developer.intra2net.com Git - imap-restore-mail/blob - file_iterator.py

   1 '''
   2 restore-mail-inject.py - Tool to inject mails via IMAP
   3
   4 Copyright (c) 2012 Intra2net AG
   5 '''
   6
   7 import os
   8 import re
   9 import logging
  10
  11 MAIL_FILENAME = re.compile("^[0-9]+\.$")
  12 MBOXFILE_LINE = re.compile("^(.*?)\t(?:\d )?default[\t ](.*)$")
  13 ACL_STRING = re.compile("^(.*?)[\t ](.*?)\t(.*)$")
  14
  15 class FileIterator:
  16     """This class iterates through the e-mail files."""
  17
  18     # class attributes
  19     # mailboxes created during file traversal
  20     created_mailboxes = None
  21     # mailboxes to update during file traversal
  22     acl_mailboxes = None
  23
  24     def __init__(self):
  25         """Creates a connection and a user session."""
  26         self.created_mailboxes = []
  27         self.acl_mailboxes = []
  28         return
  29
  30     def __del__(self):
  31         """Closes the connection and the user session."""
  32         return
  33
  34     @classmethod
  35     def _message_read(cls, filename):
  36         """Retrieves a message from the message file."""
  37         try:
  38             with open(filename, "r") as msgfile:
  39                 message = msgfile.read()
  40         except IOError:
  41             logging.warning("Could not open the e-mail file %s", filename)
  42             raise
  43         return message
  44
  45     @classmethod
  46     def load_mailbox_list(cls, mboxlistfile = ""):
  47         """Load the list of mailboxes and acl rights for each from file."""
  48         mboxdb = {}
  49         if mboxlistfile != "":
  50             try:
  51                 with open(mboxlistfile, 'r') as acl_file:
  52                     for line in acl_file:
  53                         acls = {}
  54                         linedata = MBOXFILE_LINE.match(line).groups()
  55                         if len(linedata) == 0:
  56                             logging.warning("Illegal line in mailbox list dump: %s", line)
  57                             continue
  58                         key = linedata[0]
  59                         aclstr = linedata[1]
  60
  61                         # loop through acl rights string and build dictionary of users and rights
  62                         while(aclstr != ""):
  63                             acldata = ACL_STRING.match(aclstr).groups()
  64                             if len(acldata) == 0:
  65                                 logging.error("Illegal acl string in mailbox list dump: %s", line)
  66                                 continue
  67                             aclstr = acldata[2]
  68                             acls[acldata[0]] = acldata[1]
  69
  70                         mboxdb[key] = acls
  71             except IOError:
  72                 logging.warning("Could not open mboxlist file %s", mboxlistfile)
  73         return mboxdb
  74
  75     def load_mails(self, filepath, mailpath):
  76         """Loads all e-mails from file hierarchy.
  77         This recursive generator always returns a tuple of
  78         the next found (e-mail, mailbox to store, internaldate)."""
  79         logging.debug("Entered directory %s -> %s", filepath, mailpath)
  80         try:
  81             filepath = os.path.abspath(filepath)
  82             os.chdir(filepath)
  83         except OSError:
  84             logging.warning("Can't open the directory %s", filepath)
  85             return
  86         # mark mailboxes that should be created
  87         self.created_mailboxes.append(mailpath)
  88         subpaths = os.listdir(filepath)
  89         for subpath in subpaths:
  90             if subpath == "." or subpath == "..":
  91                 continue
  92             new_filepath = filepath + "/" + subpath
  93             if (os.path.isfile(new_filepath)):
  94                 if os.path.getsize(new_filepath) == 0:
  95                     logging.info("Skipping empty file %s", subpath)
  96                 else:
  97                     if MAIL_FILENAME.match(subpath):
  98                         logging.info("Injecting file %s", subpath)
  99                         try:
 100                             message = self._message_read(new_filepath)
 101                             # suggest file modification date for internaldate
 102                             yield (message, mailpath, os.path.getmtime(new_filepath))
 103                         except IOError:
 104                             logging.warning("Could not retrieve mail from the file %s", new_filepath)
 105             else:
 106                 if os.path.isdir(new_filepath):
 107                     # cyrus ^ vs . storage replacement
 108                     subpath = subpath.replace("^", ".")
 109                     new_mailpath = mailpath + "/" + subpath
 110                     logging.debug("Inserting mails from directory %s into mailbox %s", new_filepath, new_mailpath)
 111                     # load_mails($mboxdbref, $origuser, $targetuser)
 112                     rcrs_generator = self.load_mails(new_filepath, new_mailpath)
 113                     # you enter the generator in the for loop
 114                     for rcr in rcrs_generator:
 115                         yield rcr
 116                     logging.debug("Done with directory %s and mailbox %s", new_filepath, new_mailpath)
 117         # mark mailboxes that need acl update
 118         self.acl_mailboxes.append(mailpath)
 119         return