pspg 0.5 (new formula)

[jasonmp85]

This adds a new formula for pspg, a pager intended for use by psql which adds features nice to have when viewing relational data. These include freezing columns and rows, consistent display of headers, and colorful themes.

• Have you followed the guidelines for contributing?
• Have you checked that there aren't other open pull requests for the same formula update/change?
• Have you built your formula locally with brew install --build-from-source <formula>, where <formula> is the name of the formula you're submitting?
• Does your build pass brew audit --strict <formula> (after doing brew install <formula>)?

---

[jasonmp85]

OK, so if I need to run audit with --new-formula, maybe the checklist should say that.

At any rate, how do I get around the Dependency 'ncurses' may be unnecessary as it is provided by macOS; try to build this formula without it message? I know macOS provides ncurses. The problem is that it isn't built with widechar support, which results in a warning during compilation of pspg.

Given that part of pspg's usage instructions involve enabling the unicode line drawing format for psql, I'm assuming it's best to use this software with an ncurses that supports widechars.

Can I suppress this warning?

[jasonmp85]

Hm, it appears the simplest around this would be to get the formula into homebrew and then add the dependency in a future build, as this rule only applies to new formulae?

[DomT4]

You don't need the block unless you're doing something special with head exclusively. Otherwise simply:

head "https://github.com/okbob/pspg.git"

under the sha256 line will suffice.

[jasonmp85]

Heh, I copy-pasted from another formula that had special build dependencies for head. Oops. Fixing.

[DomT4]

Any chance this can be a 2 space indent or is the 4 space indent required in ~/.psqlrc?

[jasonmp85]

It's not required at all. I think I copy-pasted this structure from asdf's formula.

[DomT4]

Alright, unless there's a reason for it let's stick to a 2 space indent here. Mixing & matching indentation where not strictly required is kinda gross 🙈.

[jasonmp85]

Done. (Though in fairness this indentation is for display purposes and not code. Homebrew might do well do establish a style for this sort of command output; right now there's vast inconsistency within the codebase (I checked postgresql for another example and it's a mess)).

[DomT4]

You're gonna get asked for a functionality test beyond a simple version check by @ilovezfs, so, if you can, start thinking of one 😉.

[jasonmp85]

It's… a pager. Kind of hard to test since most of the interaction is e.g. via keyboard (arrow keys, vim keybindings, etc.)?

I looked at e.g. nano, which calls --version and doesn't even check that its output is sane…

🤷‍♂️

[DomT4]

Yeah, appreciate that. Just thought I'd throw it out there in case you had ideas. Generally new formulae are expected to adhere to a higher standard than existing formulae.

[ilovezfs]

  test do
    ENV["TERM"] = "xterm"
    assert_match /\e\[\?1049.*\e>moo/, pipe_output("#{bin}/pspg -F", "moo", 0)
  end

maybe

[jasonmp85]

My shell-fu was too weak to get there (I was worried about repeatability). I can write something up like that (indeed I already had a mini-test that wrote out a sample header/row file and printed it using pspg).

[ilovezfs]

As long as the match is minimal it shouldn't need much tweaking over time.

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs.

[fxcoudert]

@BrewTestBot test this please

[fxcoudert]

Thanks @jasonmp85 for the pull request!