Sure, but then the pipe dream of other implementations of aiki are gone in other languages ;). I don't see a problem with just using <?php ... ?> inside of widgets.
The other option is to use smarty or other templating library that already exists and hook into it. I guess my concern is more now that we make aiki core more solid, so we could hook in different markup, etc. I have another proposal for how to handle widgets though that I think would help us, treating them more like visual flow programming language. Yeah, I guess my vote is for just allowing PHP inside of aiki, fuck it! Jon On Sat, Feb 4, 2012 at 2:41 PM, Bassel Safadi <[email protected]> wrote: > of course Jon, we are going to work together, now still discussing which is > the best approach, now let's try to think out of the box, here is a quote > from stackoverflow.com/questions/3723752/where-do-i-start-when-writing-a-new-scripting-language > >> but PHP is a "basic scripting/templating engine". Why reinvent the wheel? >> Just use regexes to strip the disallowed constructs from the source code and >> eval() it. > > > hmmm why not? instead of extensions and then markup that will not have > everything and will waste processing power and regexes. why not just do > that? maybe not eval() it but need to think of how to make use of what > exists already in php, just allow normal php in a way that can't be > dangerous. and make sure that php doesn't get processed if pulled from from > query inside the widget... what others think? > > -- > Bassel Safadi | http://bassel.ws > Skype: i.know.sy | Global: +1-323-545-3855 > > > On Sat, Feb 4, 2012 at 4:25 AM, Jon Phillips <[email protected]> wrote: >> >> Please work together. Power to the committer >> >> On Sat, Feb 4, 2012 at 7:28 AM, Bassel Safadi <[email protected]> >> wrote: >> > >> > On Fri, Feb 3, 2012 at 9:40 PM, Jakub Jankiewicz <[email protected]> wrote: >> >> >> >> >> >> On Fri, 3 Feb 2012 17:54:52 +0200 >> >> Bassel Safadi <[email protected]> wrote: >> >> >> >> > hey guys, soooooo on http://aikiframework.org/wiki/Aiki_markup_2 >> >> > I don't agree on this markup at all, using (for( )for) just because >> >> > it's easier to process than for () doesn't make since. >> >> Is not because it's easier to process but because it look like >> >> ((table_filed)) which in my opinion is the best syntax in current Aiki. >> >> I try to make all look like that. >> >> >> >> > we should keep the original plan that was something like: >> >> > <aiki >> >> > for (){ >> >> > >> >> > } >> >> > aiki> >> >> >> >> Propose complex syntax for this: >> >> http://aikiframework.org/wiki/Aiki_markup_2#Complex_example >> >> and tell me that this will be easier to parse then current Aiki markup. >> >> >> >> I want to work things like: >> >> 1. using extension in "if" statement >> >> 2. extension or aiki function inside loop. >> >> 3. sql "INSERT INTO" inside if statement so you can make conditional >> >> insert. >> >> 4. variable "covering" (when nested statement introduce new ((filed)) >> >> it >> >> will overwrite ((field)) from parent statement. >> >> (so you will have one filed instead of two like right now) >> > >> > >> > I think this is sooooo comlicated: >> > >> > >> > (for( $aiki->utils->fn() as id { >> > (sql( SELECT * FROM table WHERE id = ((id)) { >> > (if( $aiki->utls->fn( ((some_filed)) ) { >> > (*here id will be from first sql not from for loop*) >> > (sql( SELECT * FROM other_table WHERE ((other_id)) = ((id)) { >> > (if( $aiki->membership->username == '[username]' || >> > $aiki->membership->permission == 'librarian'{ >> > (*here id will be from last sql not from for loop nor >> > first >> > sql *) >> > (widget:delete_row( "other_table", ((id)) )widget) >> > })if) >> > })sql) >> > })if) >> > })sql) >> > })for) >> > >> > >> > instead it should and my implementation will make it look like this, and >> > although I don't know why are you using all those queries here. if you >> > can >> > explain your example a bit more will write you the final markup. here >> > it's >> > not clear where those queires are used. >> > >> > >> > <aiki >> > for( $aiki->utils->fn() ) as id { >> > $query = SELECT * FROM table WHERE id = ((id)) { >> > if( $aiki->utls->fn( ((some_filed)) ) { >> > (*here id will be from first sql not from for loop*) >> > $query = SELECT * FROM other_table WHERE ((other_id)) = ((id)) >> > { >> > if( $aiki->membership->username == '[username]' || >> > $aiki->membership->permission == 'librarian'{ >> > (*here id will be from last sql not from for loop nor >> > first >> > sql *) >> > (widget:delete_row( "other_table", ((id)) )widget) >> > }) >> > }) >> > }) >> > }) >> > }) >> > >> > >> > >> >> >> >> > >> >> > of course this looks less new markup and more like php and since aiki >> >> > is written in php I don't see why we need to invent yet another new >> >> > markup while the point of rewriteing the parser was not inventing new >> >> > markup. >> >> >> >> Aiki suppose to hide php from users so they can use sql, html and >> >> javascript. >> >> >> >> And I thought that we need to write parser because Aiki don't have one. >> >> http://en.wikipedia.org/wiki/Parsing#Parser >> > >> > >> > yes, and if we wrote a new parser then we should not reinvent the wheel >> > with >> > totally new language, I don't understand how Jon calls the markup >> > easier, >> > it's totally not understandable dude, if we to have markup it should be >> > closer to php than to new complicated markup, and we should not have >> > (something( )something) all over the place >> > >> >> >> >> >> >> >> >> > >> >> > sooooo and I'm making progress on making >> >> > <aiki >> >> > stuff like in php >> >> > aiki> >> >> >> >> As I post in different tread tell me how you want to look complex >> >> example like this: (and Aiki should allow for more complicated things >> >> as well) >> >> http://aikiframework.org/wiki/Aiki_markup_2#Complex_example >> >> >> >> And remember that there can be javascript inside, and other html. >> >> >> >> > >> >> > I don't think the proposal on >> >> > http://aikiframework.org/wiki/Aiki_markup_2will make aiki markup >> >> > easier instead more complicated. >> >> > >> >> >> >> And this is simple because all markup look the same, and there is very >> >> few constructs. >> >> >> >> take Python for example, python is simple, but it doesn't mean that is >> >> not powerful. >> >> >> >> If you write simple things, Aiki should be simple, but it should allow >> >> you to do complicated stuff too. And markup will be complicated only if >> >> user want to do complicated things. You can't make very complicated >> >> thing to look simple. >> >> >> >> And they should do as much as possible from Aiki using Aiki markup not >> >> to create extensions in php. >> >> >> >> > -- >> >> > Bassel Safadi | http://bassel.ws >> >> > Skype: i.know.sy | Global: +1-323-545-3855 >> >> > >> >> > >> >> > 2012/2/3 Jakub Jankiewicz <[email protected]> >> >> > >> >> > > We can keep it till 1.0 but there should be switch they should >> >> > > not coexist together IMO, we don't need to do things like languages >> >> > > (python or php), we can add to $config 'use_old_syntax' and we can >> >> > > have code that will detect old markup and show error that they use >> >> > > old markup with new Aiki with link to new documentation and that >> >> > > they should use new markup and if they really want to use old one >> >> > > they can turn it on. But if they do it new markup feature will not >> >> > > work. So those few that use old markup will be encourage to use new >> >> > > markup, and we will be able to remove that code in 1.0. >> >> > > >> >> > > There should not be in the code stuff like check for two different >> >> > > markup. Like there is with inherit right now. >> >> > > >> >> > > On Fri, 3 Feb 2012 10:31:19 +0800 >> >> > > Jon Phillips <[email protected]> wrote: >> >> > > >> >> > > > I agree. And, one thing we should keep in mind is that there are >> >> > > > still few aiki-based sites, so we have to consider: >> >> > > > >> >> > > > * do we keep support for the old markup >> >> > > > * keep it for a while >> >> > > > * or just replace it with a well thought out solution >> >> > > > * or a combination of the above >> >> > > > >> >> > > > I would lean towards abstracting the old markup if possible, and >> >> > > > maintaining support for a while, but working on the more thorough >> >> > > > plan and implement it. >> >> > > > >> >> > > > Jon >> >> > > > >> >> > > > On Fri, Feb 3, 2012 at 1:29 AM, Jakub Jankiewicz <[email protected]> >> >> > > > wrote: >> >> > > > > Sorry, I already suggest markup >> >> > > > > http://aikiframework.org/wiki/Aiki_markup_2 >> >> > > > > >> >> > > > > and I registered a blueprint for things I want to remove >> >> > > > > >> >> > > > > >> >> > > > > https://blueprints.launchpad.net/aikiframework/+spec/remove-from-aiki >> >> > > > > >> >> > > > > There was no response either from you or from others, >> >> > > > > only Jon agree with me (in response on this list not >> >> > > > > whiteboard), so I assume that this markup I propose to remove >> >> > > > > are not in use, and everybody agree with that. So I was >> >> > > > > surprise that users use it. >> >> > > > > >> >> > > > > I assume that all that markup that is not in use will go to >> >> > > > > trash, sorry if I offended you by calling it shitty. You've >> >> > > > > done great work with Aiki, I just want to make it better. >> >> > > > > >> >> > > > > If you don't agree with me about those things I want to change, >> >> > > > > then just comment that blueprint and modify that proposal. >> >> > > > > >> >> > > > > >> >> > > > > On Thu, 2 Feb 2012 17:21:39 +0200 >> >> > > > > Bassel Safadi <[email protected]> wrote: >> >> > > > > >> >> > > > >> Jakub: this markup is used a lot and very useful. and so that >> >> > > > >> you know, I don't like the word shitty to be used that much >> >> > > > >> about my code. if you think it look bad, fine go ahead and >> >> > > > >> suggest better markup. but you can't just say shit about >> >> > > > >> others work! >> >> > > > >> >> >> > > > >> >> >> > > > >> On Thu, Feb 2, 2012 at 12:46 PM, Jakub Jankiewicz >> >> > > > >> <[email protected]> wrote: >> >> > > > >> >> >> > > > >> > Do you use those shity markup I wanted to remove it from >> >> > > > >> > Aiki? >> >> > > > >> > >> >> > > > >> > And I think that this is the proper behaviour >> >> > > > >> > >> >> > > > >> > if ((post_content)) is >> >> > > > >> > >> >> > > > >> > ===0 >> >> > > > >> > this is paragraph one >> >> > > > >> > >> >> > > > >> > This is paragraph 2 >> >> > > > >> > ===0 >> >> > > > >> > >> >> > > > >> > and you put it in blockquote >> >> > > > >> > >> >> > > > >> > <blockquote>[p[ ((post_content)) ]p]</blockquote> >> >> > > > >> > >> >> > > > >> > You will have what you want. this is text sensitive not xml >> >> > > > >> > sensitive. >> >> > > > >> > >> >> > > > >> > What about if you have this >> >> > > > >> > >> >> > > > >> > ===0 >> >> > > > >> > <blockquote>this is <span>paragraph one >> >> > > > >> > >> >> > > > >> > This is</span> paragraph 2</blockquote> >> >> > > > >> > ===0 >> >> > > > >> > >> >> > > > >> > You will end with (if it work as you want): >> >> > > > >> > >> >> > > > >> > ===0 >> >> > > > >> > <blockquote><p>this is <span>paragraph one</p> >> >> > > > >> > >> >> > > > >> > <p>This is</span> paragraph 2</p></blockquote> >> >> > > > >> > ===0 >> >> > > > >> > >> >> > > > >> > which will be invalid xml too. How this simple text >> >> > > > >> > sensitive >> >> > > > >> > replacement should know how to handle this? >> >> > > > >> > >> >> > > > >> > >> >> > > > >> > On Thu, 02 Feb 2012 09:46:23 -0000 >> >> > > > >> > Christopher Adams <[email protected]> wrote: >> >> > > > >> > >> >> > > > >> > > Public bug reported: >> >> > > > >> > > >> >> > > > >> > > I noticed this while creating an RSS feed in Aiki: >> >> > > > >> > > >> >> > > > >> > > Let's say my ((post_content)) looks like this: >> >> > > > >> > > >> >> > > > >> > > ===0 >> >> > > > >> > > <blockquote>this is paragraph one >> >> > > > >> > > >> >> > > > >> > > This is paragraph 2</blockquote> >> >> > > > >> > > ===0 >> >> > > > >> > > >> >> > > > >> > > Then the output of [p[ ((post_content)) ]p] will be >> >> > > > >> > > >> >> > > > >> > > ==== >> >> > > > >> > > <p> >> >> > > > >> > > <blockquote>this is paragraph one</p><p> >> >> > > > >> > > This is paragraph 2</blockquote></p><p> >> >> > > > >> > > ==== >> >> > > > >> > > >> >> > > > >> > > You see how the </p><p> is located INISDE the >> >> > > > >> > > <blockquote>? >> >> > > > >> > > >> >> > > > >> > > This will generate an invalid XML warning: >> >> > > > >> > > >> >> > > > >> > > Opening and ending tag mismatch: blockquote line 0 and p >> >> > > > >> > > >> >> > > > >> > > My temporary fix is to use [br[ instead of [p[ >> >> > > > >> > > >> >> > > > >> > > But it would be nice if [p[ doesn't ever produce invalid >> >> > > > >> > > markup! >> >> > > > >> > > >> >> > > > >> > > ** Affects: aikiframework >> >> > > > >> > > Importance: Undecided >> >> > > > >> > > Status: New >> >> > > > >> > > >> >> > > > >> > >> >> > > > >> > -- >> >> > > > >> > Jakub Jankiewicz >> >> > > > >> > twitter: @jcubic >> >> > > > >> > www: http://jcubic.pl >> >> > > > >> > >> >> > > > >> > _______________________________________________ >> >> > > > >> > Mailing list: https://launchpad.net/~aikiframework-devel >> >> > > > >> > Post to : [email protected] >> >> > > > >> > Unsubscribe : https://launchpad.net/~aikiframework-devel >> >> > > > >> > More help : https://help.launchpad.net/ListHelp >> >> > > > >> > >> >> > > > > >> >> > > > > -- >> >> > > > > Jakub Jankiewicz >> >> > > > > twitter: @jcubic >> >> > > > > www: http://jcubic.pl >> >> > > > > >> >> > > > > _______________________________________________ >> >> > > > > Mailing list: https://launchpad.net/~aikiframework-devel >> >> > > > > Post to : [email protected] >> >> > > > > Unsubscribe : https://launchpad.net/~aikiframework-devel >> >> > > > > More help : https://help.launchpad.net/ListHelp >> >> > > > >> >> > > > >> >> > > > >> >> > > >> >> > > -- >> >> > > Jakub Jankiewicz >> >> > > twitter: @jcubic >> >> > > www: http://jcubic.pl >> >> > > >> >> >> >> -- >> >> Jakub Jankiewicz >> >> twitter: @jcubic >> >> www: http://jcubic.pl >> > >> > >> >> >> >> -- >> Jon Phillips 王✳爻气 http://fabricatorz.com ✳ skype: kidproto ✳ irc: rejon >> +1.415.830.3884 (global) ✳ +86-187-1003-9974 (beijing) > > -- Jon Phillips 王✳爻气 http://fabricatorz.com ✳ skype: kidproto ✳ irc: rejon +1.415.830.3884 (global) ✳ +86-187-1003-9974 (beijing) -- You received this bug notification because you are a member of Aiki Framework Admins, which is subscribed to aikiframework. https://bugs.launchpad.net/bugs/925356 Title: The [p[ ]p] command can create invalid markup in certain edge cases Status in Aiki Framework: New Bug description: I noticed this while creating an RSS feed in Aiki: Let's say my ((post_content)) looks like this: ===0 <blockquote>this is paragraph one This is paragraph 2</blockquote> ===0 Then the output of [p[ ((post_content)) ]p] will be ==== <p> <blockquote>this is paragraph one</p><p> This is paragraph 2</blockquote></p><p> ==== You see how the </p><p> is located INISDE the <blockquote>? This will generate an invalid XML warning: Opening and ending tag mismatch: blockquote line 0 and p My temporary fix is to use [br[ instead of [p[ But it would be nice if [p[ doesn't ever produce invalid markup! To manage notifications about this bug go to: https://bugs.launchpad.net/aikiframework/+bug/925356/+subscriptions _______________________________________________ Mailing list: https://launchpad.net/~aikiframework.admins Post to : [email protected] Unsubscribe : https://launchpad.net/~aikiframework.admins More help : https://help.launchpad.net/ListHelp
