1 #!/bin/bash

     2 

     3 # Relational pipes

     4 # Copyright © 2020 František Kučera (Frantovo.cz, GlobalCode.info)

     5 #

     6 # This program is free software: you can redistribute it and/or modify

     7 # it under the terms of the GNU General Public License as published by

     8 # the Free Software Foundation, version 3 of the License.

     9 #

    10 # This program is distributed in the hope that it will be useful,

    11 # but WITHOUT ANY WARRANTY; without even the implied warranty of

    12 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the

    13 # GNU General Public License for more details.

    14 #

    15 # You should have received a copy of the GNU General Public License

    16 # along with this program. If not, see <http://www.gnu.org/licenses/>.

    17 

    18 

    19 # This streamlet provides PDF metadata of given files. It calls the tool pdfinfo.

    20 # With no options it returns just number of pages (or 0 if the file is not a PDF).

    21 # Specific attributes can be selected using options – e.g. --option 'attribute' 'Author'

    22 # List of available attributes can be obtained by directly calling the pdfinfo command on a PDF file.

    23 

    24 

    25 . "$(dirname $0)/streamlet-common.sh"

    26 

    27 processMessage_WAITING_FOR_OUTPUT_ATTRIBUTES_METADATA() {

    28 	pdfFields=()

    29 

    30 	for (( i=0; i<${#optionNames[@]}; i++)); do

    31 		if [[ "x${optionNames[$i]}" == "xattribute" ]]; then

    32 			pdfFields+=("${optionValues[$i]}");

    33 		elif [[ "x${optionNames[$i]}" == "xprefix" ]]; then

    34 			pdfPrefix="${optionValues[$i]}";

    35 		else

    36 			echo "Unsupported option: ${optionNames[$i]}" >&2

    37 		fi

    38 	done

    39 

    40 	if [[ -z "$pdfFields" ]]; then

    41 		pdfFields=( "Pages" );

    42 	fi

    43 

    44 	for (( i=0; i<${#pdfFields[@]}; i++)); do

    45 		if [[ "x${pdfFields[$i]}" == "xPages" ]]; then local type="integer"; else local type="string"; fi

    46 		send OUTPUT_ATTRIBUTE_METADATA "${outputAttributeAliases[$i]-$pdfPrefix${pdfFields[$i]}}"    "$type"

    47 	done

    48 

    49 	send WAITING_FOR_INPUT_ATTRIBUTES

    50 }

    51 

    52 processMessage_WAITING_FOR_OUTPUT_ATTRIBUTES() {

    53 	local pdfInfo pdfValid value isNull;

    54 

    55 	[[ -d "$currentFile" ]] || pdfInfo="$(pdfinfo -isodates "$currentFile")";

    56 	pdfValid="$?";

    57 

    58 	for (( i=0; i<${#pdfFields[@]}; i++)); do

    59 		value="$(echo "$pdfInfo" | grep -P "^\Q${pdfFields[$i]}\E:" | sed -E 's/[^:]+:\s+(.*)/\1/g' | tr -d '\n';)"; # the field name must not contain "\E"

    60 

    61 		if  ([[ ! "x$pdfValid" == "x0" ]] || [[ "x$value" == "x" ]]) && [[ "x${pdfFields[$i]}" == "xPages" ]]; then value="0"; isNull="true";

    62 		elif [[ ! "x$pdfValid" == "x0" ]];                                                                     then value="";  isNull="true";

    63 		else                                                                                                                   isNull="false";

    64 		fi

    65 

    66 		send OUTPUT_ATTRIBUTE "$value"    "$isNull";

    67 	done

    68 	

    69 	send WAITING_FOR_INPUT_ATTRIBUTES;

    70 }

    71 

    72 initialize

    73 processMessages