PDFExtract.json

{
"script" : {
	"sentence_join" : "/home/usr/sentence-join/sentence-join.py",
	"kenlm_path" : "/home/usr/kenlm/bin"
}, 
"language" : 
[
	{
		"name" : "common",
		"config" : {
			"join_words" : [
				[".*[\\,\\&\\;\\:]$", "", " "],
				[".*[a-z]+\\-$", "^[a-z]+.*", ""],
				[".*[a-z]{1,}$", "^[a-z]+.*", " "],
				[".*[\\,\\;\\s][A-Z]{1,1}$", "", " "],
				[".*\\s(to|for|at|by)$", "", " "]
			],
			"absolute_eof" : [
				[".*(?<!\\,|\\&|\\;|\\:|\\s[A-Z]{1,1})$", "^[0-9 ]{0,}[A-Z]+.*"],
				[".*(\\?\\\"?|\\!\\\"?)$", ""],
				[".*\\w\\.$", ""],
				["", "^[•]+.*"]
			],
			"normalize" : [
				[ "ff", "ff"],
				["fi\\s?", "fi"],
				["fl\\s?", "fl"],
				["ffi", "ffi"],
				["ffl", "ffl"],
				["ſt", "ft"],
				["st", "st"],
				["[“”]", "\""],
				["[’´]", "'"],
				["…", "..."],
				["–", "-"],
				[" ", ""],
				["יִ", "!"],
				[" ", " "]
			],
			"repair" : [
				["\\s(\\,|\\)|\\]|\\;)", "$1"],
				["(\\(\\[)\\s", "$1"],
				["([^\\.])\\s(\\.)", "$1$2"]
			]
		}
	},
	{
		"name" : "en",
		"config" : {
			"sentencejoin_model" : "/home/usr/models/toy-model",
			"join_words" : [
			],
			"absolute_eof" : [
			],
			"normalize" : [
			],
			"repair" : [
			]
		}
	}
	
]
}