tiny util method to verify if a text had an issue with encodings. Only works for texts with a high incidence of the issue.

isProbablyWrongEncoded.java

public static boolean isProbablyWrongEncoded(String s){
        int charNum = s.length();
        float wordCount = 1.0f;
        float auxCharCount =0.0f;
        float ctrlCharCount =0.0f;

        for (int i =0; i < charNum; i++) {
            int cp = Character.codePointAt(s, i);
            if (s.charAt(i) == ' ') {
                wordCount ++;
                continue;
            }

            if (cp >= 0x00A1 && cp<=0x00FF) {
                auxCharCount ++;
            } else if (cp >= 0x0080 && cp< 0x00A1) {
                ctrlCharCount ++;
            }
        }

        float latinDensity = auxCharCount/charNum;
        boolean containsTooManyExtenderChars = 0.1 < latinDensity;
        boolean mayContainTooManyControlChars = 0.01 < (ctrlCharCount / charNum);
        boolean containsTooManyControlChars = 0.03 <= (ctrlCharCount / charNum);
        boolean isSuspiciousLen =  12 < charNum / wordCount;

        return  containsTooManyExtenderChars
                || containsTooManyControlChars
                || (mayContainTooManyControlChars && isSuspiciousLen);
    }