Maybe documenting some rationale for at least "whereso" would help
It's probably inspired by the so
function from Perl 6, which simply coerces its argument to Bool - like the not
function but without the negation.
Things like "appearing as a value in an array", and other complexities, would be better accomplished by their own operators or functions.
]]>I am confused; I don't see anywhere in this project where a web server is relevant.
As I understand it, this project is about GUIs - local, on machine, no-network programs. From the post:
"GUIDeFATE is a tool that simplifies GUI design by choosing only a few useful elements that make a usable graphical application and making those elements accessible to the beginner developer."
If I have a Perl script that takes a number of options, why not text up a description of a one-screen set of fields with dropboxes and add that to the script?
Why do I need a web environment for that?
I'm not sure what problem space you are thinking of, but I see this as the Perl equivalent/integration of https://linux.die.net/man/1/xdialog
]]>Thanks for that. I agree re-inventing the wheel is not good, particularly as it is likely the vast majority of coders have superior insight to mine. Essentially...the problem I had related specifically to Qt where the subroutines called by GUI events generally have to be pre-declared, whereas I wanted the GUI creation done by GUIDeFATE to be separate from the subroutine creation done by the developer. My limited experience in this area is the problem of-course, and I will look into Mojolicious, so many thanks for your advice.
]]>Thanks for your insight. Now that xdialog thingy sounds very interesting; presumably it is a built-in interface...
]]>You are right about the issue that the graphical front-end has been done before (different toolkits, with different paradigms, most with little documentation, most...at least for Perl, quite old). And of-course a web front end is also a comprehensive, platform independent, modern interface. You are also correct the world does not need another standard. This is merely a tool that allows that interface to be created in a simple WYSIWYG way, in Perl, abstracting away the complexities of the back-end.
Your comments have inspired me to look at Mojolicious, however, and I am learning. I expect a Mojo/HTML/CSS back-end will follow at some point ;).
]]>This is excellent...
I accept your comments about exponential increase of effort with complex scenarios. It is I guess important to try these out in "real-life" applications. A tool is only useful if it makes life easier, the effort spent on *creating* the tool should therefore reduce the effort of using it. I will use your data set for an experiment with my tool.
Thanks
]]>My two points:
1) For Perl
Remember Java 1.5 was marketed as Java 5 and it was easier to proceed from there on.
Similary Perl 6 can be skipped for Perl.
Perl has been in Perl 5 for very long time. The next major version should be released as Perl 7 which will give a fresh beginning and surely every perl programmer will leap in joy. (Perl 6 can be skipped similar to Java which skipped 3 and 4)
Language Name: Perl
What is the latest and greatest version? 7
2) For Rakudo Perl
When facebook was registered it was "theFacebook".
Why "the"?
It changed to "Facebook" and it was neater.
Similarly "Rakudo Perl 6".
Why 6?
Simply "Rakudo Perl"
Name of Language : Rakudo Perl
Name of Compiler : Rakudo
and all references internally in language, documentation and compiler should be like Rakudo is the language.
Only title of books and conferences can have "Rakudo Perl" so it will have a link with Perl name.
Also just try saying...
What language u use?
I use Perl.
What about you?
Oh. I use Rakudo Perl.
Absolutely no confusion and both will grow.
I was just thinking the other day that I hadn't seen you for a while here.
Those Levenshtein things are OK for approximate work, but they tend to give lots of false positives. I actually did extensive work on spelling correction for another related web page which converts English words into Japanese forms, but I found that the Levenshtein corrections, even with a maximum distance of one, tend to throw up so many false positives that I'm not sure it's worth the bother. This is partly because of the amount of nonsense that people type in, and partly because it tends to do things like correct "Alisdair" to "Alistair", thus annoying all the "Alisdairs" out there. Similarly with lots of "Cherril" or "Candyce" type names. I actually encourage people to type their names in, I think it could get annoying if I keep saying "do you mean 'Candice'?". I didn't put the English correction discussion into this blog post since I thought the post would get too long, but perhaps I should prepare part 2.
Anyway, actually I hand-crafted error correction based on the old inputs. Lots of "amd" for "and", etc., so I just coded that as a hand correction, like s/\bamd\b/and/, and about a thousand different ways to spell "thousand" and "million/billion". Then a few days after that I got "bollion" for "billion" which I'd never seen before.
```
use Test::File::ShareDir -share => { -dist => { 'My-Dist-Name' => 'share' } };
```