Re: sort container parameters
Inline sort param variables: If anyone files this bug report/feature request, let me know! I will sign on Best wishes, Curry Kenworthy Custom Software Development "Better Methods, Better Results" LiveCode Training and Consulting http://livecodeconsulting.com/ ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Re: sort container parameters
We used to call that, “polishing poop” which, according to Mythbusters is a thing. No, really. Bob S On Mar 3, 2021, at 2:02 PM, Paul Dupuis via use-livecode mailto:use-livecode@lists.runrev.com>> wrote: Perhaps it is a compliment to Livecode/LCS that I EXPECTED it to work and was SURPRISED when it did not :-) ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Re: sort container parameters
Craig,
It's because when you try to sort something, it's got to be a variable that's
being sorted. So all of the examples you used put "toSort" into the statement
to be "done" coming out compiled as:
sort items of "A,C,T,B" ascending
... which won't work. You need to execute:
do "sort items of toSort" && sortDir
which compiles to:
sort items of toSort ascending
... and that works.
:)
Ken Ray
Sons of Thunder Software, Inc.
Email: k...@sonsothunder.com
Website: https://www.sonsothunder.com
> On Mar 3, 2021, at 12:23 PM, Craig Newman via use-livecode
> wrote:
>
> I thought I was pretty clever at using “do”, often going down several levels,
> evaluating the whole way before trying to execute a single “do” statement.
>
> Neither of the last lines seem to work:
>
> on mouseup
>
> put "A,C,T,B" into toSort
>
> put "ascending" into sortDir
>
> get "sort items of" && quote & toSort & quote && sortDir
>
> breakpoint
>
> --do it
>
> --do "sort" && toSort && sortDir
>
> do "sort" && quote & toSort & quote && sortDir
>
> end mouseup
>
>
> I feel there must be a way…
>
> Craig
>
>
>> On Mar 2, 2021, at 7:36 PM, Alex Tweedly via use-livecode
>> wrote:
>>
>> The number of lines being sorted makes o difference.
>>
>> The number of distinct sorts will - but at a single 'do' for each of
>> "several hundreds", you'll not even notice except (maybe) if you are
>> benchmarking it (around 5ms per 1000 'do's on an elderly MacBook Pro).
>>
>> Alex.
>>
>> On 02/03/2021 22:52, Paul Dupuis via use-livecode wrote:
>>> Okay. Thank you Ralph.
>>>
>>> I was reluctant to use "do" for performance reasons. The sort could be
>>> sorting a large number of lines - as many as several hundred sorts of a
>>> thousand to ten thousand lines.
>>>
>>> I could of course do timing trials, but does anyone from the mothership (or
>>> anyone period) know if using "do" with a container sort causes any
>>> appreciable performance hit? All the data is in a variable (vs fields).
>>>
>>>
>>> On 3/2/2021 5:19 PM, Ralph DiMola via use-livecode wrote:
I found that you must us a "Do"(thank heaven for "Do"s) if you want to
make variable any LC token that colorizes like "stack", "the", "field",
"button", "put" or "ascending".
I don't think "Do" is a kludge is this case.
Ralph DiMola
IT Director
Evergreen Information Services
rdim...@evergreeninfo.net
-Original Message-
From: use-livecode [mailto:use-livecode-boun...@lists.runrev.com] On
Behalf Of Paul Dupuis via use-livecode
Sent: Tuesday, March 02, 2021 4:56 PM
To: use-livecode@lists.runrev.com
Cc: Paul Dupuis
Subject: sort container parameters
I just discovered much to my dismay that you can not execute the following:
put "ascending" into tDirection
soft lines of tContainer tDirection international
apparently neither the sort direction (ascending|descending) nor the sort
type (international|text|datetime|numeric|binary) can be variable!
That means if you want to parameterize a sort direction, you have to do
something like:
if tDirection is "ascending" then
sort lines of tContainer ascending international else
sort lines of tCOntainer descending international end if
I see this a a bug or perhaps a failure to fully robust impliment the sort
container command? Does any one else see this as a bug?
I suppose I could work around it with a "do" but that seems like a cludge
do ("sort line of tContainer" && tDirection && tSortType)
Thoughts? Comments? Opinions? Am I expecting too much of LiveCode to have
sort direction and type actual parameters?
___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your
subscription preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode
___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your
subscription preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode
>>>
>>>
>>> ___
>>> use-livecode mailing list
>>> use-livecode@lists.runrev.com
>>> Please visit this url to subscribe, unsubscribe and manage your
>>> subscription preferences:
>>> http://lists.runrev.com/mailman/listinfo/use-livecode
>>
>> ___
>> use-livecode mailing list
>> use-livecode@lists.runrev.com
>> Please visit this url to subscribe, unsubscribe and manage your subscription
>> preferences:
>> http://lists.runrev.com/mailman/listinfo/use-livecode
>
> ___
>
Re: sort container parameters
I love xTalk. While I have written code in many dozens of languages over my career, I personally will likely never code in anything other than xTalk for as long as I continue to write code. That said, one of the "problems" of a programming language that is so intuitive is that I sometime write code that I expect to work ... and it doesn't. The container sort with params was one such instance. Perhaps it is a compliment to Livecode/LCS that I EXPECTED it to work and was SURPRISED when it did not :-) On 3/3/2021 1:00 PM, Richard Gaskin via use-livecode wrote: Paul Dupuis wrote: > I just discovered much to my dismay that you can not execute the > following: > > put "ascending" into tDirection > soft lines of tContainer tDirection international > > apparently neither the sort direction (ascending|descending) nor the > sort type (international|text|datetime|numeric|binary) can be > variable! ... > I see this a a bug or perhaps a failure to fully robust impliment the > sort container command? Does any one else see this as a bug? Bug or feature? The ambiguities introduced by the HyperTalk team have made that an evergreen question. ;) Some tokens can be parameterized, others not. MC/LC goes much further than HC in allowing object references to be a mix of literal and variable expressions, e.g.: -- Works in LC, not in HC: put the long id of this cd into tCdObj get the name of btn 1 of tCdObj ...but other expressions and keywords are not so clear. Let's see what happens if we parameterize everything in your statement: put "ascending" into tDirection put "numeric" into tSortType put "items" into tChunkType put "1,2,3,4,5,6,7,8,9" into tData sort tChunkType of tData tSortType tDirection In that last line, which variables should we expect to be allowed, and which ones not? I'm certain there's a rule defining this. I'm equally certain I've never met any xTalker who was able to intuit that rule with complete confidence. ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Re: sort container parameters
I thought I was pretty clever at using “do”, often going down several levels,
evaluating the whole way before trying to execute a single “do” statement.
Neither of the last lines seem to work:
on mouseup
put "A,C,T,B" into toSort
put "ascending" into sortDir
get "sort items of" && quote & toSort & quote && sortDir
breakpoint
--do it
--do "sort" && toSort && sortDir
do "sort" && quote & toSort & quote && sortDir
end mouseup
I feel there must be a way…
Craig
> On Mar 2, 2021, at 7:36 PM, Alex Tweedly via use-livecode
> wrote:
>
> The number of lines being sorted makes o difference.
>
> The number of distinct sorts will - but at a single 'do' for each of "several
> hundreds", you'll not even notice except (maybe) if you are benchmarking it
> (around 5ms per 1000 'do's on an elderly MacBook Pro).
>
> Alex.
>
> On 02/03/2021 22:52, Paul Dupuis via use-livecode wrote:
>> Okay. Thank you Ralph.
>>
>> I was reluctant to use "do" for performance reasons. The sort could be
>> sorting a large number of lines - as many as several hundred sorts of a
>> thousand to ten thousand lines.
>>
>> I could of course do timing trials, but does anyone from the mothership (or
>> anyone period) know if using "do" with a container sort causes any
>> appreciable performance hit? All the data is in a variable (vs fields).
>>
>>
>> On 3/2/2021 5:19 PM, Ralph DiMola via use-livecode wrote:
>>> I found that you must us a "Do"(thank heaven for "Do"s) if you want to make
>>> variable any LC token that colorizes like "stack", "the", "field",
>>> "button", "put" or "ascending".
>>>
>>> I don't think "Do" is a kludge is this case.
>>>
>>> Ralph DiMola
>>> IT Director
>>> Evergreen Information Services
>>> rdim...@evergreeninfo.net
>>>
>>> -Original Message-
>>> From: use-livecode [mailto:use-livecode-boun...@lists.runrev.com] On Behalf
>>> Of Paul Dupuis via use-livecode
>>> Sent: Tuesday, March 02, 2021 4:56 PM
>>> To: use-livecode@lists.runrev.com
>>> Cc: Paul Dupuis
>>> Subject: sort container parameters
>>>
>>> I just discovered much to my dismay that you can not execute the following:
>>>
>>> put "ascending" into tDirection
>>> soft lines of tContainer tDirection international
>>>
>>> apparently neither the sort direction (ascending|descending) nor the sort
>>> type (international|text|datetime|numeric|binary) can be variable!
>>>
>>> That means if you want to parameterize a sort direction, you have to do
>>> something like:
>>>
>>> if tDirection is "ascending" then
>>> sort lines of tContainer ascending international else
>>> sort lines of tCOntainer descending international end if
>>>
>>> I see this a a bug or perhaps a failure to fully robust impliment the sort
>>> container command? Does any one else see this as a bug?
>>>
>>> I suppose I could work around it with a "do" but that seems like a cludge
>>>
>>> do ("sort line of tContainer" && tDirection && tSortType)
>>>
>>> Thoughts? Comments? Opinions? Am I expecting too much of LiveCode to have
>>> sort direction and type actual parameters?
>>>
>>>
>>> ___
>>> use-livecode mailing list
>>> use-livecode@lists.runrev.com
>>> Please visit this url to subscribe, unsubscribe and manage your
>>> subscription preferences:
>>> http://lists.runrev.com/mailman/listinfo/use-livecode
>>>
>>>
>>> ___
>>> use-livecode mailing list
>>> use-livecode@lists.runrev.com
>>> Please visit this url to subscribe, unsubscribe and manage your
>>> subscription preferences:
>>> http://lists.runrev.com/mailman/listinfo/use-livecode
>>
>>
>> ___
>> use-livecode mailing list
>> use-livecode@lists.runrev.com
>> Please visit this url to subscribe, unsubscribe and manage your subscription
>> preferences:
>> http://lists.runrev.com/mailman/listinfo/use-livecode
>
> ___
> use-livecode mailing list
> use-livecode@lists.runrev.com
> Please visit this url to subscribe, unsubscribe and manage your subscription
> preferences:
> http://lists.runrev.com/mailman/listinfo/use-livecode
___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode
Re: sort container parameters
Paul Dupuis wrote: > I just discovered much to my dismay that you can not execute the > following: > > put "ascending" into tDirection > soft lines of tContainer tDirection international > > apparently neither the sort direction (ascending|descending) nor the > sort type (international|text|datetime|numeric|binary) can be > variable! ... > I see this a a bug or perhaps a failure to fully robust impliment the > sort container command? Does any one else see this as a bug? Bug or feature? The ambiguities introduced by the HyperTalk team have made that an evergreen question. ;) Some tokens can be parameterized, others not. MC/LC goes much further than HC in allowing object references to be a mix of literal and variable expressions, e.g.: -- Works in LC, not in HC: put the long id of this cd into tCdObj get the name of btn 1 of tCdObj ...but other expressions and keywords are not so clear. Let's see what happens if we parameterize everything in your statement: put "ascending" into tDirection put "numeric" into tSortType put "items" into tChunkType put "1,2,3,4,5,6,7,8,9" into tData sort tChunkType of tData tSortType tDirection In that last line, which variables should we expect to be allowed, and which ones not? I'm certain there's a rule defining this. I'm equally certain I've never met any xTalker who was able to intuit that rule with complete confidence. -- Richard Gaskin Fourth World Systems Software Design and Development for the Desktop, Mobile, and the Web ambassa...@fourthworld.comhttp://www.FourthWorld.com ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Re: sort container parameters
Paul: > apparently neither the sort direction (ascending|descending) > nor the sort type (international|text|datetime|numeric|binary) > can be variable! > I see this a a bug or perhaps a failure to fully > robust impliment the sort container command? > Does any one else see this as a bug? Absolutely a bug or extremely annoying limitation! I've been hating this for a long time. > I suppose I could work around it with a "do" > but that seems like a cludge There has been no performance hit for "do" as a one-off. So I don't mind using "do" once. I try to avoid it in loops. Test, and you'll see that in your results. Meanwhile "do" is arguably cleaner and smaller code than if/else. But still a clumsy abomination compared to using an inline variable. Any bug report on this issue gets my vote! Yes, it's a bad thing. Best wishes, Curry Kenworthy Custom Software Development "Better Methods, Better Results" LiveCode Training and Consulting http://livecodeconsulting.com/ ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Re: sort container parameters
The number of lines being sorted makes o difference.
The number of distinct sorts will - but at a single 'do' for each of
"several hundreds", you'll not even notice except (maybe) if you are
benchmarking it (around 5ms per 1000 'do's on an elderly MacBook Pro).
Alex.
On 02/03/2021 22:52, Paul Dupuis via use-livecode wrote:
Okay. Thank you Ralph.
I was reluctant to use "do" for performance reasons. The sort could be
sorting a large number of lines - as many as several hundred sorts of
a thousand to ten thousand lines.
I could of course do timing trials, but does anyone from the
mothership (or anyone period) know if using "do" with a container sort
causes any appreciable performance hit? All the data is in a variable
(vs fields).
On 3/2/2021 5:19 PM, Ralph DiMola via use-livecode wrote:
I found that you must us a "Do"(thank heaven for "Do"s) if you want
to make variable any LC token that colorizes like "stack", "the",
"field", "button", "put" or "ascending".
I don't think "Do" is a kludge is this case.
Ralph DiMola
IT Director
Evergreen Information Services
rdim...@evergreeninfo.net
-Original Message-
From: use-livecode [mailto:use-livecode-boun...@lists.runrev.com] On
Behalf Of Paul Dupuis via use-livecode
Sent: Tuesday, March 02, 2021 4:56 PM
To: use-livecode@lists.runrev.com
Cc: Paul Dupuis
Subject: sort container parameters
I just discovered much to my dismay that you can not execute the
following:
put "ascending" into tDirection
soft lines of tContainer tDirection international
apparently neither the sort direction (ascending|descending) nor the
sort type (international|text|datetime|numeric|binary) can be variable!
That means if you want to parameterize a sort direction, you have to
do something like:
if tDirection is "ascending" then
sort lines of tContainer ascending international else
sort lines of tCOntainer descending international end if
I see this a a bug or perhaps a failure to fully robust impliment the
sort container command? Does any one else see this as a bug?
I suppose I could work around it with a "do" but that seems like a
cludge
do ("sort line of tContainer" && tDirection && tSortType)
Thoughts? Comments? Opinions? Am I expecting too much of LiveCode to
have sort direction and type actual parameters?
___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your
subscription preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode
___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your
subscription preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode
___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your
subscription preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode
___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode
Re: sort container parameters
If it is a lot to sort, I am wondering why not use a memory database with the indexes you need to sort by, or a file database if the data needs to be persistent? Bob S On Mar 2, 2021, at 2:52 PM, Paul Dupuis via use-livecode mailto:use-livecode@lists.runrev.com>> wrote: Okay. Thank you Ralph. I was reluctant to use "do" for performance reasons. The sort could be sorting a large number of lines - as many as several hundred sorts of a thousand to ten thousand lines. I could of course do timing trials, but does anyone from the mothership (or anyone period) know if using "do" with a container sort causes any appreciable performance hit? All the data is in a variable (vs fields). ___ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
Re: sort container parameters
Okay. Thank you Ralph.
I was reluctant to use "do" for performance reasons. The sort could be
sorting a large number of lines - as many as several hundred sorts of a
thousand to ten thousand lines.
I could of course do timing trials, but does anyone from the mothership
(or anyone period) know if using "do" with a container sort causes any
appreciable performance hit? All the data is in a variable (vs fields).
On 3/2/2021 5:19 PM, Ralph DiMola via use-livecode wrote:
I found that you must us a "Do"(thank heaven for "Do"s) if you want to make variable any LC token that colorizes like "stack",
"the", "field", "button", "put" or "ascending".
I don't think "Do" is a kludge is this case.
Ralph DiMola
IT Director
Evergreen Information Services
rdim...@evergreeninfo.net
-Original Message-
From: use-livecode [mailto:use-livecode-boun...@lists.runrev.com] On Behalf Of
Paul Dupuis via use-livecode
Sent: Tuesday, March 02, 2021 4:56 PM
To: use-livecode@lists.runrev.com
Cc: Paul Dupuis
Subject: sort container parameters
I just discovered much to my dismay that you can not execute the following:
put "ascending" into tDirection
soft lines of tContainer tDirection international
apparently neither the sort direction (ascending|descending) nor the sort type
(international|text|datetime|numeric|binary) can be variable!
That means if you want to parameterize a sort direction, you have to do
something like:
if tDirection is "ascending" then
sort lines of tContainer ascending international else
sort lines of tCOntainer descending international end if
I see this a a bug or perhaps a failure to fully robust impliment the sort
container command? Does any one else see this as a bug?
I suppose I could work around it with a "do" but that seems like a cludge
do ("sort line of tContainer" && tDirection && tSortType)
Thoughts? Comments? Opinions? Am I expecting too much of LiveCode to have sort
direction and type actual parameters?
___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode
___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode
___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode
RE: sort container parameters
I found that you must us a "Do"(thank heaven for "Do"s) if you want to make
variable any LC token that colorizes like "stack", "the", "field", "button",
"put" or "ascending".
I don't think "Do" is a kludge is this case.
Ralph DiMola
IT Director
Evergreen Information Services
rdim...@evergreeninfo.net
-Original Message-
From: use-livecode [mailto:use-livecode-boun...@lists.runrev.com] On Behalf Of
Paul Dupuis via use-livecode
Sent: Tuesday, March 02, 2021 4:56 PM
To: use-livecode@lists.runrev.com
Cc: Paul Dupuis
Subject: sort container parameters
I just discovered much to my dismay that you can not execute the following:
put "ascending" into tDirection
soft lines of tContainer tDirection international
apparently neither the sort direction (ascending|descending) nor the sort type
(international|text|datetime|numeric|binary) can be variable!
That means if you want to parameterize a sort direction, you have to do
something like:
if tDirection is "ascending" then
sort lines of tContainer ascending international else
sort lines of tCOntainer descending international end if
I see this a a bug or perhaps a failure to fully robust impliment the sort
container command? Does any one else see this as a bug?
I suppose I could work around it with a "do" but that seems like a cludge
do ("sort line of tContainer" && tDirection && tSortType)
Thoughts? Comments? Opinions? Am I expecting too much of LiveCode to have sort
direction and type actual parameters?
___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode
___
use-livecode mailing list
use-livecode@lists.runrev.com
Please visit this url to subscribe, unsubscribe and manage your subscription
preferences:
http://lists.runrev.com/mailman/listinfo/use-livecode
