0

I have file attached (Word Document) to the custom object record . I want read the file and get the content between some text and store it in the Custom object record field

apex:
ContentVersion con =[SELECT Id,ContentDocumentId,VersionData,
                              CreatedById ,TextPreview,Title,
                                 VersionNumber
                       from ContentVersion 
                       where ContentDocumentId = '0697j00000Xxxxxxxx' 
                       order by CreatedDate DESC];
String stringData = EncodingUtil.base64Encode(con.VersionData);
system.debug(stringData);
if(stringData.contains('Text')){
system.debug('if');
}else{
system.debug('Else');
}

the string is coming as encoded one after decoding also I was not able to get the original data from file

2
  • what type of file are you storing, is it a plain txt file? Aug 12 at 6:02
  • Its a word Document
    – sfdc1100
    Aug 12 at 6:05
1

Unfortunately, I don't think, that you will be able easily to read and edit a word file (.docx my assumption) by means of the native apex. It definitely contains some workarounds.

The reason for that is .docx is not just a plain text file, actually, it is an archive file with folders, subfolders, and mainly .xml files. Just try to open .docx file by means of any archive software, you'll be surprised to see something like this:

enter image description here

word here is the main folder, which actually stores the content: enter image description here

A big problem? Yes, very big, but not the end of the world. Here are steps which you need to perform:

  1. Get the blob of the file (no problems)
  2. Unarchive blob of docx file (problem, apex do not have native support to work with archive files)
  3. Find the correct .xml file of the word folder, where is stored text which you would like to edit. (could be very tricky, but pay attention to the document.xml file)
  4. Replace found text with desired.
  5. Archive this file back with done changes (problem).
  6. Save the file as a blob to the custom object record.

Which possible workarounds can we have:

(2) if your code is running only on the back-end without UI interaction, you can try to use Zippex (Video), (GitHub). This is an open-source library to work with archives in apex. If you do have UI-interaction, you can try to use jszip (How to use JSZip). It works with lwc.

(3) I recommend you manually to find peace of the file which should be replaced and to find some dependency, some specific combination of xml tags. You can use native xml apex parsers to find the location of text which should be replaced. You just need to find it and determine the rule.

(5) You can still try to use Zippex, or jszip for making an archive of the updated files.


That's it. Not very easy, but with big workarounds it is possible.

To be honest, I don't like this solution natively in Apex. It is better to find another one. For example, you can send this file to the external API, whereby means of other programming languages you can easily work with .docx files (there are already available libraries for that, python-docx for python), update it according to needs externally, and just save the response of this request in salesforce.

1
  • Thanks for the Suggestion
    – sfdc1100
    Aug 12 at 7:05

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.