Forums:
Taxonomy upgrade extras:
I'm not sure if it's me failing to understand or do it correctly, but I'm having an awful time getting the Category Browser to work that way I think it's supposed to.
Here's an example:
Let's assume I'm searching for any stories which contain two elements.
Under Items Containing I ensure that the all terms radio button is selected.
Under TG Elements:
I then Ctrl-LeftClick the check boxes for Long Fingernails / Manicures and for Mother-Daughter Outfits
To my way of thinking this should return only stories which contain both elements.
However - I've just run that search and the first two results are:
My Fathers Secret Part 03 which contains the TG element Mother-Daughter Outfits, and
The return of Ashley Bash. which contains the TG element Long Fingernails / Manicures.
This behavior is appropriate for an OR search, not for an AND search. The URL for the above search is http://bigclosetr.us/topshelf/taxonomy/term/84%2C87
If I carry out the same procedure as above, except I select Items Containing any terms I get the same list of stories (at least the first page is the same on the basis of a visual comparison). The URL for this search is http://bigclosetr.us/topshelf/taxonomy/term/84%2B87
How can I do a search for stories which contain all the elements I select?
Few people are aware...
...but most search boxes use common or similar routines.. Characters that refine searches are very seldom used. Try including ^ + & ( ) - in your search expressions to obtain further refinement. Google "regular expressions" to get t more information on this subject. Most search boxes use some form of UNIX regular expression which is a language in and of itself. It certainly solves the problem of google returning tens of thousands of results.
It used to work
Not sure why it isn't now. I'll take a look.
Hugs,
Erin
= Give everyone the benefit of the doubt because certainty is a fragile thing that can be shattered by one overlooked fact.
We've been working less on
We've been working less on refining the old search/display methods, and more on refining use of our general "search system"...
Take a look at this and tell me what you think?
http://bigclosetr.us/topshelf/search/apachesolr_search/?filt...
That's very cool.
That's very cool.
To be honest, I much prefer a command line to a GUI.
I have been playing with regexes as per your earlier reply, but that appears to be searching the body of the text, and leads to poor results.
This approach, however, is clearly limiting itself to the tags.
I seem to recall a reading a post asking about the tid's, however I can't find it.
So the question(s) - what's the syntax for carrying out a search and restricting the search to tags? If I require a list of tid's, where might that be located?
A couple of additional bits:
I see the list of possible search options down the right hand side after I've done a search, and from that I can ascertain the tid for several tags, but it's obviously not a complete list. That's assuming one actually needs the tid's.
It would be very nice to be able to restrict a search to forum posts. Is this possible at the moment?
Thera again...
To restrict a regexp search to tags you would have to know the layout in which the data is stored. I have not the time to explore this in detail but most browsers have an option to view page source I.e. as text only before it has been processed. If you do this you will find a way to anchor your regular expressions to extract only the data you want.
Note that this will work on pretty well any site but it uses a lot of computing power on the site. As a consequence of that you may piss a few people off and even find that a judge considers it hacking or a misuse of the site. It is not so of course, it is at most rude but when you consider that the search box is offered by the (any) site it isn't even rude really. Bringing a site down by overloading it with complex searches shouldn't be possible unless there a lot of concurrent searches. People have forgotten so much about using computers that using once common facilities (which could easily be removed) has come to be seen as an assault by some. So exercise some judgement.
If anyone wants to discuss this more then feel fee to use a pm, mail or even FaceTime.
Regexp on page source
Isn't going to work on such a site as this, where most of the content is dynamically generated.
I can almost guarantee that what gets sent to the browser bears very little relationship to how the data is stored.
Penny
I was not proposing
A reexport on the page source. I was proposing examining page sources looking for commonality in what is passed through to the page source. And it does work at least partially. I don't usually have too much trouble finding what I need to find. But I understand your point and it is not invalid.
If you and I are going to get into discussions about computing and it's history I am going to have to start logging on sober and sice I haven't been sober in eight months (since my parents died) that is going to be too hard so I hope you can accept a bit of blur in the things I say.
Piper offered the following
Piper offered the following as an example
http://bigclosetr.us/topshelf/search/apachesolr_search/?filt...
If one does an initial search for anything and then looks down the right hand side of the site there are a ton of options available to filter the results. Hovering over each of those should show you the URL that the clicking on that link would use ... a least it does in firefox.
Therefore I can determine quite easily that (for instance) the tid for the TG Element Childhood is 4102, and Halloween is 98.
So if I were searching for stories containing those two TG Elements the URL to enter would be
http://bigclosetr.us/topshelf/search/apachesolr_search/?filt...
Rather than fiddle around, however, I was hoping that there might be a list available with the tid's so as to avoid compiling such a one myself.
That and getting confirmation of my theory from Piper wouldn't go astray.
um...
Okay it's been like 8 months since this question was asked and the AND type search function still doesn't work... is there a solution to this besides going into the coding and URL? An AND type search is an extremely basic function for a site like this, and this really shouldn't be such a problem.
I mean hell, I've been trying for the past hour and I still can't figure out how to just do basic searches on this site that don't result in hundreds of pages of just random stories. I don't mean to be a dick but something as simple as accessing and finding the main content of a site should not be this unclear.
Use
Like "+hell +downstairs" as it works.
Use
Like "+hell +downstairs" as it works.
Use
Like "+hell +downstairs" as it works.