DISCUSSION: Search Engine Friendly URLs (SEF)

[de]

Maybe worth to mention a thread on how to use SEF without mod_rewrite:
http://forum.joomla.org/index.php/topic,1094.0.html

Edit: updated link

[focalguy]

Just to clarify, the htaccess.txt that you need to rename is the one in your mambo/Joomla! directory. I was trying to change the one in my root, while my install was in a sub directory. Changed that one, and everything is fine now.

[Dead Parrot]

One thing thats worth mentioning is if you are running the site off your own machine for development or whatever. Im just going to go through how to load the mod_rewrite module in apache.

Go to your apache folder and find a folder called conf. Open up the file httpd.conf, you will see lots of text that may not make much sense but heres what you have to change.

Find this line:

#LoadModule rewrite_module modules/mod_rewrite.so

Just uncomment the line (take out the #) save the file and restart apache. Its that easy.

Also to enable SEF urls you will need to have .htaccess override enabled. Again in httpd.conf head for this section of code:

# AllowOverride controls what directives may be placed in .htaccess files.
# It can be "All", "None", or any combination of the keywords:
#  Options FileInfo AuthConfig Limit
#
    AllowOverride None

#

If AllowOverride is set to none change it to All, so replace AllowOverride None to AllowOverride All. Save, restart apache. Tada you should be able to have SEF urls by following the steps that Saka posted.

[torkil]

I am developing a site and in that process I am modifying a minicalendar module to produce SEF URLs. The minicalendar (comes with the extcal2 component) displays a minicalendar (shocking!) that is basically a graphical layout of the current month with clickable dates and weeks. The problem is that you can also scroll from one month to another.

My problem is this:
I am displaying the minicalendar on a regular blogcategory page, for instance a news page. The URL is this:
http://www.kampforum.com/content/blogcategory/42/55/ (not a live site yet, so don't bother clicking the link)

What happens is that the minicalendar is displayed on this page like this:

Code: Select all

<  NOVEMBER 2005 >
1 | 2 | 3 | 4 | 5 | 6 | 7
8 | 9 | 10 | 11 | 12 | 13 | 14
(etc... you know: A calendar.)

Now the arrows at the top are clickable, so I should be able to scroll from november to october or december by clicking them. Here comes the hard stuff!

The link looks like this:

Code: Select all

<a href="index.php?option=com_content&task=blogcategory&id=42&Itemid=55&date=2005-12-01">

When I send this string to the sefRelToAbs() function, it will (as far as I can read from sef.php) enter into the if() block at line 211 and assume that this is a regular content page since it has got option=com_content in its URL. Henceforth, it will ignore the date-variable of the query string.

Had the URL contained "option=com_somethingelse", "task=new" or "task=edit", then sefRelToAbs() would have parsed all the variables in the query string as far as I can see, but wouldn't that screw up the layout on the entire blogcategory page?

So my problem is this: How do I enable SEF urls on the previous month and next month link in my calendar? Can I hack this in some way?

[torkil]

Would you call it a Joomla bug/limitation, or is the module built in the wrong way?

[secureoffice]

I see that Joomla has a built in SEF option....

So is there any point in downloading openSEF component?

What are the differences between these?

Cheers

Mochara

[Dead Parrot]

Components like openSEF allow your URL's to look a bit prettier and probably help the search spider go through your site a bit easier (supposedly). They also allow you to rename some of the standard components in your URL.

[here]

I assume the goal is to work toward Human Friendly URLs?

site/category/title/
site/section/
site/blog/year/month/day/

This is my #1 request, please do point to further relevant conversation.

[kenmcd]

Oh thank GOD for Saka and Phil-Taylor.

After 6 questions with no response it can be assumed that Saka will not respond to any questions which do not relate to his commercial component SEF Advance.
Supporting open source does not add to the bottom line profit for this "Joomla Core Developer."

I am still trying to figure out how this is supported in the Joomla community.
Given that I am not member of the "Jooomla good-old-boys-club," the only explanation I can surmise is that no one wants to support something that a core developer is selling.

All changes to the Joomla 1.0.x and 1.1.x SEF are done by Saka.
Look at the Joomla CVS:
http://developer.joomla.org/sf/scm/do/l ... joomla/scm

Heaven forbid that this commercial endeavor developer is actually supportive of the open source community.
Saka, the expert in this area, has not answered ONE post in support of non-commercial SEF.

Gee, am I just stupid or is there a problem here?
Please answer.
I will be holding my breath until this resolved.

Excuse me while I puke.
All this talk about how sleazy Miro is.
While the one expert on Joomla SEF remains silent - to support his commercial component.

Revolting.

Just another reason why I will not be allowed into the Joomla "good ole boys club."
Recently I was asked to put the Joomla CMM list into an official J site.
Yeah, right after the "Joomla Good Ole Boys Club" supports open source.
Ralph.
I got no access to the private forum because of my opinions.
Prove me wrong.
Still waiting for access to the discussion in the PRIVATE forum.

I support Joomla users - while some support the "Joomla good sole boys club."

Go ahead, delete this.
I work my ass off to support Joomla users and all I get is BS from the powers-that-be.
Go ahead, delete this as anti-Joomla rhetoric.
I have done nothing except support Joomla users with useful information.
And I am a pariah because of this.

My guess is that I am a "outsider" because I don't kiss the ass of the devs like Saka.
Sorry, I have an opinion.
I may be right, I may be wrong.
Wow, a fervent Joomla supporter and revolted by the "Good ole boys club" which supports a core developer which provide no support for non-commercial SEF.

Why bother.
This will be deleted or buried by the "good ole boys club."

Commercial developers protected by the "club."

[louis.landry]

Seriously, enough already...

Until we reach a point with the core logic where it does not take double database queries to determine the URL path for SEF urls, I also believe that the existing core solution is better.  To accomplish this with the current architecture is not very feasible while making logical sense.  All of the addons that build SEF urls that make people happy [ Xaneon/OpenSEF | sef404 | etc.. ] use the same basic logic.

When Joomla loads, they parse the query string, load some logic from whatever component is going to load, query the database for names/information to build the SEF URLs, load all the core componentry, load the component, query the database AGAIN for the same basic information that was done in the custom SEF solution, display the page.

The core solution, however, uses the existing variables in the query string to build the SEF URL which does not necessitate that it load any component logic OR MORE IMPORTANTLY query the database multiple times for the same solution.

Moving forward, this architecture will obviously change, but in the meantime... the current core system is functional (not to mention that there are other options) and there are more important underlying architectural problems to solve.

I don't know Emir, and being that I don't know him I certainly couldn't speculate on whether or not he has "held something back".  But what I DO know, is that until the control flow of the system changes so that the types of SEF URLs you are looking for don't require nearly double the work on the system, We are better off working towards a solution to the architectural problem... not adding a custom component to the core simply because it functions.

The fact is that the bickering about the issue is getting old.  I understand that tempers are high over this.  But also try to understand that while most if not all of the core team realizes that an architectural change is required... they are also trying to maintain a certain level of backward compatibility so that every little component that exists doesn't have to be rewritten just to work with the next version.  Because of this, things have to be done in steps and changes that big get phased in. 

1.1 makes some huge changes towards a new, faster, more efficient architecture and will pave the way for the changes we all want.  Please don't let tempers get the best of you... we are all on the same team

Louis

[masterchief]

Ok guys, there is *way* too much [* spam *] flying around hey.  When you have your tempers under control come back and visit, but until then go cool off.

Thanks.

[louis.landry]

Seriously, there are much more important things....

Louis

[PhilTaylor-Prazgod]

I am developing a site and in that process I am modifying a minicalendar module to produce SEF URLs. The minicalendar (comes with the extcal2 component) displays a minicalendar (shocking!) that is basically a graphical layout of the current month with clickable dates and weeks. The problem is that you can also scroll from one month to another.

My problem is this:
I am displaying the minicalendar on a regular blogcategory page, for instance a news page. The URL is this:
http://www.kampforum.com/content/blogcategory/42/55/ (not a live site yet, so don't bother clicking the link)

What happens is that the minicalendar is displayed on this page like this:

Code: Select all

<  NOVEMBER 2005 >
1 | 2 | 3 | 4 | 5 | 6 | 7
8 | 9 | 10 | 11 | 12 | 13 | 14
(etc... you know: A calendar.)

Now the arrows at the top are clickable, so I should be able to scroll from november to october or december by clicking them. Here comes the hard stuff!

The link looks like this:

Code: Select all

<a href="index.php?option=com_content&task=blogcategory&id=42&Itemid=55&date=2005-12-01">

When I send this string to the sefRelToAbs() function, it will (as far as I can read from sef.php) enter into the if() block at line 211 and assume that this is a regular content page since it has got option=com_content in its URL. Henceforth, it will ignore the date-variable of the query string.

Had the URL contained "option=com_somethingelse", "task=new" or "task=edit", then sefRelToAbs() would have parsed all the variables in the query string as far as I can see, but wouldn't that screw up the layout on the entire blogcategory page?

So my problem is this: How do I enable SEF urls on the previous month and next month link in my calendar? Can I hack this in some way?

Hello.

This is quite simple to hack.  Here is a step by step guide:

Open up includes/sef.php in a text editor.

at around line 18 after

Code: Select all

if ($mosConfig_sef) {

add this code block

Code: Select all

//custom
	if (ereg('date,',$_SERVER['REQUEST_URI'])){
		$temp = split('date,', $_SERVER['REQUEST_URI']);
		$temp = split('/', $temp[1]);
		$_GET['date'] 			= $temp[0];
		$_REQUEST['date'] 		= $temp[0];
		$_SERVER['REQUEST_URI'] = str_replace('/date,'.$temp[0].'/','',$_SERVER['REQUEST_URI']);
	}

Then scroll down to the bottom of the file (almost) and on line 273 before the line:

Code: Select all

	$string = $sefstring;

add the folowwing code block:

Code: Select all

//custom
			if (eregi('&date=',$string)) {
				$temp = split('&date=', $string);
				$temp = split('&', $temp[1]);

				$sefstring .= 'date,'.$temp[0].'/';
			}

save and upload the file.

Now when you pass the string:
index.php?option=com_content&task=blogcategory&id=42&Itemid=55&date=2005-12-01

to sefRelToAbs you wll get a SEF URL like:
http://homework/content/blogcategory/42 ... 005-12-01/

and to access the date in your component set it using standard techniques

Code: Select all

 $date = mosGetParam($_REQUEST,'date');
 echo $date;

Notes:
1. This will only work with built in SEF and not any other Commercial or open source SEF solution.
2. Dont forget to backup your sef.php file in case you want to revert back
3. dont forget to backup your modified sef.php before you upgrade to any new version of Joomla.
5. This should work for Mambo 4.5.2 as well.

[adam_nz]

kenmcd ... you should be ashamed of yourself. Did your dog eat your favourite toy? (Admin - I think the earlier post by ken is begging the SNIP function as well).

Believe it or not, I've worked on the Internet since the day it started. Over the years, I've often seen projects grow from friendly, enthusiastic team efforts to ego-touting pi*sing competitions.

Though not new to PHP, I'm a Mambo/Joomla noob - and have been (up to now) quite impressed at how helpful these forums generally are. I'm happy to wait for these [talented] developers (whoever they are) to come up with a decent solution - *in their own time*. I've beta-tested software in the past and there's nothing worse than a *half*-solution.

Adam.

[kenmcd]

Thousands of Joomla users waiting for support from Saka.

I will not be holding my breath.
Not one response to user needs since this heated discussion.
When will SEO be integrated?
http://forum.joomla.org/index.php/topic,427.0.html

No one will suspect that his lack of support for open source users could possibly be related to his commmercial component.

Yeah, right.

[kenmcd]

kenmcd ... you should be ashamed of yourself. Did your dog eat your favourite toy? (Admin - I think the earlier post by ken is begging the SNIP function as well).

Believe it or not, I've worked on the Internet since the day it started. Over the years, I've often seen projects grow from friendly, enthusiastic team efforts to ego-touting pi*sing competitions.

Adam,

I am glad you are a founder of the internet, but I don't see the connection.

Read the other thread and you may be abreast of the issues.

The lead Core developer of the most sought after Joomla feature provides no support to average Joomla users because he sells a component which capitalizes on the short comings of Joomla SEF capabilities.

What part of this do you not understand?

[masterchief]

Ken, yelling at us won't get the problem solved or the code written any faster.  All that does is give you an ulcer.

I have personally put a lot of thought into techniques for improving the way urls can be human readable, pathway lookups, flexibility, scalability etc.  The pathway has also been heavily refactored over the last few days to remove problems on that side things.  Work is moving on it.  Sorry if it's not as fast as you'd like.

Maybe your involvement in OpenSEF will be able to contribute back to the project somehow and help us work thru some of the gotcha's?

[kenmcd]

I know work has been progressing on the SEF issue.
It is not my intention to detract from that generous, very useful, and on-target work.
Some of the core team are obviously supportive of the need of better SEF support in Joomla.
Everyone, including me, are thankful for the attention to this issue and all of your hard work.
You are all doing a fantastic job for thousands of grateful users.
This includes me.
Thank you all.

But, I watch the SVN updates, and I watch the lack of response in the forum.
Not one response from the SEF expert to numerous problems.
What else is to be the conclusion?
My challenge from the previous post stands.
Prove me wrong Joomla SEF expert.
Provide some open source support which supports other solutions than the commercial component.
Thousands of grateful Joomla users will be waiting . . . 

[adam_nz]

Though I'm not actually adding anything new to the thread by posting this  ... I appreciate your your post, Ken. 

[Saka]

I know work has been progressing on the SEF issue.
It is not my intention to detract from that generous, very useful, and on-target work.
Some of the core team are obviously supportive of the need of better SEF support in Joomla.
Everyone, including me, are thankful for the attention to this issue and all of your hard work.
You are all doing a fantastic job for thousands of grateful users.
This includes me.
Thank you all.

But, I watch the SVN updates, and I watch the lack of response in the forum.
Not one response from the SEF expert to numerous problems.
What else is to be the conclusion?
My challenge from the previous post stands.
Prove me wrong Joomla SEF expert.
Provide some open source support which supports other solutions than the commercial component.
Thousands of grateful Joomla users will be waiting . . . 

No I won't respond to any of your queries as long as you are attacking me. Your behavior is so arrogant that it doesn't deserve any reply.

I already explained 100 times that human readable SEF depends highly on the internal DB structure which must be changed in order to create stable SEF in core.
All current solutions (including my commercial one) are patches that require constant support work.

Until DB is structured in a more flexible way, use my core SEF (which is solid good and stable) or some 3rd party solution. There are tons of free ones, so what's your problem?

BTW if Joomla users so desparatly need my input on stuff, they can go back to you and thank you for the lack of it.

[torkil]

People, please. Let me step in with one of my OL's

1) I appreciate the SEF support that Joomla has at the moment. It might be just search engine friendly and not too human eyes friendly, but still... So thanks to Sakic for his input and work on that point.

2) Thanks to Ken McDonald and his teams work on openSEF, and the Xaneon Extensions before them. If it had not been for their free software, http://www.kampforum.com would not work as smoothly as it does today.

3) I think you should stop bickering. I can see that Ken brings up a valid point with a core member being involved in the core of Joomla on one area and then making commercial extensions on a private basis. Still, I think Sakic has explained himself thoroughly here, and I see no need to stress this point further. No good will come out of it anyways. But Ken: You have made your point and your point has been read by everyone. If Sakic, for some strange reason, was to try to lead everyone on and not tell the truth here, it would hurt himself and the product Joomla!, and in concequence he would only be making problems for his own business of selling commercial addons.

4) I believe that what Sakic is saying about the core structure is true, and I do so without having looked at the code. I base my assumption on the fact that the original Xaneon Extensions team left Mambo to develop for Drupal because of exactly this. I know: When I assume I make an ASS out of U and ME, but still...

5) To conclude: Arguing gets us nowhere. Constructive critizism might followed up by brilliant suggestions on the other hand...

[Elpie]

There are some good issues raised in this thread however there is also a lot of frustration being expressed.  It seems that what began as an FAQ has become a mix of general discussion on the topic plus wishlist/feature request.

I note that Ken stated the purpose of the FAQ board as follows:

The purpose of this forum is to provide answers to frequently asked questions.
Please do post your helpful FAQs - with the answers.
Please post plain questions in the appropriate forums.

This discussion has digressed from helpful FAQ's and answers into something more.

Would the moderators please consider moving the posts to another area of the forum while leaving a copy of Saka's original FAQ in place?  The FAQ's are a resource for the whole community and will soon loose their effectiveness if discussion and arguments take the place of FAQ's and answers.

[Saka]

Oh one more thing: anyone who can see beyond his nose and may think there is any substance in Ken's accusations/attacks, go re-read webImagery's post:
http://forum.joomla.org/index.php/topic ... #msg142102
This post will give you an unbiased view of a knowledgable person.

And Ken, reason why I haven't been active on this forums is YOU. I quit coming here because all I could see were your personal attacks and accusations.

Thank God (administrator) you are not moderator any more so I can come here again in peace and without disturbance to help people.
Just like I've helped for 3 years developing OPEN SOURCE Mambo / Joomla core and OPEN SOURCE components and answering user questions before you knew anything about it.

To finish with this year's Nobel prize winner in Literature Harold Pinter's words: "This forum would not allow me to say what I would say to you. It's very short.".

[PhilTaylor-Prazgod]

http://en.wikipedia.org/wiki/Internet_troll

In Internet terminology, a troll is a person who posts inflammatory messages on the internet, such as on online discussion forums, to disrupt discussion or to upset its participants. "Troll" can also mean the inflammatory message itself posted by a troll or be a verb meaning to post such messages. "Trolling" (the gerund) is also commonly used to describe the activity.

Users may have various motives for trolling. A common factor for many is the desire to draw attention to the troll

The more attention the troll's activities draw from users, the more persistent the troll's behavior in the forum. This gives rise to the often repeated protocol in internet culture: "Do not feed the trolls."

Often, a person will post a sincere message about which they are emotionally sensitive. Skillful trolls know that an easy way to upset them is to disingenuously claim that the person is a "troll".

Suppression of information: A particularly nihilistic troll often aims to curb the sharing of helpful information between forum participants. For example, the skilled troll can turn an informative discussion about tips and techniques on coping with disease X... into a completely useless flame fest. This can keep essential information out of the hands of those who need it most, thus proliferating human suffering.

Harassment: following a person — who has been targetted for harassment in one forum, but who has chosen to escape being victimized by moving on — and trolling the forum as a means of making that new "home" an uncomfortable place for that person to be online.

Literature on conflict resolution suggests that labeling participants in internet discussions as "trolls" can perpetuate the unwanted behaviors. A person rejected by a social group, both online and offline, may assume an antagonistic role toward it, and seek to further annoy or anger members of the group. The "troll" label, often a sign of social rejection, may therefore perpetuate trolling.

Better results normally ensue when users take the moderator role and describe more constructive behaviors in a non-judgmental, non-confrontational way. Trolls are excited by trollhunters, and frustrated by "ignorers", and neither of these emotions produce positive results for the forum. Engaging trolls results in "flame wars". Trolls frustrated by the "ignore strategy" may leave the forum (and either troll elsewhere, or become constructive users) or may become progressively more inflammatory until they get a response.

Novice trolls may experience serious "troll's remorse", a feeling of great regret after losing their account (whether it be from an Internet service provider or from a website) as a consequence of their reckless trolling.

There are those who argue that a lack of response to trolling can also inspire trolling, a "Damned if you do, damned if you don't" result. Particularly fanatical or irrational commentators will respond to a forum that irks them largely independent of responses. Trolls also often continue to post, taking umbrage with peripheral arguments or arguments that were less well-founded, until their positions become untenable, then turning either to insults or moving to another topic. By this logic, relentless confrontation through argument of trolls (when such argument is to be found) can be vital.

Point made.

[gram]

This is a touchy subject, and clearly Ken is manifesting the frustration of many other users.  We have sites that do not properly present consistant and singular urls for the search engines, unless we choose NOT to use the SEF funtionality in Joomla.

Our database is not properly constructed to implement an easy solution.

Existing solutions (3pd) require inordinate amounts of maintenance to resolve continuing issues with new components and applications.

Instead of proper 404 errors for pages not found, search engines and users are presented with the front page (another duplicate  page).

We have no internal (script based) knoweldge of the architecture and structure of our sites.

We don't have a solution.  And developing one will require creating new standards for 3pds and modules that encourage consitant usage of command line parameters, as well as grouping of command line parameters that constitute a true page view, and not a component query (not ALL component/page queries should be SEF url formed, at least not all of their parameters).

How can we help?  And can we?  There are many biting into this problem in a hard way, and I can assure you we would like to help.

The best effort I have seen yet, with regard to solving all of these problems lies in the OpenSEF solution, which is committed to being available for Core inclusion at some point but needs more work to mature, I have not seen Sakars solution.

The most appropriate core modifications (db) would not be backwards compatible, requiring some form of import filter to upgrade to, but the solution would be relatively straightforward.

ANDREW:  Expand your core team.  I realize the logistics get terrible, but create and assign a TEAM to this so we can get a solution on its way.  We all understand development life cycles, and the longer this gets put off, the more difficult these types of dicussions are going to become.

Forgive my $.02, it has been painful for me too.

Almost forgot, Thanks for the great CMS guys/girls.  They all have problems, this is simply one of our challenges.  If I didn't think it was the best option out there, I would not weigh into this discussion at all.  Lets include those who can help.

GRAM

[masterchief]

ANDREW:  Expand your core team.  I realize the logistics get terrible, but create and assign a TEAM to this so we can get a solution on its way.  We all understand development life cycles, and the longer this gets put off, the more difficult these types of dicussions are going to become.

I wish everything in an OS project were that simple

Yes there are serious logisitics involved and many, many competing priorities - ACL, SEF, multi-language, 'just make it easier to use', flexibility, scalability, increased security from attack... the list goes on and there is a contingent of zealots for each.  Our challenge is to address all those issues as a project as a whole, not in isolation.

I *know* SEF is a pain - it's a pain for me.

I *know* ACL is a pain - it's holding me back in my commercial work (yes, I do do commercial work, and a lot of it has and does end up back in the core).

We can continuing patching things until the whole wagon falls apart, or we can plan things the correct way and make sure that when we do it, it works, it scales well, it can be used by 3rd party devs as well, etc.

Solutions to these big issues don't just materialise and don't necessarily get fixed by throwing a whole team of resources at them.  Sometimes it just takes one bright spark in the community to look at the framework as a whole and produce some seriously cool code.

Please be understanding in these matters.  We don't deliberately choose not to do things to make life miserable for users as some would think.

[gram]

Thank you for the response Andrew.  I have stood in your shoes (not an OS project, but similiar in size and complexity).  It can be both incredibly rewarding and incredibly painful.  You are doing a great job, and we all appreciate it.

Forgive the blocked shout, I figured it would get your attention only.  Not meant as a real shout.

Your solution must be whatever works for you and your team.  We all want to support that effort, and what we have now for SEF is miles better than what we had without it  .  Easy to forget that sometimes.

If we knew what direction the team was headed that would help tremendously.  Even knowing there was no direction would help.  None of us want to invest our time in improving code that is headed the wrong way.  Can someone help us on this?

It seems that we have an intrinsic need at two base levels, and combining them into one could have great results.

Joomla has no internal reference for site structure (a sitemap).  And we need a consistant strategy for SEF.  The two are very closely related, and it would make sense to reference the same data set for both.  Something that could be cached and or loaded upon script execution, and then used to create either.  I know I am talking out my ears here, but just trying to move forward.  There are members of this community who would work hard to move the code base forward, independent of the team, if we knew a little more.  If we get lucky, someone will do as you suggested, and come up with something really cool.

Thanks,

GRAM

[masterchief]

If we knew what direction the team was headed that would help tremendously.  Even knowing there was no direction would help.  None of us want to invest our time in improving code that is headed the wrong way.  Can someone help us on this?

My currest 'musings' are to attack it on a few fronts (note, this is just thinking out loud)

One is to require a field like the infamous title_alias in the content table.  Actually, step back a sec.

What would be handy is to have two common fields in all data tables:

name - this is the text using in SEF url's, eg "spacenews"
title - this is the 'real' title, eg "News About Space"

Name would need to be held constant to maintain URL stability but the actual title can change to whatever you like.  They could be the same.

Now, let's start with the menu table.  Add a field called pathway, which would be a delimited lookmup field, eg:

id | parent_id | title | name | pathway
1 | 0 | Home | home | /home
2 | 1 | - News | news | /home/news

Now, regardless of the component, you can find the itemid by looking up the pathway (mostly)

I am then thinking you do the same thing for the sections and categories table (actually, they should merge).

id | parent_id | title | name | pathway
1 | 0 | News | news | /news
2 | 1 | - Space | space | /news/space

With that, somehow you should be able to marry the two with some very simple lookup queries.  In fact, you might end up deciding to have on really big pathway lookup table.  Not sure yet.

The next step is to quasi build the sef into the url by default, eg:

SEF Off:

http://localhost/index.php?sef=/home/ne ... e_elevator

SEF On:

http://localhost/home/news/space/space_elevator

That's a very rough specification for a solution.

Now, if there is anyone out there with a client willing to commission me to do this and allow me to feed it back into the core I'm all ears (and if you can get a little extra to do the ACL as well I'll be on cloud nine).

[DoctorFox]

Honestly stop the bickering!

If you have read down this far then I am glad.  If you are the administrator of this forum move all the other bickering to an SEO issues thread, because out of all the messages i read here maybe 4 or 5 had to do with the actual FAQ.

That being said i have some information.

RE: SEF on FreeBSD and Apache 2.2

mod_rewrite when compiled in the server did not work correctly.  I had to install it as a shared object. with

--enable-rewrite=shared

if you are having problems with SEF/SEO on Apache 2.2.0 then you'll want to:

./configure (your options here) --enable_rewrite=shared\
make
make install

[ChuckTrukk]

I am then thinking you do the same thing for the sections and categories table (actually, they should merge).

id | parent_id | title | name | pathway
1 | 0 | News | news | /news
2 | 1 | - Space | space | /news/space

With that, somehow you should be able to marry the two with some very simple lookup queries.  In fact, you might end up deciding to have on really big pathway lookup table.  Not sure yet.

The next step is to quasi build the sef into the url by default, eg:

SEF Off:

http://localhost/index.php?sef=/home/ne ... e_elevator

SEF On:

http://localhost/home/news/space/space_elevator

That's a very rough specification for a solution.

Would this work with the new work that is being done with the pathway? I noticed in a Useability forum a lot of work is being done on the pathway by webImagery (I think I'm right) here-http://forum.joomla.org/index.php/topic ... #msg142423 (he's got some great ideas)... not that you dont...

Would these work hand-in-hand? (I know he is just proposing this).

Chuck