view test/memorydb.py @ 5710:0b79bfcb3312

Add support for making an idempotent POST. This allows retrying a POST that was interrupted. It involves creating a post once only (poe) url /rest/data/<class>/@poe/<random_token>. This url acts the same as a post to /rest/data/<class>. However once the @poe url is used, it can't be used for a second POST. To make these changes: 1) Take the body of post_collection into a new post_collection_inner function. Have post_collection call post_collection_inner. 2) Add a handler for POST to rest/data/class/@poe. This will return a unique POE url. By default the url expires after 30 minutes. The POE random token is only good for a specific user and is stored in the session db. 3) Add a handler for POST to rest/data/<class>/@poe/<random token>. The random token generated in 2 is validated for proper class (if token is not generic) and proper user and must not have expired. If everything is valid, call post_collection_inner to process the input and generate the new entry. To make recognition of 2 stable (so it's not confused with rest/data/<:class_name>/<:item_id>), removed @ from Routing::url_to_regex. The current Routing.execute method stops on the first regular expression to match the URL. Since item_id doesn't accept a POST, I was getting 405 bad method sometimes. My guess is the order of the regular expressions is not stable, so sometime I would get the right regexp for /data/<class>/@poe and sometime I would get the one for /data/<class>/<item_id>. By removing the @ from the url_to_regexp, there was no way for the item_id case to match @poe. There are alternate fixes we may need to look at. If a regexp matches but the method does not, return to the regexp matching loop in execute() looking for another match. Only once every possible match has failed should the code return a 405 method failure. Another fix is to implement a more sophisticated mechanism so that @Routing.route("/data/<:class_name>/<:item_id>/<:attr_name>", 'PATCH') has different regexps for matching <:class_name> <:item_id> and <:attr_name>. Currently the regexp specified by url_to_regex is used for every component. Other fixes: Made failure to find any props in props_from_args return an empty dict rather than throwing an unhandled error. Make __init__ for SimulateFieldStorageFromJson handle an empty json doc. Useful for POSTing to rest/data/class/@poe with an empty document. Testing: added testPostPOE to test/rest_common.py that I think covers all the code that was added. Documentation: Add doc to rest.txt in the "Client API" section titled: Safely Re-sending POST". Move existing section "Adding new rest endpoints" in "Client API" to a new second level section called "Programming the REST API". Also a minor change to the simple rest client moving the header setting to continuation lines rather than showing one long line.
author John Rouillard <rouilj@ieee.org>
date Sun, 14 Apr 2019 21:07:11 -0400
parents c4f9a152258d
children ae35daa5baab
line wrap: on
line source

'''Implement an in-memory hyperdb for testing purposes.
'''

import shutil
import os

from roundup import date
from roundup import hyperdb
from roundup import roundupdb
from roundup import security
from roundup import password
from roundup import configuration
from roundup.backends import back_anydbm
from roundup.backends import indexer_dbm
from roundup.backends import sessions_dbm
from roundup.backends import indexer_common
from roundup.support import ensureParentsExist
from roundup.anypy.strings import s2b

def new_config(debug=False):
    config = configuration.CoreConfig()
    config.DATABASE = "db"
    #config.logging = MockNull()
    # these TRACKER_WEB and MAIL_DOMAIN values are used in mailgw tests
    if debug:
        config.LOGGING_LEVEL = "DEBUG"
    config.MAIL_DOMAIN = "your.tracker.email.domain.example"
    config.TRACKER_WEB = "http://tracker.example/cgi-bin/roundup.cgi/bugs/"
    return config

def create(journaltag, create=True, debug=False):
    db = Database(new_config(debug), journaltag)

    # load standard schema
    schema = os.path.join(os.path.dirname(__file__),
        '../share/roundup/templates/classic/schema.py')
    vars = hyperdb.__dict__
    vars['Class'] = Class
    vars['FileClass'] = FileClass
    vars['IssueClass'] = IssueClass
    vars['db'] = db
    exec(compile(open(schema).read(), schema, 'exec'), vars)
    initial_data = os.path.join(os.path.dirname(__file__),
        '../share/roundup/templates/classic/initial_data.py')
    vars = dict(db=db, admin_email='admin@test.com',
        adminpw=password.Password('sekrit'))
    exec(compile(open(initial_data).read(), initial_data, 'exec'), vars)

    # load standard detectors
    thisdir = os.path.dirname(__file__)
    dirname = os.path.join(thisdir,
        '../share/roundup/templates/classic/detectors')
    for fn in os.listdir(dirname):
        if not fn.endswith('.py'): continue
        vars = {}
        exec(compile(open(os.path.join(dirname, fn)).read(),
                     os.path.join(dirname, fn), 'exec'), vars)
        vars['init'](db)

    vars = {}
    exec(compile(open(os.path.join(thisdir, "tx_Source_detector.py")).read(),
                 os.path.join(thisdir, "tx_Source_detector.py"), 'exec'), vars)
    vars['init'](db)

    '''
    status = Class(db, "status", name=String())
    status.setkey("name")
    priority = Class(db, "priority", name=String(), order=String())
    priority.setkey("name")
    keyword = Class(db, "keyword", name=String(), order=String())
    keyword.setkey("name")
    user = Class(db, "user", username=String(), password=Password(),
        assignable=Boolean(), age=Number(), roles=String(), address=String(),
        supervisor=Link('user'),realname=String(),alternate_addresses=String())
    user.setkey("username")
    file = FileClass(db, "file", name=String(), type=String(),
        comment=String(indexme="yes"), fooz=Password())
    file_nidx = FileClass(db, "file_nidx", content=String(indexme='no'))
    issue = IssueClass(db, "issue", title=String(indexme="yes"),
        status=Link("status"), nosy=Multilink("user"), deadline=Date(),
        foo=Interval(), files=Multilink("file"), assignedto=Link('user'),
        priority=Link('priority'), spam=Multilink('msg'),
        feedback=Link('msg'))
    stuff = Class(db, "stuff", stuff=String())
    session = Class(db, 'session', title=String())
    msg = FileClass(db, "msg", date=Date(),
                           author=Link("user", do_journal='no'),
                           files=Multilink('file'), inreplyto=String(),
                           messageid=String(), summary=String(),
                           content=String(),
                           recipients=Multilink("user", do_journal='no')
                           )
    '''
    if create:
        db.user.create(username="fred", roles='User',
            password=password.Password('sekrit'), address='fred@example.com')

    db.security.addPermissionToRole('User', 'Email Access')
    '''
    db.security.addPermission(name='Register', klass='user')
    db.security.addPermissionToRole('User', 'Web Access')
    db.security.addPermissionToRole('Anonymous', 'Email Access')
    db.security.addPermissionToRole('Anonymous', 'Register', 'user')
    for cl in 'issue', 'file', 'msg', 'keyword':
        db.security.addPermissionToRole('User', 'View', cl)
        db.security.addPermissionToRole('User', 'Edit', cl)
        db.security.addPermissionToRole('User', 'Create', cl)
    for cl in 'priority', 'status':
        db.security.addPermissionToRole('User', 'View', cl)
    '''
    return db

class cldb(dict):
    def __init__(self, **values):
        super(cldb, self).__init__()
        for key, value in values.items():
            super(cldb, self).__setitem__(s2b(key), value)
    def __getitem__(self, key):
        return super(cldb, self).__getitem__(s2b(key))
    def __setitem__(self, key, value):
        return super(cldb, self).__setitem__(s2b(key), value)
    def __delitem__(self, key):
        return super(cldb, self).__delitem__(s2b(key))
    def __contains__(self, key):
        return super(cldb, self).__contains__(s2b(key))
    def close(self):
        pass

class BasicDatabase(dict):
    ''' Provide a nice encapsulation of an anydbm store.

        Keys are id strings, values are automatically marshalled data.
    '''
    def __init__(self, **values):
        super(BasicDatabase, self).__init__()
        for k, v in values.items():
            super(BasicDatabase, self).__setitem__(s2b(k), v)
    def __getitem__(self, key):
        if key not in self:
            d = self[key] = {}
            return d
        return super(BasicDatabase, self).__getitem__(s2b(key))
    def __setitem__(self, key, value):
        return super(BasicDatabase, self).__setitem__(s2b(key), value)
    def __delitem__(self, key):
        return super(BasicDatabase, self).__delitem__(s2b(key))
    def __contains__(self, key):
        return super(BasicDatabase, self).__contains__(s2b(key))
    def exists(self, infoid):
        return infoid in self
    def get(self, infoid, value, default=None):
        return self[infoid].get(value, default)
    def getall(self, infoid):
        if infoid not in self:
            raise KeyError(infoid)
        return self[infoid]
    def set(self, infoid, **newvalues):
        self[infoid].update(newvalues)
    def list(self):
        return list(self.keys())
    def destroy(self, infoid):
        del self[infoid]
    def commit(self):
        pass
    def close(self):
        pass
    def updateTimestamp(self, sessid):
        pass
    def clean(self):
        pass

class Sessions(BasicDatabase, sessions_dbm.Sessions):
    name = 'sessions'

class OneTimeKeys(BasicDatabase, sessions_dbm.Sessions):
    name = 'otks'

class Indexer(indexer_dbm.Indexer):
    def __init__(self, db):
        indexer_common.Indexer.__init__(self, db)
        self.reindex = 0
        self.quiet = 9
        self.changed = 0

    def load_index(self, reload=0, wordlist=None):
        # Unless reload is indicated, do not load twice
        if self.index_loaded() and not reload:
            return 0
        self.words = {}
        self.files = {'_TOP':(0,None)}
        self.fileids = {}
        self.changed = 0

    def save_index(self):
        pass
    def force_reindex(self):
        # TODO I'm concerned that force_reindex may not be tested by
        # testForcedReindexing if the functionality can just be removed
        pass

class Database(back_anydbm.Database):
    """A database for storing records containing flexible data types.

    Transaction stuff TODO:

    - check the timestamp of the class file and nuke the cache if it's
      modified. Do some sort of conflict checking on the dirty stuff.
    - perhaps detect write collisions (related to above)?
    """

    dbtype = "memorydb"

    def __init__(self, config, journaltag=None):
        self.config, self.journaltag = config, journaltag
        self.classes = {}
        self.items = {}
        self.ids = {}
        self.journals = {}
        self.files = {}
        self.tx_files = {}
        self.security = security.Security(self)
        self.stats = {'cache_hits': 0, 'cache_misses': 0, 'get_items': 0,
            'filtering': 0}
        self.sessions = Sessions()
        self.otks = OneTimeKeys()
        self.indexer = Indexer(self)

        # anydbm bits
        self.cache = {}         # cache of nodes loaded or created
        self.dirtynodes = {}    # keep track of the dirty nodes by class
        self.newnodes = {}      # keep track of the new nodes by class
        self.destroyednodes = {}# keep track of the destroyed nodes by class
        self.transactions = []
        self.tx_Source = None

    def filename(self, classname, nodeid, property=None, create=0):
        shutil.copyfile(__file__, __file__+'.dummy')
        return __file__+'.dummy'

    def filesize(self, classname, nodeid, property=None, create=0):
        return len(self.getfile(classname, nodeid, property))

    def post_init(self):
        pass

    def refresh_database(self):
        pass

    def getSessionManager(self):
        return self.sessions

    def getOTKManager(self):
        return self.otks

    def reindex(self, classname=None, show_progress=False):
        pass

    def __repr__(self):
        return '<memorydb instance at %x>'%id(self)

    def storefile(self, classname, nodeid, property, content):
        if isinstance(content, str):
            content = s2b(content)
        self.tx_files[classname, nodeid, property] = content
        self.transactions.append((self.doStoreFile, (classname, nodeid,
            property)))

    def getfile(self, classname, nodeid, property):
        if (classname, nodeid, property) in self.tx_files:
            return self.tx_files[classname, nodeid, property]
        return self.files[classname, nodeid, property]

    def doStoreFile(self, classname, nodeid, property, **databases):
        self.files[classname, nodeid, property] = self.tx_files[classname, nodeid, property]
        return (classname, nodeid)

    def rollbackStoreFile(self, classname, nodeid, property, **databases):
        del self.tx_files[classname, nodeid, property]

    def numfiles(self):
        return len(self.files) + len(self.tx_files)

    def close(self):
        self.clearCache()
        self.tx_files = {}
        # kill the schema too
        self.classes = {}
        # just keep the .items

    #
    # Classes
    #
    def __getattr__(self, classname):
        """A convenient way of calling self.getclass(classname)."""
        if classname in self.classes:
            return self.classes[classname]
        raise AttributeError(classname)

    def addclass(self, cl):
        cn = cl.classname
        if cn in self.classes:
            raise ValueError(cn)
        self.classes[cn] = cl
        if cn not in self.items:
            self.items[cn] = cldb()
            self.ids[cn] = 0

        # add default Edit and View permissions
        self.security.addPermission(name="Create", klass=cn,
            description="User is allowed to create "+cn)
        self.security.addPermission(name="Edit", klass=cn,
            description="User is allowed to edit "+cn)
        self.security.addPermission(name="View", klass=cn,
            description="User is allowed to access "+cn)

    def getclasses(self):
        """Return a list of the names of all existing classes."""
        return sorted(self.classes.keys())

    def getclass(self, classname):
        """Get the Class object representing a particular class.

        If 'classname' is not a valid class name, a KeyError is raised.
        """
        try:
            return self.classes[classname]
        except KeyError:
            raise KeyError('There is no class called "%s"'%classname)

    #
    # Class DBs
    #
    def clear(self):
        self.items = {}

    def getclassdb(self, classname, mode='r'):
        """ grab a connection to the class db that will be used for
            multiple actions
        """
        return self.items[classname]

    def getCachedJournalDB(self, classname):
        return self.journals.setdefault(classname, {})

    #
    # Node IDs
    #
    def newid(self, classname):
        self.ids[classname] += 1
        return str(self.ids[classname])
    def setid(self, classname, id):
        self.ids[classname] = int(id)

    #
    # Journal
    #
    def doSaveJournal(self, classname, nodeid, action, params, creator,
            creation):
        if creator is None:
            creator = self.getuid()
        if creation is None:
            creation = date.Date()
        self.journals.setdefault(classname, {}).setdefault(nodeid,
            []).append((nodeid, creation, creator, action, params))

    def doSetJournal(self, classname, nodeid, journal):
        self.journals.setdefault(classname, {})[nodeid] = journal

    def getjournal(self, classname, nodeid):
        # our journal result
        res = []

        # add any journal entries for transactions not committed to the
        # database
        for method, args in self.transactions:
            if method != self.doSaveJournal:
                continue
            (cache_classname, cache_nodeid, cache_action, cache_params,
                cache_creator, cache_creation) = args
            if cache_classname == classname and cache_nodeid == nodeid:
                if not cache_creator:
                    cache_creator = self.getuid()
                if not cache_creation:
                    cache_creation = date.Date()
                res.append((cache_nodeid, cache_creation, cache_creator,
                    cache_action, cache_params))
        try:
            res += self.journals.get(classname, {})[nodeid]
        except KeyError:
            if res: return res
            raise IndexError(nodeid)
        return res

    def pack(self, pack_before):
        """ Delete all journal entries except "create" before 'pack_before'.
        """
        pack_before = pack_before.serialise()
        for classname in self.journals:
            db = self.journals[classname]
            for key in db:
                # get the journal for this db entry
                l = []
                last_set_entry = None
                for entry in db[key]:
                    # unpack the entry
                    (nodeid, date_stamp, self.journaltag, action,
                        params) = entry
                    date_stamp = date_stamp.serialise()
                    # if the entry is after the pack date, _or_ the initial
                    # create entry, then it stays
                    if date_stamp > pack_before or action == 'create':
                        l.append(entry)
                db[key] = l

class Class(back_anydbm.Class):
    pass

class FileClass(back_anydbm.FileClass):
    def __init__(self, db, classname, **properties):
        if 'content' not in properties:
            properties['content'] = hyperdb.String(indexme='yes')
        if 'type' not in properties:
            properties['type'] = hyperdb.String()
        back_anydbm.Class.__init__(self, db, classname, **properties)

    def export_files(self, dirname, nodeid):
        dest = self.exportFilename(dirname, nodeid)
        ensureParentsExist(dest)
        f = open(dest, 'wb')
        f.write(self.db.files[self.classname, nodeid, None])
        f.close()

    def import_files(self, dirname, nodeid):
        source = self.exportFilename(dirname, nodeid)
        f = open(source, 'rb')
        self.db.files[self.classname, nodeid, None] = f.read()
        f.close()
        mime_type = None
        props = self.getprops()
        if 'type' in props:
            mime_type = self.get(nodeid, 'type')
        if not mime_type:
            mime_type = self.default_mime_type
        if props['content'].indexme:
            self.db.indexer.add_text((self.classname, nodeid, 'content'),
                self.get(nodeid, 'content'), mime_type)

# deviation from spec - was called ItemClass
class IssueClass(Class, roundupdb.IssueClass):
    # Overridden methods:
    def __init__(self, db, classname, **properties):
        """The newly-created class automatically includes the "messages",
        "files", "nosy", and "superseder" properties.  If the 'properties'
        dictionary attempts to specify any of these properties or a
        "creation" or "activity" property, a ValueError is raised.
        """
        if 'title' not in properties:
            properties['title'] = hyperdb.String(indexme='yes')
        if 'messages' not in properties:
            properties['messages'] = hyperdb.Multilink("msg")
        if 'files' not in properties:
            properties['files'] = hyperdb.Multilink("file")
        if 'nosy' not in properties:
            # note: journalling is turned off as it really just wastes
            # space. this behaviour may be overridden in an instance
            properties['nosy'] = hyperdb.Multilink("user", do_journal="no")
        if 'superseder' not in properties:
            properties['superseder'] = hyperdb.Multilink(classname)
        Class.__init__(self, db, classname, **properties)

# vim: set et sts=4 sw=4 :

Roundup Issue Tracker: http://roundup-tracker.org/