[tw5] Text-Slicer Plugin

1,494 views
Skip to first unread message

Jeremy Ruston

unread,
Aug 1, 2015, 8:31:06 AM8/1/15
to TiddlyWiki
I've just pushed a new prerelease that includes an early cut of a tool to slice longer texts into individual tiddlers based on headings and lists. It's based on ideas that have come up in previous discussions about dealing with long, structured tiddlers.

You can try it out at:


You'll need to carefully follow the instructions in the "HelloThere" tiddler:

* Scroll down to the "Sample Text" tiddler and click on the "text slicer" icon
* Click the "import" button in the resulting import listing
* Open the tiddler "Sliced up Sample Text"

You should see a copy of the original text, but you can explore the table of contents to see how it is composed of individual tiddlers that are threaded together by tags.

I'd welcome any feedback on the tool. I would also love some help in finding a better sample text, something public domain that we can re-distribute.

Best wishes

Jeremy


--
Jeremy Ruston
mailto:jeremy...@gmail.com

jayfresh

unread,
Aug 1, 2015, 9:57:04 AM8/1/15
to TiddlyWiki
Oh that's jolly good. I'd probably use it to slice up meeting minutes, project plans, to-do lists etc.

It's an example of allowing free-text to be the basic unit of entry, with the structure coming later and broadly though automated means. Sadly this is an insight most, if not all, productivity tools have failed to grasp - instead they make the unit the "task" which is a crazy constraint and assumption about how people think.

Birthe C

unread,
Aug 1, 2015, 10:36:53 AM8/1/15
to TiddlyWiki, Tiddl...@googlegroups.com, jeremy...@gmail.com
Hi Jeremy

I have really longed for a tool like this. I couldn't resist testing it with some of my own far too long tiddlers. All went well when I realised the tricky part being $:/import open.


Birthe


Mat

unread,
Aug 1, 2015, 3:52:41 PM8/1/15
to TiddlyWiki, Tiddl...@googlegroups.com, jeremy...@gmail.com
Now THIS is great news for usability! Thank you Jeremy, wonderful!

Here are some first spontaneous thoughts;

It was surprising to see how the tiddler in edit mode also appears in the assembled "document" - I'm not sure this is intentional... but it is actually really good! I would prefer editing it at that local place before having it appear separately at some other place. But with the option to edit it separately (Ctrl+click?) But editing it in-doc would of course require some way to access that tiddlers toolbar via the assembled doc, at that point. Could there perhaps be a toolbar appearing on-hover, in doc, for the relevant tiddler? Maybe with some kind of marking to show which exact section of the assembled doc it concerns, such as a differently colored background when hovering or a frame? I have some further thoughts if this at all sounds interesting.

A much welcome addition!

<:-)

Spangenhelm

unread,
Aug 1, 2015, 7:13:07 PM8/1/15
to TiddlyWiki, Tiddl...@googlegroups.com, jeremy...@gmail.com
Hi, even if the idea is good (please think of adding a "select all/none" option in the import process) there are hundreds of opened issues (some since 2013!) that needs your attention way more than this don't you think?

Michael Wiktowy

unread,
Aug 2, 2015, 2:22:28 AM8/2/15
to TiddlyWiki, Tiddl...@googlegroups.com, jeremy...@gmail.com
On Saturday, August 1, 2015 at 8:31:06 AM UTC-4, Jeremy Ruston wrote:

I'd welcome any feedback on the tool. I would also love some help in finding a better sample text, something public domain that we can re-distribute.


I love it!

I have been chopping up CFRs from the eCFR to build up searchable regulation tiddlywikis in much the same way to help me with my job. I have been doing them manually so far and I can see this speeding things up quite a bit.

Feedback:
1) I think slicing down to the paragraph level might be too far. It adds a lot of autogenerated cruft in the TOCs. There may be some need to go down to this level but I would make the delimiter for levels beyond just headings configurable/regexp-able. In the eCFR example above. It would be useful to go down to the paragraph denoted with a lowercase letter in braces and the subparagraph denoted by a number ... end even sub-subparagraph denoted by lower case roman numerals ... but I don't know what a good title for the paragraphs would be other than their typical legal reference (i.e. §###.## (a)(1)(i)). I have not tackled splitting thing up to that level manually mainly due to the work it would entail and the difficulty in piecing together a tiddler name that won't be error-prone.

2) In the eCFR example, there are a lot of annoying "back to top" links that I have to manually remove. It would be nice if some configurable/regexp text could be excluded from the slicing import.

I'll have to exercise this more (particularly slicing updated regs that supersede previous ones and linked graphics)  since it looks like a great tool for me.

Thanks!
/Mike

PMario

unread,
Aug 2, 2015, 5:17:52 AM8/2/15
to TiddlyWiki, Tiddl...@googlegroups.com, jeremy...@gmail.com
Hi Jeremy,

Interesting approach.

Is there a white paper, that allows us to see the concept, that you implemented, without the need of reverse engineering?

There is an issue. If I import the sliced tiddlers, they don't show up in the recent tab. So after closing the import tiddler, it is impossible to open the sliced tiddlers.

There are some questions, for me that come up immediately.

 1) What happens I change the title of a sliced tiddler. Does the structure fall apart?
 2) Are the auto created tiddlers still usable for tools like TiddlyMap, in a sane way?
 3) What if I need to translate the content of individual tiddlers. Do I need to recreate the structure, or can I use the existing one? Especially the heading tiddlers.
 
I have a vague idea, how it works, so it would be useful to see the tiddlers in the recent tab.
I think, the document abstraction doesn't go far enough for translation atm but I'm not sure about this.

 4) Is there a way to undo the slicing?
 5) How can I export one eg heading with all its content, without forgetting some content tiddlers?

---------------

Just a remark.

I'd label this plugin as "highly experimental", since it completely changes the structure of a tiddlywiki and there is no undo atm. ...

I did read the highly enthusiastic comments, which is great.
So testing and feedback is good but I'd not recommend to use it with production TWs yet. Because it removes the possibility for Jeremy to make breaking changes to the plugin, without causing a lot of pain.

I personally would like to get the "best possible functionality" and not a plugin the halve backed and can't be changed anymore, because it needs to be backwards compatible to a beta version.

have fun!
mario

Jeremy Ruston

unread,
Aug 2, 2015, 6:28:05 AM8/2/15
to PMario, TiddlyWiki
Thanks for the feedback everyone.

> Hi, even if the idea is good (please think of adding a "select all/none" option in the import process) there are hundreds of opened issues (some since 2013!) that needs your attention way more than this don't you think?

The reason I'm working on this now is because I need it for my day job -- bear in mind that nobody pays me to work on TiddlyWiki :)

It might be worth exploring your thoughts about the state of the TW5 issues list in a new thread. For me, I certainly don't see it as my todo list.

>It was surprising to see how the tiddler in edit mode also appears in the assembled "document" - I'm not sure this is intentional... but it is actually really good! I would prefer editing it at that local place before having it appear separately at some other place. But with the option to edit it separately (Ctrl+click?) But editing it in-doc would of course require some way to access that tiddlers toolbar via the assembled doc, at that point. Could there perhaps be a toolbar appearing on-hover, in doc, for the relevant tiddler? Maybe with some kind of marking to show which exact section of the assembled doc it concerns, such as a differently colored background when hovering or a frame? I have some further thoughts if this at all sounds interesting.

The idea is indeed to have in-place editing of content. There are several steps:

* Support for edit mode, which makes the tiddler toolbars appear
* Support for inserting new tiddlers under existing ones
* Drag and drop to reorder items

The current state of affairs is, as you've discovered, that switching a tiddler into edit mode will display the draft within the document.

> Is there a white paper, that allows us to see the concept, that you implemented, without the need of reverse engineering?

It's pretty basic, if you understand the existing table of contents mechanism. It's based on child entries being tagged by their parent title.

> There is an issue. If I import the sliced tiddlers, they don't show up in the recent tab. So after closing the import tiddler, it is impossible to open the sliced tiddlers. 

I'll fix that.

>  1) What happens I change the title of a sliced tiddler. Does the structure fall apart?

My next plan is to support renaming tiddlers properly by changing tag and list references to the tiddler (as per Danielo's plugin).

>  2) Are the auto created tiddlers still usable for tools like TiddlyMap, in a sane way?

Hopefully. The idea is that by reusing the TOC structure it should fit well with existing tools.

> 3) What if I need to translate the content of individual tiddlers. Do I need to recreate the structure, or can I use the existing one? Especially the heading tiddlers. 

The plan is for each tiddler to have a language cascade, so that for example it will display de-AT content, falling back to de-DE and then en-GB
 
> 4) Is there a way to undo the slicing?

The slicing is non-destructive, in that the tiddler you slice up is unchanged. Currently it will overwrite existing tiddlers if the titles of the extracted tiddler clash.

> 5) How can I export one eg heading with all its content, without forgetting some content tiddlers?

When we support dragging multiple tiddlers in one go we should be able to support a drag handle that says "drag this tiddler and all dependents".

Alternatively, you can export the rendered version of the tiddler, which will include the transclusions.

> I'd label this plugin as "highly experimental", since it completely changes the structure of a tiddlywiki and there is no undo atm. ... 

I'll clarify the labelling to indicate that it is experimental and subject to change, but I'm unsure what you mean by changing the structure of a TW. As I said above, the slicing operation is non-destructive.

> 1) I think slicing down to the paragraph level might be too far. It adds a lot of autogenerated cruft in the TOCs. There may be some need to go down to this level but I would make the delimiter for levels beyond just headings configurable/regexp-able. In the eCFR example above. It would be useful to go down to the paragraph denoted with a lowercase letter in braces and the subparagraph denoted by a number ... end even sub-subparagraph denoted by lower case roman numerals ... but I don't know what a good title for the paragraphs would be other than their typical legal reference (i.e. §###.## (a)(1)(i)). I have not tackled splitting thing up to that level manually mainly due to the work it would entail and the difficulty in piecing together a tiddler name that won't be error-prone.

Yes, I think the tiddler-per-paragraph approach is fairly extreme. I'd like to support other options through configuration options.

I'd also like to support numbered headings properly.

> 2) In the eCFR example, there are a lot of annoying "back to top" links that I have to manually remove. It would be nice if some configurable/regexp text could be excluded from the slicing import.

That would be useful.

Best wishes

Jeremy.


Alex Hough

unread,
Aug 2, 2015, 8:02:35 AM8/2/15
to TiddlyWiki
@JayFresh makes a good point

"It's an example of allowing free-text to be the basic unit of entry, with the structure coming later and broadly though automated means. Sadly this is an insight most, if not all, productivity tools have failed to grasp - instead they make the unit the "task" which is a crazy constraint and assumption about how people think."

I think this is a great feature. A use case arose for me:

I created a missing tiddler, i copied and pasted the title of the tiddler into the search box. The search returns a tiddler -- a copy of an email: it is long and linear, structured in paragraphs. One of the paragraphs was relevant to my missing tiddler. Manually I'll create a new tiddler containing an edited version of the useful paragraph with links to the parent and the missing.
 

Alex

--
You received this message because you are subscribed to the Google Groups "TiddlyWiki" group.
To unsubscribe from this group and stop receiving emails from it, send an email to tiddlywiki+...@googlegroups.com.
To post to this group, send email to tiddl...@googlegroups.com.
Visit this group at http://groups.google.com/group/tiddlywiki.
To view this discussion on the web visit https://groups.google.com/d/msgid/tiddlywiki/CAPKKYJbE8ii9UhT8nX_-YwgkW7Rvu0vnP9Y0LwVOdfd5vLCMmQ%40mail.gmail.com.

For more options, visit https://groups.google.com/d/optout.

Alex Hough

unread,
Aug 2, 2015, 8:41:27 AM8/2/15
to TiddlyWiki
This idea, @Mario's point, about reverse engineering and Steve's "off-the-cuff description" of a potential conference theme (see Missing Tiddler below)  sparked a thought.

Many of the tasks I use TW for is to convert chunks of text into hypertext. I am at a curious stage with a text, there are various activities I find myself performing: 1) I am reverse engineering text a produced as a hypertext which I then removed all hypertext links to get the text into Harvard referencing format. 2) I find some text composed in an email and cut and pasted into my TW. It contains some references which don't exist in my TW....

The world of the Day Job consists of writing email, word docs and reading pdfs -- this is pretty much a given for many people. TW is an adaptable hypertext tool and as such brings all the benefits the hypertext pioneers  write about. Converting texts (with combersome references, footnotes, endnotes etc) to hypertext is part of a wider ambition to make reading complex interconnected texts less of a burden on the mind. The conventions associated with referencing texts produces a cognitive overhead for the reader. 

Hypertext removes the need for pointing, establishing context and referencing, TW is a tool which helps the thinker manage these for him / her self

------- Missing Tiddler -----

Draft conference theme

"Tiddlywiki can be envisioned and described within the broad vison of "hypertext" imagined by pioneers Vannevar Bush, Ted Nelson and Doug Englebart.* This conference will bring together users and developers within the Tiddlywiki ecosystem to consider, demonstrate and document  the relationship between the Bush/Nelson/Englebart vision, and the Tiddlywiki (TW5) community's implementation, practices and techniques." 


---------

PMario

unread,
Aug 2, 2015, 1:46:52 PM8/2/15
to TiddlyWiki, pmar...@gmail.com, jeremy...@gmail.com

Thanks for your info!


On Sunday, August 2, 2015 at 12:28:05 PM UTC+2, Jeremy Ruston wrote:

> I'd label this plugin as "highly experimental", since it completely changes the structure of a tiddlywiki and there is no undo atm. ... 

I'll clarify the labelling to indicate that it is experimental and subject to change, but I'm unsure what you mean by changing the structure of a TW. As I said above, the slicing operation is non-destructive.

Hmmm,
So slicing basically means, that some content is duplicated and the user is responsible for the "destructive delete" action. That's ok. "Your content is yours and you are responsible for your own actions :)"

But still, if I as a user, delete the "source" tiddler, I'm bound to the new structure and converting it back may be a lot of work, depending on the original tiddler size. That's why I think it is important, that the user understands the new structure, that is created by the plugin. ... So in the case, that I need to "reunite" the pieces, this should be also easy.

All in all, I'm very pleased, with the: "The Tiddler is the Thing" approach and the new possibilities that I can see with this extension.

have fun!
mario


Mark S.

unread,
Aug 2, 2015, 4:08:25 PM8/2/15
to TiddlyWiki, Tiddl...@googlegroups.com, jeremy...@gmail.com
Minor thing -- hovering over the pair of scissors the tip text says "Clone this tiddler".

What do you not like in the current sample text? Is it the Latin?

Mark

Alex Hough

unread,
Aug 2, 2015, 4:36:55 PM8/2/15
to TiddlyWiki
This is slightly off topic....

I wanted to take a list of references from a paper (in Pdf format)

Then I cut and paste the text into a tiddler, the paragraph formating is lost.

Can anyone suggest a quick way of adding back the spaces between the lines to that I can then split the tiddler using the new tool.

I think this could be a great tool for the education tiddlywiki project. With a few adaptions, you could cut and paste a list of references and get a set of tiddlers pre-populated with date, author and paper title fields. they could then be displayed on a timeline. It wouild be useful for a literature review


Alex

--
You received this message because you are subscribed to the Google Groups "TiddlyWiki" group.
To unsubscribe from this group and stop receiving emails from it, send an email to tiddlywiki+...@googlegroups.com.
To post to this group, send email to tiddl...@googlegroups.com.
Visit this group at http://groups.google.com/group/tiddlywiki.

Mark S.

unread,
Aug 2, 2015, 8:06:03 PM8/2/15
to TiddlyWiki
Everyone should have a favorite editor for just such emergencies. You need an editor that can see and replace a single line return with a double line return. I usually use Emacs for this kind of thing. In the past, I've also used Notetab for Windows. I believe Word will let you search for special codes.

Stuff coming off of PDF files often doesn't have quite the same arrangement as what you see on screen. It may help to try different PDF readers (older versions of Adobe had more copy options). There are also converters to turn PDF into text. Calibre is the only one I know of that's free. You will probably always be stuck doing a little fix-it by hand.

-- Mark

Jeremy Ruston

unread,
Aug 4, 2015, 12:53:28 PM8/4/15
to TiddlyWiki
I've posted an update that includes a first pass at the ability to rename tiddlers, while keeping the tagging relationships intact. I've also dispensed with the use of an import tiddler:


Use the "show toolbar" checkbox at the top of the sliced output to access the renaming feature.

Best wishes

Jeremy.



For more options, visit https://groups.google.com/d/optout.

PMario

unread,
Aug 4, 2015, 1:36:20 PM8/4/15
to TiddlyWiki, jeremy...@gmail.com
On Tuesday, August 4, 2015 at 6:53:28 PM UTC+2, Jeremy Ruston wrote:
I've posted an update that includes a first pass at the ability to rename tiddlers, while keeping the tagging relationships intact. I've also dispensed with the use of an import tiddler:


Use the "show toolbar" checkbox at the top of the sliced output to access the renaming feature.

\o/ I see "In Place Editing" coming.

looks good.
-m

Michael Wiktowy

unread,
Aug 4, 2015, 1:58:16 PM8/4/15
to tiddl...@googlegroups.com
Really cool!

There is a funny quirk ... If you rename the same field a few times and then rename it to the original name again, it works properly but autofills the input field with previous iterations of the rename ... which isn't a problem per-se but a bit confusing.
For instance:
Rename Sample -> Sample1
Rename Sample1 -> Sample 12
Rename Sample12 -> Sample

Sample1 will be in the input field so if you keep on pressing the rename button it will cycle through all those fields again. I am not sure if this is Chrome autofill magic of Tiddlywiki keeping some circular buffer around.

/Mike

--
You received this message because you are subscribed to a topic in the Google Groups "TiddlyWiki" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/tiddlywiki/vp62YKOsE54/unsubscribe.
To unsubscribe from this group and all its topics, send an email to tiddlywiki+...@googlegroups.com.

To post to this group, send email to tiddl...@googlegroups.com.
Visit this group at http://groups.google.com/group/tiddlywiki.

For more options, visit https://groups.google.com/d/optout.



--
Michael Wiktowy mwik...@gmail.com

PMario

unread,
Aug 4, 2015, 6:32:39 PM8/4/15
to TiddlyWiki
On Tuesday, August 4, 2015 at 7:58:16 PM UTC+2, Michael Wiktowy wrote:
There is a funny quirk ... If you rename the same field a few times and then rename it to the original name again, it works properly but autofills the input field with previous iterations of the rename ... which isn't a problem per-se but a bit confusing.

IMO this is browser standard input field behaviour.
-m

Jeremy Ruston

unread,
Aug 5, 2015, 10:18:09 AM8/5/15
to PMario, TiddlyWiki
\o/ I see "In Place Editing" coming.

Well sleuthed! Here it is:


I've pushed a new prerelease build to:


Best wishes

Jeremy
 

looks good.
-m

Jeremy Ruston

unread,
Aug 5, 2015, 10:19:19 AM8/5/15
to TiddlyWiki, Michael Wiktowy
Hi Michael

There is a funny quirk ... If you rename the same field a few times and then rename it to the original name again, it works properly but autofills the input field with previous iterations of the rename ... which isn't a problem per-se but a bit confusing.
For instance:
Rename Sample -> Sample1
Rename Sample1 -> Sample 12
Rename Sample12 -> Sample

Sample1 will be in the input field so if you keep on pressing the rename button it will cycle through all those fields again. I am not sure if this is Chrome autofill magic of Tiddlywiki keeping some circular buffer around.

Thanks, I'll fix that shortly,

Best wishes

Jeremy

 

For more options, visit https://groups.google.com/d/optout.

Mark S.

unread,
Aug 5, 2015, 12:56:31 PM8/5/15
to TiddlyWiki, Tiddl...@googlegroups.com, jeremy...@gmail.com
This is great stuff, but I'm wondering it it wouldn't be better to have a separate field (e.g.: connect; linkup; follows; sliceof) rather than use the tag space. To me tags are meant to add semantic context to tiddlers. Here they're used for connecting a set of tiddlers. A single slice instantly pollutes tag-space with tags that have no use except to connect a very specific set of tiddlers to each other.

Thanks!

Mark

On Saturday, August 1, 2015 at 5:31:06 AM UTC-7, Jeremy Ruston wrote:

Jeremy Ruston

unread,
Aug 5, 2015, 1:08:30 PM8/5/15
to TiddlyWiki
Hi Mark

On Wed, Aug 5, 2015 at 5:56 PM, 'Mark S.' via TiddlyWiki <tiddl...@googlegroups.com> wrote:
This is great stuff, but I'm wondering it it wouldn't be better to have a separate field (e.g.: connect; linkup; follows; sliceof) rather than use the tag space. To me tags are meant to add semantic context to tiddlers. Here they're used for connecting a set of tiddlers. A single slice instantly pollutes tag-space with tags that have no use except to connect a very specific set of tiddlers to each other.

I like the simplicity of using tags, but would like to support user defined fields instead of "tags" as an option; the idea of multiple tags fields is something that has been discussed before: there's only a few enhancements needed to make it possible.

Best wishes

Jeremy.


 

Thanks!
Mark

On Saturday, August 1, 2015 at 5:31:06 AM UTC-7, Jeremy Ruston wrote:
I've just pushed a new prerelease that includes an early cut of a tool to slice longer texts into individual tiddlers based on headings and lists. It's based on ideas that have come up in previous discussions about dealing with long, structured tiddlers.

You can try it out at:


You'll need to carefully follow the instructions in the "HelloThere" tiddler:

* Scroll down to the "Sample Text" tiddler and click on the "text slicer" icon
* Click the "import" button in the resulting import listing
* Open the tiddler "Sliced up Sample Text"

You should see a copy of the original text, but you can explore the table of contents to see how it is composed of individual tiddlers that are threaded together by tags.

I'd welcome any feedback on the tool. I would also love some help in finding a better sample text, something public domain that we can re-distribute.

Best wishes

Jeremy


--
Jeremy Ruston
mailto:jeremy...@gmail.com

--
You received this message because you are subscribed to the Google Groups "TiddlyWiki" group.
To unsubscribe from this group and stop receiving emails from it, send an email to tiddlywiki+...@googlegroups.com.
To post to this group, send email to tiddl...@googlegroups.com.
Visit this group at http://groups.google.com/group/tiddlywiki.

For more options, visit https://groups.google.com/d/optout.

PMario

unread,
Aug 5, 2015, 3:33:33 PM8/5/15
to TiddlyWiki, jeremy...@gmail.com
On Wednesday, August 5, 2015 at 7:08:30 PM UTC+2, Jeremy Ruston wrote:
... the idea of multiple tags fields is something that has been discussed before: there's only a few enhancements needed to make it possible.

This would be a really great enhancement.
-m

Mat

unread,
Aug 15, 2015, 4:50:06 AM8/15/15
to TiddlyWiki, jeremy...@gmail.com
 the idea of multiple tags fields is something that has been discussed before: there's only a few enhancements needed to make it possible.

+1 very much so! This would open up for a lot of wonderful stuff!

<:-)


 

cmari

unread,
Aug 15, 2015, 1:38:01 PM8/15/15
to TiddlyWiki, Tiddl...@googlegroups.com, jeremy...@gmail.com
This is not actually about the slicer but an extremely minor question about the fold/unfold feature, which I really like - thanks!

I wanted to add color to the new buttons, and was tripped up temporarily because the class format is slightly different. For consistency, could the class for the new images include the word "button"? 

Compare:
svg class="tc-image-edit-button tc-image-button"
to:
svg class="tc-image-fold-others tc-image-button"

cmari


On Saturday, August 1, 2015 at 5:31:06 AM UTC-7, Jeremy Ruston wrote:

Danielo Rodríguez

unread,
Aug 20, 2015, 4:31:30 AM8/20/15
to TiddlyWiki, Tiddl...@googlegroups.com, jeremy...@gmail.com
I like the Idea, and I like the inline editing.
Here are my "critics"
  1. I don't understand what those titles that contains title-para-number means
  2. The auto-generated tiddlers with that weird names fills my tiddler name space very fast. It would be cool to be able to "hide" them.
  3. The "edit mode" it's a bit uncomfortable to reach. Maybe adding a double click-to edit widget to each section title simplifies this.

This is looking cool.

Regards

Jeremy Ruston

unread,
Aug 24, 2015, 3:47:07 PM8/24/15
to cmari, TiddlyWiki
Hi cmari

I wanted to add color to the new buttons, and was tripped up temporarily because the class format is slightly different. For consistency, could the class for the new images include the word "button"? 

The button classes used at the moment are indeed inconsistent in whether they include the word "button". To fix it and maintain backwards compatibility we'd have to use both variants: "tc-image-fold-others" and "tc-image-fold-others-button". I'd be happy to accept a pull request if anyone is interested in doing it.

Best wishes

Jeremy
 

Compare:
svg class="tc-image-edit-button tc-image-button"
to:
svg class="tc-image-fold-others tc-image-button"

cmari

On Saturday, August 1, 2015 at 5:31:06 AM UTC-7, Jeremy Ruston wrote:
I've just pushed a new prerelease that includes an early cut of a tool to slice longer texts into individual tiddlers based on headings and lists. It's based on ideas that have come up in previous discussions about dealing with long, structured tiddlers.

You can try it out at:


You'll need to carefully follow the instructions in the "HelloThere" tiddler:

* Scroll down to the "Sample Text" tiddler and click on the "text slicer" icon
* Click the "import" button in the resulting import listing
* Open the tiddler "Sliced up Sample Text"

You should see a copy of the original text, but you can explore the table of contents to see how it is composed of individual tiddlers that are threaded together by tags.

I'd welcome any feedback on the tool. I would also love some help in finding a better sample text, something public domain that we can re-distribute.

Best wishes

Jeremy


--
Jeremy Ruston
mailto:jeremy...@gmail.com

Jeremy Ruston

unread,
Aug 24, 2015, 3:50:17 PM8/24/15
to Danielo Rodríguez, TiddlyWiki
Hi Danielo

I like the Idea, and I like the inline editing.

Thanks.
  1. I don't understand what those titles that contains title-para-number means
We need to generate titles for paragraphs (headings use the heading text as the title). The current algorithm uses the first few words of the paragraph plus a number if needed to make the title unique. The plan is to let the user specify their preferred prefix for the autogenerated titles. I'm open to suggestions as to other ways of titling these tiddlers.
  1. The auto-generated tiddlers with that weird names fills my tiddler name space very fast. It would be cool to be able to "hide" them.
At the moment the only approach that would hide these paragraphs would be to give them system titles, but that doesn't seem appropriate. If we implemented custom prefixes then the user could of course opt to use system titles.
  1. The "edit mode" it's a bit uncomfortable to reach. Maybe adding a double click-to edit widget to each section title simplifies this.
I agree that double-click to edit would be useful,

Best wishes

Jeremy.

 

This is looking cool.

Regards

Evolena

unread,
Sep 5, 2015, 6:35:04 AM9/5/15
to TiddlyWiki, rdan...@gmail.com, jeremy...@gmail.com
I'm following this plugin with interest, especially with the recent commit to use list field instead of tags for the parental relation (not yet available on the prerelease) :)

It doesn"t manage all wikitext for the moment, ok. But it recognize only one level of list (I tried to slice the HelloThere tiddler), is it intended?

Jeremy Ruston

unread,
Sep 7, 2015, 8:52:15 AM9/7/15
to Evolena, TiddlyWiki, Danielo Rodríguez
Hi Evolena

On Sat, Sep 5, 2015 at 11:35 AM, Evolena <julie.st...@gmail.com> wrote:
I'm following this plugin with interest,

Thanks for the comments (and good to see you posting again!).
 
especially with the recent commit to use list field instead of tags for the parental relation

Yes, it makes much more sense. It's funny: I've been thinking about this approach since TiddlyDocs back in 2008 at Osmosoft, but it's only as I do the programming myself that the pieces have really started to come together.
 
(not yet available on the prerelease) :)

I've now updated the prerelease.
 
It doesn"t manage all wikitext for the moment, ok.

I will be doing some of the simple formatting soon, but in my dayjob I don't need complex things like illustrations or tables at the moment.
 
But it recognize only one level of list (I tried to slice the HelloThere tiddler), is it intended?

It looks like the lists aren't coming through. I'll investigate.

Best wishes

Jeremy.
 


 

Evolena

unread,
Sep 8, 2015, 3:06:53 PM9/8/15
to TiddlyWiki, julie.st...@gmail.com, rdan...@gmail.com, jeremy...@gmail.com
Le lundi 7 septembre 2015 14:52:15 UTC+2, Jeremy Ruston a écrit :
Thanks for the comments (and good to see you posting again!).

I was busy spreading the word by using Tiddlywiki in a project with others people (so I went further in interface customization than I ever did before, with forms to create tiddlers or selectable fields in edit mode), but I've always kept an eye on the Google Group and on the new things on GitHub (except maybe between around the v1.0.8).

The more I practice Tiddlywiki, the more I think that tags should remain for user/content-oriented structure (I can accept system tags since they are less visible), and that fields are the best way to manage all that is "below" the application (all that "make it work", the interface, etc.). Because we can give a semantic to the values of fields through the field name, but it is not as easy for tags.
I've once reuse the whole toc-macro recursive logic to use a "parent" field instead of tags (+ some cosmetic changes).

Jeremy Ruston

unread,
Sep 9, 2015, 8:05:29 AM9/9/15
to Evolena, TiddlyWiki, Danielo Rodríguez
Hi Evolena

I was busy spreading the word by using Tiddlywiki in a project with others people (so I went further in interface customization than I ever did before, with forms to create tiddlers or selectable fields in edit mode), but I've always kept an eye on the Google Group and on the new things on GitHub (except maybe between around the v1.0.8).

Great, I hope you'll be able to share some of the fruits in due course. 

The more I practice Tiddlywiki, the more I think that tags should remain for user/content-oriented structure (I can accept system tags since they are less visible), and that fields are the best way to manage all that is "below" the application (all that "make it work", the interface, etc.). Because we can give a semantic to the values of fields through the field name, but it is not as easy for tags.
I've once reuse the whole toc-macro recursive logic to use a "parent" field instead of tags (+ some cosmetic changes).

Interesting. Would you think that we should migrate away from the current system tag mechanism. For instance with a new "system-tags" field that contains a list. We could support the new schema alongside the old.

Best wishes

Jeremy

Danielo Rodríguez

unread,
Sep 9, 2015, 8:12:58 AM9/9/15
to Jeremy Ruston, TiddlyWiki, Evolena

Hello Jeremy,
Why adding another mechanism instead of using the ones that exists already?

Jeremy Ruston

unread,
Sep 9, 2015, 8:14:45 AM9/9/15
to Danielo Rodríguez, TiddlyWiki, Evolena
Hi Danielo

Why adding another mechanism instead of using the ones that exists already?

I think that Evolena's point is that system tags and using tags for structure get in the way of using tags semantically. But nothing new is being invented here: the core already extensively uses fields to identify tiddlers and their characteristics.

Best wishes

Jeremy

Danielo Rodríguez

unread,
Sep 9, 2015, 9:44:23 AM9/9/15
to TiddlyWiki, rdan...@gmail.com, julie.st...@gmail.com, jeremy...@gmail.com
Hello Jeremy, 

Sorry, I have not explained myself well.

I agree with Evolena suggestion, I was talking about your proposal of creating a parallel tag system.

Regards

Jeremy Ruston

unread,
Sep 9, 2015, 10:16:04 AM9/9/15
to Danielo Rodríguez, TiddlyWiki, Julie S.
Hi Danielo

I agree with Evolena suggestion, I was talking about your proposal of creating a parallel tag system.

Well, the proposal is to shift something from being an ordinary tag to being otherwise encoded via fields; the aim of the proposal is thus explicitly to create a parallel tag system so that the ordinary tag space isn't polluted.

The idea of being able to apply tag semantics to any field has come up before, and aroused interest, so it's definitely on my list.

Best wishes

Jeremy.

 

Regards

Alex Hough

unread,
Oct 14, 2015, 7:29:49 AM10/14/15
to TiddlyWiki
Dear All,

I thought I'd share an "ah-ah" moment appertaining to text slicer plugin.

I receive emails alerts from Google Scholar
I started to make notes on a reply the the alert
* deleted the "no-relpy [...]" subject line gmail automatically gives
* added "tags" to the "To" field in gmail (the field does not insist that the text entered is an email address. 
** I used the To field to direct my notes on the alert towards various projects and groups of people.
* the notes get saved as a draft. the notes can also be given a label and a quick link to all drafts added to gmail's sidebar

Then I copy and pasted my notes into the text slicer TW.

I then edited my slices, downloaded the TW and then attached the file to the draft email.

The draft email is now "tagged" with the phrases I added the the "To" field, and I can search gmail using these terms. It is also a place where I have saved a TW.

All of this gave rise to the idea of a workflow about receiving information, editing it and annotating it, connecting it, categorising it. 

I can begin to see this process taking place in several steps using combinations of media. Where as previously I'd be wanting to get things into my "Node TW" as quickly as possible, I now see the benefits of steps which involve a certain messiness. I am happy with the messiness as I've got it stored as a draft in gmail, and the quality of search features in gmail remove the fear thay my precious insights are not lost. 



Alex

--
You received this message because you are subscribed to the Google Groups "TiddlyWiki" group.
To unsubscribe from this group and stop receiving emails from it, send an email to tiddlywiki+...@googlegroups.com.
To post to this group, send email to tiddl...@googlegroups.com.
Visit this group at http://groups.google.com/group/tiddlywiki.

Alex Hough

unread,
Oct 14, 2015, 11:10:23 AM10/14/15
to TiddlyWiki
Another thought,

I was cutting and pasting from a PDF, line breaks are askew. I used a tool to remove them [1], it would be useful if text slicer could remove line breaks, and add paragraph breaks so that slicing would be easier





Alex

Matabele

unread,
Oct 14, 2015, 12:13:21 PM10/14/15
to tiddl...@googlegroups.com, rdan...@gmail.com, julie.st...@gmail.com, jeremy...@gmail.com
Hi Jeremy

I am still working on user list fields -- your suggestion to extend my MangleTags widget to any field was easily implemented. However, this option opened a can-of-worms. It seems that one of the primary uses of user lists involves maintaining a particular order; and this necessitates management of that order (rather than simply appending the added string.)

My current approach is to make use the existing library of filters to maintain order (sort[], !sort[] etc.), rather than the more direct approach of specifying the position of insertion with attributes (before, after, first, last etc.) I have made some progress -- strings can be added and removed, sorted in various ways, prefix's added and removed, and so on. The one limitation is that only single instances of a string can exist in the list, where some believe lists should permit multiple instances (arbitrary lists.)

My current problem is specifying a point of insertion for an added string using filters (before[mymarker] and after[mymarker].) I now have something working (using a pair of new filters), but the user syntax is horrible. May have to revert to the more direct approach for this particular problem. 


Demo here.

P.S.

The ActionListops widget has now been merged into the core -- details here.

regards

Evolena

unread,
Dec 9, 2015, 9:00:15 AM12/9/15
to tiddl...@googlegroups.com, Tiddl...@googlegroups.com, jeremy...@gmail.com
Hi Jeremy,

Here are my long due returns on this plugin.

First, I'll present my use-case.
I've some PDF files, containing text with chapters, subsections, paragraphs, lists, inserts, icons, images, etc., that I want to turn into a TW. I want to present all these elements in a way that allows easy navigation in the table of content, and with some respect to the original layout. Here is an example of these PDF files (in French): http://issuu.com/lesombresdesteren/docs/esteren_livre_0_prologue/1.
My first attempt, before the text-slicer plugin, satisfied me, but was quite complex. I used fields to store a reference for each tiddler (in a way that an alphanumeric sort gives the orer I want), and the parent reference, which were used to build a recursive representation inspired by the toc-macro. I also used a lot of macros for styling the inserts or some paragraphs that come with icons. I liked the result, but building the structure (references, ...) was quite time-consuming, and I made my test on the shorter of the PDF I want to transform...

So, my hope for the text-slicer plugin is that it can make this building process faster, and I can focus on the presentation layer. And it seems very promising.

With the two means (manually or with text-slicer), I must first transform the PDF into a text file (I'll manage images later, as they need some work). Then I insert some Wikitext formatting (mainly headings and the various sorts of lists).


Here are some of my thought about the text-slicer plugin.

I need a lot of CSS styling. And not tags for "How to apply custom styles by tag", because it only applies to tiddlers, not to transclusions (or I need more CSS-magic). So I modify the js code of each slicer by replacing the "tags" field by a "class" field to be valuated with the classes extracted from the DOM nodes (sorry if I don't have the appropriate vocabulary, I'm neither english-native nor web programmer).
this.addToList(parentTitle,this.addTiddler({
 
"toc-type": "paragraph",
 title
: title,
 text
: "",
 
class: tags
 
}));
I also modify each (interactive, the only ones that I have interest in) template to add a class attribute with the value of this "class" field, for example the list interactive template :
\define body(type:"ul")
<$type$ class={{!!class}}>
<$list filter="""[all[current]] $(tv-exclude-filter)$ +[limit[1]]""" variable="item">
<$list filter={{!!toc-list-filter}} template="$:/plugins/tiddlywiki/text-slicer/templates/interactive/tiddler"/>
</$list>
</
$type$>
\end

<$macrocall $name="body" type={{!!toc-list-type}}/>
Note that I could use the "tags" field for that... but I will end with a lot of CSS classes, and I want to reserve tags for thematic tags. Later.


It would be extremely usefull to be able to easily modify the order of children in a tiddler list. And also to modify the parent of a tiddler (for example, if I have a "conclusion" paragraph of a h2 section which also contains h3 sections, it must be a child of the h2 tiddler, not of the last h3 tiddler). The new listops widget from matabele recently merged on github could be useful here.


I will need a prefix for tiddlers titles. I can modify each js slicer to add it, but as you mentionned that the choice of a prefix was in your list of future enhancement, I would be glad to see it.


There are too much margins the way lists are displayed. So I remove the block mode from the item transclusion in $:/plugins/tiddlywiki/text-slicer/templates/interactive/tiddler and I remove the double linebreaks in $:/plugins/tiddlywiki/text-slicer/templates/interactive/item.


I don't want duplicated text when I open a tiddler created by the slicing, only its recursive content. So I add to all these tiddlers the field "hide-body" to "yes" (modification in each js slicer; this could maybe be an built-in option?), and I remove in $:/plugins/tiddlywiki/text-slicer/ui/view-template-segment this part:
<div class="tc-view-template-document-tiddler-heading">

//This tiddler is part of a document. The content as it appears in the document appears below.//

</div>


I will certainly need at least a support for <cite> elements, and maybe also <blockquote>. I've already done a test to add a js slicer for that, it seems to work.


Finally, as I need to have some styling (background color and border) over several headings (for example 2 headings of a total a 4, with all their descendance), I think I will need support for <div> elements.


I may forget some other things... but I think these are the main points that I've found the need to modify for my first test of text-slicer usage.

Evolena

unread,
Dec 10, 2015, 11:43:41 AM12/10/15
to TiddlyWiki, Tiddl...@googlegroups.com, jeremy...@gmail.com
I've just noticed that state tiddlers for slider are stored in "$:/config/plugins/tiddlywiki/text-slicer/heading-status/..."
I think it would be better to use the same prefix as for other state tiddler: "$:/state/...". This way, if someone don't want to keep state tiddlers on saving, it would be consistent, instead of having to add another bit in the save filter(s).

Matthew Petty

unread,
Apr 28, 2016, 3:07:47 AM4/28/16
to TiddlyWiki, Tiddl...@googlegroups.com, jeremy...@gmail.com
I was also doing stuff with the eCFR. Nasty structure, isn't it?

On Sunday, 2 August 2015 10:22:28 UTC+4, Michael Wiktowy wrote:
On Saturday, August 1, 2015 at 8:31:06 AM UTC-4, Jeremy Ruston wrote:

I'd welcome any feedback on the tool. I would also love some help in finding a better sample text, something public domain that we can re-distribute.


I love it!

I have been chopping up CFRs from the eCFR to build up searchable regulation tiddlywikis in much the same way to help me with my job. I have been doing them manually so far and I can see this speeding things up quite a bit.

Feedback:
1) I think slicing down to the paragraph level might be too far. It adds a lot of autogenerated cruft in the TOCs. There may be some need to go down to this level but I would make the delimiter for levels beyond just headings configurable/regexp-able. In the eCFR example above. It would be useful to go down to the paragraph denoted with a lowercase letter in braces and the subparagraph denoted by a number ... end even sub-subparagraph denoted by lower case roman numerals ... but I don't know what a good title for the paragraphs would be other than their typical legal reference (i.e. §###.## (a)(1)(i)). I have not tackled splitting thing up to that level manually mainly due to the work it would entail and the difficulty in piecing together a tiddler name that won't be error-prone.

2) In the eCFR example, there are a lot of annoying "back to top" links that I have to manually remove. It would be nice if some configurable/regexp text could be excluded from the slicing import.

I'll have to exercise this more (particularly slicing updated regs that supersede previous ones and linked graphics)  since it looks like a great tool for me.

Thanks!
/Mike

Michael Wiktowy

unread,
Apr 28, 2016, 9:20:31 AM4/28/16
to tiddl...@googlegroups.com, jeremy...@gmail.com

A little bit, yes. But it didn't take an inordinate time to parse out manually. I see the potential of the autoslicing making it *way* easier with some custom delimiters and maybe a sed-type filter to filter out all the "back to top" links.

I dread handling updates ... I don't know of a good efficient way to diff eCFR regs other than semi-manually.

/Mike

LG

unread,
May 8, 2016, 5:02:22 PM5/8/16
to TiddlyWiki, Tiddl...@googlegroups.com, jeremy...@gmail.com
Great tool! Thank you. 

I copied Word into Mammoth.js (on a WP site) then into a text editor to remove all the paragraph tags. When I pasted the newly cleaned text over into TW and sliced it, I ran into the million or so new tiddlers at the paragraph level. Has there been any progress on keeping the texts at the header level chunking? 

I have 13.5 years worth of text I'd like to put into TW. Having every single line/paragraph broken into a tiddler would surely break the system (I generate about 300 pages of text every 4 months x 13.5 years). 

Thanks, 

LG

On Saturday, August 1, 2015 at 8:31:06 AM UTC-4, Jeremy Ruston wrote:
I've just pushed a new prerelease that includes an early cut of a tool to slice longer texts into individual tiddlers based on headings and lists. It's based on ideas that have come up in previous discussions about dealing with long, structured tiddlers.

You can try it out at:


You'll need to carefully follow the instructions in the "HelloThere" tiddler:

* Scroll down to the "Sample Text" tiddler and click on the "text slicer" icon
* Click the "import" button in the resulting import listing
* Open the tiddler "Sliced up Sample Text"

You should see a copy of the original text, but you can explore the table of contents to see how it is composed of individual tiddlers that are threaded together by tags.

I'd welcome any feedback on the tool. I would also love some help in finding a better sample text, something public domain that we can re-distribute.

Best wishes

Jeremy

Jeremy Ruston

unread,
May 9, 2016, 11:58:06 AM5/9/16
to LG, TiddlyWiki
Hi LG

Interesting, can you share an excerpt of some of your text?

It sounds like you’d benefit from finer control over the tiddlerisation; perhaps making a tiddler for heading, rather than for each paragraph.

Best wishes

Jeremy

LG

unread,
May 9, 2016, 7:54:08 PM5/9/16
to TiddlyWiki
Dear Jeremy, 

Here's a completely random bit of text I just generated which matches the format of my real document. It has the headers + plain paragraphs with line breaks in-between. This was created in Sublime text editor. I then pasted it and split it. As expected, each paragraph was its own tiddler. 

Btw, I must say I'm overwhelmed and impressed by the community. I posted some questions in a few places online regarding other plugins. Everyone answered within 24 hours. It just blew me away. I'm used to the less friendly communities of major CMS software apps...

Best, 

LG
sample_text_for_slicer.txt

Jeremy Ruston

unread,
May 10, 2016, 2:32:24 PM5/10/16
to tiddl...@googlegroups.com
Hi Amanda

Here's a completely random bit of text I just generated

Terrific random text if I may say so.

which matches the format of my real document. It has the headers + plain paragraphs with line breaks in-between. This was created in Sublime text editor. I then pasted it and split it. As expected, each paragraph was its own tiddler. 

The sample lacks <p> tags around the paragraphs. I tried pasting the text as it is into both a text/html tiddler and an ordinary wikitext tiddler. In both cases, the broken markup prevents things from working properly.

I added <p> tags around the paragraphs (attached). Processing the result as text/html does have the expected output of a separate tiddler for each heading and for paragraph.

If that’s looking like it’s going to generate too many tiddlers for your texts, then the best approach may be to extend the text-slicer plugin with more options, so that we could have a tiddler for each heading plus it’s immediate text.

I won’t have time to explore that for a while. The other option would be to preprocess your texts to merge contiguous paragraphs, putting a couple of <br>s in between.

Btw, I must say I'm overwhelmed and impressed by the community. I posted some questions in a few places online regarding other plugins. Everyone answered within 24 hours. It just blew me away. I'm used to the less friendly communities of major CMS software apps…

Thank you — from my perspective the community is also what makes doing this such fun.

Best wishes

Jeremy


Best, 

JR_edited_sample_text_for_slicer copy.txt

Michael Wiktowy

unread,
May 11, 2016, 1:50:27 AM5/11/16
to tiddl...@googlegroups.com
Hi,

The sample text that I have been parsing up is found in the eCFR site. I take pages like this:
http://www.ecfr.gov/cgi-bin/text-idx?node=pt14.4.413&rgn=div5
Copy all the regulation text into a tiddler as plain text and clean out the "back to top" cruft. Then I cut and paste each section into a separate tiddler using the "New Here" option.

At that point, I have a root tiddler that I can just add a toc macro and a transclusion through a template to display the full regulation or be able to search and display easily a specific section.

Things are difficult in a few spots:
* Parsing it up in the first place ... it's quite labour-intensive
** it would be a nice feature to be able to text-slice using the section symbol (but only when it starts a line)
** a tiddler per paragraph is not terribly useful in the case, especially if the titles are just meaningless sequential IDs
** it is convenient in this case that, due to the section numbering, each tiddler name is unique. I can see how a more flexible slicer would be tricky for more texts to keep the tiddler names unique.

* ordering the resulting tiddlers is tricky ... §413.11 gets ordered before §413.2
** I have used some plugins to made a utility-tiddler to parse tiddler titles to extract out the section and subsection in order to add a section and subsection field to each tiddler to sort numerically on.
*** I have run into situations where i would like to stack up the sort filters ... nsort(section)nsort(subsection) but it seems that only one gets parsed.

I dread updating my regulations tiddlywiki when there are updates. Luckily rulemaking runs at a glacial pace and I haven't had to do so so far ... but that will change eventually.

/Mike


--
You received this message because you are subscribed to a topic in the Google Groups "TiddlyWiki" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/tiddlywiki/vp62YKOsE54/unsubscribe.
To unsubscribe from this group and all its topics, send an email to tiddlywiki+...@googlegroups.com.

To post to this group, send email to tiddl...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

-- 
You received this message because you are subscribed to the Google Groups "TiddlyWiki" group.
To unsubscribe from this group and stop receiving emails from it, send an email to tiddlywiki+...@googlegroups.com.
To post to this group, send email to tiddl...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.
<sample_text_for_slicer.txt>

--
You received this message because you are subscribed to a topic in the Google Groups "TiddlyWiki" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/tiddlywiki/vp62YKOsE54/unsubscribe.
To unsubscribe from this group and all its topics, send an email to tiddlywiki+...@googlegroups.com.

To post to this group, send email to tiddl...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.




--
Michael Wiktowy mwik...@gmail.com

LG

unread,
May 11, 2016, 11:57:31 PM5/11/16
to TiddlyWiki
Dear Jeremy, 

Thanks again for your insight. Just as a follow-up as to what I found. 

I took the sample text I posted previously and used Sublime to add break tags between each paragraph. When outputted in slicer, I got: 

  • All the text which is not a header in one chunk
  • Heading 1
  • Heading 2
  • So on...
Best, 

LG

Jeremy Ruston

unread,
May 12, 2016, 12:07:44 PM5/12/16
to tiddl...@googlegroups.com
Hi LG

I took the sample text I posted previously and used Sublime to add break tags between each paragraph. When outputted in slicer, I got: 

<br> tags are not recognised as paragraph markers; it’s intentional because the role of <br> in HTML is essentially to permit one to type a linebreak within a paragraph. You’ll need to wrap the paragraphs in <p> tags.

Best wishes

Jeremy.


  • All the text which is not a header in one chunk
  • Heading 1
  • Heading 2
  • So on...
Best, 

LG

On Tuesday, May 10, 2016 at 2:32:24 PM UTC-4, Jeremy Ruston wrote:
Hi Amanda

Here's a completely random bit of text I just generated

Terrific random text if I may say so.

which matches the format of my real document. It has the headers + plain paragraphs with line breaks in-between. This was created in Sublime text editor. I then pasted it and split it. As expected, each paragraph was its own tiddler. 

The sample lacks <p> tags around the paragraphs. I tried pasting the text as it is into both a text/html tiddler and an ordinary wikitext tiddler. In both cases, the broken markup prevents things from working properly.

I added <p> tags around the paragraphs (attached). Processing the result as text/html does have the expected output of a separate tiddler for each heading and for paragraph.

If that’s looking like it’s going to generate too many tiddlers for your texts, then the best approach may be to extend the text-slicer plugin with more options, so that we could have a tiddler for each heading plus it’s immediate text.

I won’t have time to explore that for a while. The other option would be to preprocess your texts to merge contiguous paragraphs, putting a couple of <br>s in between.

Btw, I must say I'm overwhelmed and impressed by the community. I posted some questions in a few places online regarding other plugins. Everyone answered within 24 hours. It just blew me away. I'm used to the less friendly communities of major CMS software apps…

Thank you — from my perspective the community is also what makes doing this such fun.

Best wishes

Jeremy


Best, 


-- 
You received this message because you are subscribed to the Google Groups "TiddlyWiki" group.
To unsubscribe from this group and stop receiving emails from it, send an email to tiddlywiki+...@googlegroups.com.
To post to this group, send email to tiddl...@googlegroups.com.
Visit this group at https://groups.google.com/group/tiddlywiki.

Ákos Szederjei

unread,
May 12, 2016, 1:49:02 PM5/12/16
to tiddl...@googlegroups.com
Hello TW users and creators!

In a tiddler I would like to have the following text:

Line1
Line2
Line3
Line4

Is there any other solution besides <br> at the end of the line?

Thank you for the help!

Ákos, wanna be tiddler master :)

Ton Gerner

unread,
May 12, 2016, 1:58:05 PM5/12/16
to TiddlyWiki, ak...@szederjei.eu

Matabele

unread,
May 12, 2016, 3:08:29 PM5/12/16
to TiddlyWiki, ak...@szederjei.eu
Hi Ákos Szederjei 

I often misuse the definition syntax for this purpose -- it may be used to create a title followed by a number of indented lines, or the title may be left blank -- like so.

; Title
: Line 1
: Line 2
: Line 3
: Line 4

 -- or like so.

;
: Line 1
: Line 2
: Line 3
: Line 4

The ';' will carry over -- only one is required and may be followed by indented lines interspersed with other text -- like so.

; Title
: Line 1
: Line 2
And another paragraph another paragraph another paragraph another paragraph another paragraph another paragraph another paragraph another paragraph another paragraph another paragraph another paragraph another paragraph

: Line 3
: Line 4

regards

Ákos Szederjei

unread,
May 12, 2016, 5:44:34 PM5/12/16
to tiddl...@googlegroups.com
&Ton Garner

Thank you, I missed that, or more to my embarrassment, I forgot that option.

@Matabele
Thanks, that is even better. Never though of that!

Ákos

On 5/12/2016 9:08 PM, Matabele wrote:
> Hi Ákos Szederjei
>
> I often misuse the definition syntax for this purpose -- it may be used
> to create a title followed by a number of indented lines, or the title
> may be left blank -- like so.
>
> |
> ;Title
> :Line1
> :Line2
> :Line3
> :Line4
> |
>
> -- or like so.
>
> |
> ;
> :Line1
> :Line2
> :Line3
> :Line4
> |
>
> The ';' will carry over -- only one is required and may be followed by
> indented lines interspersed with other text -- like so.
>
> |
> ;Title
> :Line1
> :Line2
> Andanother paragraph another paragraph another paragraph another
> paragraph another paragraph another paragraph another paragraph another
> paragraph another paragraph another paragraph another paragraph another
> paragraph
>
> :Line3
> :Line4
> |
>
> regards
>
> On Thursday, 12 May 2016 19:49:02 UTC+2, Ákos Szederjei wrote:
>
> Hello TW users and creators!
>
> In a tiddler I would like to have the following text:
>
> Line1
> Line2
> Line3
> Line4
>
> Is there any other solution besides <br> at the end of the line?
>
> Thank you for the help!
>
> Ákos, wanna be tiddler master :)
>
> --
> You received this message because you are subscribed to the Google
> Groups "TiddlyWiki" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to tiddlywiki+...@googlegroups.com
> <mailto:tiddlywiki+...@googlegroups.com>.
> To post to this group, send email to tiddl...@googlegroups.com
> <mailto:tiddl...@googlegroups.com>.
> Visit this group at https://groups.google.com/group/tiddlywiki.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/tiddlywiki/ce46ae4e-ec15-4ca4-91b3-9d31dec96e9f%40googlegroups.com
> <https://groups.google.com/d/msgid/tiddlywiki/ce46ae4e-ec15-4ca4-91b3-9d31dec96e9f%40googlegroups.com?utm_medium=email&utm_source=footer>.
Reply all
Reply to author
Forward
0 new messages