corona

#!/bin/bash

POSIXLY_CORRECT=yes

filters() {
    sed -r 's/\s+//g' | sed '/^[[:space:]]*$/d' | validation | awk -F, -v DATEAFTER=$DATEAFTER -v DATEBEFORE=$DATEBEFORE -v GENDER=$GENDER '{if (substr($1,1,7)=="Invalid") {print; next;}}{   
    if (DATEAFTER != "") {
        if ($2 >= DATEAFTER) {
            print;
            next;
        } else {
            next;
        }
    }
    if (DATEBEFORE != "") {
        if ($2 <= DATEBEFORE) {
            print;
            next;
        } else {
            next;
        }
    }        
    if (GENDER != "") {
        if($4 == GENDER) {
            print;
            next;
        } else {
            next;
        }
    }    

    {print}}'
}

correctDATE() {
    if [[ $1 =~ ^[0-9]{4}-[0-1]{1}[0-9]{1}-[0-3]{1}[0-9]{1}$ ]] && (date "+%Y-%m-%d" -d "$1") 2> /dev/null 1> /dev/null ; then 
            return 0
        else 
            return 1
    fi
}

removeHeaders() {
    sed '/'"id,datum,vek,pohlavi,kraj_nuts_kod,okres_lau_kod,nakaza_v_zahranici,nakaza_zeme_csu_kod,reportovano_khs"'/d'
}

detectNonExistentFile() {
    
    if ! [ -f $1 ] ;
        then
            echo "Neexistující soubor $1";
            exit 1;
    fi        
} 

validation() {
    awk -F, '{
    if ($3 !~ /^[0-9]*$/) {
        print "Invalid age: " $0;
        next;
    }
    if ($2 !~ /^[0-9][0-9][0-9][0-9]-[0-1][0-9]-[0-3][0-9]$/ && $2 != "") {
        print "Invalid date: " $0;
        next;
    }
    month = substr($2, 6, 2);
    day = substr($2, 9, 10);
    if (month != "00" && month <= 12 && day != "00" && day <= 31) {
        if (month == "04" || month == "06" || month == "09" || month == "11") {
            if (day == "31") {
                print "Invalid date: " $0;
                next;
            }
        } else if (month == "02") {
                if (day > 29) {
                    print "Invalid date: " $0;
                    next;
                }
                year = substr($2, 1, 4);
                if (year % 400 == 0) {

                    } else if (year % 100 == 0) {
                        if (day > 28) {
                            print "Invalid date: " $0;
                            next;
                            }
                        } else if (year % 4 == 0) {

                            } else {
                                if (day > 28) {
                                print "Invalid date: " $0;
                                 next;
                                 }
                            }
            }
    } else {print "Invalid date: " $0;
       next;}
print}'
}

InvalidDataToError() {
    awk -F: '{if (substr($1,1,7)=="Invalid") {print > "/dev/stderr"; next;}}{print}'
}

histogram() {
    if [ $WIDTH == 0 ] ; then
    awk -F: -v HASH_COUNT=$1 '{if (substr($1,1,7)=="Invalid") {print; next;}}
    {printf $1": ";
    VALUE=$2; 
    HASH=VALUE/HASH_COUNT;
    for (i=1; i <= HASH; i++)
        printf "#";
    print "";}'
    else
            awk -F: -v WIDTH=$WIDTH '{if (substr($1,1,7)=="Invalid") {print; next;}}
    { if (NR ==1) MAX_VALUE=substr($2,2);
        printf $1": ";
    VALUE=$2; 
    HASH=WIDTH/(MAX_VALUE/VALUE);
    for (i=1; i <= HASH; i++)
        printf "#";
    print "";}'
    fi
}

gender() {
    filters | awk -F, '{if (substr($1,1,7)=="Invalid") {print; next;}}
    BEGIN{count_M=0; count_Z=0; count_NONE=0;} {
    if ($4 == "M") {count_M+=1} else if ($4 == "Z") {count_Z+=1} else if ($4 == "") {count_None+=1}
    }END{{print "M: "count_M"\nZ: "count_Z} if (count_NONE != 0) {print "None: "count_NONE}}'
}

age() {
    filters | awk -F, '{if (substr($1,1,7)=="Invalid") {print; next;}}
    BEGIN{C_5=0; C_15=0; C_25=0; C_35=0; C_45=0; C_55=0; C_65=0; C_75=0; C_85=0; C_95=0; C_105=0; C_N=0; C_NONE=0;} {
    if ($3 >= 0 && $3 <= 5) {C_5+=1} else if ($3 >= 6 && $3 <= 15) {C_15+=1} else if ($3 >= 16 && $3 <= 25) {C_25+=1} else 
    if ($3 >= 26 && $3 <= 35) {C_35+=1} else if ($3 >= 36 && $3 <= 45) {C_45+=1} else if ($3 >= 46 && $3 <= 55) {C_55+=1} else 
    if ($3 >= 56 && $3 <= 65) {C_65+=1} else if ($3 >= 66 && $3 <= 75) {C_75+=1} else if ($3 >= 76 && $3 <= 85) {C_85+=1} else 
    if ($3 >= 86 && $3 <= 95) {C_95+=1} else if ($3 >= 96 && $3 <= 105) {C_105+=1} else if ($3 >= 106) {C_N+=1} else if ($3 >= "") {C_NONE+=1} 
    }END{{print "0-5   : "C_5"\n6-15  : "C_15"\n16-25 : "C_25"\n26-35 : "C_35"\n36-45 : "C_45"\n46-55 : "C_55"\n56-65 : "C_65"\n66-75 : "C_75"\n76-85 : "C_85"\n86-95 : "C_95"\n96-105: "C_105"\n>105  : "C_N} if (C_NONE != 0) {print "None  : "C_NONE}}'
}

daily() {
    filters | sort -t, -k2 | awk -F, '{if (substr($1,1,7)=="Invalid") {print; next;}}
    BEGIN{count_DAY=0; day=""; count_NONE=0;}
    {if (NR==1) {day=$2}}
    {if ($2=="") {count_NONE+=1;}}

    {if (day==$2) {count_DAY+=1;}
    else { {if (day != "") print day ": " count_DAY;} count_DAY=1; day=$2;}} 

    END{{if (day != "") {print day ": " count_DAY}} if (count_NONE > 0) print "None: " count_NONE;}'
}

monthly() {
    filters | sort -t, -k2 | awk -F, '{if (substr($1,1,7)=="Invalid") {print; next;}}
    BEGIN{count_MONTH=0; month=""; count_NONE=0;}
    {if (NR==1) {month=substr($2,1,7)}}
    {if ($2=="") {count_NONE+=1;}}

    {if (month==substr($2,1,7)) {count_MONTH+=1;}
    else { {if (month != "") print month ": " count_MONTH;} count_MONTH=1; month=substr($2,1,7);}} 

    END{{if (month != "") {print month ": " count_MONTH}} if (count_NONE > 0) print "None: " count_NONE;}'
}

yearly() {
    filters | sort -t, -k2 | awk -F, '{if (substr($1,1,7)=="Invalid") {print; next;}}
    BEGIN{count_YEAR=0; year=""; count_NONE=0;}
    {if (NR==1) {year=substr($2,1,4)}}
    {if ($2=="") {count_NONE+=1;}}

    {if (year==substr($2,1,4)) {count_YEAR+=1;}
    else { {if (year != "") print year ": " count_YEAR;} count_YEAR=1; year=substr($2,1,4);}} 

    END{{if (year != "") {print year ": " count_YEAR}} if (count_NONE > 0) print "None: " count_NONE;}'
}

countries() {
    filters | sort -t, -k8 | awk -F, '{if (substr($1,1,7)=="Invalid") {print; next;}}
    BEGIN{count_COUNTRY=0; country=""; count_NONE=0;}
    {if (NR==1) {country=$8}}
    {if ($8=="") {count_NONE+=1;}}

    {if (country==$8) {count_COUNTRY+=1;}
    else { {if (country != "" && country != "CZ") print country ": " count_COUNTRY;} count_COUNTRY=1; country=$8;}} 

    END{if (country != "")  print country ": " count_COUNTRY;}'
}

districts() {
    filters | sort -t, -k6 | awk -F, '{if (substr($1,1,7)=="Invalid") {print; next;}}
    BEGIN{count_DISTRICT=0; district=""; count_NONE=0;}
    {if (NR==1) {district=$6}}
    {if ($6=="") {count_NONE+=1;}}

    {if (district==$6) {count_DISTRICT+=1;}
    else { {if (district != "") print district ": " count_DISTRICT;} count_DISTRICT=1; district=$6;}} 

    END{{if (district != "") {print district ": " count_DISTRICT}} if (count_NONE > 0) print "None: " count_NONE;}'
}

regions() {
    filters | sort -t, -k5 | awk -F, '{if (substr($1,1,7)=="Invalid") {print; next;}}
    BEGIN{count_REGION=0; region=""; count_NONE=0;}
    {if (NR==1) {region=$5}}
    {if ($5=="") {count_NONE+=1;}}

    {if (region==$5) {count_REGION+=1;}
    else { {if (region != "") print region ": " count_REGION;} count_REGION=1; region=$5;}} 

    END{{if (region != "") {print region ": " count_REGION}} if (count_NONE > 0) print "None: " count_NONE;}'
}

unzip() {

    if [[ $1 =~ ^.*\.gz$ ]] ;
        then
            gunzip -c "$1"
            return
    fi

        if [[ $1 =~ ^.*\.bz2$ ]] ;
        then
            bunzip2 -c "$1"
            return
    fi

    if [[ $1 =~ ^.*\.csv$ && $1!=""  ]] ;
        then
            cat $1
            return
    fi

    if [[ $1 == "" ]] ;
        then
            cat $1
            return
    fi        
}

while getopts :ha:b:g:s o
do 

case "$o" in
		h) echo "NÁPOVĚDA:"
        echo ""
        echo "Použití: corona [-h] [FILTERS] [COMMAND] [LOG [LOG2 [...]]"
        echo ""
        echo "FILTERS může být kombinace následujících (každý maximálně jednou):
-a DATETIME — after: jsou uvažovány pouze záznamy PO tomto datu (včetně tohoto data). DATETIME je formátu YYYY-MM-DD.
-b DATETIME — before: jsou uvažovány pouze záznamy PŘED tímto datem (včetně tohoto data).
-g GENDER — jsou uvažovány pouze záznamy nakažených osob daného pohlaví. GENDER může být M (muži) nebo Z (ženy).
-s [WIDTH] u příkazů gender, age, daily, monthly, yearly, countries, districts a regions vypisuje data ne číselně, ale graficky v podobě histogramů. Nepovinný parametr WIDTH nastavuje šířku histogramů, tedy délku nejdelšího řádku, na WIDTH. Tedy, WIDTH musí být kladné celé číslo. Pokud není parametr WIDTH uveden, řídí se šířky řádků požadavky uvedenými níže."
echo ""
echo "COMMAND může být jeden z:
infected — spočítá počet nakažených.
merge — sloučí několik souborů se záznamy do jednoho, zachovávající původní pořadí (hlavička bude ve výstupu jen jednou).
gender — vypíše počet nakažených pro jednotlivá pohlaví.
age — vypíše statistiku počtu nakažených osob dle věku (bližší popis je níže).
daily — vypíše statistiku nakažených osob pro jednotlivé dny.
monthly — vypíše statistiku nakažených osob pro jednotlivé měsíce.
yearly — vypíše statistiku nakažených osob pro jednotlivé roky.
countries — vypíše statistiku nakažených osob pro jednotlivé země nákazy (bez ČR, tj. kódu CZ).
districts — vypíše statistiku nakažených osob pro jednotlivé okresy.
regions — vypíše statistiku nakažených osob pro jednotlivé kraje."
        exit;;

		a)
        correctDATE "$OPTARG";
        if [ $? == 1 ] ; then
            echo "DATE \"$OPTARG\" is invalid in filter -a! | CORRECT DATE FORMAT: YYYY-MM-DD" 1>&2
            exit
        else
            DATEAFTER=$OPTARG
        fi;;

		b)
        correctDATE "$OPTARG";
        if [ $? == 1 ] ; then
            echo "DATE \"$OPTARG\" is invalid in filter -b! | CORRECT DATE FORMAT: YYYY-MM-DD" 1>&2
            exit
        else
            DATEBEFORE=$OPTARG
        fi;;

		g)
        if [[ $OPTARG == "M" || $OPTARG == "Z" ]] ; then
            GENDER=$OPTARG
        else     
            echo "GENDER \"$OPTARG\" is invalid in filter -g! | AVAILABLE GENDERS: M/Z" 1>&2
            exit
        fi;;   

		s) HISTOGRAMS=1
        eval WIDTH=\${$OPTIND}
            if [[ $WIDTH != -* ]] ; then
                if [[ -n $WIDTH ]] ; then
                    if [[ !($WIDTH =~ ^[0-9]+$) ]] ; then
                        WIDTH=0
                        else OPTIND=$((OPTIND + 1))
                    fi
                else
                    WIDTH=0
                fi
            fi;;
		*) echo "Neznámý přepínač, viz ./corona -h" 1>&2
        exit;;
	esac

done

((OPTIND--))
shift $OPTIND

if [ "$1" = "infected" ] || [ "$1" = "merge" ] || [ "$1" = "gender" ] || [ "$1" = "age" ] || [ "$1" = "daily" ] || [ "$1" = "monthly" ] || [ "$1" = "yearly" ] || [ "$1" = "countries" ] || [ "$1" = "districts" ] || [ "$1" = "regions" ]
then
    COMMAND=$1
fi

if [ "$2" = "infected" ] || [ "$2" = "merge" ] || [ "$2" = "gender" ] || [ "$2" = "age" ] || [ "$2" = "daily" ] || [ "$2" = "monthly" ] || [ "$2" = "yearly" ] || [ "$2" = "countries" ] || [ "$2" = "districts" ] || [ "$2" = "regions" ]
then
    echo "WRONG INPUT! USAGE: corona [-h] [FILTERS] [COMMAND] [LOG [LOG2 [...]]" 1>&2
    exit 1
fi

if [[ "$COMMAND" == merge || -z "$COMMAND" ]] ;
then
    
    if [[ -z "$COMMAND" ]] ;
    then 
        FILE="$1"
    else
        FILE="$2"   
    fi    

    if [[ $FILE == "" ]] ; then
    TEXT="unzip $FILE"
    fi


    while [[ $FILE != "" ]]
    do
        if [[ !($FILE =~ ^.*\.csv$) && !($FILE =~ ^.*\.gz$) && !($FILE =~ ^.*\.bz2$) && $FILE != "" ]] ;
            then
                echo "File \"$FILE\" has wrong format!" 1>&2
                exit
        fi

        detectNonExistentFile "$FILE";

        if [[ -z "$TEXT" ]] ;
        then
            TEXT="unzip $FILE"
        else 
            TEXT="$TEXT ; unzip $FILE"
        fi


        shift    
    if [[ -z "$COMMAND" ]] ;
    then 
        FILE="$1"
    else
        FILE="$2"   
    fi    
    done

    echo "id,datum,vek,pohlavi,kraj_nuts_kod,okres_lau_kod,nakaza_v_zahranici,nakaza_zeme_csu_kod,reportovano_khs";
    eval $TEXT | removeHeaders | filters | awk -F, '{if (substr($1,1,7)=="Invalid") {print > "/dev/stderr"; next;}}{print}'

else
    FILE="$2"

    if [[ $FILE == "" ]] ; then
    TEXT="unzip $FILE"
    fi

    while [[ $FILE != "" ]]
    do
        if [[ !($FILE =~ ^.*\.csv$) && !($FILE =~ ^.*\.gz$) && !($FILE =~ ^.*\.bz2$) && $FILE != "" ]] ;
            then
                echo "File \"$FILE\" has wrong format!" 1>&2
                exit
        fi

        detectNonExistentFile "$FILE";

        if [[ -z "$TEXT" ]] ;
        then
            TEXT="unzip $2"
        else 
            TEXT="$TEXT ; unzip $2"
        fi

        shift
        FILE="$2"
    done 

    case "$COMMAND" in
    infected)
    eval $TEXT | removeHeaders | filters | awk -F, 'BEGIN{count=0}{if (substr($1,1,7)=="Invalid") {print; next;}}{count++}END{print count}' | InvalidDataToError;;

    gender) 
    if [[ $HISTOGRAMS == 1 ]] ; then
        eval $TEXT | removeHeaders | gender | sort -t: -k2 -g -r | histogram "100000" | sort -t: -k1 | InvalidDataToError
    else
        eval $TEXT | removeHeaders | gender | InvalidDataToError
    fi;;

    age) 
    if [[ $HISTOGRAMS == 1 ]] ; then
       eval $TEXT | removeHeaders | age | awk -F: '{print NR+9 $0}' | sort -t: -k2 -g -r | histogram "10000" | sort -t: -k1 | awk -F: '{print substr($0,3)}' | InvalidDataToError
    else
        eval $TEXT | removeHeaders | age | InvalidDataToError
    fi;;         

    daily)
    if [[ $HISTOGRAMS == 1 ]] ; then
        eval $TEXT | removeHeaders | daily | sort -t: -k2 -g -r | histogram "500" | sort -t: -k1 | InvalidDataToError
    else
        eval $TEXT | removeHeaders | daily | InvalidDataToError
    fi;;

    monthly) 
    if [[ $HISTOGRAMS == 1 ]] ; then
        eval $TEXT | removeHeaders | monthly | sort -t: -k2 -g -r | histogram "10000" | sort -t: -k1 | InvalidDataToError
    else
        eval $TEXT | removeHeaders | monthly | InvalidDataToError
    fi;;

    yearly)
    if [[ $HISTOGRAMS == 1 ]] ; then
        eval $TEXT | removeHeaders | yearly | sort -t: -k2 -g -r | histogram "100000" | sort -t: -k1 | InvalidDataToError
    else
        eval $TEXT | removeHeaders | yearly | InvalidDataToError
    fi;;

    countries)
    if [[ $HISTOGRAMS == 1 ]] ; then
        eval $TEXT | removeHeaders | countries | sort -t: -k2 -g -r | histogram "100" | sort -t: -k1 | InvalidDataToError
    else
        eval $TEXT | removeHeaders | countries | InvalidDataToError
    fi;;

    districts) 
    if [[ $HISTOGRAMS == 1 ]] ; then
        eval $TEXT | removeHeaders | districts | sort -t: -k2 -g -r | histogram "1000" | sort -t: -k1 | InvalidDataToError
    else
        eval $TEXT | removeHeaders | districts | InvalidDataToError
    fi;;

    regions) 
    if [[ $HISTOGRAMS == 1 ]] ; then
        eval $TEXT | removeHeaders | regions | sort -t: -k2 -g -r | histogram "10000" | sort -t: -k1 | InvalidDataToError
    else
        eval $TEXT | removeHeaders | regions | InvalidDataToError
    fi;;


esac
fi