Mark, This is highly speculative!
I think it is the correct time to mention it because you are suggesting a similar mechanism that I wonder if it could be generalised further. Interesting your talk about load the HTML into it's own DOM tree, I do not have the skills to do it but I have being wondering if there is a way we could do this to use HTML forms and the like with a tiddlywiki widget pulling the values entered out of its own DOM as variables or into fields or tiddlers rather than a submit to php etc.. I know this sounds a little odd and obscure but actually I believe it may be a method not only to enable and "make use of html solutions" that proliferate on the web to be integrated with tiddlywiki, but also by operating in a separate DOM such interactions will not trigger Tiddlywiki refresh until the activity is ended. Why would this be important? As far as I understand most html solutions have multiple pages with their own head and body and movement between pages loads a whole new page. tiddlywiki in effect lets us stay inside a single page, and as a result this integrates all the features we love tiddlywiki for, including its quine nature, single file option and more including its refresh tree. All brilliant stuff. This advantage also makes it difficult to introduce some functionality because various coded solutions must participate in the tiddlywiki tree, via widgets and plugins, this forces the need to integratie into tiddlywiki. My thought is if other websites can move from page to page introducing their own javascript and css as needed is there a way we can emulate this with a generic solution that transfers an independant DOM into tiddlywiki objects, for example I imagine a replacement for the HTML submit that rather than post to a php script, or run javascript it posts the results into a tiddler, variables and or fields, which the tiddlywiki can then respond to. Enabling this would allow additional solutions normaly hosted outside tiddlywiki to be brought inside the tiddlywiki and further strengthen the single file model. This would allow more sophisticated ways to interact with the tiddlywiki's hosts or other websites. I have already experimented with some html/javascript and PHP solutions partially embedded inside tiddlywiki with some success, but they often involve multiple files in the same folder as the tiddlywiki. Many of these files can be hosted inside the tiddlywiki such as CSS, data and html all that I can see that remains is to deal with the following; - pass the results of such interactions into tiddlywiki rather than needing to post outside - Some issues with a shared html head/body Please forgive my speculation, I think I have enough knowledge to see the possibilities but not to ask someone in technical terms so they understand what I am pointing towards. So I am trying to voice this when the opportunity arises, and your comment suggested, that perhaps you could help consider these ideas because you said; *A more ambitious approach would be to load the HTML into it's own DOM > tree, find and parse the elements, and convert to TW text.* Regards Tony On Tuesday, October 15, 2019 at 1:02:06 AM UTC+11, Mark S. wrote: > > Somewhere, lost in this forum, I made such a cleaner/converter that works > with BJ's tiddlyclip. > It uses regular expression to do a usually-but-not-always successful clean > up. It uses JS, > so purists may not like it. > > A more ambitious approach would be to load the HTML into it's own DOM > tree, find and parse > the elements, and convert to TW text. > > Also in the forum is something we did (I believe you helped) with pandoc. > > For tasks like these, it would be handy if TW spoke markdown natively. > There are already > libraries and tools for HTML-to-markdown. > > On Monday, October 14, 2019 at 2:41:51 AM UTC-7, @TiddlyTweeter wrote: >> >> bimlas >> >> https://html-cleaner.com/ >> <https://www.google.com/url?q=https%3A%2F%2Fhtml-cleaner.com%2F&sa=D&sntz=1&usg=AFQjCNEP74kld8-w4No2ohdB_D45-ygRQg> >> >> Nice tool. Useful. >> >> FWIW, it should be possible to make a tool in TW to do that. Plus >> optionally convert HTML to WikiText. >> >> When I get time I'll make a protoype. >> Also there has been some work (I can't find it at the moment that does >> conversion, & I think somone did it for WikiPedia pages?) >> >> Best wishes >> TT >> >> On Monday, 14 October 2019 10:38:02 UTC+2, bimlas wrote: >>> >>> Dear all, >>> >>> When you copy text from a web page and paste it into the wiki, it >>> usually appears in the style of the web page (for example, it has a white >>> background or appears in a different font). >>> >>> I just found an online tool that makes it easier to import HTML texts >>> into TiddlyWiki: Select and copy the desired section from a web page, paste >>> it into this tool and press the "Clean HTML" button to delete the >>> unnecessary parts (inline style, classes). You can paste the stripped text >>> into the wiki without any problems, and it will have the same appearance as >>> other tiddlers. >>> >>> Different options can be set for what to delete. >>> >>> https://html-cleaner.com/ >>> >>> It is also possible to delete all HTML markup. >>> >>> https://html-cleaner.com/features/remove-html-tags/ >>> >> -- 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 [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/tiddlywiki/4617fd08-40f7-45fc-a612-391510ae1c2e%40googlegroups.com.
