1 #!/usr/bin/env python3

     2 

     3 from pydbkit import toolbase

     4 

     5 

     6 class BatchQueryTool(toolbase.SimpleTool):

     7     def __init__(self):

     8         toolbase.SimpleTool.__init__(self, name='batchquery', desc='Run a query using columns from CSV file as arguments.')

     9         self.parser.add_argument('--query', dest='query', type=str, help='Query to run. Use %%s for arguments, or %%(name)s for named arguments (see --header).')

    10         self.parser.add_argument('--file', dest='file', type=str, help='CSV file with data to use as arguments.')

    11         self.parser.add_argument('--init', dest='init', type=str, help='Query which initialize database session (eg. temporary function).')

    12         self.parser.add_argument('--output', dest='output', type=str, help='File name for results.')

    13         self.parser.add_argument('--outputfunc', dest='outputfunc', type=str, help='Python function which will format results (format_row(args, rows)).')

    14         self.parser.add_argument('--header', dest='header', action='store_true', help='First line of CSV is header with names for columns. These name can be used in query.')

    15 

    16     def _split_line(self, line):

    17         return [x.strip() for x in line.split(',')]

    18 

    19     def main(self):

    20         results = []

    21         # load query from file

    22         with open(self.args.query, 'r', encoding='utf8') as f:

    23             query = f.read()

    24         # connect DB

    25         with self.pgm.cursor('target') as curs:

    26             # run init query

    27             if self.args.init:

    28                 with open(self.args.init, 'r', encoding='utf8') as f:

    29                     curs.execute(f.read(), [])

    30             # read CSV file

    31             with open(self.args.file, 'r', encoding='utf8') as f:

    32                 # read header

    33                 names = None

    34                 if self.args.header:

    35                     line = f.readline()

    36                     names = self._split_line(line)

    37                 # read and process lines

    38                 for line in f:

    39                     args = self._split_line(line)

    40                     if names:

    41                         args = dict(zip(names, args))

    42                     curs.execute(query, args)

    43                     rows = curs.fetchall()

    44                     results.append((args, rows))

    45             curs.connection.commit()

    46         # write results to output file

    47         if self.args.output:

    48             format_row = None

    49             if self.args.outputfunc:

    50                 with open(self.args.outputfunc, 'r', encoding='utf8') as f:

    51                     d = dict()

    52                     exec(f.read(), d)

    53                     format_row = d['format_row']

    54 

    55             with open(self.args.output, 'w', encoding='utf8') as f:

    56                 for args, rows in results:

    57                     if format_row:

    58                         f.write(format_row(args, rows))

    59                     else:

    60                         f.write(repr(args))

    61                         f.write(' -> ')

    62                         f.write(repr(rows))

    63                     f.write('\n')

    64 

    65 

    66 tool = BatchQueryTool()

    67 tool.setup()

    68 tool.main()

    69