jsonofabitch 0.3.1

JSOB: Fault tolerant JSON parser. A standardization of low standards. Be a JSLOB.

JSONofabitch

• fault tolerant json parser
• JSON correction/formatting engine
• stochastic ascii structured data format
• a standard for low standards
• the retard proof multi-parameter, multi-type input parser
• a troll of api scraping reverse engineers
• a JSON of a bitch!

functions

• loads() - loads JSLOB compliant text
• dumps() - dumps regular JSON
• correct() - turns JSLOB to JSON
• dumpslob() - dumps stochastic JSLOB

install

pip3 install jsonofabitch

build requires lark but jsonofabitch is stand-alone

JSLOB syntax

• keypairs - : or = separate key value pairs, whitespace is ignored 'g=5' -> {"g": 5} and 'g:5' -> {"g": 5}, such keypairs are implicitly assumed to be in a JSLOB object which resolves to a python dict
• keys - double quoted string or alphanumeric plus underscore, non-numeric first character(same as python vars).
• values - another JSLOB object, a tupple, a list, alphanumeric string without quotes, quote-enclosed string, boolean true false , null , integers, floats
• 'true' -> True 'false' -> False 'null' -> None case sensitive
• tuples in ( ) lists in [ ] objects separated by comma , or semicolon ; . trailing comma/semicolon okay
• dicts/JSLOB objects have keypairs separated by semicolon or comma, nested jslob objects(in lists, tuples or jslob objects) are terminated and separated simultaneously by ;; or ,, , IE: '[g=5,h=3]' -> [{'g': 5, 'h': 3}] and '[g=5;;h=3]' -> [{'g': 5}, {'h': 3}]
• other - unquoted strings must be alphanumeric plus _ non-numeric first character. only double quotes can be used
• scientific notation 2e4 resolves to $2.0 * 10^4$, always resolve to float
• numbers with decimal present resolve to floats while others resolve to ints IE '3.0' -> 3.0 and '3' -> 3 '3.' -> 3.0

about/conventions

• the parser is called JSOB(jay-sawb) in polite company
• the standardized syntax is called JSLOB (jay-slawb)
• suggested file extension .jslob or .slob or .jsob
• the underlying parser is built using lark which generates stand-alone python from the lark meta-langauge that defines the grammar

changelog

• version 0.3.0 transitioned from using the parseing library lark at runtime to a stand-alone parser generated by lark at build

examples

import jsonofabitch as jsob
d = jsob.loads("f=5, size=big, style=SLOB, coord=[(3,4),(1.2,3)]")
#{'f': 5, 'size': 'big', 'style': 'SLOB', 'coord': [(3, 4), (1.2, 3)]}

jsob.dumps(d)
#Out[3]: '{"f": 5, "size": "big", "style": "SLOB", "coord": [[3, 4], [1.2, 3]]}'

jsob.dumpslob(d)
#Out[4]: ' "f"=  5;  "size"="big";style=  "SLOB";  coord: [ (3 ,4 ), (1.2, 3,)  ];'

jsob.dumpslob(d)
#Out[5]: ' "f"= 5;size="big", style: SLOB, "coord": [ (3, 4 ),(1.2 , 3  ) ];'

jsob.dumpslob(d)
#Out[6]: ' "f"= 5,  size=  "big",style=  "SLOB";  coord:  [ (3,4 ,) ,(1.2 , 3,)  ];'