Search and find headings in a tiddler using TW 5.1.20 core features

[Mohammad]

With the new features like match and improved search in Tiddlywiki 5 who can I search and list all headings in a Tiddler?

I mean, If I have a tiddler with a text like below

!h1-1
text
...

!h2-1
text
...


!h1-2
text
...

!h2-2
text
....

!h3-2
text
...

etc

How a simple filter can return

• h1-1
• h2-1
• h1-2
• h2-2
• h3-2

I know there is plugins to do this for example those by Tobias Beer, but I want to only use the core features of TW  5.1.20

--Mohammad

[Mat]

\define extract()
<$list filter="$(t)$ +[splitregexp[!]prefix[h]]">

</$list>
\end

<$set filter="[[mytid]get[text]]" name=t>
<<extract>>
</$set>

<:-)

[Mat]

Ah, you should probably make those quote triple quotes since you're get'ing the whole text which could contain single quotes.

<:-)

[Mat]

Somewhat slimmer

\define extract()
<$list filter="""$(t)$ +[splitregexp[!]prefix[h]]""">

<

/$list>
\end

<$set value={{mytid}} name=t>
<<extract>>
</$set>

<:-)

[Mohammad]

Thank Mat that works!

Another question, If I mean by h1 and h2 and h3, heading 1 and 2 and 3 like

! some text

!! some text

!!some text

How the filter shall be revised to cover this new form?

[Mat]

Somewhat to my surprise this turned out to be very concise:

<$list filter="""[{mytid}split[
]prefix[!]splitregexp[!]]""">

</$list>

(Yes, that line break in the middle of the split brackets should be there.)

<:-)

[Mohammad]

This is really amazing!

A KISS solution :-)

[Mohammad]

Mat you match the line breaks that way! Very smart solution!

So, no matter which browser I use because you did not use \n

Am I right?

--Mohammad

[Mohammad]

one small issue:

You know both of below lines works as valid wikitext

!  heading one
   !  another heading one

The second line cases the filter does not match heading

--Mohammad

[Mohammad]

What do you think if we use trim[]?

--Mohammad

[Mat]

I learnt the "split[

]" trick several years ago from someone, possibly PMario.

Trim seems to work, yes.

Also, there should be a removal of blanks. So result is:

<$list filter="""[{mytid}split[
]trim[]prefix[!]splitregexp[!]!is[blank]]""">

</$list>

<:-)

[Mohammad]

Added to TW-Scripts.

Thank you Mat! I love this short and Smart solution!

I will use this as a macro to create small TOC for long tiddlers! Of course it is not clickable but very useful to show what is discussed at a glance.

--Mohammad

[Mat]

You can actually make the titles clickable, especially if it is for a single long tiddler, even if it not a perfect solution:

http://intlinks.tiddlyspot.com/

<:-)

[Mohammad]

Thanks for the reference!

Cheers

Mohammad