[PD-dev] tcl 8.5 help browser bug and a fix

Hans-Christoph Steiner hans at eds.org
Mon Feb 26 04:30:17 CET 2007


IIRC, I believe it's a change in syntax, so that 8.4 understands it  
as a syntax error.

.hc

On Feb 25, 2007, at 11:20 AM, Ivica Ico Bukvic wrote:

> Interesting. Even if the line is preceded with an if statement  
> which checks
> for tclversion 8.5?
>
> Ico
>
>> -----Original Message-----
>> From: Hans-Christoph Steiner [mailto:hans at eds.org]
>> Sent: Sunday, February 25, 2007 1:47 AM
>> To: Miller Puckette
>> Cc: ico at vt.edu; pd-dev at iem.at
>> Subject: Re: [PD-dev] tcl 8.5 help browser bug and a fix
>>
>>
>> That line causes Tcl/Tk 8.4 to choke, since it uses special 8.5
>> syntax.  Someone would need to figure out how to make it only visible
>> to 8.5, otherwise it needs to remain commented out until Pd switches
>> to 8.5.
>>
>> .hc
>>
>> On Feb 21, 2007, at 2:34 PM, Miller Puckette wrote:
>>
>>> Hi all,
>>>
>>> I don't know how the help browser works (Hans-Christophe wrote  
>>> it) and
>>> don't know whether that comment character is in there for some  
>>> reason
>>> or not.  So I'm scared to fix this without hearing from HC.
>>>
>>> I wonder if the old-fashioned idea of just using the file browser
>>> should
>>> be available as at least an option...
>>>
>>> cheers
>>> Miller
>>>
>>> On Sun, Feb 18, 2007 at 10:58:32PM -0500, ico at vt.edu wrote:
>>>> Hi all,
>>>>
>>>> Tried sending a couple of e-mails not realizing I am not
>>>> subscribed to dev
>>>> list... DOH! So, please prepare for a small salvo of e-mails :-)
>>>>
>>>> Topic #1: tcl/tk 8.5 help browser bug
>>>>
>>>> This help browser bug using CVS and tcl/tk 8.5 has been present
>>>> since at least
>>>> last fall. After digging through the code, I found the following:
>>>>
>>>> proc doc_make_listbox {base dir count} {
>>>>     # check for [file readable]?
>>>>     if { [info tclversion] >= 8.5 } {
>>>>         # requires Tcl 8.5 but probably deals with special chars
>>>> better
>>>> #        destroy {expand}[lrange [winfo children $base] [expr {2 *
>>>> $count}] end]
>>>>     } else {
>>>>
>>>> The comment before "destroy" in effect disables destroying of
>>>> children widgets
>>>> and causes errors whenever I have more than one column open and
>>>> try to click on
>>>> a column preceding one that is already created. The error reported
>>>> is that the
>>>> "listbox1-list already exists" and the only way around is to close
>>>> the help
>>>> browser and reopen it. This obviously prevents going up the file
>>>> tree once a
>>>> particular subfolder has been opened which obviously means it is a
>>>> non-lethal
>>>> bug, but nonetheless annoying.
>>>>
>>>> The fix is obvious, remove the comment from the offending line and
>>>> everything
>>>> works like a charm:
>>>>
>>>> proc doc_make_listbox {base dir count} {
>>>>     # check for [file readable]?
>>>>     if { [info tclversion] >= 8.5 } {
>>>>         # requires Tcl 8.5 but probably deals with special chars
>>>> better
>>>>         destroy {expand}[lrange [winfo children $base] [expr {2 *
>>>> $count}] end]
>>>>     } else {
>>>>
>>>> It would be great if one of the devs/maintiners would please
>>>> commit this change
>>>> (I'd do it but I have not bothered to get the CVS dev access).
>>>>
>>>> Many thanks!
>>>>
>>>> Best wishes,
>>>>
>>>> Ico
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> PD-dev mailing list
>>>> PD-dev at iem.at
>>>> http://lists.puredata.info/listinfo/pd-dev
>>>
>>> _______________________________________________
>>> PD-dev mailing list
>>> PD-dev at iem.at
>>> http://lists.puredata.info/listinfo/pd-dev
>>
>>
>> --------------------------------------------------------------------- 
>> ---
>>
>>                                                http://at.or.at/hans/
>


------------------------------------------------------------------------

Man has survived hitherto because he was too ignorant to know how to  
realize his wishes.  Now that he can realize them, he must either  
change them, or perish.    -William Carlos Williams






More information about the Pd-dev mailing list