entry: an individual list entry

[ Previous: list: a data source ] [ Top:  ] [ Next: cli: a command-line interface definer ]
The entry is the wftk's basic unit of activity. From a workflow perspective, it is a process which may have active tasks assigned to users, roles to be filled, interested users, and so forth; from a data management perspective it is a set of named values which may also include large binary attachments, undergo version control, be published to or listed on a Web page, or represent a report or query of an external system. The entry is a complex entity with a lot of functionality packed into it.

Each entry in a repository belongs to one named list of entries, and each entry has a unique key in that list. Configuration is done on a listwise basis, with each list knowing where and how its entries are stored, default values for fields, where its entries are indexed, and actions to take on particular events. The entry object thus needs to know which list it belongs to, which repository it's stored in (or intends to be stored in at some point), and its key. The key is a readonly value, however, as it is generated from the values stored in the entry's fields. (If you need to store specific keys, the key is simply stored in a named field...)

Since a entry is first and foremost a collection of named values, it presents a handy dictionary interface to the world, which is implemented in the xmlobj object.

The _key attribute stores the key of the entry in its list -- if "None" it signifies an entry that hasn't yet been stored at all.
  
class entry(xmlobj):
   """Implements an active object in a repository.
   """

   def __init__ (self, repository, list, key=None):
      self._repository = repository
      self._listid = list
      self._list = repository.defn (list)
      self._list = self._list._xml
      self._key = key
The chief thing we do with entries is retrieve them, add them, modify them, delete them. However, I suspect that the extension of the wftk into the object-oriented realm is going to make it easier for me to conceptualize and prototype additional functionality, and some of those added functions will be things like status management, activation of tasks and such, and things like that. It's going to be interesting to see where we can go with this whole thing.

2005 March 08: Oh, yeah, attachments.
  
   def load (self):
      if not self._key: return None
      self._xml = repmgr.get (self._repository.repos, self._listid, self._key)
      if not self._xml: return None
      self._key = repmgr.getkey (self._repository.repos, self._listid, self._xml)
      return self._key
   def save (self):
      if not self._xml: return None
      if self._key == None:
         repmgr.add (self._repository.repos, self._listid, self._xml)
         self._key = repmgr.getkey (self._repository.repos, self._listid, self._xml)
      else:
         self._key = repmgr.getkey (self._repository.repos, self._listid, self._xml)
         repmgr.mod (self._repository.repos, self._listid, self._xml, self._key)
      return self._key
   def getkey (self):
      if not self._xml: return None
      self._key = repmgr.getkey (self._repository.repos, self._listid, self._xml)
      return self._key
   def setxml (self, xml):
      self.copy (xml)
   def attach_file (self, fldname, file):
      a = open (file)
      self.attach_stream (fldname, a)
   def attach_stream (self, fldname, str):
      a = str.read()
      self.attach (fldname, a)
   def attach (self, fldname, a):
      repmgr.attach (self._repository.repos, self._listid, self._key, fldname, a)
      self.load()
   def retrieve (self, fldname, a=None):
      return repmgr.retrieve (self._repository.repos, self._listid, self._key, fldname, a)
[ Previous: list: a data source ] [ Top:  ] [ Next: cli: a command-line interface definer ]

This code and documentation are released under the terms of the GNU license. They are additionally copyright (c) 2001, Vivtek. All rights reserved except those explicitly granted under the terms of the GNU license.