(anonymous guest) (logged out)

Copyright (C) by the contributors. Some rights reserved, license BY-SA.

Sponsored by the Wiki Symposium and the Nuveon GmbH.

Add new attachment

Only authorized users are allowed to upload new attachments.

This page (revision-19) was last changed on 10-Oct-2007 01:19 by YvesPiguet

This page was created on 06-Mar-2007 18:47 by RadomirDopieralski

Only authorized users are allowed to rename pages.

Only authorized users are allowed to delete pages.

Difference between version and

At line 144 added 164 lines
Ok, my engine is now opensource, under the new BSD license. The
[[http://nyctergatis.com/creole/SimpleMarkup/html/|doxygen documentation]]
covers the whole source code, including the command-line application. I'll add a downloadable
archive very soon. Feedback welcome, of course.
-- [[YvesPiguet]], 2007-Mar-20
I've renamed the library "Nyctergatis Markup Engine" (NME), made its source code available in a downloadable
archive and rewritten the pages @nyctergatis.com. I hope I haven't broken anything.
-- [[YvesPiguet]], 2007-Mar-20
Just wanted to mention that I haven't forgotten NME, but plan to test it as soon as I have time.
-- [[SteffenSchramm]], 2007-Mar-28
No problem! I'll continue improving it, so you'd better download it right before taking a look.
-- [[YvesPiguet]], 2007-Mar-28
----
//Discussion begun by email//
I am afraid, I find that the orderd/unordered list bug I reported earlier seems not to be fixed.
Rationale: When in list mode (ordered and unordered list) the parser interprets double {{{**}}} or {{{##}}} at the beginning of the next line as the beginning of a new list, when it should be just {{{<b>}}} or {{{<tt>}}}. It also writes empty {{{<b></b>}}} and places the contents before it (last example).
In my interpretation, this is not correct. I added two plain paragraph examples which show the correct behaviour, IMO.
Please find a few input and output examples below. I am using [[http://nyctergatis.com/creole/|NME-071004.zip]].
Here is the Creole code:
{{{
a b
##c##
a b
**c**
* ##a## b
##c##
# ##a## b
##c##
* ##a## b
**c**
# a
##b##
* a
**b**
}}}
My HTML output looks like this:
{{{
<!-- Generated by Nyctergatis Markup Engine, Oct 5 2007 19:27:19 -->
<html><body>
<p>a b <tt>c</tt></p>
<p>a b <b>c</b></p>
<ul>
<li><tt>a</tt> b</li>
<ol>
<li>c<tt></tt></li>
</ol>
</ul>
<ol>
<li><tt>a</tt> b</li>
<ol>
<li>c<tt></tt></li>
</ol>
</ol>
<ul>
<li><tt>a</tt> b</li>
<ul>
<li>c<b></b></li>
</ul>
</ul>
<ol>
<li>a</li>
<ol>
<li>b<tt></tt></li>
</ol>
</ol>
<ul>
<li>a</li>
<ul>
<li>b<b></b></li>
</ul>
</ul>
</body></html>
}}}
-- RJ, 2007-Oct-5
It isn't a bug, imo, it's Creole specifications and implementation choices.
Nested lists are supported in Creole, so double-stars and double-sharps should
begin sublists when they're at the beginning of a line in a list. A design
choice which can be criticized is that item mark mismatches (such as {{{#}}}
following {{{*}}}, or {{{##}}} following *) are ignored. It's documented, though
("For clarity, list markers should be used in a consistent way; but only the
first item of each list fixes the kind of the whole list").
I've "fixed" the problem with {{{##}}} in readme.nme you mentionned in a previous
message by moving it so that it doesn't appear at the beginning of a line.
My error is a proof that my design choice leads too easily to ambiguities.
I'm probably going to change it.
Finally, concerning the empty {{{<b></b>}}}, it's caused by a trailing style marker
at the end of a paragraph (the first occurence of {{{**}}} is the sublist item marker).
Not really a problem, I think: it reflects accurately the source, even if it's
useless.
If you generate automatically Creole markup, a simple way to avoid the
ambiguities of {{{**}}} and {{{##}}} in lists is to avoid line breaks in lists items,
and in paragraphs for the sake of consistency. That's how we've converted
our help files for Sysquake from XML.
-- [[YvesPiguet]], 2007-Oct-5
(...)
From my novice common sense understanding I believe that the following could make sense:
New list items of any kind ({{{*}}}, {{{**}}}, {{{#}}}, {{{##}}}) are only started if there is at least one white space chacter after the list characters (all list examples in the documentation also do it this way):
{{{
##text## -- <tt>text</tt>
**text** -- <b>text</b>
}}}
but
{{{
## text -- <ol><ol><li>text</li></ol></ol>
** text -- <ul><ul><li>text</li></ul></ul>
}}}
It would also solve the dilemma I reported previously, and it would also allow to start 2nd level nested cells with no prior 1st level nesting.
The rationale is that most list markers are white-space separated from the list in the final output. Because this is so common, I believe that it also makes the Wiki Syntax easier to read. And it there is not much sense in bold whitespace like {{{** text**}}} anyway (it would not be rendered by HTML for sure).
If I was to change the documentation, it would read like:
"List items begin with a {{{*}}} or a {{{#}}} at the beginning of a line. Whitespace is optional before the {{{*}}} or {{{#}}} characters, but at least one space is required to separate it from the item's text. A list item ends at the line which begins with a new list or sublist item (* or # character followed by a space), blank line, heading, table, or nowiki block; like paragraphs, it can span multiple lines and contain line breaks forced with {{{\\}}}."
I can see now why the {{{<b></b>}}} is there. But from a common sense persepective, I still believe it should not.
-- RJ, 2007-Oct-5
Your suggestion was already discussed here: [[Require Space After Bullet Proposal]]. It wasn't retained...
-- [[YvesPiguet]], 2007-Oct-6
In NME-071009, list markup must be consistent: in the example below, the first item spans two lines with //bar// displayed in monospace, and a one-item numbered sublist.
{{{
* foo
## bar
*# baz
}}}
-- [[YvesPiguet]], 2007-Oct-10
Version Date Modified Size Author Changes ... Change note
19 10-Oct-2007 01:19 14.408 kB YvesPiguet to previous NME-071009
18 06-Oct-2007 10:56 14.168 kB YvesPiguet to previous | to last Star and sharp ambiguities
17 28-Mar-2007 20:44 9.609 kB YvesPiguet to previous | to last Reply
16 28-Mar-2007 19:46 9.476 kB SteffenSchramm to previous | to last
15 20-Mar-2007 23:38 9.338 kB YvesPiguet to previous | to last Source code released
14 20-Mar-2007 18:05 9.107 kB YvesPiguet to previous | to last Opensourced
13 20-Mar-2007 11:18 8.786 kB YvesPiguet to previous | to last Answers
12 20-Mar-2007 10:14 7.25 kB SteffenSchramm to previous | to last
11 20-Mar-2007 01:14 5.678 kB YvesPiguet to previous | to last On opensourcing
10 18-Mar-2007 17:30 5.024 kB SteffenSchramm to previous | to last Why not focus our work to build the one and only Creole reference parser?
9 06-Mar-2007 22:31 3.438 kB RadomirDopieralski to previous | to last
8 06-Mar-2007 22:02 3.201 kB YvesPiguet to previous | to last Parser overview
7 06-Mar-2007 21:12 1.962 kB RadomirDopieralski to previous | to last sorry and details
6 06-Mar-2007 20:56 1.634 kB YvesPiguet to previous | to last Reply (clarification)
5 06-Mar-2007 19:40 1.633 kB YvesPiguet to previous | to last Reply
4 06-Mar-2007 19:23 0.9 kB RadomirDopieralski to previous | to last more errors :)
3 06-Mar-2007 19:20 0.694 kB RadomirDopieralski to previous | to last the ~| doesn't work in tables as expected
2 06-Mar-2007 19:14 0.61 kB YvesPiguet to previous | to last Fixed
1 06-Mar-2007 18:47 0.45 kB RadomirDopieralski to last error
« This page (revision-19) was last changed on 10-Okt-2007 01:19 by YvesPiguet