tru_tags: A tagging plugin for textpattern

kevinpotts · 2010-03-21 17:05:54

@truist — The problem was indeed the future dating of the articles. A single time attribute fixed everything. Thanks for the speedy response, as always. (And fantastic plugin … it’s been a key piece of my arsenal for years.)

truist · 2010-03-22 17:34:44

Markus, I believe that setting it back to ‘0’ will revert the plugin back to the behavior before I implement the 404 feature – namely that the user will just get an empty page, with no articles.

merz1 · 2010-03-22 18:06:05

Nathan, thanks for the answer. Well, I am not using it anyway, but…

If you find time you might add a funny (4 line?) matrix table into the tru_tags documentation which will precisely show what to expect :-)

truist · 2010-05-21 19:54:36

Survey / Poll:

tru_tags currently has a problem where keywords with “hyphens-and spaces” don’t work. (Keywords with just hyphens or just spaces do work.) I have a user who has offered to implement a fix and submit a patch, but it will be cumbersome to fix it in a backwards-compatible way.

The way tru_tags currently handles spaces is by converting them to hyphens in URLs, and then converting hyphens back to spaces when querying the Textpattern database. The ideal fix would be to either use a less-likely character (like “+” or “—”), or to URLencode the keywords before putting them in URLs (so “with spaces” will become “with%20spaces” in URLs). The problem is that this would break bookmarks and any external links into your sites, which I have very carefully avoided doing in the past.

So, a two-part poll:

1) Which solution would you prefer: (a) a clean-looking but still non-standard character like “+” or “—” or (b) URLencoded keywords in URLs?

2) How important is it to you that we provide a “do it the old way” setting, so your old URLs still work the same way?

I can’t make a real poll here, so please just reply by making comments. Thanks!

alanfluff · 2010-05-21 20:34:03

Hi Nathan,

1) If you’re going to bother to do this, then I assume someone might want a “+” or “—” as part of a tag name(!), so for those (edge case) people, URLencoded would be best. URLencoded seems the most complete solution if I understand this right.

2) I would say it was very important to provide a “do it the old way” facility for sites that are up and that spidered by the SE’s.

Cheers, -Alan

Last edited by alanfluff (2010-05-21 20:35:00)

alanfluff · 2010-05-21 20:37:34

PS: It seems a chunk of work for you for no big gain. Presumably people are keep to loose the visual look of dashes in a tag cloud? jQuery etc could re-write those on the fly, as it’s arguably a presentational request.

Again, if I’ve understood this right ;)

merz1 · 2010-05-22 10:25:35

1) I don’t care

2) ~~Very~~ Not so important (because I have plenty of articles where I use tag links inside the article body. But on the other hand) I have no idea how many ‘n word’ tags I have used in the past inside article bodies but not too many.

2a) Give me a database filter and I clean up the mess :)
2b) SEO wise it’s the same. ‘N word tags’ are seldom and search engines will adopt in a matter of days.

3) But I like the ‘hyphen way’ tru_tags is working now. Does that count as an argument?
3a) Building external URLs with tru_tags (eg Amazon widgets) is safe with hyphens. I am not sure about other characters.
3a1) For Amazon search links I already use striphyphens="1" urlencode="1".
3a2) Because of bad performance all Amazon widgets are already deleted.

Go ahead as you like :)

pfctdayelise · 2010-06-03 13:56:38

Hi,

I’ve been using tru_tags on a few Textpattern sites for as long as I knew about it.
I recently redid my friend’s site and noticed the snippets on tag pages were displaying “special characters” poorly. Mainly I think characters are escaped that shouldn’t be.
I saw that in tru_tags 3.5 there is a comment that “Fixed problems with multi-byte (utf-8, unicode, international) characters.” so I upgraded. But I still have bad output:
compare
http://www.clntranslations.org/tag/social-security/
to
http://www.clntranslations.org/search-results/?q=social+security

I have unticked the box “Convert tags to lowercase by default” under Extensions. I’m running tru_tags 3.6 on Textpattern 4.2.0.

On my tag page I have

<txp:tru_tags_handler form="search_results"/>

this form is also called by my search page, although slightly differently:

<txp:article limit="100" searchform="search_results" />

I tried changing the tru_tags_handler attribute to searchform, but it didn’t fix the smart quotes problem.

Any ideas?

thanks,
Brianna

truist · 2010-06-03 14:54:39

pfctdayelise – the problem is probably with a tag in your search_results form. I’ll bet you’re using something like the TXP tag search_term which won’t work with tru_tags (because it specifically looks for the ‘q’ parameter. Can you paste the contents of your search_results form?

pfctdayelise · 2010-06-04 00:23:59

truist wrote:

pfctdayelise – the problem is probably with a tag in your search_results form. I’ll bet you’re using something like the TXP tag search_term which won’t work with tru_tags (because it specifically looks for the ‘q’ parameter. Can you paste the contents of your search_results form?

Sure, here it is:

I use txp:search_term in another form that the search results wrap around this form. maybe search_result_excerpt is bad? In actual search results, the search term is highlighted, so maybe a bit of magic is going on there. what should I use instead?

jens31 · 2010-07-16 11:01:16

high

i have a little request regarding the sorting of the tags:

i now have following code:
<txp:tru_tags_if_has_tags>
<txp:tru_tags_from_article texttransform=“capitalize” sort=“count” />
</txp:tru_tags_if_has_tags>

~~the one problem is, that its not sorting at all, tags are listed alphabetically (i tried both ways, desc & asc, nothing happens)~~
reading helps: forgot the useoverallcounts=“1” attribute.

PLUS id like to propose another setting: sort=“fixed”, so the plug-in could pull out the tags, as they are inserted in the article..?

of course this would only make sense in the tru_tags_from_article tag

im running txp 4.2 + tru_tags 3.6

thanks in advance and i must say: love the plug-in. google goes nuts, whenever i install it :)

Last edited by jens31 (2010-07-16 11:06:56)

Clayton · 2010-09-04 11:20:42

Using version 3.6, I’m having problems with trailing slashes:

<txp:tru_tags_handler linkpath="http://my.com/tag/" linkpathtail="" form="myform" />

In http://my.com/tag no errors are shown and all seems to work as expected. But when I go to http://my.com/tag/anyoldtag, txp sends up an error that linkpath and linkpathtail are unknown attributes.

Note that I am also using tag cloud without any problems:

<txp:tru_tags_cloud showcounts="title" linkpath="http://my.com/tag/" linkpathtail="" />

merz1 · 2010-09-05 07:40:04

Both attributes are for tru_tags_from_article and not for tru_tags_handler.

The /tag/ sub-directory must be defined via section ‘tag’.

And yes, tru_tags_cloud allows those attributes :)

Edit: Ok, OK, I’ll add a few words :)

tru_tags_from_article and tru_tags_cloud allow to link to external sources.
tru_tags_handler is the txp:tag to handle internal keyword searches. Those search results will be shown (as a list) on the internal section defined for keyword search result pages via tru_tags.
To distinguish between ‘show all tags’ and ‘show keyword search results’ the tru_tags_if_tag_search txp:tag is used on the section (eg /tag/) page template reserved for tru_tags.

Last edited by merz1 (2010-09-05 07:54:07)

Clayton · 2010-09-05 13:46:26

Thanks Markus! I was missing the if_search on my tag page. It all works fine now :) BTW, tru_tags_handler “will also accept all of the attributes used by tru_tags_cloud” according to rainskit.com. It almost did :)

I now have:

<txp:tru_tags_if_tag_search>
<txp:tru_tags_handler form="myform" />
<txp:tru_tags_cloud showcounts="title" linkpath="http://my.com/tag/" linkpathtail="" />
<txp:else />
<txp:tru_tags_cloud showcounts="title" linkpath="http://my.com/tag/" linkpathtail="" />
</txp:tru_tags_if_tag_search>

Clayton · 2010-09-05 14:05:48

I spoke too soon. Now tru_tags_from_article which I have in sidebar on article pages, is not accepting linkpath, linkpathtail or wraptag and break. Very strange. It worked fine before. No error is showing but it just does not strip the tail or wrap the tags.

EDIT: IT WAS ME, DOH! It’s fine.

Last edited by Clayton (2010-09-05 14:12:18)

Textpattern CMS

Textpattern CMS support forum

#676 2010-03-21 17:05:54

Re: tru_tags: A tagging plugin for textpattern

#677 2010-03-22 17:34:44

Re: tru_tags: A tagging plugin for textpattern

#678 2010-03-22 18:06:05

Re: tru_tags: A tagging plugin for textpattern

#679 2010-05-21 19:54:36

Re: tru_tags: A tagging plugin for textpattern

#680 2010-05-21 20:34:03

Re: tru_tags: A tagging plugin for textpattern

#681 2010-05-21 20:37:34

Re: tru_tags: A tagging plugin for textpattern

#682 2010-05-22 10:25:35

Re: tru_tags: A tagging plugin for textpattern

#683 2010-06-03 13:56:38

Re: tru_tags: A tagging plugin for textpattern

#684 2010-06-03 14:54:39

Re: tru_tags: A tagging plugin for textpattern

#685 2010-06-04 00:23:59

Re: tru_tags: A tagging plugin for textpattern

#686 2010-07-16 11:01:16

Re: tru_tags: A tagging plugin for textpattern

#687 2010-09-04 11:20:42

Re: tru_tags: A tagging plugin for textpattern

#688 2010-09-05 07:40:04

Re: tru_tags: A tagging plugin for textpattern

#689 2010-09-05 13:46:26

Re: tru_tags: A tagging plugin for textpattern

#690 2010-09-05 14:05:48

Re: tru_tags: A tagging plugin for textpattern

Board footer