Zero percent coverage

---

node: 10.1.0 npm: 5.6.0 truffle: 4.1.8

Unable to get any reported coverage of a contract implementation, although it does show 100% of the interface is covered.

---

Contract is setup similar to:

    ┌───────────┐  ┌─────────┐
    │ Interface │  │ Library │
    ├───────────┤  └─────────┘
    │ +methodA  │        │
    └───────────┘        │
         │               │
         │               │
         │               │
    ┌──────────────┐     │
    │ Contract     │     │
    ├──────────────┤     │
    │ +libraryProp │─────┘
    │ +methodA     │
    └──────────────┘

After installing, and running, solidity-coverage we get the following:

---------------------|----------|----------|----------|----------|----------------|
File                 |  % Stmts | % Branch |  % Funcs |  % Lines |Uncovered Lines |
---------------------|----------|----------|----------|----------|----------------|
 contracts/          |        0 |        0 |        0 |        0 |                |
  Contract.sol       |        0 |        0 |        0 |        0 |... 411,412,414 |
  Interface.sol      |      100 |      100 |      100 |      100 |                |
---------------------|----------|----------|----------|----------|----------------|
All files            |        0 |        0 |        0 |        0 |                |
---------------------|----------|----------|----------|----------|----------------|

On the first run we’re also experiencing the following error:

Event trace could not be read.
Error: ENOENT: no such file or directory, open './allFiredEvents'
Exiting without generating coverage...

To work around this we do a % touch ./allFiredEvents, and then re-run ./node_modules/.bin/solidity-coverage.

Also, after adding some more in-depth code/tests, we also start running into:

Error: VM Exception while processing transaction: out of gas
...

although regular unit tests run just fine. Also, the coverage is still generated

[cgewecke]

Ok great - you just need to capture the process ID of the testrpc-sc instance you’re launching in the container and kill it explicitly when the test run stops (or detect an existing instance and use that instead of launching a new one). There’s an example of doing this in a shell script at zeppelin-solidity here.

Oh sorry - I just read the whole note - solidity-coverage should be killing the testrpc successfully if you’re not launching it separately. Are you using the latest version?

[cgewecke]

@juztin Sweet, thanks.