Website import doesn't work sometimes

When I try to import a web page that contains certain characters like an umlaut (de.wikipedia.org/wiki/Carl_F%C3%BCrstenberg) I get the error message “URL import failed”. Scrivener turns “ü” into “%C3%BC”. Even if I make it a “ü” again, the import fails.
Do others have the same problem?

“File->Import->Web page…” is grayed-out for me, and I cannot import web pages at all (which is a different bug, I guess). I did copy/paste from that webpage, and it seemed to work okay.

However, in “Options → Sharing → Import/Export”, the line “Encoding for plain text” might affect the import? Default is Windows-1252, which should do umlauts, but you might try UTF-8 instead. Not sure how well it works at the moment.

I changed the setting, but unfortunately it has no effect. Nevertheless, thank you for the proposal.
By the way: Importing web pages works very well otherwise. However, when I try to copy/paste parts of a web page, I get content that I can hardly use, But one problem after another;-)

Scriv 3 won’t import web pages into manuscript documents. Pity. I could use that feature.

Also, Scrivener will not “Web import” from locally-saved web pages. If I have C:\Downloads\Somedoc.html, Scriv will NOT import it using Web Import, no matter what the name. I can import it using Import->Files. Haven’t tried starting up a server to see if it will import from there.

The keyboard shortcut for Web Import also does not seem to work.

And, I managed to break Scrivener, it seems, playing with this. All instances frozen.

More coming on this topic.

1 Like

Ctrl-shift-W does not open the same Web Page Import screen. Instead, it opens up a dialog for a Firewall exception (I run TinyWall, and Scrivener is whitelisted). To me, that says something odd is happening with the shortcut.

However… if I try to import a WEB page using Import → file…? Scrivener locks up. It doesn’t refuse to import; it just stops working.

Ctrl-shift-W works fine for me. Mostly the import of a website works fine too - but not if certain characters like umlauts are included. Maybe an error with the URL encoding?
A little additional request: The default “http://” for the website import seems superfluous to me. Most of the time I copy the URL directly - and the default should be “https://” by now :wink:

Turns out, TinyWall enables several “global hotkeys” by default, and guess what one of them was?

Indeed, the default should be “https://” because that’s normal nowadays.

Interestingly, Firefox understands “Carl Fürstenberg – Wikipedia” perfectly well. As a workaround, if I save it locally and Import File, the import seems to work just fine. But a workaround should not be needed.

I did find a mention of umlaut issues with QT5: forum.qt.io/topic/34306/qt5-and … %BC-umlaut

At any rate, I’m pretty sure this is a bug.

Thank you so much for your help. Unfortunately the workaround only works partially. The layout and also the pictures are not taken over. It would be nice if my post would lead to a revision of the tool.
Since probably most users insert the URL via copy/paste, in my opinion a default “http(s)://” could be omitted completely.

1 Like