Import of Word Documents to ServiceNow Knowledge base

[welterp]

I am trying to import Word Documents to ServiceNow Knowledge Articles with out losing any images. I have tried using the Word Cleaner and SED installs without any success. Is there any fix that works?

[Goran WitchDoc]

Hi,

Im trying to get this to work in fuji as well. We have eureka now and its working like a charm there. I have my dev-instance with fuji where I try to get it to work as well. I stumple into an error that is above my knowledge and looking for some help.

If I Use the code and comment the whole part with images it works like it should, but if I try to uncomment it and run it, it doesnt work.

I get on the inbound mail: Skipping 'Create KB', did not create or update kb_knowledge

And looking in the log it says:

worker.0 WARNING *** WARNING *** Javascript compiler exception: The undefined value has no properties. (sysevent_in_email_action.f591746d4f6f4600028f7ab28110c7a6; line 25) in:

var gdt = new GlideDate();

  gdt.addDays(2);

  //current.valid_to = gdt;

  var desc = email.subject;

  desc = desc.replace("Knowledge:","");

current.kb_knowledge_base = 'a7e8a78bff0221009b20ffffffffff17';  

  current.short_description = desc.trim();

  current.workflow_state = "draft";

  current.topic = "General";

  current.roles = "knowledge";

  //Find and replace the image tags with the proper source.  

  //Get the number of attachments so the loop can be ex...

Here is the script:

var gdt = new GlideDate();

      gdt.addDays(2);

      //current.valid_to = gdt;

      var desc = email.subject;

      desc = desc.replace("Knowledge:","");

      current.kb_knowledge_base = 'a7e8a78bff0221009b20ffffffffff17';    

      current.short_description = desc.trim();

      current.workflow_state = "draft";

      current.topic = "General";

      current.roles = "knowledge";

      //Find and replace the image tags with the proper source.      

      //Get the number of attachments so the loop can be exited      

      //so it will stop no mater what.      

      var currentCount = 0;

      var newBody = email.body_html;

      var emailuid = email.uid;

      var searchBody = email.body_html.replace(/\n/g, " ");

//If I comment the all the below beside the last 3 lines it works.

      var regex = /<img(.*?)>/ig;

      var match;

    var match2;

              while(match == regex.exec(searchBody)){

          //Add a style float tag next to the align tag.

          var alignText = match[0].replace(/align=['"]?left['"]?/gi, 'align="left" style="FLOAT: left"');

          alignText = alignText.replace(/align=['"]?right['"]?/gi, 'align="right" style="FLOAT: right"');

          searchBody = searchBody.replace(match[0], alignText);

          var regex2 = /src=("(.)*?"|'(.)*?'|(.)*?\s+$)?/ig;

          currentCount += 1;

          if(currentCount >= 100)

          break;

      }

      searchBody = searchBody.replace(/<o:p>(.*?)<\/o:p>/ig, "");

      currentCount = 0;

      var regex2 = /<!--\[if(.*?)<!\[endif\]-->/ig;

              while(match == regex2.exec(searchBody)){

          searchBody = searchBody.replace(match[0], "");

          currentCount += 1;

          if(currentCount >= 100)

          break;

      }

      currentCount = 0;

      var regex2 = /<!\[if !vml\]>(.*?)<!\[endif\]>/ig;

              while(match == regex2.exec(searchBody)){

          searchBody = searchBody.replace(match[0], match[1]);

          currentCount += 1;

          if(currentCount >= 100)

          break;

      }

      current.text = searchBody;

      current.insert();

      event.state="stop_processing";

[treed]

Where did your sys_id come from in the line:

current.kb_knowledge_base = 'a7e8a78bff0221009b20ffffffffff17';

I ran into the same error and was using the wrong ID.   The key I used is under Knowledge Base -> Administration -> Properties

[Goran WitchDoc]

I tried both that sys_id which is the "knowledge" knowledge base and I tried testing with another one as well.

the error lies must somewhere in the attachment code part since if I comment out the part below. the script works and I get knowledge documents made. But if I uncomment this part the error appears.

      var regex = /<img(.*?)>/ig;

      var match;

    var match2;

              while(match == regex.exec(searchBody)){

          //Add a style float tag next to the align tag.

          var alignText = match[0].replace(/align=['"]?left['"]?/gi, 'align="left" style="FLOAT: left"');

          alignText = alignText.replace(/align=['"]?right['"]?/gi, 'align="right" style="FLOAT: right"');

          searchBody = searchBody.replace(match[0], alignText);

          var regex2 = /src=("(.)*?"|'(.)*?'|(.)*?\s+$)?/ig;

          currentCount += 1;

          if(currentCount >= 100)

          break;

      }

      searchBody = searchBody.replace(/<o:p>(.*?)<\/o:p>/ig, "");

      currentCount = 0;

      var regex2 = /<!--\[if(.*?)<!\[endif\]-->/ig;

              while(match == regex2.exec(searchBody)){

          searchBody = searchBody.replace(match[0], "");

          currentCount += 1;

          if(currentCount >= 100)

          break;

      }

      currentCount = 0;

      var regex2 = /<!\[if !vml\]>(.*?)<!\[endif\]>/ig;

              while(match == regex2.exec(searchBody)){

          searchBody = searchBody.replace(match[0], match[1]);

          currentCount += 1;

          if(currentCount >= 100)

          break;

      }

[Goran WitchDoc]

I found the error.

I needed to put these:

var match;

var match2;

as

var match = [];

    var match2 = [];

[DrewW]

The reason you had to do this is because of this line

while(match == regex.exec(searchBody)){

The original is

while(match = regex.exec(searchBody)){

Only 1 equals sign.