chore(instrumentation-hapi): use exported strings for attributes

[maryliag]

Which problem is this PR solving?

• Part Of Update usage of semantic conventions to new exported strings #2025

Short description of the changes

On package opentelemetry-instrumentation-hapi:

• Update @opentelemetry/semantic-conventions ^1.23.0
• Use exported strings for Semantic Attributes.

[trentm]

Whoa, what's going on with the 19k line change to package-lock.json?

[maryliag]

I hadn't even noticed that... what... how... but... why 😅

[trentm]

Some of the changes like:

     "node_modules/@aws-crypto/ie11-detection": {
       "version": "2.0.2",
-      "resolved": "https://registry.npmjs.org/@aws-crypto/ie11-detection/-/ie11-detection-2.0.2.tgz",
-      "integrity": "sha512-5XDMQY98gMAf/WRTic5G++jfmS/VLM0rwpiOpaainKi4L0nqWMSB1SzsrEG5rjFZGYN6ZAefO+/Yta2dFM0kMw==",
       "dev": true,
+      "license": "Apache-2.0",
       "dependencies": {
         "tslib": "^1.11.1"
       }
     },

This happens sometimes for reasons I don't yet understand. I am beyond annoyed that npm is so fluid with its package-lock included fields that you can sometimes get monster changes like this. I haven't been able to attribute this inclusion/exclusion of fields like "license", "integrity", "resolved" to a particular npm version, or whether there was a local cache, or the order of running npm ci or npm install or npm update or ... I don't know.

Part of the change updating 1.24.0 to 1.24.1 and 0.51.0 to 0.51.1 are handled in #2180

My total guess is that you ran some sort of npm update or related in your working tree.
I think you'll need to regenerate the package-lock change.

[maryliag]

oh yes, I did run npm install after updating the version, my response was more that I didn't expect that amount of change from one version change. I have a feeling no one has run a install or update on this package for awhile.
Something on a new version is breaking one of the tests, so I would also need to check that

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 95.12%. Comparing base (dfb2dff) to head (bb3cc46).
Report is 121 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #2190      +/-   ##
==========================================
+ Coverage   90.97%   95.12%   +4.14%     
==========================================
  Files         146       10     -136     
  Lines        7492      697    -6795     
  Branches     1502      143    -1359     
==========================================
- Hits         6816      663    -6153     
+ Misses        676       34     -642     

see 137 files with indirect coverage changes

[maryliag]

I merged the main branch since the PR you mentioned was merged, and the lock file went form 19k to 17k, which is still a lot, but maybe this is just expected and we should merge it (after I figure out the test failure)

[JamieDanielson]

I merged the main branch since the PR you mentioned was merged, and the lock file went form 19k to 17k, which is still a lot, but maybe this is just expected and we should merge it (after I figure out the test failure)

I believe the test failure is related to the package-lock change, and if that can get regenerated then the tests will pass.

Locally I pulled down main, went to the hapi directory, and ran npm i @opentelemetry/semantic-conventions@1.22.0 and then npm run compile. Then from the root directory I ran npm ci, npm run compile, and npm run test. Aside from the mongoose test timeout errors I always get locally and the aws test error (new? unrelated and not a problem in CI), other tests passed and my package-lock change is minimal.

[trentm]

I pushed a small tweak to package-lock to get instr-hapi to use v1.24.1 in the package-lock, even though it is ^1.22.0 in the package.json. I did this by running npm update @opentelemetry/semantic-conventions in the instr-hapi dir.