Various trials I've run to confirm various aspects of TWiki behavior.

Contents

to disable TWiki markup

Apparently <nop>, %NOP%, and ! will work under varying circumstances, need to do some testing.

Nesting and precedence of highlighting markup (e.g., bold and italic)

Update: I have done all I need to for now. At some point I should come back and do similar testing for underlined and struck-out text, as Scintilla can handle those (with styles, iirc), and I would like to have those working in Scintilla based editors. I don't know if kwrite/kate editors support underlining and/or struck-out text, but I don't care too much because, as it stands now, for me, the use of kwrite/kate for my askRhk thingie is deprecated -- I don't plan to update its syntax highlighting capabilities anytime in the near future -- at best, maybe sometime after I get Scintilla highlighting working as desired.

The testing in this section is intended to compare the behavior of syntax highlighting on a TWiki vs. on my [T]Wikilike / askSam worksimilar thingie (code named askRhk). It will use two different "sets" of editors (and more are possible) -- one is the kate/kwrite family of editors, the other will be the Scintilla family of editors (after I get a working lexer / folder for Scintilla).

Already, I see some differences in highlighting between a TWiki and a kate/kwrite editor -- I don't know that I can reasonably / easily change the kate/kwrite behavior to match the TWiki behavior, but it is a low priority for me as, for me, the kate/kwrite editors are deprecated -- my goal is to get the Scintilla family of editors to work which will provide support to a large number of editors (on LInux, Windows, and Apple).

Note: "/" is not valid markup on TWiki, it is valid only on my home askRhk thingie.

In general, I intend to create two lines with the same basic markup -- in the first I will try to show the raw markup (by disabling it), and in the second, show the results of that markup. Then, if it differs from results on the kate/kwrite editors, I will add a note.

Other notes: Some differences in -behavior- may result from different font choices on TWiki vs. askRhk, for example, some fonts have / support an italic bold option, some don't. Somewhat similarly, afaik, most fonts do not support a strikeout option. In some cases there is a completely different but similar looking font that shows strikeout, in other cases, strikeout must be obtained by doing an "overtype" operation. (The same may apply for underlining.)

Markup: *copying or /creating some/ text for testing, then trying some markup.

Result: *copying or /creating some/ text for testing, then trying some markup.

Note: / does not work in TWiki to mark as italic (I want to use it in askRhk so that I can use _ for underlining.

Markup: *copying or _creating some_ text for testing, then trying some markup.

Result: *copying or creating some text for testing, then trying some markup.

Note:

Markup: *copying or _creating some_ text* for testing, then trying some markup.

Result: copying or creating some text for testing, then trying some markup.

Note: TWiki can handle (properly) nested sections of italic areas within bold areas (with the nested area in bold italics). kate/kwrite cannot -- it makes the entire area bold, no italics.

Markup: *copying or /creating some/ text* for testing, then trying some markup.

Result copying or /creating some/ text for testing, then trying some markup.

Note: As stated earlier, TWiki does not recognize / as markup for italics, my kate/kwrite highlighter does.

Markup: _copying or *creating some* text_ for testing, then trying some markup.

Result: copying or creating some text for testing, then trying some markup.

Note: TWiki can (also) handle (properly) nested sections of bold areas within italic areas (with the nested area in bold italics). kate/kwrite cannot -- it makes the entire area italic, no bold.

Markup: __copying or creating some text__ for testing, then trying some markup.

Result: copying or creating some text for testing, then trying some markup.

Note: TWiki can use double underbar markup to get bold italic. My kate/kwrite highlighter cannot.

Markup: <u>copying or creating some text</u> for testing, then trying some markup.

Result: copying or creating some text for testing, then trying some markup.

Note: TWiki can use the HTML u markup to get underlined text. Note however that the u markup is (was?) deprecated, and there is another way to get underlined markup (using the style attribut) that will probably work in TWiki -- I don't plan to test atm. My kate/kwrite highlighter cannot display underlined text. Scintilla can underline text, so this is a capability I do plan to implement in my Scintilla highlighter, and at that time should do some tests with respect to nesting and precedence (as I've done somewhat up above for other markup).

Markup: <strike>copying or creating some text</strike> for testing, then trying some markup.

Result: copying or creating some text for testing, then trying some markup.

Note: TWiki can use the HTML strike markup to get struck out text. My kate/kwrite highlighter cannot display struck out text. Scintilla can strike out text, so this is a capability I do plan to implement in my Scintilla highlighter, and at that time should do some tests with respect to nesting and precedence (as I've done somewhat up above for other markup).

<possible future testing -- should also add tests with the nesting reversed / inverted/>

Markup: <u>copying or *creating some* text</u> for testing, then trying some markup.

Result: copying or creating some text for testing, then trying some markup.

Markup: <strike>copying or *creating some* text</strike> for testing, then trying some markup.

Result: copying or creating some text for testing, then trying some markup.

Markup: <u>copying or _creating some_ text</u> for testing, then trying some markup.

Result: copying or creating some text for testing, then trying some markup.

Markup: <strike>copying or _creating some_ text</strike> for testing, then trying some markup.

Result: copying or creating some text for testing, then trying some markup.

<"inverting" the nesting />

Markup: *copying or <u>creating some</u> text* for testing, then trying some markup.

Result: copying or creating some text for testing, then trying some markup

Markup: *copying or <strike>creating some</strike> text* for testing, then trying some markup.

Result: copying or creating some text for testing, then trying some markup.

Markup: _copying or <u>creating some</u> text_ for testing, then trying some markup.

Result: copying or creating some text for testing, then trying some markup

Markup: _copying or <strike>creating some</strike> text_ for testing, then trying some markup.

Result: copying or creating some text for testing, then trying some markup.

Using the automatic anchor created by %TOC%

The automatic anchor created by the %TOC% "command" consists of the words in a heading separated by underbars ("_") up to a total of 32 characters.

To use the automatic anchor, use syntax as shown in Links to Anchors.

No spaces around emdash

Now that I know the proper (English) syntax for an emdash, I want to see if it works on TWiki:

trying--two dashes

trying—HTML entity

Seems fine for normal text—don't know if there'll be some gotchas elsewhere.

\xa0 as a Non-Breaking Space

1.2.�Declaring

This is a test of a \xa0 character to see if it is a non-breaking space.

Yup, it is, as this shows up as a "definition" item. The \xa0 character shows up as a space between the "1.2." and "Declaring".

20200927 (rhk): Something apparently changed from the time I created this entry -- the \xa0 now shows up as a strange "character" (a white "?" inside a black diamond) -- not sure what the genesis / intent of that is.

I forget, how did I insert the \xa0 character? Maybe cut and paste from something like Nedit?

20200927 (rhk): Some further testing (below) -- I forgot (forget) why the non-breaking space was required, so trying a few things:

With an ordinary space:

1. 3 Declaring: Test

With no space:

1.3Declaring

Test

TWiki Name Including Underbar

Testing the ability to link to A TWiki name including an underbar:

TildeSlashDotBash_history TildeSlashDotBash _history

[[TildeSlashDotBash_history]] TildeSlashDotBash_history

It works with the square bracket markup, not otherwise. Interesting thing to note: some people call the links with square brackets "free links" (I've seen this on other wikis if not on TWiki), I tend to call them "forced links". Interesting difference in perspective (and am I always odd man out?).

Links to Anchors

Testing the ability to link to an anchor on another TWiki page without the entire URL, e.g.:

• [[Acronyms#Other_Resources]] square bracket markup
• Acronyms#Other_Resources square bracket markup
• WebChanges#TWiki_Searches not square bracket markup
• WebChanges not square bracket markup

OK, those work, and I'm sure the variations with a plain text label will work as well.

Interwiki Link to a TWiki including an Anchor

not tested, so far

So, an Interwiki link to a TWiki could have a syntax like this?:

[[<InterTWiki>:<Web>.<TWikiPage>#<anchor>]]

Testing Punctuation in Headings / Anchors

AFAICT:

• most punctuation is replaced with "_"
• consecutive (adjacent) "_"s are replaced with a single "_"

Although I could look at the TWiki source to see what they do (and how it is done), testing here first seems somewhat easy.

Test~`!@#$%^&*()+={}[]|\:;"'<,>.?/Heading

Update: I realized I can use an RE like [^a-zA-Z0-9]+ which pretty well matches all punctuation. (If I really wanted to, I could add \r, \n, \t, and ?? to avoid matching them.)

The resulting anchor is Test#Test_Heading which indicates that all the punctuation (that I can think of, or find on the keyboard) is replaced with "_" and multiple "_"s are supressed.

I could paste the above string of punctuation in my macro, but I'm sure some of those characters have to be escaped. I'm not immediately sure which do and which don't. Surely / will have to be escaped.

Change Detection

Attempting to add a ChangeDetection.com panel

| Monitor page for changes | | <form action="http://www.ChangeDetection.com/detect.html" method="get" target=ChangeDetectionWiz> <INPUT TYPE="TEXT" NAME="email" SIZE="12" value="enter email"> | <input type="submit" name="enter" value=" OK " onclick="window.open('', 'ChangeDetectionWiz', 'resizable=yes,scrollbars=yes,width=624,height=460' );return true"> | <a href="http://www.ChangeDetection.com/privacy.html" target=ChangeDetectionPrivacy onclick="window.open('http://www.ChangeDetection.com/privacy.html', 'ChangeDetectionPrivacy', 'resizable=yes,scrollbars=yes,width=624,height=400');return false"> it's private | | | | by ChangeDetection |

Change Notifications

Today (July 4, 2022) I got another notification that the page changed, and again (like on May 5, 2004), I could find no changes in the actual text, even though the TWiki comparison tool showed quite a few changes (in each changed area, the text seemed to be the same before and after the change). It just sort of worried me -- I guess if somebody put some kind of malicious hidden text on the page I'd actually see it on the change comparison tool, but, just in the hope of staying safe, I reverted the change. The worrisome change was noted as "r16 - 2022-07-04 - TWikiGuest"

Just a note, on May 5, 2004 I received a notice that the page changed. I can't (readily) see anything that changed, I'm assuming there was a change and it was more like something the core developers did to the template or something similar.

I wonder if I'll get notification of this change?

Contributors

• RandyKramer
• RandyKramer - 2020-09-26 -- reverted R9 (by "TWikiGuest"): I don't see the benefit of explicitly showing the <nop>s which are inserted to make the page display the "raw text" of the various markups, further, that revision deleted something about the Change Notification attempt, and finally, I wish that people that edit the page would use a real TWikiName and explain the purpose / intent of their changes so that they can be considered and/or discussed. (Aside: I'm not sure I got the notification email when R9 was "created" -- maybe I just overlooked it somehow.) * RandyKramer - 2022-0704 -- reverted r16 - 2022-07-04 - TWikiGuest as I could see no changes -- see the section Change Notifications, above.

PageRating