Several URL in sources

LegacyUser

Ludovic LEDIEU said: Could you please add a way to set several URL in a source ? For exemple, a permanent link to Pas-de-Calais archives only display 1 page. When the document is more than one page we need to keep several URL, one per page, to be able to access each scan. So far I put the supplementary URL in description but they are not clickable.

LegacyUser

Paul said: Showing how computer-illiterate I really am by this question, so please don't shoot me down! But, isn't it possible to create a routine whereby a typed URL is recognised as such and a direct link automatically created? I realise the dedicated field against the created source is programmed to do this, but assume the URL must be recorded in that field for the direct link function to work.

(I'm thinking along the lines of how the text turns blue when I type a URL in Word - not that it provides a direct link by clicking from there, but at least Word "knows" a URL has been typed.)

LegacyUser

Ludovic LEDIEU said: Such a solution with embedded URL in text description is nice too. Benefit is that there is no change on data model.

LegacyUser

David Newton said: Why would this be useful? Elimination of multiple source citations for the same archival document online at different providers. Attach the Familysearch 1911 England and Wales census and then have the ability to put in hyperlinks to the Ancestry version, the Findmypast version and the My Heritage version for example.

LegacyUser

Juli said: The problem is that this would require that a text field (the Notes/Description) be parsed as HTML -- and that way lies madness. There have been several periods where an error was introduced because this text field was being parsed as something other than plain text. (Most commonly, it develops an appetite for whitespace: it eats all line breaks and other attempts at formatting.)

LegacyUser

David Newton said: In other words such URLs need their own text field so the whitespace goblin doesn't start joining all the text together.

LegacyUser

joe martel said: I'm not sure the added complexity for multiple fields is in the plans. I have created Sources that reference multiple URL as you are requesting, for the multi-page artifact where each page has a separate URL, but is treated as one artifact. Today I just paste in the separate URLs, separated by two new lines in the Notes field. Yes it takes more select control to get the full URL but it's there. I also do this when I need an extra citation (less dependent on the query URL I use at other sites).

If there are Sources where the whitespace has been lost, please let us know. That would be a data-loss bug that needs to be attended to. Thanks

LegacyUser

Adrian Bruce said: Paul it's quite reasonable to expect text beginning "www." to be treated as a URL. Unfortunately while it is reasonable to people like you and I (I only use text, I've never tried to program its interpretation) it does look like it's a bit more of a job than it looks like at first glance and I'm grateful to Juli for her neat summary of parsing. And to David for the amusement of his "whitespace goblin"!

As a different but interlinked issue - just think how difficult it is to use "greater than" and "less than" signs in GetSatisfaction.

LegacyUser

David Newton said: Problem is that you absolutly cannot assume URLs will start with www.. After all look at the URL of this very site for an example.

LegacyUser

Paul said: Juli, David and Adrian

Thank you for your explanation of why "idea" could cause problems, regardless of its suitability relating to Ludovic's suggestion.

LegacyUser

David Newton said: Look at the very post you replied to for an example of problems encountered with parsers. I posted the pattern stereotypically associated with the start of a URL, the parser triggered and now there's a useless, dead-end hyperlink in that post.