Re: [web2py] Re: Error in templating system

2011-03-05 Thread Jonathan Lundell
On Mar 5, 2011, at 4:38 PM, Thadeus Burgess wrote: > Ok, when put into this context I agree, it should be fixed. It should all > behave the same. I'll create some more unit tests to expose the current behavior and then remove the special case. In the meantime, the workaround is to observe this

Re: [web2py] Re: Error in templating system

2011-03-05 Thread Thadeus Burgess
Ok, when put into this context I agree, it should be fixed. It should all behave the same. -- Thadeus On Fri, Mar 4, 2011 at 8:22 PM, Jonathan Lundell wrote: > s a strange set of rules, if you ask me. And confusing, as we've already > seen, especially since the error message is not hel

Re: [web2py] Re: Error in templating system

2011-03-05 Thread Jonathan Lundell
On Mar 5, 2011, at 12:16 PM, szimszon wrote: > > It was on purbose. If so, it has only a cosmetic effect; the parser strips the spaces. > > On márc. 5, 20:18, pbreit wrote: >> There was also a suggestion that the space before the = might have been an >> issue: {{ =XML(msg)}} Is that space in

[web2py] Re: Error in templating system

2011-03-05 Thread szimszon
It was on purbose. On márc. 5, 20:18, pbreit wrote: > There was also a suggestion that the space before the = might have been an > issue: {{ =XML(msg)}}  Is that space in there on purpose or by accident?

Re: [web2py] Re: Error in templating system

2011-03-05 Thread Jonathan Lundell
On Mar 5, 2011, at 11:18 AM, pbreit wrote: > There was also a suggestion that the space before the = might have been an > issue: {{ =XML(msg)}} Is that space in there on purpose or by accident? Leading spaces get stripped (between {{ and =), so that wasn't relevant (in this case).

[web2py] Re: Error in templating system

2011-03-05 Thread pbreit
There was also a suggestion that the space before the = might have been an issue: {{ =XML(msg)}} Is that space in there on purpose or by accident?

[web2py] Re: Error in templating system

2011-03-05 Thread szimszon
I thought that after the "=" I could only write one string/variable/ function with string output which is wrapped in write... Anything else I could put in non-python block... Never mind. I can live with what you choose as solution :) Just I say it was working before :-o I will stay with web2py an

Re: [web2py] Re: Error in templating system

2011-03-04 Thread Jonathan Lundell
On Mar 4, 2011, at 6:03 PM, villas wrote: > > On Mar 4, 10:22 pm, Thadeus Burgess wrote: >> I don't think you should have code included in a {{= block. > > +1 Explicit is better > > {{if abc:}} > {{=xyz}} > {{pass}} > > Yes, there are extra brackets, but no one's going to have a problem > wi

[web2py] Re: Error in templating system

2011-03-04 Thread villas
On Mar 4, 10:22 pm, Thadeus Burgess wrote: > I don't think you should have code included in a {{= block. +1 Explicit is better {{if abc:}} {{=xyz}} {{pass}} Yes, there are extra brackets, but no one's going to have a problem with that.

Re: [web2py] Re: Error in templating system

2011-03-04 Thread Thadeus Burgess
I don't think you should have code included in a {{= block. {{= should ONLY be used to wrap anything in its code block with a response.write(). Exactly how it does it now. Doing otherwise is simply bad template design. I consider it a bug if it ever worked before. Allowing this syntax encourages

Re: [web2py] Re: Error in templating system

2011-03-04 Thread Jonathan Lundell
On Mar 4, 2011, at 12:54 PM, Massimo Di Pierro wrote: > > All I meant is that > > {{=msg > pass}} > > should become: > > response.write(msg) > pass > > and not > > response.write(msg > pass) > > since there is no ambiguity. Right. The question the code has to answer, implicitly, is: exactly

[web2py] Re: Error in templating system

2011-03-04 Thread Massimo Di Pierro
All I meant is that {{=msg pass}} should become: response.write(msg) pass and not response.write(msg pass) since there is no ambiguity. On Mar 4, 2:25 pm, Jonathan Lundell wrote: > On Mar 4, 2011, at 12:19 PM, Massimo Di Pierro wrote: > > > > > I agree that that is what it should do. Please

Re: [web2py] Re: Error in templating system

2011-03-04 Thread Jonathan Lundell
On Mar 4, 2011, at 12:19 PM, Massimo Di Pierro wrote: > > I agree that that is what it should do. Please open a ticket about > this. Are you saying (I think) that both cases should terminate on a newline (if present)? > > On Mar 4, 11:20 am, Jonathan Lundell wrote: >> On Mar 4, 2011, at 1:52

[web2py] Re: Error in templating system

2011-03-04 Thread Massimo Di Pierro
I agree that that is what it should do. Please open a ticket about this. On Mar 4, 11:20 am, Jonathan Lundell wrote: > On Mar 4, 2011, at 1:52 AM, szimszon wrote: > > > > > > > > > > > > > web2py™  Version 1.92.1 (2011-02-16 15:04:40) > > Python     Python 2.5.2: /usr/bin/python > > > Traceback (

[web2py] Re: Error in templating system

2011-03-04 Thread villas
The code you posted was clearly badly formed. You need to correct it first. The 'pass' function should always look like this: {{pass}} If something only works in a strange, undocumented way, it is likely to be broken in the future. This is nothing to do with backwards compatibility, it is simply

[web2py] Re: Error in templating system

2011-03-04 Thread szimszon
Don't know could be but The interpreter which checks the syntax about all the pass-es are in right place give me the syntax error. What I try to say that first I visit the page the msg is None so the "if" branch isn't executed at all. If I just byte compile the app I have the same syntax error me

[web2py] Re: Error in templating system

2011-03-04 Thread pbreit
Oh, ok, I see, sorry about that. So it somehow dropped the closing ) after =XML(msg). Could your msg have any special characters in it? I see some UTF-8 related diffs in the code for html.py recently.

[web2py] Re: Error in templating system

2011-03-04 Thread szimszon
Yes I already did it and I mentioned it in the code above but the point is that it was working until I upgraded the framework. The original syntax was good too I mean I can't see what is wrong with it. So it's a backward compatibility issue :( On márc. 4, 10:58, pbreit wrote: > Maybe try closin