abs-lang / abs

72%

master: 71%

LAST BUILD ON BRANCH negative-indexes

branch: negative-indexes

CHANGE BRANCH

x

Reset

• negative-indexes
• badges
• 1.1.x
• make-hashes-iterable
• destructuring-hashes
• issue-144-cd-only
• key-in-hash
• sleep
• contributing
• release-1.1.0
• 1.1.0
• master
• 1.2.x
• issue-144-v2
• issue-166
• issue-164
• issue-164-v2
• parse-exponential-notation
• absrc-source-file
• background-commands
• issue-171
• issue-171-alex
• mount-abs-history-in-dev
• better-illegal-tokens-when-parsing-numbers
• number-math-rounding
• issue-190
• doc-command-executor
• add-code-of-conduct-1
• 1.2.0
• 1.3.x
• bats-testing
• go-1.12
• unicode
• issue-199-exec
• 1.3.0
• 1.4.x
• underscores-in-numeric-literals
• go-mod
• hash-str-should-convert-to-valid-json
• implicit-return-value
• fix-hash-to-json
• 1.3.1
• benchmark-for-lazy-evaluation
• lazy-evaluation-in-1.3.x
• 1.3.2
• repl-without-terminal-no-panic
• eval
• allow-digits-in-vars
• 1.4.0
• 1.5.x
• negate-in
• 1.4.1
• break-continue
• higher-file-writers
• stack-overflow
• unique
• 1.5.0
• numeric-bash-arguments-support
• 1.6.x
• 1.5.1
• deprecate-legacy-command-syntax
• index-ranges
• 1.6.0
• 1.7.x
• return-semicolon
• 1.6.1
• playground
• reverse-search
• echo-dying-message
• reverse-search-2
• replace-array
• replace-short-form
• number-abbr
• index-ranges-bug
• 1.6.2
• 1.7.0
• 1.8.x
• source-vs-require
• bash-style-interpolation-in-strings
• go-prompt-master
• array-tsv
• go-1.13
• abs-installer-fixes

Build # 871

Build Type

Pull #276

travis-ci-com

Committed by web-flow

Commit Message

Negative indexes, closes #274

This PR adds support for negative indexes in arrays / strings.

``` bash
[1,2,3][-1] #3
```

There is one change that might alter existing ABS scripts, and that's
non-existing indexes, for strings, returning an empty string rather than
`null`.

I believe it's an ok thing to break as both will evaluate to `false`
when casted to boolean, and to check whether an index exists one can
simply:

* check the length of the string (`s.len()`)
* check the boolean value of the index (`!!s[idx]`)

both these examples do not break with these changes. I instead admit
that it would be weird to see code such as:

``` bash
if s[idx] == null {
  ...
}
```

rather than

```
if !s[idx] {
  ...
}
```

So I guess this is a change that can go through, as it really shouldn't
impact much of the userbase.

Pull Request Pull Request #276: Negative indexes, closes #274

Run Details

22 of 22 new or added lines in 1 file covered. (100.0%)

26 existing lines in 1 file now uncovered.

2739 of 3800 relevant lines covered (72.08%)

84.27 hits per line