Skip to content

decide what to do with underscores in section labels #2609

Closed

Description

@zygoloid

Continuing from #1498, we have 81 section labels containing underscores. These look extremely ugly: the font we use to display section labels does not have a reasonable underscore character.

We should either:

  • change the font used to render section labels (both in the body text and in xrefindex and xrefdelta) to have a less ugly underscore, or
  • revise the section labels to avoid using underscores

Generally when we form a section label from an identifier, we change underscores into periods (eg, [unord.map.syn] not [unordered_map.syn]); fixing these section labels to do the same may be the most consistent option.

Activity

  1. added
    decision-requiredA decision of the editorial group (or the Project Editor) is required.
    on Dec 21, 2018
  2. jensmaurer commented on Dec 21, 2018

    @jensmaurer
    Member

    Let's revise the section labels, avoiding the underscores.
    @tkoeppe consistently talks us out of having more font creativity in the standard.

  3. tkoeppe commented on Dec 21, 2018

    @tkoeppe
    Contributor

    That's right :-) I wouldn't be opposed to finding a solution that simply redefines this particular symbol, globally, to a somewhat more size-appropriate underscore, though. As long as we just have one kind of underscore... @godbyk: Any idea how to do that? It's the bold-roman body font that needs fixing.

  4. godbyk commented on Dec 21, 2018

    @godbyk
    Contributor

    @tkoeppe Yes, we can redefine \textunderscore (which is used behind the scenes when you type _ in non-math mode) to draw our own underscore instead of using one from the font if we don't like the one from the font. By default, a 'fake' underscore is drawn if one isn't available in the font. It may be that the default fake one doesn't fit the design well and we should tweak it.

  5. timsong-cpp commented on Jan 4, 2019

    @timsong-cpp
    Contributor

    Speaking of ugly underscores, maybe we can do something about static_assert-declaration too.

  6. removed
    decision-requiredA decision of the editorial group (or the Project Editor) is required.
    on Sep 10, 2019
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions