vector40

Thanks! I guess I’ll cool my heels (or try to avoid italicized quotes) until somebody fixes it.

Aaah, brilliant! You’re a wizard. It was imported from HTML.

Tobias,

I’m stumped — one of my columns is again failing to sort numerically, and I can’t see why. All I have are numbers in there from 0 through some various thousands — the header alone is alphanumeric, but that’s not causing problems in other columns.

See: https://emsbasics.com/digital-research-library/spinal-immobilization/

The “Size” column is sorting as strings.

Thoughts?

Thank you — I think that would help. And thanks in general; despite the non-step demands, I know we’re all deeply thankful for your generous work with this project.

Jeepers. You’re right, it’s working for me now as well. Suffice to say that it’s an inconsistent issue.

Incidentally, for anybody who’s also working with generated HTML, the Textism Wordcleaner does a quite good job of getting your code into a workable state (https://textism.com/wordcleaner/).

As a long term solution, what’s the chance of a more robust HTML import somewhere in the future? There are clearly a number of us having issues.

Tobias,

After several exhausting days, I think I’ve pinned down an isolated bug in the HTML import. I’m sure there are other reasons it will kick up an error, but in this case, it’s related to length: if I keep the import under a certain size (not clear if it’s file size, number of characters, or if characters in-cell vs markup makes a difference — I’ve been adding in-cell content to test) it imports fine, if it’s any bigger it fails with a “Data for the import is invalid” message.

Here’s a painstakingly derived test case: https://degreesofclarity.com/emsbasics/goodshort.html

This imports without a problem. However, if you add a single additional character to any cell (spaces count), it fails.

Thoughts?

If I may spin this onto a related topic… the search function seems to be limited to a basic, “find all listed terms” search. Is there any avenue you know of, whether by an extension or whatever, to expand this into a more powerful search function? Whether with basic modifiers like AND and OR, or even just the ability to search for quoted phrases, Google-style. That would add hugely to the capabilities of the tool. Is this possible?

I think I’ve found a workaround. If I export from Docs in HTML, I’m able to open that file using Excel, preserving the links… and I can then export from Excel to HTML, and TablePress successfully opens this with links intact. Presumably Excel is exporting cleaner HTML.

I’m encountering another problem now, however. One of my columns is a numerical count, and when you attempt a user-initiated sort from the output table, it seems to be sorting that column alphabetically rather than numerically (that is, 382 is “before” 43 which is before 9). Is there something I need to do in order to instruct DataTables to interpret this correctly? A few rows in the column do have text in them instead (NA or the like); if that could be causing the problem I can probably find a way to avoid that.

Thanks for all the help.

Sadly I only get errors when I try to import HTML from Docs — I don’t think the formats get along.

At this point I’m resigned to manually linkifying, but there’s another problem. This will be a table that frequently gets updated. If each time, I export from Docs and reimport the table into TablePress, it will overwrite any links I’ve manually created, and I’ll have to go redo them all, won’t it? Is there some way when I import to have the contents added rather than completely overwrite the previous table?

Yep, I know, but there’s no other format that’s cross-compatible (any chance of supporting HTML imports?)

The problem with the conversion extension is that I don’t just have URLS that I want to linkify; I have English text linking to a different URL (i.e. This is the link), and I don’t know of a way to have that convert automatically.