Support multi-values (expect_one_of) in jsonpath tolerance expect

[saravanan30erd]

Added support for multi-values (expect_one_of) in jsonpath tolerance expect field.

Example:

"tolerance": {
                    "type": "jsonpath",
                    "path": "$.status",
                    "expect": ["yellow"],
                    "target": "body"
                },

In this case, path $.status provides value either ["green"] or ["yellow"] or ["red"]. Both yellow and green considered as healthy status. Example: In Elastic-search cluster, If you remove or shutdown one or two nodes in cluster, the status will be green before action and yellow after action. But both should be considered as healthy state in ES.

Currently it doesn't support or / expect_one_of feature in expect value.

so added support for expect_one_of feature in expect value.

  "tolerance": {
                    "type": "jsonpath",
                    "path": "$.status",
                    "expect": [["yellow", "green"]],
                    "target": "body"
                },

If we want to validate the output against multiple expected values, then we can use list of values instead of string.

Output value: ["green", "completed"] OR sometimes  ["yellow", "completed"]

**Default**:
Expect:  ["green", "completed"]
**expect_one_of**
Expect:  [["green", "yellow"], "completed"]

[saravanan30erd]

@Lawouach I did a mistake, thought to use list inside list to mention both expect_one_of values bcoz I used one sample json payload which is straight forward. But later realized this won't work for other payloads so tried some other method but it didn't suit for some complex payloads.

At last, I used another expect field called expect_alt to mention the alternative expected values.

Output value: ["green", "completed"] OR sometimes  ["yellow", "completed"]

Default:
Expect:  ["green", "completed"]

expect_one_of:
Expect:  ["green", "completed"],
Expect_alt:  ["yellow", "completed"],

[Lawouach]

Hello, for a better review, could you please squahs your commits and force push?

[Lawouach]

Hey @saravanan30erd I appreciate the effort and thank you for the tests but I think this needs a bit more work.

The naming expect_alt is confusing to me, I don't understand what it does from its name.

Side note, could you add a changelog entry too? Also there will be a need for a spec PR too (on chaostoolkit-documentation). Maybe we should start there to clarify the goal?

[saravanan30erd]

@Lawouach

expect_alt- its alternative expect field for expect.
since we are adding one more expect field for using alternative values, I named it as expect_alt
Please provide your suggestion for naming convention. I will use that name.

Sure, I will raise the PR for documentation once the naming convention is finalized.

[Lawouach]

I thought expect_one_of was more direct and descriptive. Unless expecte_alt is behaving differentlty, I'd go for expect_one_of.

[saravanan30erd]

@Lawouach Renamed the new field as expect_one_of and also added change log.

[codecov-io]

Codecov Report

Merging #191 into master will decrease coverage by 0.04%.
The diff coverage is 77.77%.

@@            Coverage Diff             @@
##           master     #191      +/-   ##
==========================================
- Coverage   88.02%   87.97%   -0.05%     
==========================================
  Files          25       25              
  Lines        2113     2121       +8     
==========================================
+ Hits         1860     1866       +6     
- Misses        253      255       +2     

Impacted Files	Coverage Δ
chaoslib/hypothesis.py	89.37% <77.77%> (-0.58%)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update f781dfe...e196962. Read the comment docs.

[Lawouach]

LGTM

But needs a couple of minor changes.

We'll also need to update the specification at https://docs.chaostoolkit.org/reference/api/experiment/#steady-state-probe-tolerance

[saravanan30erd]

@Lawouach Did changes based on your comments. Also added the information in documentation and raised PR.
chaostoolkit/chaostoolkit-documentation#103