: %s\nURL: %s\nError: %s" % (e.code, e.msg, e.geturl(),
raise TrackerApiException(message)
def _ApiQueryStories(self, query=None):
if query:
output = self._Api('stories?filter=' + urllib.quote_plus(query),
output = self._Api('stories', 'GET')
# Hack: throw an exception if we didn't get valid XML.
xml.parsers.expat.ParserCreate('utf-8').Parse(output, True)
return output
def GetStoriesXml(self):
return self._ApiQueryStories()
def GetReleaseStoriesXml(self):
return self._ApiQueryStories('type:release')
def GetIterationStories(self, iteration=None, offset=None, limit=None):
iteration = ('/%s' % iteration) if iteration else ''
params = []
if offset:
params.append('offset=%s' % urllib.quote_plus(str(offset)))
if limit:
params.append('limit=%s' % urllib.quote_plus(str(limit)))
response = self._Api('iterations%s?%s' % (iteration, '&'.join(params)), 'GET')
# Hack: throw an exception if we didn't get valid XML.
xml.parsers.expat.ParserCreate('utf-8').Parse(response, True)
parsed = xml.dom.minidom.parseString(response)
els = parsed.getElementsByTagName('story')
lst = []
for el in els:
return lst
def GetStories(self, filt=None):
"""Fetch all Stories that satisfy the filter.
filt: a Tracker search filter.
List of Story().
stories = self._ApiQueryStories(filt)
parsed = xml.dom.minidom.parseString(stories)
els = parsed.getElementsByTagName('story')
lst = []
for el in els:
return lst
def GetStory(self, story_id):
story_xml = self._Api('stories/%d' % story_id, 'GET')
return Story.FromXml(story_xml)
def AddComment(self, story_id, comment):
comment = '<note><text>%s</text></note>' % xml.sax.saxutils.escape(comment)
self._Api('stories/%d/notes' % story_id, 'POST', comment)
def AddNewStory(self, story):
"""Persists a new story to Tracker and returns the new Story."""
story_xml = story.ToXml()
res = self._Api('stories', 'POST', story_xml)
story = Story.FromXml(res)
return story
def UpdateStoryById(self, story_id, story):
"""Persist changes to an existing story to Tracker.
Use this method if you are changing a story without first retreiving the
story_id: The ID of the story to mutate
story: The Story containing values to change.
The updated Story().
story_xml = story.ToXml()
res = self._Api('stories/%d' % story_id, 'PUT', story_xml)
return Story.FromXml(res)
def UpdateStory(self, story):
"""Persists changes to an existing story to Tracker.
Use this method if you have a full Story object created by one of the query
story: a Story()
The updated Story().
story_xml = story.ToXml()
res = self._Api('stories/%d' % story.GetStoryId(), 'PUT', story_xml)
return Story.FromXml(res)
def DeleteStory(self, story_id):
"""Deletes a story by story ID."""
self._Api('stories/%d' % story_id, 'DELETE', '')
class TrackerAuth(object):
"""Abstract base class for establishing credentials for pytracker."""
def __init__(self, username, password):
self.username = username
self.password = password
def EstablishAuthToken(self, opener):
"""Returns the value for use as the X-TrackerToken HTTP header, or None.
This method may mutate the cookie jar via opener.
opener: a urllib2.OpenerDirector instance that will be used for
subsequent HTTP API calls.
raise NotImplementedError()
class TrackerAuthException(Exception):
"""Raised when something goes wrong with authentication."""
class NoTokensAvailableException(Exception):
"""Raised when HostedTrackerAuth can't find any tokens for this user."""
class TrackerApiException(Exception):
"""Raised when Tracker returns an error."""
class HostedTrackerAuth(TrackerAuth):
"""Authentication rules for hosted Tracker instances."""
def EstablishAuthToken(self, opener):
"""Returns the first auth token returned by /services/tokens/active."""
url = ''
data = urllib.urlencode((('username', self.username),
('password', self.password)))
req =, data)
except urllib2.HTTPError, e:
if e.code == 404:
raise NoTokensAvailableException(
'Did you create any? Check')
res =
dom = minidom.parseString(res)
token = dom.getElementsByTagName('guid')[0]
return token
class Story(object):
"""Represents a Story.
This class can be used to represent a complete Story (generally queried from
the Tracker class), or can contain partial information for update or create
operations (constructed with default constructor).
Internally, Story uses None to indicate that the client has not specified a
value for the field or that it has not been parsed from XML. This enables us
to use the same Story object to define an update to multiple stories, without
requiring that the client first fetch, parse, and update an existing story.
This is supported by all mutable fields except for labels, which are
represented by Tracker as a comma-separated list of strings in a single tag
body. For label operations on existing stories to be performed correctly,
the Story must first be fetched from the server so that the existing labels
are not lost.
# Fields that can be treated as strings when embedding in XML.
UPDATE_FIELDS = ('story_type', 'current_state', 'name',
'description', 'estimate', 'requested_by', 'owned_by')
# Type: immutable ints.
story_id = None
iteration_number = None
# Type: immutable times (secs since epoch)
created_at = None
# Type: mutable time (secs since epoch)
deadline = None
# Type: mutable set (API methods expose as string)
labels = None
# Type: immutable strings
url = None
# Type: mutable strings
requested_by = None
owned_by = None
story_type = None
current_state = None
description = None
name = None
estimate = None
def __str__(self):
return "Story(%r)" % self.__dict__
def FromXml(as_xml):
"""Parses an XML string into a Story.
as_xml: a full XML document from the Tracker API.
parsed = minidom.parseString(as_xml.encode('utf-8'))
story = Story()
story.story_id = int(parsed.getElementsByTagName('id')[0]
story.url = parsed.getElementsByTagName('url')[0]
story.owned_by = Story._GetDataFromTag(parsed, 'owned_by')
story.created_at = Story._ParseDatetimeIntoSecs(parsed, 'created_at')
story.requested_by = Story._GetDataFromTag(parsed, 'requested_by')
iteration = Story._GetDataFromTag(parsed, 'number')
if iteration:
story.iteration_number = int(iteration)
story.SetName(Story._GetDataFromTag(parsed, 'name'))
story.SetDescription(Story._GetDataFromTag(parsed, 'description'))
story.SetDeadline(Story._ParseDatetimeIntoSecs(parsed, 'deadline'))
estimate = Story._GetDataFromTag(parsed, 'estimate')
if estimate is not None:
story.estimate = estimate
labels = Story._GetDataFromTag(parsed, 'labels')
if labels is not None:
return story
def _GetDataFromTag(dom, tag):
"""Retrieve value associated with the tag, if any.
dom: XML DOM object
tag: name of the desired tag
None (if tag doesn't exist), empty string (if tag exists, but body is
empty), or the tag body.
tags = dom.getElementsByTagName(tag)
if not tags:
return None
elif tags[0].hasChildNodes():
return tags[0]
return ''
def _ParseDatetimeIntoSecs(dom, tag):
"""Returns the tag body parsed into seconds-since-epoch."""
el = dom.getElementsByTagName(tag)
if not el:
return None
assert el[0].getAttribute('type') == 'datetime'
data = el[0]
# Tracker emits datetime strings in UTC or GMT.
# The [:-4] strips the timezone indicator
when = time.strptime(data[:-4], '%Y/%m/%d %H:%M:%S')
# calendar.timegm treats the tuple as GMT
return calendar.timegm(when)
# Immutable fields
def GetStoryId(self):
return self.story_id
def GetIteration(self):
return self.iteration_number
def GetUrl(self):
return self.url
# Mutable fields
def GetRequestedBy(self):
return self.requested_by
def SetRequestedBy(self, requested_by):
self.requested_by = requested_by
def GetOwnedBy(self):
return self.owned_by
def SetOwnedBy(self, owned_by):
self.owned_by = owned_by
def GetStoryType(self):
return self.story_type
def SetStoryType(self, story_type):
assert story_type in ['bug', 'chore', 'release', 'feature']
self.story_type = story_type
def GetCurrentState(self):
return self.current_state
def SetCurrentState(self, current_state):
self.current_state = current_state
def GetName(self):
def SetName(self, name): = name
def GetEstimate(self):
return self.estimate
def SetEstimate(self, estimate):
self.estimate = estimate
def GetDescription(self):
return self.description
def SetDescription(self, description):
self.description = description
def GetDeadline(self):
return self.deadline
def SetDeadline(self, secs_since_epoch):
self.deadline = secs_since_epoch
def GetCreatedAt(self):
return self.created_at
def SetCreatedAt(self, secs_since_epoch):
self.created_at = secs_since_epoch
def AddLabel(self, label):
"""Adds a label (see caveat in class comment)."""
if self.labels is None:
self.labels = set()
def RemoveLabel(self, label):
"""Removes a label (see caveat in class comment)."""
if self.labels is None:
self.labels = set()
except KeyError:
def AddLabelsFromString(self, labels):
"""Adds a set of labels from a comma-delimited string (see class caveat)."""
if self.labels is None:
self.labels = set()
self.labels = self.labels.union([x.strip() for x in labels.split(',')])
def GetLabelsAsString(self):
"""Returns the labels as a comma delimited list of strings."""
if self.labels is None:
return None
lst = list(self.labels)
return ','.join(lst)
def ToXml(self):
"""Converts this Story to an XML string."""
doc = xml.dom.getDOMImplementation().createDocument(None, 'story', None)
story = doc.getElementsByTagName('story')[0]
# Most fields are just simple strings or ints, so we treat them all in the
# same way.
for field_name in self.UPDATE_FIELDS:
v = getattr(self, field_name)
if v is not None:
new_tag = doc.createElement(field_name)
# Labels are represented internally as sets.
if self.labels:
labels_tag = doc.createElement('labels')
# Dates are special
DATE_FORMAT = '%Y/%m/%d %H:%M:%S UTC'
if self.deadline:
formatted = time.strftime(DATE_FORMAT, time.gmtime(self.deadline))
deadline_tag = doc.createElement('deadline')
deadline_tag.setAttribute('type', 'datetime')
if self.created_at:
formatted = time.strftime(DATE_FORMAT, time.gmtime(self.created_at))
created_at_tag = doc.createElement('created_at')
created_at_tag.setAttribute('type', 'datetime')
return doc.toxml('utf-8')