The template in this sample is for JS2, but can be used for any JS step from 1 through to 7.
This template also includes code for debugging in Microsoft Visual Studio.
//Language Studio Pre-Processing Script
//Copyright 2008-2016 Omniscien Technologies Pte Ltd. All Rights Reserved.
//NOTE: This script is designed to be debugged in Microsoft Visual Studio.
// To run without the Microsoft JavaScript debugger type following command line:
// cscript default.pre.3.js
//
// To run with the Microsoft JavaScript debugger type the following command line:
// cscript default.pre.3.js //D //X
//
// No changes are necessary to run in production, simply upload and the code will autodetect that it is in the Language Studio production environment.
//IMPORTANT: There are very minor differences in RegEx between Microsoft and Oracle Java implementations.
//
//Description:
//Template - Below are the typical tasks performed in JS2
//1. Document level normalization of encoding and content formats (i.e. csv, html, ANSI in UTF-8, etc.)
//2. Basic repairs of things that would possibly cause issues in translation quality (NOTE: may be better to do most in JS3 if they do not impact TU Extract)
//3. Apply any special cases for splitting segments. (e.g. rules to handle exceptions in splitting segments)
//4. Mark content that should not be translated which will then be excluded from the TU extract step.
//
var iJSNo = 2; //The number of the JS Script
var sSourceLang = "en"; //SET YOUR SOURCE LANGUAGE
var sTargetLang = "ja"; //SET YOUR TARGET LANGUAGE
//DO NOT CHANGE
var isDebug = 0;
var isExternalTest = 0;
var logText = "";
var sLangPair = "";
//Detects if we are running in the Language Studio Enterprise runtime or development environment
var bIsDev = ((typeof AOSession) == "undefined");
if (bIsDev) {
isDebug = 1;
isExternalTest = 1;
//Set the paths to load the input files from for testing in Microsoft Visual Studio debugger.
var sTestInFilePath = currentPath() + "\\In" + iJSNo + "." + sSourceLang + ".txt";
var sTestOutFilePath = currentPath() + "\\Out" + iJSNo + "." + sSourceLang + ".txt";
}
//USER VARIABLES
//Declare your global variables here
//Add prototypes if needed
if (!String.prototype.trim) {
String.prototype.trim = function () {
return this.replace(/^\s+|\s+$/g, '');
}
}
function main(sAllSourceSegments) {
//Initialize
log("main (start)");
initializeLang();
//Initialize
setScriptNo();
//Process the all segments as 1 block of text
sAllSourceSegments = processAllSourceSegments(sAllSourceSegments);
//Process each segment line one at a time - MUCH SLOWER - TRY TO AVOID UNLESS ABSOLUTELY NECESSARY
sAllSourceSegments = processEachSegment(sAllSourceSegments);
//Cleanup and make sure none of the formatting has intruduced avoidable errors.
sAllSourceSegments = processFinalCleanup(sAllSourceSegments)
log("main (exit)");
if (logText.length > 0)
sAllSourceSegments = sAllSourceSegments + "\r\n<notran>AOSPLITLOGAO\r\n" + logText + "</notran>";
// Return output
return sAllSourceSegments;
}
//Process the entire source as 1 block of data - faster than single Segment
function processAllSourceSegments(sAllSourceSegments) {
log("processAllSourceSegments (start)");
//YOUR CODE HERE
log("processAllSourceSegments (exit)");
return sAllSourceSegments;
}
//Processes a single segment - slower as it loops through each segment
function processSegment(sSegment) {
//YOUR CODE HERE
return sSegment;
}
//-----------------------------------------------------------------
//Standard functions do not modify below this point
//-----------------------------------------------------------------
function processEachSegment(sAllSourceSegments) {
//NOTE: If at all possible, put your code in processAllSegments
//Line break must be \n not \r\n
var aSourceSegments = sAllSourceSegments.split("\n");
for (var i in aSourceSegments) {
aSourceSegments[i] = processSegment(aSourceSegments[i]);
}
sAllSourceSegments = aSourceSegments.join("\n");
return sAllSourceSegments;
}
function removeBOM(sIn) {
var sBOM = "\xef\xbb\xbf";
if (sIn.indexOf(sBOM) == 0) {
return sIn.substring(sBOM.length, sIn.length - sBOM.length);
}
else {
return sIn;
}
}
function stringCount(sIn, sCount) {
return sIn.split(sCount).length - 1;
}
function Upper(input) {
var sOut = input.toUpperCase();
return sOut;
}
function Lower(input) {
var sOut = input.toLowerCase();
return sOut;
}
function initializeLang() {
sSourceLang = sSourceLang.toLowerCase();
sTargetLang = sTargetLang.toLowerCase();
sLangPair = sSourceLang + '-' + sTargetLang;
}
function processFinalCleanup(sAllSourceSegments) {
log("processFinalCleanup (start)");
regexp = new RegExp(/(<\/notran><notran>)/g); //
sAllSourceSegments = sAllSourceSegments.replace(regexp, "");
if (!isDebug) {
//Cleanup markers
sAllSourceSegments = sAllSourceSegments.replace(RegExp(/AOFORWARDSLASHAO/gmi), '/');
sAllSourceSegments = sAllSourceSegments.replace(RegExp(/AOBACKSLASHAO/gmi), '\\');
sAllSourceSegments = sAllSourceSegments.replace(RegExp(/AOXAO/gmi), 'x');
sAllSourceSegments = sAllSourceSegments.replace(RegExp(/AOATAO/gmi), '@');
sAllSourceSegments = sAllSourceSegments.replace(RegExp(/AOATDASHATAO/gmi), '@-@');
sAllSourceSegments = sAllSourceSegments.replace(RegExp(/AODASHAO/gmi), '-');
sAllSourceSegments = sAllSourceSegments.replace(RegExp(/AOPLUSAO/gmi), '+');
sAllSourceSegments = sAllSourceSegments.replace(RegExp(/AOQUOTEAO|AODOUBLEQUOTEAO/gmi), '"');
sAllSourceSegments = sAllSourceSegments.replace(RegExp(/AOSINGLEQUOTEAO/gmi), "'");
}
sAllSourceSegments = sAllSourceSegments.replace(RegExp(/(<notran>[ ]{0,}(\-|\+)[ ]{0,}<\/notran>)/gmi), "<notran> $2 </notran>");
if (!isExternalTest) {
log("processFinalCleanup (Internal adjustments)");
//Change the content to internal formats if past JS2
if (iJSNo > 2) {
//Control XML
sAllSourceSegments = sAllSourceSegments.replace(RegExp(/<wall\/>/g), "<wall/>");
sAllSourceSegments = sAllSourceSegments.replace(RegExp(/<\/zone>/g), "</zone>");
sAllSourceSegments = sAllSourceSegments.replace(RegExp(/<zone>/g), "<zone>");
sAllSourceSegments = sAllSourceSegments.replace(RegExp(/<notrani>(.*?)<\/notrani>/g), "<aotran type=\"notran\" translation=\"$1\">aonotranao</aotran>");
sAllSourceSegments = sAllSourceSegments.replace(RegExp(/<notran>(.*?)<\/notran>/g), "<aotran type=\"notran\" translation=\"$1\">aonotranao</aotran>");
sAllSourceSegments = sAllSourceSegments.replace(RegExp(/<notran>(.*?)<\/notran>/g), "<aotran type=\"notran\" translation=\"$1\">aonotranao</aotran>");
//Multiple White Space
sAllSourceSegments = sAllSourceSegments.replace(RegExp(/[ ]{2,}/g), " ");
}
}
else {
//Check for test case markers and mark them to not be translated
//regexp = new RegExp(/(##TESTCASE)([^\r\n]*)/g);
sAllSourceSegments = sAllSourceSegments.replace(RegExp(/(##TESTCASE)([^\r\n]*)/g), "<notran>$1$2</notran>");
}
//BOM safety check
sAllSourceSegments = removeBOM(sAllSourceSegments);
log("processFinalCleanup (exit)");
return sAllSourceSegments;
}
function log(sText) {
if (isDebug == 1) {
var now = new Date();
logText += "\r\n" + formatLogDate(now) + "\t" + sText;
}
}
function formatLogDate(dDate) {
//zero-pad a single zero if needed
var zp = function (val) {
return (val <= 9 ? '0' + val : '' + val);
}
//zero-pad up to two zeroes if needed
var zp2 = function (val) {
return val <= 99 ? (val <= 9 ? '00' + val : '0' + val) : ('' + val);
}
var d = dDate.getDate();
var m = dDate.getMonth() + 1;
var y = dDate.getFullYear();
var h = dDate.getHours();
var min = dDate.getMinutes();
var s = dDate.getSeconds();
var ms = dDate.getMilliseconds();
return '' + y + '-' + zp(m) + '-' + zp(d) + ' ' + zp(h) + ':' + zp(min) + ':' + zp(s) + '.' + zp2(ms);
}
function setScriptNo() {
//Autodetect the script number from the name
if (bIsDev) {
var sNum = WScript.ScriptName.toString().replace(RegExp(/(.{0,})(\d{1,})(.{0,})/gmi), "$2");
iJSNo = parseInt(sNum);
}
else {
if (iJSNo = 0) {
iJSNo = parseInt(AOSession.getValue("ScriptNo"));
}
}
}
//WINDOWS TEST FUNCTIONS - For use with WScript
if (isExternalTest) ExternalTest();
function ExternalTest() {
if (isDebug) WScript.Echo('--Start Test--');
//Read input from a file
var sAllSourceSegments = fileToString(sTestInFilePath);
//Normalize CRLF to \n
sAllSourceSegments = sAllSourceSegments.replace(RegExp(/\r\n|\r/g), "\n");
//Execute test as a normal pre-processing step would execute it.
var sOut = main(sAllSourceSegments);
//Write the output to a file
stringToFile(sTestOutFilePath, sOut);
if (isDebug) {
if (logText.length > 0) {
WScript.Echo(logText);
}
WScript.Echo('--End Test--');
}
}
function currentPath() {
return new ActiveXObject("WScript.Shell").CurrentDirectory;
}
//Load the UTF-8 file into a string
function fileToString(sFilePath) {
//Relies on Microsoft ActiveX Data Objects (ADO) which much be installed -
var oStream = new ActiveXObject("ADODB.Stream");
oStream.CharSet = "utf-8";
oStream.Open;
oStream.LoadFromFile(sFilePath);
var sData = removeBOM(oStream.ReadText());
oStream.Close();
return sData;
}
//Write a string out to a UTF-8 file
function stringToFile(sFilePath, sData) {
//Relies on Microsoft ActiveX Data Objects (ADO) which much be installed -
var oStream = new ActiveXObject("ADODB.Stream");
oStream.CharSet = "utf-8";
oStream.Open
oStream.WriteText(sData);
oStream.SaveToFile(sFilePath, 2);
oStream.Flush();
oStream.Close();
}