autocomplete in gst REPL

> While I tried out Nico's sample code, I noticed that
> the tab-auto-complete in gst's REPL is not as all-knowing
> as it could be:
>
> st>  error sign<TAB>
> signal:               signalOn:             signalWithArguments:
> signalClass:          signalWith:           signed:
>
> although error signalingContext is defined in my gst.
>
> Bug or feature?

> On 07/02/2009 08:34 PM, Stefan Schmiedl wrote:
> > While I tried out Nico's sample code, I noticed that
> > the tab-auto-complete in gst's REPL is not as all-knowing
> > as it could be:
> >
> > st>  error sign<TAB>
> > signal:               signalOn:             signalWithArguments:
> > signalClass:          signalWith:           signed:
> >
> > although error signalingContext is defined in my gst.
> >
> > Bug or feature?
>
> Half-half.  The completion list is built from the symbol table, which
> includes both variables and methods.  Including all methods including
> unary methods would have meant having variables too in the completion
> list.  I can change it if people prefer so.

> On 07/02/2009 08:34 PM, Stefan Schmiedl wrote:
>>
>> While I tried out Nico's sample code, I noticed that
>> the tab-auto-complete in gst's REPL is not as all-knowing
>> as it could be:
>>
>> st>  error sign<TAB>
>> signal:               signalOn:             signalWithArguments:
>> signalClass:          signalWith:           signed:
>>
>> although error signalingContext is defined in my gst.
>>
>> Bug or feature?
>
> Half-half.  The completion list is built from the symbol table, which
> includes both variables and methods.  Including all methods including unary
> methods would have meant having variables too in the completion list.  I can
> change it if people prefer so.
>
> Paolo
>
>
> _______________________________________________
> help-smalltalk mailing list
> [hidden email]
> http://lists.gnu.org/mailman/listinfo/help-smalltalk
>

> On Thu, 2 Jul 2009 22:45:16 -0400
>
> Tim Menzies <[hidden email]> wrote:
> > i do a lot of teaching with GST smalltalk and the auto-completion is a
> > cool tool for making beginners more comfortable with the environment.
>
> Especially when they see their self-defined methods pop up in there :-)
>
> > So, unless there is some bad memory or runtime hit, I'd vote that:
> >
> > Dictionary at<TAB>
> >
> > returns  (at least)
> >
> > at:
> > at:put:
>
> and since you bring it up, here is what I sent to Paolo late last night:
>
> libgst/input.c
>
> void
> _gst_add_symbol_completion (const char *str,
>    int len)
> {
>   const char *base = str;
>   const char *p = str;
>
>   if (completions_enabled < 1)
>     return;
>
>   /* Everything goes in ... what can happen? */
>   add_completion (str, len);
> }
>
> st> aDict at<TAB><TAB>
> at:                          atPutSubpart:
> at:ifAbsent:                 atRandom
> at:ifAbsentPut:              atSubpart:
> at:ifPresent:                atSynonym:put:
> at:noCObjectsPut:type:       atime
> at:put:                      atimeSec
> at:put:type:                 attr
> at:splitAndPut:decrementBy:  attrArray
> at:type:                     attributeAt:
> atAll:                       attributeAt:ifAbsent:
> atAll:put:                   attributes
> atAllPut:                    attributesArray
> atAllSynonyms:put:           attributesDo:
> atEnd
>
> st> aThing print<TAB><TAB>
> print                              printOn:indent:
> print:                             printOn:line:
> printAsAttributeOn:                printOn:special:
> printByteCodesOn:                  printOn:tag:indent:
> printCodePointOn:                  printString
> printHeaderOn:                     printString:
> printHierarchy                     printStringRadix:
> printNl                            printSubclasses:using:
> printOn:                           printXmlOn:collection:tag:indent:
> printOn:base:                      printedFileName
> printOn:in:
>
> You'll have to back up into the suggestion though (alt-b comes
> in handy).
>
> s.
>
>
> _______________________________________________
> help-smalltalk mailing list
> [hidden email]
> http://lists.gnu.org/mailman/listinfo/help-smalltalk