Page Index Toggle Pages: 1 Send TopicPrint
Normal Topic Recognizing duplicates (Read 8615 times)
Jerry Krinock
Admin
Offline
Posts: 1666
Location: San Jose, California, USA
Joined: Sep 29th, 2008
Re: Recognizing duplicates
Reply #4 - Jun 25th, 2013 at 9:37pm
Print Post  
Dave wrote on Jun 25th, 2013 at 9:16pm:
my understand has always been that the trailing slash in the case of foo.com/bar and foo.com/bar/ is redundant.  Both specify the same path component, a directory not a file.
That's not what I read.  Indeed, my Mac has many regular files on it which do not have file extensions.  Granted, I don't publish my Mac on the web, but I think I could.

Dave wrote on Jun 25th, 2013 at 9:16pm:
I think the preferred construction would include the trailing slash.  There is no doubt what that means.
That would probably be a "best practice" for web admins.  But BookMacster needs to deal with both good and bad practices Smiley

In any case, we've spent enough time on this.  Let me know if the new preference gives you any trouble.
  

Jerry Krinock
Back to top
IP Logged
 
Dave
New Member
Offline
Posts: 2
Joined: Jun 13th, 2013
Re: Recognizing duplicates
Reply #3 - Jun 25th, 2013 at 9:16pm
Print Post  
Thanks Jerry.  I will check out the alpha right now.

Setting aside the domain name case,  my understand has always been that the trailing slash in the case of foo.com/bar and foo.com/bar/ is redundant.  Both specify the same path component, a directory not a file.  Only the addition of an explicit file type extension to "bar" (e.g., bar.html) makes them different.   Maybe I'm out of date.  If in the absence of a trailing slash or an extension if the protocol is to assume some extension (e.g., html) then the two are different.

For the sake of clarity, I think the preferred construction would include the trailing slash.  There is no doubt what that means.
Undecided
  
Back to top
 
IP Logged
 
Jerry Krinock
Admin
Offline
Posts: 1666
Location: San Jose, California, USA
Joined: Sep 29th, 2008
Re: Recognizing duplicates
Reply #2 - Jun 13th, 2013 at 7:33pm
Print Post  
The feature you need has now been added, to a new Alpha build of BookMacster.  You can get the new version with these instructions.  Be sure to switch on the Alpha Versions checkbox.

After you get BookMacster 1.15.5 (or later) installed, activate your document and click in the menu: Bookmarkshelf > Aggressively Normalize URLs.  Click the round "?" Help Button in the sheet which appears to read about how it works.

Let us know if you have any trouble.
« Last Edit: Jun 15th, 2013 at 4:33am by Jerry Krinock »  

Jerry Krinock
Back to top
IP Logged
 
Jerry Krinock
Admin
Offline
Posts: 1666
Location: San Jose, California, USA
Joined: Sep 29th, 2008
Re: Recognizing duplicates
Reply #1 - Jun 13th, 2013 at 3:10pm
Print Post  
Hello, Dave.

Thank you for the question.  Trailing slashes are controversial.

Until May 2012, BookMacster flagged foo.com/bar and foo.com/bar/ as duplicates, as you would like.  But, technically these two bookmarks could refer to different web pages, and a user who did have bookmarks to one of these rare sites where the "trailing slash on a path mattered" complained about it.  After reviewing the IETF standards ("RFC" pages), we concluded that this user was correct, and updated BookMacster to correctly treat trailing slashes on path components as different URLs.

(Note that a trailing slash on the end of the host component is different, and BookMacster will normalize, for example, http://apple.com to http://apple.com/, because they will always refer to the same site.  The "/", in this case, is actually the path component.)

However, you're not the first person since then to request a more relaxed treatment of trailing slashes on paths, and we have an open trouble ticket to provide this feature.  Tell me this:  Given those two bookmarks, foo.com/bar and foo.com/bar/, if BookMacster were to normalize them, which one is correct?  With the slash or without the slash?
  

Jerry Krinock
Back to top
IP Logged
 
Dave
New Member
Offline
Posts: 2
Joined: Jun 13th, 2013
Recognizing duplicates
Jun 13th, 2013 at 2:07pm
Print Post  
I have many instances of duplicates in my bookshelf that are not recognized by BM.  My question is whether there is a setting I'm missing somewhere or is what I need not yet implemented. 

The duplicate URLs are of the form shown in the attached image.

These are not recognized as duplicates but they should be.  What can I do?  Undecided

(please don't suggest manual removal - there are thousands)

Thanks!

* Dave
  
Back to top
 
IP Logged
 
Page Index Toggle Pages: 1
Send TopicPrint