Parser for printf-style format strings

Question

Is there a good way to parse a printf-style format string and get the various bits (literal text & format points with processable options) as part of the standard library, or failing that as a third-party library? I could not find one in the stdlib, and a web search does not seem to turn up anything useful.

Python 2.6 added string.Formatter which allows processing and customising format-style strings, but I did not manage to find a similar tool for printf-style format strings.

I'm looking for it to do things like validate translations, and possibly convert printf-style format strings to format-style ones (or even f-strings).

So, you are asking if there is a printf-style formatting for stdlib or I am lost in the question? — vahdet
– vahdet, Commented Jul 11, 2018 at 8:36
Yeah I kinda forgot putting the question in the clear, I've added it and reorganised the post: as the title hints I'm looking for a parser of printf-style string. — Masklinn
– Masklinn, Commented Jul 11, 2018 at 9:21

PaulMcG · Accepted Answer · 2018-07-11 12:25:01Z

Not hard to slap one together with pyparsing, after looking at the docs:

import pyparsing as pp

# from https://docs.python.org/3/library/stdtypes.html?highlight=string%20interpolation#printf-style-string-formatting
PCT,LPAREN,RPAREN,DOT = map(pp.Literal, '%().')
conversion_flag_expr = pp.oneOf(list("#0- +"))
conversion_type_expr = pp.oneOf(list("diouxXeEfFgGcrsa%"))
length_mod_expr = pp.oneOf(list("hlL"))
interp_expr = (PCT
               + pp.Optional(LPAREN + pp.Word(pp.printables, excludeChars=")")("mapping_key") + RPAREN)
               + pp.Optional(conversion_flag_expr("conversion_flag"))
               + pp.Optional(('*' | pp.pyparsing_common.integer)("min_width"))
               + pp.Optional(DOT + pp.pyparsing_common.integer("max_width"))
               + pp.Optional(length_mod_expr("length_modifier"))
               + conversion_type_expr("conversion_type")
               )

tests = """
    Now is the winter of our %s made %(adjective)s %(season)s by this %(relative)s of %(city)s
    %(name)s, %(name)s, %(name)s, a %4.8d deaths were not enough for %(name)s 
""".splitlines()

for t in tests:
    print(t)
    for t,s,e in interp_expr.scanString(t):
        print("start: {} end: {}\n{}".format(s, e, t.dump()))
    print()

Prints:

Now is the winter of our %s made %(adjective)s %(season)s by this %(relative)s of %(city)s
start: 29 end: 31
['%', 's']
- conversion_type: 's'
start: 37 end: 50
['%', '(', 'adjective', ')', 's']
- conversion_type: 's'
- mapping_key: 'adjective'
start: 51 end: 61
['%', '(', 'season', ')', 's']
- conversion_type: 's'
- mapping_key: 'season'
start: 70 end: 82
['%', '(', 'relative', ')', 's']
- conversion_type: 's'
- mapping_key: 'relative'
start: 86 end: 94
['%', '(', 'city', ')', 's']
- conversion_type: 's'
- mapping_key: 'city'

%(name)s, %(name)s, %(name)s, a %4.8d deaths were not enough for %(name)s 
start: 4 end: 12
['%', '(', 'name', ')', 's']
- conversion_type: 's'
- mapping_key: 'name'
start: 14 end: 22
['%', '(', 'name', ')', 's']
- conversion_type: 's'
- mapping_key: 'name'
start: 24 end: 32
['%', '(', 'name', ')', 's']
- conversion_type: 's'
- mapping_key: 'name'
start: 36 end: 41
['%', 4, '.', 8, 'd']
- conversion_type: 'd'
- max_width: 8
- min_width: 4
start: 68 end: 76
['%', '(', 'name', ')', 's']
- conversion_type: 's'
- mapping_key: 'name'

If you prefer regex, it should not be hard to convert this prototype to something workable.

Collectives™ on Stack Overflow

Parser for printf-style format strings

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related