Re: [PATCH 3/4] strbuf.h: format asciidoc code blocks as 4-space indent
- From
Michael Haggerty <mhagger@alum.mit.edu>
- Date
- Dec 14, 2014, 17:42 UTC
- Message-ID
- <548DCC02.60100@alum.mit.edu>
- In-Reply-To
- <20141212223903.GE29365@google.com>
On 12/12/2014 11:39 PM, Jonathan Nieder wrote:
Show 15 quoted lines
> Jeff King wrote: > >> This is much easier to read when the whole thing is stuffed >> inside a comment block. And there is precedent for this >> convention in markdown (and just in general ascii text). >> >> Signed-off-by: Jeff King <peff@peff.net> >> --- > > Reviewed-by: Jonathan Nieder <jrnieder@gmail.com> > >> As a side note, I actually find markdown much more pleasant to read and >> write than asciidoc. > > I do, too. Quoting in asciidoc is a nightmare.
Peff, thanks for working on this. I think it is a definite improvement.
I suggest that we accept the use of asciidoc/markdown's convention of using backwards quotes to mark code snippets (especially identifier names) within comments *anywhere* in our code base. For example, this appears in refs.c:
/*
* Create a struct ref_entry object for the specified dirname.
* dirname is the name of the directory with a trailing slash
* (e.g., "refs/heads/") or "" for the top-level directory.
*/I claim that it is more readable with a tiny bit of markup:
/*
* Create a `struct ref_entry` object for the specified `dirname`.
* `dirname` is the name of the directory with a trailing slash
* (e.g., "refs/heads/") or "" for the top-level directory.
*/Marking up `struct ref_entry` helps make it clear that the two words belong together, and marking up `dirname` makes it clear that we are talking about a specific identifier (in this case, a function parameter).
Currently, comments use a mix of unadorned text, single-quoted text, and double-quoted text when talking about code. I think the asciidoc/markdown convention is clearer [1].
I think we shouldn't be pedantic about this. When a comment is readable with no markup, there's no need to add markup. And "incorrect" markup shouldn't by itself be reason to reject a patch. But in many examples, a little bit of markup makes the text less ambiguous and easier to read.
Michael
[1] Yes, I see the irony in trying to improve a mixture of three
conventions by adding a fourth one.-- Michael Haggerty mhagger@alum.mit.edu