Markdown Discuss - Oct 2014 - (text/markdown) link label vs. link identifier and last-one-wins

In working on the text/markdown spec, I am making a reference to the 
syntax of links, specifically the things [1] and [Google] used in this 
construct:

    Hello I am some [Markdown][1] and I use [Google][].

    [1]: http://daringfireball.net/projects/markdown/
    [Google]: http://www.google.com/ "This is Google"


What is the common Markdown way of identifying this syntax element?

stmd (CommonMark) consistently calls it the "link label". (To be
clear,
it also calls [Markdown] a link label, even though that element does not 
have the same behavior as [1].)

However, [Markdown Syntax][MDSYNTAX] refers to it once as a "link 
identifier" in the bullet point: "Square brackets containing the *link
identifier* (optionally indented from the left margin using up to three 
spaces)". Elsewhere it refers to it as a label. For example: 
"Reference-style links use a second set of square brackets, inside which 
you place a label of your choosing to identify the link" and "Then, 
anywhere in the document, you define your link label like this, on a 
line by itself".

I reviewed Markdown.pl, which consistently uses the variable name 
$link_id (see also $g_urls, and comments such as # Link defs are in the 
form: ^[id]: url "optional title").

On this basis, I am going to call it "link identifier". Questions?

Also, Markdown.pl seems to define last-wins behavior: the last link 
definition is indexed as the definition in $g_urls. (See 
_StripLinkDefinitions.) Older ones get overwritten by newer ones. Is 
this common or normative behavior? How do other implementations do it?

It's important that I keep the original reference list short; I would 
rather not refer normatively to documents other than Gruber's own 
Markdown rules.

Sean

[MDSYNTAX]: http://daringfireball.net/projects/markdown/syntax
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<https://pairlist6.pair.net/pipermail/markdown-discuss/attachments/20141010/bae430cf/attachment.html>

Le 10-oct.-2014 ? 11:56, Sean Leonard <dev+ietf at seantek.com> a ?crit :
> On this basis, I am going to call it "link identifier".
Questions?
It is called both a "label", an "identifier", and a
"name" under the [link][] section of the syntax description on Daring
Fireball. Choose as you like.

 [link]: http://daringfireball.net/projects/markdown/syntax#link

> Also, Markdown.pl seems to define last-wins behavior: the last link
definition is indexed as the definition in $g_urls. (See _StripLinkDefinitions.)
Older ones get overwritten by newer ones. Is this common or normative behavior?
How do other implementations do it?
Well, it is not specified in the official documentation. You can always check on
Babelmark 2 to see how various implementations are doing it:
<http://johnmacfarlane.net/babelmark2/?normalize=1&text=%5Blink%5D%5B%5D%0A%0A%5Blink%5D%3A+1%0A%5Blink%5D%3A+2>

> It's important that I keep the original reference list short; I would
rather not refer normatively to documents other than Gruber's own Markdown
rules.
I don't understand what all this has to do with a spec for a MIME type.


-- 
Michel Fortin
michel.fortin at michelf.ca
http://michelf.ca

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 4154 bytes
Desc: not available
URL:
<https://pairlist6.pair.net/pipermail/markdown-discuss/attachments/20141010/9cbdd430/attachment.bin>