130 replies to this topic

[uramagget]

I have not used osCommerce for long, but I find the system quite inconsistent in the area of templating. There are a number of reasons why osCommerce should use a templating system.

The foremost reason being that it makes the process of users setting up a base layout much faster, since they do not have to hunt around the entire source to change the way the output for X feature should look. This would also make templating and distribution so much easier than it is now. Application logic and template logic would then be separated, which again sets apart designers and programmers and it makes upgrading less of a hassle. The most user-friendly and flexible system I can think of is SUIT: suitframework.com.

Factoring in that the source is messy enough as is, I am not a fan of DB API calls being made in the same place where a table tag is created. These things require a fair enough level of abstraction. If it is being done for languages, there is no reason why not for templates as well. All in all, there doesn't seem to be any disadvantage toward using one.

Please discuss.

Edited by Jan Zonjee, 27 July 2010, 19:42.
commercial interest?

[Jan Zonjee]

uramagget, on 27 July 2010, 19:34, said:

I have not used osCommerce for long, but I find the system quite inconsistent in the area of templating. There are a number of reasons why osCommerce should use a templating system.

The upcoming version 3 uses templates. For osC 2.2 there is the STS contribution.

[uramagget]

Forgive me, although I'm not sure if I understand why a STS would be necessary if there are other systems that accomplish this but in a more proven true and elegant way. It seems like all of the work put into STS would have been better directed towards generally cleaning OSC's source code and addressing other important issues.

[DunWeb]

Arnold,

The development team did not create STS, it was created for use by other users as with most of the other contributions. The development team is......let's say.....not very efficient when it comes to keeping up with the times so it is generally others who provide contributions and support.


Chris

ps. Jan, looks like a definite commercial interest.

[Jan Zonjee]

uramagget, on 27 July 2010, 19:48, said:

Forgive me, although I'm not sure if I understand why a STS would be necessary if there are other systems that accomplish this but in a more proven true and elegant way. It seems like all of the work put into STS would have been better directed towards generally cleaning OSC's source code and addressing other important issues.

I wouldn't know because I have never used the STS contribution myself. However the first uploads of the STS contribution are from 2003. The framework you linked to has a copyright 2008-2010.... osC 2 is pretty old.

[uramagget]

DunWeb, on 27 July 2010, 20:09, said:

Arnold,

The development team did not create STS, it was created for use by other users as with most of the other contributions. The development team is......let's say.....not very efficient when it comes to keeping up with the times so it is generally others who provide contributions and support.


Chris

ps. Jan, looks like a definite commercial interest.

That may be true, however it is being included by default in OSC, so it can be equally said that the community should prioritize the importance of a templating system. Granted, STS does not seem to fulfill much; a quick look through the actual templates files:

<?php DEFINE ('STS_END_CHAR', '$'); ?>

<?php DEFINE ('STS_CONTENT_END_CHAR', '$'); ?>

This should not be there. I do not think it would be a bad idea at all to invest time in having SUIT support built-in to OSC. If it is dependent on a contribution, I have no issue in modifying the core files to make use of the framework.

Jan Zonjee, on 27 July 2010, 20:09, said:

I wouldn't know because I have never used the STS contribution myself. However the first uploads of the STS contribution are from 2003. The framework you linked to has a copyright 2008-2010.... osC 2 is pretty old.

I'm assuming that by referring to the year, you are debating the maturity of both. As far as maturity goes, "age is not an accomplishment and youth is not a sin." - SUIT is years of research. It is very stable, suitable for production use, and has the attributes of a true templating system (in my opinion.) I use it on at least three of my own websites.

[Jan Zonjee]

uramagget, on 27 July 2010, 20:27, said:

'm assuming that by referring to the year, you are debating the maturity of both.

What I mean is that you cannot blame an open source project for not using a system that was developed seven+ years later.

[uramagget]

Jan Zonjee, on 27 July 2010, 20:36, said:

What I mean is that you cannot blame an open source project for not using a system that was developed seven+ years later.

If that is the case, then now that you know of it, I do not see why not. Based on the clean syntax and the flexibility of the rules system it uses, these could be adapted to "suit" ( ) OSC. I don't particularly see why not.

By the way, to DunWeb, I find it rude that you assume that I am doing this for my own gain ("commercial interest") when I hold no affiliation with any of the people involved with the project. I find it more appropriate to cite a specific system rather than vaguely recommend using a template system, because not all of them are particularly good. Not everyone who recommends specific software is doing it for their own gain; I am not a drone promoting shamelessly. I am genuinely interested in the quality of the software I use, and the components of it are no exception.

Edited by uramagget, 27 July 2010, 21:45.

[Jan Zonjee]

uramagget, on 27 July 2010, 21:39, said:

By the way, to DunWeb, I find it rude that you assume that I am doing this for my own gain ("commercial interest")

Chris alluded to the reason why I edited your first post. First posts immediately linking to external websites are highly suspicious in my experience. Lots of spam gets deleted every day from this forum...

Of course it is not always spam, only 98% or so of the cases

[uramagget]

Jan Zonjee, on 27 July 2010, 21:52, said:

Chris alluded to the reason why I edited your first post. First posts immediately linking to external websites are highly suspicious in my experience. Lots of spam gets deleted every day from this forum...

Of course it is not always spam, only 98% or so of the cases

My case is definitely a part of the 2% minority, then, considering what has gone down as of now.

[jcall]

There are dozens upon dozens of conversations on these forums on the use of template systems, for/against, whether or not the dev team should use templating systems, the current state of templating in osC v3, STS vs BTS, STS vs nothing, etc. There is a lot of material that you can use to get further educated on this topic. It'll be faster than re-hashing it.

-jared

[uramagget]

jcall, on 28 July 2010, 05:10, said:

There are dozens upon dozens of conversations on these forums on the use of template systems, for/against, whether or not the dev team should use templating systems, the current state of templating in osC v3, STS vs BTS, STS vs nothing, etc. There is a lot of material that you can use to get further educated on this topic. It'll be faster than re-hashing it. :)

-jared

It seems more to me like the development team has been ignoring this because they believe that over-simplifying these issues with watered down libraries is the best of ideas. You're right; I should stop re-hashing these conversations and in comparison to the others that have attempted, but failed to get them to use a real templating system, actually get the conversion done. <_<

[burt]

There is nothing to stop you from forking on github, then integrating suit (or any other templating system).

Once complete, come back and let us know, then we can merge the changes on our own github forks if we feel your changes are good.

Easy as 123. I look forward to seeing it, in fact, I'd be happy to help in my spare time.

[Debs]

You like templates. You have many choices for that. It is very shallow to think we all like template systems.

[uramagget]

burt, on 28 July 2010, 08:58, said:

There is nothing to stop you from forking on github, then integrating suit (or any other templating system).

Once complete, come back and let us know, then we can merge the changes on our own github forks if we feel your changes are good.

Easy as 123. I look forward to seeing it, in fact, I'd be happy to help in my spare time.

I will do that, seeing as I've found the github repository for OSC v3. Although I'm a bit doubtful that you would just help a random stranger with a fork. ;)

Debs, on 28 July 2010, 19:55, said:

You like templates. You have many choices for that. It is very shallow to think we all like template systems.

Irrelevant; this is not a question of liking them, because that does not change the fact that a good web app. separates its concerns, templates being no exception to this. It is not shallow to be considerate of the people that do not want to sort through spaghetti code and markup just to change a few elements of the catalog. You've been here for six years and this sounds like someone who has no idea of how web apps. should be structured.

[jcall]

uramagget, on 28 July 2010, 23:56, said:

Irrelevant; this is not a question of liking them, because that does not change the fact that a good web app. separates its concerns, templates being no exception to this. It is not shallow to be considerate of the people that do not want to sort through spaghetti code and markup just to change a few elements of the catalog. You've been here for six years and this sounds like someone who has no idea of how web apps. should be structured.

I apologize if I've misread the tone of your post, but you've been here for less than 6 months and apparently haven't read any of the hordes of similar email threads started other folk who are equally well-intentioned but also equally underinformed about the history of the project. You make good points, but they are points that have been made many times before. Have you looked at v3?

In many ways, you're preachin' to the choir. Spend some time searching the forums on this topic. Look at the project release history and timeline. Look at the coding style and structure of other projects that were released around the same time as the 2.2 code base was initially released. Look at the comments that have been made by the actual development team (present and past) on this subject. You'll understand more after that.

One of the beauties of FOSS is that you can put your own resources behind it, forking it if you want, to make the project any way you like. In the meantime, you're not going to endear yourself to potentially helpful community members by complaining. If you don't like osC, don't use it. You didn't pay for it either in $$ or in community forum assistance (judging solely on your account history).

We're not the devs, but we know the project. If you have an actual question, feel free to ask and we'll help how/where we can. If you just want to complain, please take your complaints elsewhere.

Enjoy!

-jared

[uramagget]

jcall, on 29 July 2010, 01:15, said:

I apologize if I've misread the tone of your post, but you've been here for less than 6 months and apparently haven't read any of the hordes of similar email threads started other folk who are equally well-intentioned but also equally underinformed about the history of the project. You make good points, but they are points that have been made many times before. Have you looked at v3?

In many ways, you're preachin' to the choir. Spend some time searching the forums on this topic. Look at the project release history and timeline. Look at the coding style and structure of other projects that were released around the same time as the 2.2 code base was initially released. Look at the comments that have been made by the actual development team (present and past) on this subject. You'll understand more after that.

One of the beauties of FOSS is that you can put your own resources behind it, forking it if you want, to make the project any way you like. In the meantime, you're not going to endear yourself to potentially helpful community members by complaining. If you don't like osC, don't use it. You didn't pay for it either in $$ or in community forum assistance (judging solely on your account history).

We're not the devs, but we know the project. If you have an actual question, feel free to ask and we'll help how/where we can. If you just want to complain, please take your complaints elsewhere.

Enjoy!

-jared

Allow me to clarify; this is not a complaint topic. This is a discussion topic, as I had clearly pointed out in my first post. As I have stated to the posters that you have missed in your displacement with me, I have seen the topics, and I cannot see how they make my words have any less merit. I have seen STS implemented in v3 through the latest revision, and I find the actual STS system unsuitable for what the needs of an e-commerce application are. Hence, I volunteered to fork it and implement a more SUITable alternative, to which a veteran responded may be included if they like the changes. Thus, I would not be complaining; I would be solving. Seen in this light, your conclusion of my intentions is based on a misunderstanding.

As such, I apologize if my original post came off as exceedingly rude.

Edited by uramagget, 29 July 2010, 04:29.

[jcall]

I'm looking forward to seeing the fork! If it takes you less than a few years, then it will be faster than the current development cycle.

-jared

[14steve14]

Surely oscommerce with a template system would end up like zencart. Then you would have two different products using exactly the same code.Os commerce is not that hard to alter, so why is a template system needed. I say personnally reduce the code, make it easier. if version 3 is ever finished it may be worth a look, but so far there are tooo few contributions to make it worth while using.

[uramagget]

14steve14, on 29 July 2010, 16:50, said:

Surely oscommerce with a template system would end up like zencart. Then you would have two different products using exactly the same code.Os commerce is not that hard to alter, so why is a template system needed. I say personnally reduce the code, make it easier. if version 3 is ever finished it may be worth a look, but so far there are tooo few contributions to make it worth while using.

From my experience with Zencart, it is very sub par in contrast to OSC. Your point in 'two different products using exact same code' is also rather fallacious, as having a templating system does not necessarily make you "exactly the same as other software." Plus, Zencart uses Smarty, which is not even comparable. I concede the point that OSC is not very hard to alter, but that is only true if you are used to reading your application logic mangle with your markup, which can get really nasty and confusing when it reaches the size of OSC, because you just may end up accidentally breaking your application altogether when all you were trying to do was move x component to y place. OSCv3 seems to be a step in the right direction, so I will attempt to work with what's there to get this conversion done.