m_librarian/inp.py

   1
   2 import os
   3 from zipfile import ZipFile
   4 from sqlobject import sqlhub
   5 from sqlobject.sqlbuilder import Select
   6 from .db import Author, Book, Extension, Genre, Language, \
   7     insert_name, insert_author
   8
   9 __all__ = ['import_inpx']
  10
  11
  12 EOT = chr(4)  # INP field separator
  13
  14
  15 def split_line(line):
  16     parts = line.strip().split(EOT)
  17     _l = len(parts)
  18     if _l < 11:
  19         raise ValueError('Unknown INP structure: "%s"' % line)
  20     if _l == 11:  # Standard structure
  21         parts.append(None)  # Emulate lang
  22     else:  # New structure
  23         parts = parts[:12]
  24     return parts
  25
  26
  27 def import_inp_line(archive, parts):
  28     authors, genres, title, series, ser_no, file, size, lib_id, deleted, \
  29         extension, date, language = parts
  30     try:
  31         ser_no = int(ser_no)
  32     except ValueError:
  33         ser_no = None
  34     size = int(size)
  35     deleted = deleted == '1'
  36     extension_row = insert_name(Extension, extension)
  37     language_row = insert_name(Language, language)
  38     book = Book(title=title, series=series, ser_no=ser_no,
  39                 archive=archive, file=file, size=size,
  40                 lib_id=lib_id, deleted=deleted,
  41                 extension=extension_row, date=date,
  42                 language=language_row)
  43     authors = authors.split(':')
  44     seen_authors = set()
  45     for author in authors:
  46         if author:
  47             if author in seen_authors:
  48                 continue
  49             seen_authors.add(author)
  50             alist = author.split(',', 2)
  51             surname = alist[0]
  52             if len(alist) > 1:
  53                 name = alist[1]
  54                 if len(alist) == 3:
  55                     misc_name = alist[2]
  56                 else:
  57                     misc_name = ''
  58             else:
  59                 name = misc_name = ''
  60             author_row = insert_author(surname, name, misc_name)
  61             book.addAuthor(author_row)
  62     for genre in genres.split(':'):
  63         if genre:
  64             genre_row = insert_name(Genre, genre, title=genre)
  65             book.addGenre(genre_row)
  66
  67
  68 def import_inp(archive, inp):
  69     files = set()
  70     connection = sqlhub.processConnection
  71     for file, in connection.queryAll(connection.sqlrepr(
  72             Select(Book.q.file, Book.q.archive == archive))):
  73         files.add(file)
  74     for line in inp:
  75         line = line.decode('utf-8')
  76         parts = split_line(line)
  77         file = parts[5]
  78         if file not in files:
  79             files.add(file)
  80             import_inp_line(archive, parts)
  81
  82
  83 def import_inpx(path):
  84     inpx = ZipFile(path)
  85     for name in inpx.namelist():
  86         archive, ext = os.path.splitext(name)
  87         if ext != '.inp':
  88             continue
  89         inp = inpx.open(name)
  90         sqlhub.doInTransaction(import_inp, archive + '.zip', inp)
  91         inp.close()
  92     connection = sqlhub.processConnection
  93     if connection.dbName == 'postgres':
  94         for table in Author, Book, Extension, Genre, Language:
  95             connection.query("VACUUM %s" % table.sqlmeta.table)
  96     elif connection.dbName == 'sqlite':
  97         connection.query("VACUUM")