Web Page Import keeps failing

I have recently downloaded Scrivener for a free trial. However, the “Import Web Page” function is refusing to work, saying “The web page [web address] could not be imported.” This happens even on web addresses I know are good ones. Please advise.

Try this: From your web browser, drag a link from the page into the Binder (there are other ways to do it, but this seems to be the most universal method that works). You should get an import according to your preferences. If that still does not work, do you have something like Little Snitch installed, by any chance?

Close. It turned out to be our firewall’s anti-spyware component. Once I told the firewall Scrivener was allowed internet access things went without a hitch.

Excellent. Glad you got it working, `tis a nice feature!

Glad you got it working, and thanks for taking the time to post to say that it wasn’t a problem with the program, which saves me fretting over it. :slight_smile:
Best,
Keith

Amber, are you saying that dragging the URL FROM the address bar in e.g. Safari TO the binder should automatically import an archive of that page? Because it doesn’t. I have to

(1) Hit Add Web Page
(2) Drag the URL from the address bar
(3) Clean up the duplicated http:// which Scr already pre-populates the address field with
(4) Name the web page
(5) hit OK.

Now, gosh, it would be wonderful (and how most apps that do this, do it) if I could grap the URL from the address bar, dump it in the Binder, and, Lo!, here’s my webArchive scrivened away… but I can’t. Not anyhow.

EDIT: If I drag a NAMED link from a webpage (so what I’m dragging is Blah blah ) I’m fine. If I just drag “http://blah.blah.com” – which is what the URL is going to be most of the time – nada.

This is a problem, because it means that, before I can get a webarchive of a page into Scr. by drag-and-drop, I have to find a LINK TO THAT PAGE. No good being on the page itself. Cumbersome – about the only thing which is cumbersome in Scr., but my goodness it really is a bi of a palaver… 8)

Oddly enough, the problem is with certain browsers. You should be able to drag a link from a page in any browser, but if you want to drag the address from its bar, you’ll have to use Camino --as Keith has explained elsewhere. God (or Keith) knows why.
Cheers,
E

Which is precisely what I meant by dragging a URL from the page itself – the site content – not any of the Browser bits, such as the URL bar. Sorry for the lack of clarity. The reason I recommended that method for testing is that, as everyone has joined in with, it is the only safe way to do it in multiple browsers, and I did not know which browser you were using. Some, such as Camino, work no matter how you drag it – I suspect this has something to do with the fact that Keith happens to love Camino. cough

So, to get around that clumsiness with a method that is considerably less clumsy than finding a page that references the page you want to archive (ack!), just press Cmd-L in Safari, Cmd-C to copy the URL – switch over to Scrivener, press Cmd-Opt-W (making sure the Binder is focussed and not in the Draft) and paste it in to that dialogue. It seems like a lot of finger dancing, but once you do it a lot, it really disappears.

Holy Mother of God. If only I’d known that before… 8)

Up to a point, but you still have to name the page before Scr. will accept it. The absolute desideratum here is to be able to drag from the address bar, and have the page appear, complete with page name, as a web archive.

A brief survey (in case it’s useful to anyone) of my current browser stocks. (Ideal in green, okay in orange, bad in red)

Dragging from the browser address bar to Scrivener behaves as follows:

[color=green]DevonAgent: Web archive created, with proper page title too!

[color=orange]Camino: creates a new page whose title is “Web Page” and whose content is a web archive of the relevant page.

[color=orange]OmniWeb: as for Camino

[color=orange]NetNewsWire: As for Camino.

[color=red]Firefox: drag and drop from the address bar creates a new page whose title is the URL and whose content is the URL. I.e. as much use as a one-legged man at an arse-kicking party.

[color=red]Safari: Error message: “Some files could not be imported because they were of an unsupported type” (well, who’d a thunk it? But a shame, IMO, because Safari is, after all, the default Mac browser… might put off otherwise enthusiastic Scrivener users.)

[color=red]DevonThink: (Drag URL from DevonThink browser) As for Safari (unsurprisingly). (Drag URL from page header): Nothing.

[color=red]Yojimbo (Drag bookmark) As for Safari.

[color=black]Internet Explorer: who gives a damn?

Hope this helps others.

(EDIT: List ordered; rant about Safari added.)

Thanks for the list, Michael. I’ve been meaning to test them all myself. But of course something more interesting -like the dishes- somehow always comes up.:slight_smile:
Nice to see DevonAgent (same engine as Safari) works.
As for why Camino (wheeze) works, while a real browser like Firefox (based on the same engine) doesn’t?
Again, only God (Keith) knows.

cheers,

E

I suspect the reason DevonAgent works while Safari & DevonThink don’t (all three are WebKit, after all) is that, when you drag a search result from Agent into Scrivener, what you are actually doing is dragging a PATH to DevonAgent’s own temporary WebArchive (stored in /tmp/DevonAgent), rather than an href string.

Looks to me like the order Scrivener is doing things is

(1) Is this a fully-formatted URL (i.e. <A href="http://www.blah.com"Wazoo? If so, name the new page Web Page, grab the bit in quotes, fetch it and archive it.

(2) Is this just a string beginning http://? In which case, um, give up.

(3) Is this a web archive? In which case, load it.

So when using DevonAgent for drag-and-drop, all Scrivener has to do is wait for DevonAgent to deal with the URL, then load DevonAgent’s web archive from the tmp/ file. I.e., Scrivener is cheating. :smiling_imp:

Apparently, God and Keith have company. :slight_smile:

Here are some more, just for the record:

NewsFire: Works just like Camino.

DEVONagent 1.x: Seems to work just like 2.x, you even get a proper title.

And more:

Like Omniweb and Camino [color=orange]Opera properly loads the page but the name is always “Web Page”

I found early on that I couldn’t drag a Safari URL to Scrivener, and, um, I decided that’s a feature. I drag web pages to DevonNote for reference and keep Scrivener for writing. Via its Services menu, Scrivener accepts web text clips. It’s a writing tool, not a browser or a word processor. Let’s keep the focus on basic form and function.

Oy! I’m not getting the web page title in Devonagent (2.1) either: just “Web Page”.

The problem with that philosophy, howarth, is that a lot of writers are using the web for substantial portions of their research these days. I am not just talking about the information available on the WWW, but subscriber databases as well, where immense amounts of information are stored. Now, you can ferret out bits of text from a page (hoping that the web designer did not resort to horrible design practices, resulting in the “forking path of text”) and Service it in – assuming your browser supports it, or you could just swipe the whole page and be done with it. I’ll take option B in most circumstances. It isn’t about making Scrivener a “web browser,” its about getting it to smoothly interface with the most convenient and instantaneously transmitted body of information in the world.

Okay, here’s the science bit:

The standard Cocoa way of dragging a URL from one application to another is via NSURLPboardType. That is, the URL pasteboard type. The responsibility for putting this particular information on the dragging pasteboard lies with the browser. All Scrivener does is look for this information. If it’s there, Scrivener can use it to import a web page.

Now, it’s no secret that I love Camino. It trounces Safari for me. But one of the reasons for this is that it is in many ways a better Cocoa app than Safari, and this is just one example of how. Camino places the URL on the pasteboard as an NSURLPboardType when you try dragging it from its address bar. For some reason, Safari does not. Frankly, the reason for this is beyond me. Safari is Apple’s Cocoa-based browser, and yet it does not place the URL on the pasteboard in the standard Cocoa way. So, it’s between God, Guatama and Apple - nothing to do with me, I’m afraid. Or rather, glad. To say, that is.

Yup.
Best,
Keith

Thank you, Amber, for explaining. I use the web extensively for research and need to store some of this in Scr. along with all my other research material for a particular project. So it IS part of Scr.'s “form and function”–that is why Keith included a ‘Research’ folder. :slight_smile:

Ah. Apple. Yes. “Do as we say, not as we do.” Quite so. All is now clear.