Monospaced output in Console on macOS

[RandyMcMillan]

This PR addresses issue #273
A monospace font is used on Linux and Windows for the console output - but not on MacOS.
This change forces the MacOS GUI to use the embedded RobotoMono-Bold.ttf font,
which is defined as the GUIUtil::fixedPitchFont()

[RandyMcMillan]

REF: Issue: #273

Per @hebasto 's suggestion

Before:

After:

[RandyMcMillan]

A few more examples:

[RandyMcMillan]

[shaavan]

Concept ACK

This is a definite improvement over the currently displayed font.
However, I think the newly displayed font is too bold and looks very bulky when there is a lot of text on the screen. Like in this screenshot:

What do you think about experimenting with a lighter version of the monospace font?

For example, in #273, the screenshot of Linux GUI RPC Console displays a lighter version of the monospace font. That text looks very light while each word is also easily readable.
Screenshot:

[RandyMcMillan]

This issue has been posted since April (7 months) - if you want to focus on a bunch of minutia - feel free to on your own time. I am not burning a bunch of time on a 30 second edit that anybody else could have done.

[shaavan]

Well, I was not expecting such an aggressive response. I was under the impression that the goal of this PR was to fix the problem entirely and not just to fix a tiny issue that anybody could have fixed.

[RandyMcMillan]

@shaavan - may apologizes for the tone of my previous response...

The whole embedding fonts approach opens up a whole can of worms - which I prefer to avoid.
This PR is a simple "fix" avoiding all that controversy.

There is clearly a more elegant fix which you may want to implement and I would be open to reviewing and contributing to.

With that said - this PR is a minimal fix which shouldn't evoke the font issues.
Feel free to review previous discussions on the topic.

[shaavan]

may apologizes for the tone of my previous response...

It's okay. Sometimes all of us feel a little out of place and may lose our cool. The good thing is to accept our mistakes humbly. So I appreciate that.

With that said - this PR is a minimal fix which shouldn't evoke the font issues.

I understand the context of this PR now. And IMO this PR accomplishes the goal it set out to achieve.

ACK 4516e39

[katesalazar]

ACK 4516e39.

[katesalazar]

I would be happier enclosing the new code under __APPLE__ ifdefs though.

[hebasto]

Concept ACK.

I would be happier enclosing the new code under __APPLE__ ifdefs though.

Agree that there are no reasons to change look in systems that do not require any change.

Would you mind to provide the pr description? And refer to #273 in it?

Also please use qt: prefix in the commit message instead of rcpconsole.cpp:.

[katesalazar]

ACK 9e6775d.

I'd be happier with a commit message bringing
a thorough description copying bits from here and the parent issue.
Maybe in the future Apple fixes this and the two lines do the same,
I wouldn't want to check the issue tracker to know why this code,
only VCS history.
Lossless VCS history migration is far easier than lossless issue tracker migration.
In other words, GitHub is temporary, VCS history could be forever.

[hebasto]

Approach ACK 9e6775d.

[hebasto]

ACK b9f0aff, Tested on macOS Big Sur 11.6.1 (20G224) + Homebrew's Qt 5.15.2:

[shaavan]

reACK b9f0aff

Changes since my last review:

• Applied ifdef, endif conditional statement to prevent this change from affecting other Operating Systems. This change prevents the risk of having unintentional behavior changes on some OS.
• Added /*use_embedded_font=*/ in the argument of fixedPitchFont function. This is done to make the code more readable.

[jarolrod]

tACK b9f0aff

[katesalazar]

Well a couple of late points... I'm always swamped y'know.

• Won't ACK b9f0aff only because
  I'm not familiar enough with Q_OS_MAC.

• I tested this, and (only after testing) I have to agree to
  the boldness point written by shaavan, I can't avoid
  writing it, but

  • in my system the boldness could be made away with
    this patch immediately on top of b9f0aff:

    diff --git a/src/qt/guiutil.cpp b/src/qt/guiutil.cpp
    index 4262866f32..847e73c7ae 100644
    --- a/src/qt/guiutil.cpp
    +++ b/src/qt/guiutil.cpp
    @@ -88,7 +88,11 @@ QString dateTimeStr(qint64 nTime)
     QFont fixedPitchFont(bool use_embedded_font)
     {
         if (use_embedded_font) {
    +#ifdef __APPLE__
    +        return {"PT Mono"};
    +#else
             return {"Roboto Mono"};
    +#endif
         }
         return QFontDatabase::systemFont(QFontDatabase::FixedFont);
     }
    

    I think this is strongly more "UI-idiomatic" than the
    b9f0aff state of things, I'd recommend macOS
    users to try it.

    I'm afraid it's not the actual end form the patch should have,
    if the effect were to be desirable and wanted in!

  I'm not familiar enough with macOS development as to know
  what's the more appropriate monospaced font on macOS
  (or if such thing depends on macOS versions). But this one
  would be definitely an improvement...

Cheers!

[RandyMcMillan]

I will take another look - to see how the boldness can be addressed. :)

[luke-jr]

I will take another look - to see how the boldness can be addressed. :)

After #497, it should be simpler to make the console font user-configurable.

[katesalazar]

current Roboto Mono on Dark Appearance:

available PT Mono on Dark Appearance: