@Dscho pointed out I should reply to the list not GGG. Hope this works -------- Forwarded Message -------- Subject: Re: [gitgitgadget/git] README.md: add CodingGuidelines and a link for Translators (PR #1115) Date: Sat, 15 Jan 2022 10:45:43 +0000 From: Philip Oakley <philipoakley@iee.email> To: gitgitgadget/git <reply+AAG7HJ4LJ4YFCNJJ6NT7PPF75X5VTEVBNHHEDSVTJA@xxxxxxxxxxxxxxxx>, gitgitgadget[bot] <notifications@xxxxxxxxxx>, gitgitgadget/git <git@xxxxxxxxxxxxxxxxxx> CC: Mention <mention@xxxxxxxxxxxxxxxxxx> On 15/01/2022 00:50, gitgitgadget[bot] wrote: > On the Git mailing list > <https://lore.kernel.org/git/xmqqv8yl6d0f.fsf@gitster.g>, Junio C > Hamano wrote (reply to this > <https://github.com/gitgitgadget/gitgitgadget/wiki/ReplyToThis>): > > Philip Oakley <philipoakley@iee.email> writes: > Unfortunately it `[]` > is not working for me for the new additions. > GitHub magic? Let's > step back a bit. Is there somebody reading README.md as a more generic > MarkDown? I do not think our build procedure creates any rendered > document from it. GitHub will auto render the README.md at the bottom of the git.git home page, and will render any .md files when they are visited. I just checked GitLab which looks to do the same rendering. I prepared the changes because of a user question on the "Git for Human Beings" git-users list https://groups.google.com/d/msgid/git-users/aa68b9c8-4cf4-4193-8af3-79d7e3feafbbn%40googlegroups.com about some linter issues with C99 and an offer to help. Despite having answered such questions before it still took a while to find the right pointers for the user, hence the proposed clarifications. > My recollection was that we moved from README.txt to README.md for the > sole purpose of showing it https://github.com/git/git/ as more than > just a plain text, so it is a hard requirement that GitHub flavored > MarkDown renderer is happy with whatever we write in the file. If > GitHub flavored MarkDown only supports [] and breaks with (), then > there is no need for further discussion. We must stick to [] that has > been working for us in the file. On the other hand, if both () and [] > work well at GitHub, I am OK to replace [] to (), so that it shows > well at GitHub and also it can be used with other MarkDown renderers. Oddly it appears that the [] was only working for the original text, but that () was needed for my new text. I'm not able to explain why, but didn't want too much churn within the patch. > I prefer, from the organizational point of view, to see two patches, > in this order, if we were to go that way. [PATCH 1/2] README: link to > more contributor documents which uses [] (not ()). After this, > README.md uses only [], including the two links you add in this step. > And then [PATCH 2/2] README: use "()", not "[]", to please more > MarkDown processors which replaces all uses of [] with (). That would > allow us to queue both, try them out, and we can back out [2/2] if > GitHub's renderer does not like the result. Already been there. GitHub would only render the old text as links when I pushed to my repo and viewed the README.md in my browser, so 2/2 would need to go first, but without any obvious need... I was hoping someone might know why? It feels very screwy to me. --- follow up I'll redo the series as 1/2 : use "()", not "[]", to please Markdown processors 2/2 : README: link to more contributor documents
