Fix RSpec/ExampleWording in escaping and % string literals

[r7kamura]

This change resolves some of the following Issue:

• RSpec/ExampleWording unintentionally removes escapes #1799

It still does not handle quotes well for dstr node, but at least it will handle them well for str node.

---

Before submitting the PR make sure the following are checked:

• Feature branch is up-to-date with master (if not - rebase it).
• Squashed related commits together.
• Added tests.
• Updated documentation.
• Added an entry to the CHANGELOG.md if the new code introduces user-observable changes.
• The build (bundle exec rake) passes (be sure to run this locally, since it may produce updated documentation that you will need to commit).

If you have created a new cop:

• Added the new cop to config/default.yml.
• The cop is configured as Enabled: pending in config/default.yml.
• The cop is configured as Enabled: true in .rubocop.yml.
• The cop documents examples of good and bad code.
• The tests assert both that bad code is reported and that good code is not reported.
• Set VersionAdded: "<<next>>" in default/config.yml.

If you have modified an existing cop's configuration options:

• Set VersionChanged: "<<next>>" in config/default.yml.

[r7kamura]

Not related to this issue, but if string literals using %q, %, etc. are used, this cop may not be able to handle that well. It bothered me a little while reading the code. I haven't addressed it directly, but I have taken care to use node.loc.begin and node.loc.end insetad of ' or " so it will be somewhat better when we try to address it in the future.``

This cop has not been able to handle %(...) and other string literals that use %, I decided to take this opportunity to fix this problem as well.

[ydah]

Looks good! Thank you.

[pirj]

For quoted strings, both begin and end are the same. Do we need to check twice?

[r7kamura]

This is for %q(...), %Q(...), or %<...> etc.

irb(#<RuboCop::Cop::RSpec::Exampl...):001> node.loc.begin.source
=> "%q("
irb(#<RuboCop::Cop::RSpec::Exampl...):002> node.loc.end.source
=> ")"

But on second thought, we must check against node.loc.begin.source[-1], not node.loc.begin.source. I'll try to fix that.

[r7kamura]

hmm, since escaping should only be necessary if begin and end are different (e.g. %!...! may require escaping like \!, while %(...) does not), it seems that only one of them needs to be checked after all.

[pirj]

Thank you!

I’d love to extensively test this, but only the quoting autocorrection part. Wondering how to do that.

[r7kamura]

I've added a few more improvements:

• changed #docstring for more accurate % string literals handling
• removed unnecessary node.loc.begin checking from #needs_escape?
• added test cases about %(...) and %!...!

Since the handling of % string literals has been improved and the scope of change is no longer only autocorrection, the title and changelog have been slightly changed accordingly.

[bquorning]

I think these examples are in the wrong context ('when DisallowedExamples: Workz').

[bquorning]

Maybe it’s a silly question, but why are we changing the type of quotes from ' to " here? If we kept the same quote type, we wouldn’t need to concern ourselves with escaping, would we? (unless we insert an apostrophe/single quote)

[r7kamura]

This is for implementation convenience. If there is an implementation like string.inspect(quote_type: 'single_quote'), I would like to use it. If there is a way to rewrite only substrings while preserving quoting and escaping, that would be great, but do you have any other ideas?

[r7kamura]

This is a just idea, but, if the offense range had been like this, we might not have to go through all this trouble?

it "should return foo's bar"
    ^^^^^^^^^^^^^

then we would only need to replace this range without changing the unrelated part that may contain troublesome
escapes.

[pirj]

Fantastic idea!