Web link mask malfunctions

[Canada]

4.1.6 Windows 10
I want to set a link in a writer document to a website incl. a jump target there. I understand how to manage this in accordance with help, but the link mask doesn't offer any space for setting the jump target. I cannot insert the jump target, because a pre-set variety of possibilities appears in the box and cannot be deleted (see att.). Of course, I can set the link for the website in general, works fine, but not to a particular sentence there. Any ideas ? Tks in adv. Gustav

[Hagar Delest]

Don't know very much about HTML but basically, the url should be a string that that leads your web browser to a specific target. So if that string exists, it should be enough to paste it in the target field. Else, you may want a feature that will pass to the web browser a string that it cannot understand.
Or is there a more clever process to process targets in a page?

[keme]

Append the target name to the url, separated only by #.

Example: Assuming there is a ChapterThree target anchor defined in the accessibility file, you can use http://<•••>/accessibility.html#ChapterThree

https://www.w3schools.com/html/html_links.asp

[Canada]

Tks, Keme, for this hint. I kept trying to solve the problem myself and did exactly just this, too. But it didn't work. With or without the # attachment, I always can jump to the requested website in general. Even now, that I managed to add the jump target to the url by using the proper way, it doesn't work. It must be a malfuncture or wrong programming. Maybe, someone with much more knowledge than I have can fix it generally.

[robleyd]

You’ll need to look at the source of the web page you want to link and see if there is an existing link target to which you can point the link. If no link exists then you can only link to the page.

[Canada]

Hi David, I'm sorry to tell you that nothing works. I opened the source of the website and copied at least the data for the paragraph that contains the sentence I'm going to use as the jump target, but I cannot convince the program to follow suit. The website itself to reach - no problem. But nothing else. Might be something for a programer to follow up with.

Greeting to down under.

Gustav

[robleyd]

Does the paragraph have a target link? It will look something like <a name=“xxx”>. If not as I said you won’t be able to link to that paragraph.

[keme]

As others have also stated: to jump to specific content within a web page, the page creator must first enable this by inserting a jump target in the page.

If it is the mathematikum page which shows in the picture, there are no <a name="..."> tags, but there are five <a id="..."> tags which also constitute link targets. All those targets are in the page footer and probably not very useful.

The "id" property is in use for a few other entities, so automated extraction of content should be possible. I haven't tried this, and I am not certain that it is possible within OpenOffice apps.

[Canada]

Wow, Keme, it looks like you took this problem personally, with all the search you did. Many thanks for. Well, I have given up to find a way to jump target anywhere on this site. I came across this site accidentally and noticed, that its English version shows a very basic English, to say the least. And while I originally come from the same city in Germany and have a lot of time on my hands, I decided to make them a better translation. Page by page I translate the German version into English and want to insert jump targets in my drafts to make it easier for these guys to place my translations on the appropriate page. Now I have to set a link to the right page only. Can't help it. However, something's wrong with this link feature in writer. Might be a task for the OO developer to fix it. I definitely don't have the necessary skills.

Thanks to all who have tried help me with this issue.

rgds

Gustav

[keme]

Wow, Keme, it looks like you took this problem personally, with all the search you did.
...

Not a lot of work, really.

...
However, something's wrong with this link feature in writer. Might be a task for the OO developer to fix it. I definitely don't have the necessary skills.
...

What in particular do you see as wrong?

[Canada]

It just doesn't work. Inside the web link mask there is a space supposed to open to paste a reference to a jump target on the website (accordingly to "help"), but it never appears. When I hit the button "jump target" after having inserting the website link already, a sub-mask appears, but only if I hit this button twice. This mask offers some opportunities, but nothing to insert a customs choosed jump target. And even if I choose one of the displayed headlines from the page under "headlines", the jump will not happen. Again, only the link to the page works, not the jump target. The needed space for the jump target only appears if I choose "document", but it doesn't work either, of course, even if I combine the web link there by using # with the jump target.

[keme]

Targets should be defined with the anchor tag, and will appear under "bookmarks" in the target list. I can apply those, and they are simply appended to the URL in the target field (same as when you type it in, as per my previous suggestion). It works fine on my install (using AOO version 4.1.6).

There is a lot going on in the background (dialog updates automatically from document content, previous link, and external resource), and it is easy for the dialog to pick up unwanted parts. You need to stay on the narrow, winding path:

• Click the target button once to open the target (or "fragments") list box.
• Click the target button again to download the identifiers.
• Expand the Bookmarks branch.
  This will sometimes initiate an action that looks to be another identifier reload. Bug of sorts, perhaps. Just wait.
• Select the identifier to use.
• Click Apply in the list box
• Click Apply in the hyperlink dialog
• Close the dialog and test your link

Looks like the automated fetching is not supported for https services. You can still type it into the target field manually if you want to link into a https-served page.

It is also possible that the name/ID property is not well formed in the source (e.g. contains spaces or other characters that should be %-encoded). Using an invalid name will make the page open in the beginning.