# Does ETag in composition response header really need quotes?

**Category:** [ITS](https://discourse.openehr.org/c/its/41)
**Created:** 2020-03-03 11:38 UTC
**Views:** 3024
**Replies:** 10
**URL:** https://discourse.openehr.org/t/does-etag-in-composition-response-header-really-need-quotes/404

---

## Post #1 by @johnmeredith

It's all in the title but I note that the compositionUID comes back within quotes e.g. `"acf0a908-71b0-4e17-9d77-005cca119672::local.ehrbase.org::1"`. What's the reason behind the quotes as this obviously puts another step into converting this into a variable for further use.

---

## Post #2 by @sebastian.iancu

Hmm, good catch @johnmeredith . 
Actually I just realize it now, our specification is wrong, ETag should have quotes (sorry!). See https://developer.mozilla.org/en-US/docs/Web/HTTP/Headers/ETag. I will make a correction for v1.1.0 release.

Where do you saw it with quotes? Was it in a response of a specific implementation?

---

## Post #3 by @birger.haarbrandt

"acf0a908-71b0-4e17-9d77-005cca119672::**local.ehrbase.org**::1"

:wink:

---

## Post #4 by @sebastian.iancu

I thought the question was about openehr-specification. I was not sure if @johnmeredith found it somewhere in our (openEHR) or in an implementation specification. The example is obviously from EHRbase :slight_smile:
In any case my response was related to changes that needs to be done to openehr specs.

---

## Post #5 by @johnmeredith

By including the quotes in the code block I thought it was obvious that the quotes are presetn in the ETag ;-)

So ETag with quotes is to spec?

---

## Post #6 by @sebastian.iancu

Yes, there is a change request proposed to add missing quotes. That has to go through SEC workflow and eventually will be published.

---

## Post #7 by @ian.mcnicoll

It is  slightly annoying as this is the kind of data that it is nice to pull back easily in the response - perhaps also make ehr_id/value part of the minimal return (without the "")!

Ian

---

## Post #8 by @sebastian.iancu

What do you mean - can you give an example?

---

## Post #9 by @birger.haarbrandt

Hi Ian,

ETags are part of HTTP. We cannot remove the double quotes without breaking HTTP rules.

---

## Post #10 by @ian.mcnicoll

Sure - I got that. I was suggesting that the ehr_id might be a part of the representation=minimal response itself, not the headers.

TBH I did find it quite confusing that successful response can return a no content 204 message after a succesful template upload- one place where returning te templateId in the response is useful - I use returned ehrIds, composition Is and templateIds quite a bit when using Postman in trainig/ demos./ testing etc.

Not a major issue, just feels we have lost a little helpfulness!!

---

## Post #11 by @johnmeredith

This will strip out the extras as a test in postman:

    pm.test("CompositionId in header is present", function () {
    pm.response.to.have.header("ETag");
    var originalETag = postman.getResponseHeader("ETag");
    var newETag = originalETag.replace(/\"/g,"");
    postman.setEnvironmentVariable("compositionId",newETag);
    postman.setEnvironmentVariable("versioned_object_uid",newETag.split("::", 1));
    });

---


**Canonical:** https://discourse.openehr.org/t/does-etag-in-composition-response-header-really-need-quotes/404
**Original content:** https://discourse.openehr.org/t/does-etag-in-composition-response-header-really-need-quotes/404