World: r3wp
[Core] Discuss core issues
| older newer | first last |
| Ladislav 22-May-2006 [4635] | (section 15) |
| Gregg 22-May-2006 [4636] | local without defining it in the spec -- I don't think that's a good way to think about modeling it. |
| Joe 22-May-2006 [4637] | I see. This is a major limitation. I see the rebol3 has a blog entry that "bind expands contexts" How could I approach this with the current rebol ? |
| Anton 22-May-2006 [4638] | No, that's right. Contexts cannot be extended with new words (at this time). I would pass a context to your function with the template and all the words in it. This context will have to be built at the beginning. |
| Gregg 22-May-2006 [4639] | It's still not clear to me what the real goal is. i.e. where values come from and how eval-template may be reused without giving it some parameters to guide it. |
| Anton 22-May-2006 [4640] | Sorry, I mean you eval-template function can create a context on the fly with the words in it. |
| Joe 22-May-2006 [4641] | ok Anton, I will try to implement this with your approach |
| Anton 22-May-2006 [4642] | It's not a terribly easy thing to do, but can be done. |
| Joe 22-May-2006 [4643] | Gregg, the GOAL is to be able to have a generic eval-template function that can evaluate multiple templates without having to include hundreds of locals in eval template. The values are generated dynamically with a code-block e.g. eval-template template code-block where code-block is [title: func-title "aha" tag1: emit-whatever ... |
| Ladislav 22-May-2006 [4644] | hmm, I may not understand your wish, Joe, but don't you want to have it this way? |
| Pekr 22-May-2006 [4645] | Today I reread the blog, and I was not comfort with the idea that simple bind should auto-extend context ... maybe just my internal feeling, dunno .... just wanted to state opinion of less experienced reboller :-) |
| Ladislav 22-May-2006 [4646] | eval-template: func [
template
/local res
] [do something with the template...]
|
| Anton 22-May-2006 [4647] | Eval-template will accept a template string 1) convert to block 2) extract the words 3) create a context with these words 4) unset all words in the context 5) do your code bound to the context ( do bind your-code the-context ) 6) handle errors 7) return results |
| Joe 22-May-2006 [4648] | What I've been using in the past is different versions of eval-template for each template and I found the typical bug to be when missing a local given that the templates are evaluated multiple times |
| Anton 22-May-2006 [4649] | After extracting the words you need to convert them all to set-words and put them in a spec block, eg: spec: copy [] foreach word words [append spec to-set-word word] append spec none ; now you can create the context using the spec the-context: context spec ; now unset each word in the context, etc.. |
| Joe 22-May-2006 [4650] | Yes anton, I will code it this way. I am worried that unbinding the words every time is a performance hit |
| Anton 22-May-2006 [4651] | Possibly, we can optimize it later. Binding is supposed to be pretty fast in rebol. |
| Joe 22-May-2006 [4652] | Ladislav, I will try to communicate better. What do you mean by "don't want to have it this way ?" |
| Ladislav 22-May-2006 [4653x2] | eval-template: func [
template
/local res
] [do something with the template...]
|
I just thought, that you wanted to have a function accepting and processing templates | |
| Joe 22-May-2006 [4655] | Yes, my problem is an error handling problem. So far the template blocks are global, and I want to have functions that can build the templates without actually passing the template as a parameter |
| Ladislav 22-May-2006 [4656] | it looks, that you actually don't need UNBIND |
| Gregg 22-May-2006 [4657x2] | Could I ask again for pseudo-code? I'm having the same problem as Ladislav I think. |
If we could step back and look at the problem, without implementation details, it would help me to understand. | |
| Anton 22-May-2006 [4659] | Yes, it's a bit confusing jumping straight into it like this. |
| Gregg 22-May-2006 [4660x2] | I'm only asking because I think you're making this a lot harder than it needs to be, but I can't be sure if I don't understand the problem. |
I've found that I can make things hard for myself, if I fight REBOL, or easy, if I don't. :-) | |
| Anton 22-May-2006 [4662] | Yes, I can't say I find myself needing to do such an operation very often. Been a while since I generated html though. |
| Joe 22-May-2006 [4663x4] | a new attempt. The code that follows works except that I have to define the tags as locals, which is a pain when there are lots of them |
template: to block! "<html><head><title> title </title></head><body>tag1 tag2 tag3</body></html>" blk: remove-each val copy template [tag? val] ?? blk ; == [title tag1 tag2 tag3] eval-template: func [ /local res title tag1 tag2 tag3 ][ res: copy "" bind/copy blk 'res title: "hey" tag1: "this is tag1^/" tag2: "i am tag 2^/" ;tag3: "might not be set^/" repend res bind/copy template 'res res ] probe eval-template | |
so in this example tag3 is none because it was not set in the body but this is clearly an error. I am trying to solve 2 things: | |
1) Not having to declare the locals title tag1 tag2 tag3 in the function spec but have them as locals (might not be possible) | |
| Ladislav 22-May-2006 [4667] | that *is* possible if needed |
| Joe 22-May-2006 [4668x2] | 2) detect when one of the variables has not been set . In the above example with locals defined this can easily be fixed by checking for tags that are none |
Ladislav, then if there is an elegant way to define the locals inside the function body given blk: [title tag1 tag2 tag3] my problem is solved | |
| Gregg 22-May-2006 [4670] | Why not just use FUNCTION with your existing code body and spec; make eval-template funcs on the fly? |
| Anton 22-May-2006 [4671x2] | You would have to define a function every time, yes. |
(for the same reason - contexts can't be extended with new words). | |
| Joe 22-May-2006 [4673] | Yes, but how costly is this ? ( I currently can generate 1000 pages/minute) |
| Gregg 22-May-2006 [4674] | I would still lean towards a higher level solution at some point, but this makes more sense to me than messing with bindings. |
| Ladislav 22-May-2006 [4675] | ;This is one way how to do it: eval-template: lfunc [template [string!]][] [ res: copy "" bind/copy blk 'res title: "hey" tag1: "this is tag1^/" tag2: "i am tag 2^/" ;tag3: "might not be set^/" repend res bind/copy to block! template 'res res ] see http://www.fm.tul.cz/~ladislav/rebol/lfunc.rThis would throw an error if the tag3: line is commented |
| Gregg 22-May-2006 [4676] | Profile it. :-) How many do you *need* to generate, how much data, how many tags, etc. There are a lot of variables here (no pun intended). |
| Joe 22-May-2006 [4677x2] | there are three approaches so far: 1) building function everytime with locals (anton/gregg) 2) template object (anton) and 3) lfunc |
I am going to code this using lfunc for the simplicity it brings and see what I learn Thanks all | |
| Ladislav 22-May-2006 [4679] | the LFUNC approach would cause an error if the 'TAG3 variable were used in the template, otherwise it wouldn't cause an error. Is that what you wanted? |
| Anton 22-May-2006 [4680] | (if your templates are not changing, the corresponding template objects can be made once and reused many times). |
| Joe 22-May-2006 [4681x3] | Ladislav, yes |
Anton, can you code the above example using the template object you proposed | |
Maybe lfunc should be something part of Rebol3 . | |
| Ladislav 22-May-2006 [4684] | :-) |
| older newer | first last |