In a mail merge application I use the .replace() method to replace field identifiers by custom values and also in a reverse process to get the identifiers back.
The first way works every time since the replace first argument is a pretty normal string that I have chosen on purpose... but when I reverse the process it happens sometimes that the string contains incorrect regular expression characters.
This happens mainly on phone numbers in the form +32 2 345 345 or even with some accentuated characters.
Given I can't prevent this from happening and that I have little hope that my endusers won't use this phone number format I was wondering if someone could suggest a workaround to escape illegal characters when they come up ? note : it can be at any place in the string.
below is the code for both functions.
... (partial code)
var newField = ChampSpecial(curData,realIdx,fctSpe);// returns the value from the database
if(newField!=''){replacements.push(newField+'∏'+'#ch'+(n+1)+'#')};
//Logger.log('value in '+n+'='+realIdx+' >> '+Headers[realIdx]+' = '+ChampSpecial(curData,realIdx,fctSpe))
app.getElementById('textField'+(n+1)).setHTML(ChampSpecial(curData,realIdx,fctSpe));
if(e.parameter.source=='insertInText'){
body.replaceText('#ch'+(n+1)+'#',newField);
}
}
UserProperties.setProperty('replacements',replacements.join('|'));
cloakOn();
colorize('#ffff44');
return app;
}
function fieldsInDoc(e){
cloakOff();// remet d'abord les champs vides
var replacements = UserProperties.getProperty('replacements').split('|');
var doc = DocumentApp.getActiveDocument();
var body = doc.getBody();
for(var n=0;n<replacements.length;++n){
var field = replacements[n].split('∏')[1];
var testVal = replacements[n].split('∏')[0];
body.replaceText(testVal,field);
}
colorize('#ffff44');
}
In the reverse process you are using the fieldvalues provided that can include regex special characters. you have to escape them before replacing:
body.replaceText(field.replace(/[[\]{}()*-+?.,\\^$|#\s]/, '\\$&'), '#ch'+(n+1)+'#');
This said, the "replace back the markers" a bad idea. What happens if two fields of the mail merge have the same value or the replacement text is already present in the document template...
One possible solution was to prevent the example fields in the doc from containing regex special characters so the replace had to occur in the forward process, not in the reverse (as suggested in the other answer).
Escaping these character in the fields values didn't work* so I ended up with a simple replacement by a hyphen (which make sense in most cases to replace a slash or a '+').
(*) the reverse process uses the value kept in memory so the escape sign was disturbing the replace in that function, preventing it to work properly.
the final working code goes simply like this :
//(in the first function)
var newField = ChampSpecial(curData,realIdx,fctSpe).replace(/([*+?^=!:${}()|\[\]\/\\])/g, "-");// replace every occurrence of *+?^... by '-' (global search)
About the comment stating that this approach is a bad idea I can only say that I'm afraid there is not really other ways to get that behavior and that the probability to get errors if finally quite low since the main usage of mail merge is to insert proper names, adresses, emails and phone numbers that are rarely in the template itself.
As for the field indicators they will never have the same name since they are numerically indexed (#chXX#).
EDIT : following Taras's comment I'll try another solution, will update later if it works as expected.
EDIT June 19 , Yesssss... found it.
I finally found a far better solution that doesn't use regular expression so I'm not forced to escape special characters ... the .find() method accepts any string.
The code is a bit more complex but the results is worth the pain :-))
here is the full code in 2 functions if ever someone looks for something similar.
function valuesInDoc(e){
var lock = LockService.getPrivateLock(); // just in case one clicks the second button before this one ends
var success = lock.tryLock(5000);
if (!success) {
Logger.log('tryLock failed to get the lock');
return
}
colorize('#ffffff');// this function removes the color tags on the field marlers
var app = UiApp.getActiveApplication();
var listVal = UserProperties.getProperty('listSel').split(',');
var replacements = [];
var doc = DocumentApp.getActiveDocument();
var body = doc.getBody();
var find = body.findText('#ch');
if(find == null){return app };
var curData = UserProperties.getProperty('selItem').split('|');
var Headers = [];
var OriHeaders = UserProperties.getProperty('Headers').split('|');
for(n=0;n<OriHeaders.length;++n){
Headers.push('#'+OriHeaders[n]+'#');
}
var fctSpe = 0 ;
for(var i in Headers){if(Headers[i].indexOf('SS')>-1){fctSpe = i}}
for(var n=0;n<listVal.length;++n){
var realIdx = Number(listVal[n]);
Logger.log(n);
var newField = ChampSpecial(curData,realIdx,fctSpe);
//Logger.log(newField);
app.getElementById('textField'+(n+1)).setHTML(ChampSpecial(curData,realIdx,fctSpe));
if(e.parameter.source=='insertInText'){
var found = body.findText('#ch'+(n+1)+'#');// look for every field markers in the whole doc
while(found!=null){
var elemTxt = found.getElement().asText();
var startOffset = found.getStartOffset();
var len = ('#ch'+(n+1)+'#').length;
elemTxt.deleteText(startOffset, found.getEndOffsetInclusive())
elemTxt.insertText(startOffset,newField);// remove the marker and write the sample value in place
Logger.log('n='+n+' newField = '+newField+' for '+'#ch'+(n+1)+'#'+' at position '+startOffset)
replacements.push(newField+'∏'+'#ch'+(n+1)+'#'+'∏'+startOffset);// memorize the change that just occured
found = body.findText('#ch'+(n+1)+'#',found); //loop until all markers are replaced
}
}
}
UserProperties.setProperty('replacements',replacements.join('|'));
cloakOn();
colorize('#ffff44');// colorize the markers if ever one is left but it shouldn't happen
lock.releaseLock();
return app;
}
function fieldsInDoc(e){
var lock = LockService.getPrivateLock();
var success = lock.tryLock(5000);
if (!success) {
Logger.log('tryLock failed to get the lock');
return
}
cloakOff();// remet d'abord les champs vides > shows the hidden fields (markers that had no sample velue in the first function
var replacements = UserProperties.getProperty('replacements').split('|');// recover replacement data as an array
Logger.log(replacements)
var doc = DocumentApp.getActiveDocument();
var body = doc.getBody();
for(var n=replacements.length-1;n>=0;n--){ // for each replacement find the data in doc and write a field marker in place
var testVal = replacements[n].split('∏')[0]; // [0] is the sample value
if(body.findText(testVal)==null){break};// this is only to handle the case one click on the wrong button trying to place markers again when they are already there ;-)
var field = replacements[n].split('∏')[1];
var testValLength = testVal.length;
var found = body.findText(testVal);
var startOffset = found.getStartOffset();
Logger.log(testVal+' = '+field+' / start: '+startOffset+' / Length: '+ testValLength)
var elemTxt = found.getElement().asText();
elemTxt.deleteText(startOffset, startOffset+testValLength-1);// remove the text
// elemTxt.deleteText(startOffset, found.getEndOffsetInclusive() )
elemTxt.insertText(startOffset,field);// and write the marker
}
colorize('#ffff44'); // colorize the marker
lock.releaseLock();// and release the lock
}
