Change test strings for clearer picture of what's going on#48
Conversation
rwjblue
left a comment
rwjblue
left a comment
There was a problem hiding this comment.
Seems good to me, mind poking at why CI is failing?
|
@rwjblue ✅ |
| language: node_js | ||
| node_js: | ||
| - "stable" | ||
| - "4" |
There was a problem hiding this comment.
I think we should make this change independently, and also update the engines in package.json. It'll require a major version bump.
I think we should support 8, 10, and >= 12.
|
@rwjblue master has been failing since @kratiahuja merged the update to Fastboot 2.x because Fastboot 2.x doesn't support Node 4.x How would you like to proceed? |
|
Basically like I mentioned above #48 (comment) bump .travis.yml and package.json's engines config in a separate PR (for easier inclusion in the changelog) |
|
OK I can do that. Should I bump the major version as well, considering that fastboot went from 1.x to 2.x as a result of "updating node.js support matrix"? |
|
No, I'll bump major's when I release |
It was unclear to me what scenarios would legitimately render a body that matched "error" vs when "hello world" was rendered because index.html was rendered or because the middleware was rendering that string directly. This changes index.html to contain the string "Original body" and the middleware to send the string "special error handling" to disambiguate the two. Additionally it adds some assertions around what the body _should_ match for examples that only had what it _should not_ match.
|
@rwjblue I've removed the commit that changes travis yml, but that means CI will fail until we merge #49 And I assume you're already aware but looks like @kratiahuja has already bumped major version yesterday |
|
@rwjblue I also just opened issue #50 to try to figure out why a couple of these test cases pass when afaict they shouldn't Also, I'm not sure how these fixtures work... I'd like to spruce up the |
ghost
closed this by deleting the head repository
2 participants
It was unclear to me what scenarios would legitimately render a body that matched "error" vs when "hello world" was rendered because index.html was rendered or because the middleware was rendering that string directly.
This changes index.html to contain the string "Original body" and the middleware to send the string "special error handling" to disambiguate the two.
Additionally it adds some assertions around what the body should match for examples that only had what it should not match.
On a related note: I'm still trying to figure out why index.html is rendered fine with
resilientenabled in these tests but not infastboot-app-servertests (even after updating fastboot-app-server to properly passresilientthrough).