Re: [Pharo-users] About neoCSV

[Sven Van Caekenberghe]

I am not sure I understand 100% what you want, but consider this example:

(NeoCSVReader on: 'foo bar baz
1 2 3
11 22 33' readStream)
separator: $ ;
namedColumnsConfiguration;
upToEnd.

=> an Array(a Dictionary(#bar->'2' #baz->'3' #foo->'1' ) a 
Dictionary(#bar->'22' #baz->'33' #foo->'11' ))

Since you did not use $, but space I had to set that using #separator:

#namedColumnsConfiguration uses the header as keys and reads dictionaries 
instead of arrays.

Instead of #upToEnd you can also iterate with #do:

Is this what you are looking for ?

> On 12 Sep 2017, at 08:23, Stephane Ducasse  wrote:
> 
> Hi sven
> 
> I have a question with NeoCVS
> Do you offer the possibility to manage the data of one line based on
> the description given by the first line
> 
> Ex
> 
> foo bar baz
> 1 2 3
> 11 22 33
> 
> =>
> 
> res doWith: [:tag :data |
> Transcript show: tag ; space ; show: data printString ;cr
> ]
> 
> foo 1 bar 2 baz 3
> foo 11 bar 22 baz 33
> 
> Stef
> 

[Pharo-users] DataFrame inspector issues

[Alistair Grant]

Hi Everyone,

I've started using DataFrame, it's great to have this functionality
easily available, and noticed a few issues with the inspector:

1. The inspector always shows row numbers and not the supplied name.

2. The last column label isn't displayed.

If I remember correctly, there has been some discussion about making the
column widths draggable in the inspector.  I would prefer to have a
context menu that is "size to fit".  This could be potentially expensive
to calculate, so it would be user-initiated.

Where should I report issues?
https://github.com/PolyMathOrg/DataFrame/issues

I have a patch that addresses the first issue, although I'm not
completely happy with it, and I'll take a look at the second issue.


Thanks very much,
Alistair

Re: [Pharo-users] DataFrame inspector issues

[Serge Stinckwich]

On Tue, Sep 12, 2017 at 9:22 AM, Alistair Grant 
wrote:

> Hi Everyone,
>
>
​Hi Alistair,
​

> I've started using DataFrame, it's great to have this functionality
> easily available, and noticed a few issues with the inspector:
>
> 1. The inspector always shows row numbers and not the supplied name.
>
> 2. The last column label isn't displayed.
>
> If I remember correctly, there has been some discussion about making the
> column widths draggable in the inspector.  I would prefer to have a
> context menu that is "size to fit".  This could be potentially expensive
> to calculate, so it would be user-initiated.
>
> Where should I report issues?
> https://github.com/PolyMathOrg/DataFrame/issues
>
>
​Thank you for your report.​

​At the moment, you can report issues here. I guess later, we will
intregrate
DataFrame in the main PolyMath repository.
​

> I have a patch that addresses the first issue, although I'm not
> completely happy with it, and I'll take a look at the second issue.
>

​This is maybe better to discuss about DataFrame​ and other PolyMath issues
on the polymath mailing-list:
https://groups.google.com/forum/#!forum/scismalltalk

​See you there !​

-- 
Serge Stinckwich
UCN & UMI UMMISCO 209 (IRD/UPMC)
Every DSL ends up being Smalltalk
http://www.doesnotunderstand.org/

Re: [Pharo-users] Who is maintaining of Artefact?

[Denis Kudriashov]

There is also PDFTalk from Christian Haider https://wiki.pdftalk.de/doku.php.
But it needs to be ported into Pharo.

2017-09-11 21:03 GMT+02:00 Alejandro Infante :

> Hi!
> I would like to explore having support for Unicode characters. I want to
> help for it.
>
> Who is the right person to talk about it?
>
> Cheers!
> Alejandro
>

Re: [Pharo-users] Who is maintaining of Artefact?

[H. Hirzel]

Looks like a worthwhile exercise

PDFTalk https://wiki.pdftalk.de/doku.php

"The syntax is covered comprehensively, which means that any PDF file
can be read and any PDF object can be written in a conforming way.

Typical PDF objects are dictionaries of which many are specialized as
Smalltalk objects.

The library code is annotated with the original descriptions and
definitions from the specification, so that learning about PDF itself
is well supported. "

I wonder what the main issues would be  when porting this to Pharo.

--Hannes

On 9/12/17, Denis Kudriashov  wrote:
> There is also PDFTalk from Christian Haider
> https://wiki.pdftalk.de/doku.php.
> But it needs to be ported into Pharo.
>
> 2017-09-11 21:03 GMT+02:00 Alejandro Infante
> :
>
>> Hi!
>> I would like to explore having support for Unicode characters. I want to
>> help for it.
>>
>> Who is the right person to talk about it?
>>
>> Cheers!
>> Alejandro
>>
>

Re: [Pharo-users] Who is maintaining of Artefact?

[Denis Kudriashov]

Look at slides PDFtalk for Gemstone
 from last ESUG about
porting it to Gemstone (which is done).
Maybe it is similar as porting GLORP into Pharo. Somebody can comment on
this process.

2017-09-12 11:04 GMT+02:00 H. Hirzel :

> Looks like a worthwhile exercise
>
> PDFTalk https://wiki.pdftalk.de/doku.php
>
> "The syntax is covered comprehensively, which means that any PDF file
> can be read and any PDF object can be written in a conforming way.
>
> Typical PDF objects are dictionaries of which many are specialized as
> Smalltalk objects.
>
> The library code is annotated with the original descriptions and
> definitions from the specification, so that learning about PDF itself
> is well supported. "
>
> I wonder what the main issues would be  when porting this to Pharo.
>
> --Hannes
>
> On 9/12/17, Denis Kudriashov  wrote:
> > There is also PDFTalk from Christian Haider
> > https://wiki.pdftalk.de/doku.php.
> > But it needs to be ported into Pharo.
> >
> > 2017-09-11 21:03 GMT+02:00 Alejandro Infante
> > :
> >
> >> Hi!
> >> I would like to explore having support for Unicode characters. I want to
> >> help for it.
> >>
> >> Who is the right person to talk about it?
> >>
> >> Cheers!
> >> Alejandro
> >>
> >
>
>

Re: [Pharo-users] Who is maintaining of Artefact?

[Christian Haider]

Moin,

 

the port is almost done. Then I will add documentation and publish the code 
under MIT licence.

 

Including the code used to port the library from VW to Gemstone using a novel 
approach.

 

The heard of the “PDF engine” is the type system which allows the assignment of 
Smalltalk classes to raw PDF objects.

In the new version, the PDF types are decoupled from the Smalltalk classes, so 
that classes can be renamed freely (i.e. adding prefixes).

 

I think that porting the new version to Pharo (and VA and …) has gotten much 
easier. 

I hope that someone would take the challenge :).

 

Cheers,

Christian

 

 

Von: Pharo-users [mailto:pharo-users-boun...@lists.pharo.org] Im Auftrag von 
Denis Kudriashov
Gesendet: Dienstag, 12. September 2017 11:12
An: Any question about pharo is welcome 
Betreff: Re: [Pharo-users] Who is maintaining of Artefact?

 

Look at slides PDFtalk for Gemstone 
  from last ESUG about 
porting it to Gemstone (which is done).

Maybe it is similar as porting GLORP into Pharo. Somebody can comment on this 
process.

 

2017-09-12 11:04 GMT+02:00 H. Hirzel mailto:hannes.hir...@gmail.com> >:

Looks like a worthwhile exercise

PDFTalk https://wiki.pdftalk.de/doku.php

"The syntax is covered comprehensively, which means that any PDF file
can be read and any PDF object can be written in a conforming way.

Typical PDF objects are dictionaries of which many are specialized as
Smalltalk objects.

The library code is annotated with the original descriptions and
definitions from the specification, so that learning about PDF itself
is well supported. "

I wonder what the main issues would be  when porting this to Pharo.

--Hannes


On 9/12/17, Denis Kudriashov mailto:dionisi...@gmail.com> > wrote:
> There is also PDFTalk from Christian Haider
> https://wiki.pdftalk.de/doku.php.
> But it needs to be ported into Pharo.
>
> 2017-09-11 21:03 GMT+02:00 Alejandro Infante
> mailto:alejandroinfant...@gmail.com> >:
>
>> Hi!
>> I would like to explore having support for Unicode characters. I want to
>> help for it.
>>
>> Who is the right person to talk about it?
>>
>> Cheers!
>> Alejandro
>>
>

 

Re: [Pharo-users] DataFrame inspector issues

[Alistair Grant]

On Tue, Sep 12, 2017 at 09:28:13AM +0200, Serge Stinckwich wrote:
> 
> On Tue, Sep 12, 2017 at 9:22 AM, Alistair Grant  wrote:
> 
> Hi Everyone,
> 
> ?Hi Alistair,
> 
> I've started using DataFrame, it's great to have this functionality
> easily available, and noticed a few issues with the inspector:
> 
> 1. The inspector always shows row numbers and not the supplied name.
> 
> 2. The last column label isn't displayed.
> 
> 
> Where should I report issues?
> https://github.com/PolyMathOrg/DataFrame/issues
> 
> 
> 
> ?Thank you for your report.?
> 
> ?At the moment, you can report issues here. I guess later, we will intregrate
> DataFrame in the main PolyMath repository.
> ?
> 
> I have a patch that addresses the first issue, although I'm not
> completely happy with it, and I'll take a look at the second issue.
> 
> 
> ?This is maybe better to discuss about DataFrame? and other PolyMath issues 
> on the polymath mailing-list: 
> https://groups.google.com/forum/#!forum/scismalltalk
> 
> ?See you there !?

Hi Serge,

Thanks, I've joined the group and reposted the message there.

I wrote:
> If I remember correctly, there has been some discussion about making the
> column widths draggable in the inspector.  I would prefer to have a
> context menu that is "size to fit".  This could be potentially expensive
> to calculate, so it would be user-initiated.

I think this is more of a FastTable / GT issue than DataFrame, so maybe
the discussion should stay here?  (if anyone is interested :-)).

Thanks,
Alistair

Re: [Pharo-users] Who is maintaining of Artefact?

[Denis Kudriashov]

Hi Christian.

2017-09-12 11:38 GMT+02:00 Christian Haider <
christian.hai...@smalltalked-visuals.com>:

> Moin,
>
>
>
> the port is almost done. Then I will add documentation and publish the
> code under MIT licence.
>

Cool. And code repo is not public yet?


>
>
> Including the code used to port the library from VW to Gemstone using a
> novel approach.
>
>
>
> The heard of the “PDF engine” is the type system which allows the
> assignment of Smalltalk classes to raw PDF objects.
>
> In the new version, the PDF types are decoupled from the Smalltalk
> classes, so that classes can be renamed freely (i.e. adding prefixes).
>
>
>
> I think that porting the new version to Pharo (and VA and …) has gotten
> much easier.
>
> I hope that someone would take the challenge J.
>

Definitely


>
>
> Cheers,
>
> Christian
>
>
>
>
>
> *Von:* Pharo-users [mailto:pharo-users-boun...@lists.pharo.org] *Im
> Auftrag von *Denis Kudriashov
> *Gesendet:* Dienstag, 12. September 2017 11:12
> *An:* Any question about pharo is welcome 
> *Betreff:* Re: [Pharo-users] Who is maintaining of Artefact?
>
>
>
> Look at slides PDFtalk for Gemstone
>  from last ESUG
> about porting it to Gemstone (which is done).
>
> Maybe it is similar as porting GLORP into Pharo. Somebody can comment on
> this process.
>
>
>
> 2017-09-12 11:04 GMT+02:00 H. Hirzel :
>
> Looks like a worthwhile exercise
>
> PDFTalk https://wiki.pdftalk.de/doku.php
>
> "The syntax is covered comprehensively, which means that any PDF file
> can be read and any PDF object can be written in a conforming way.
>
> Typical PDF objects are dictionaries of which many are specialized as
> Smalltalk objects.
>
> The library code is annotated with the original descriptions and
> definitions from the specification, so that learning about PDF itself
> is well supported. "
>
> I wonder what the main issues would be  when porting this to Pharo.
>
> --Hannes
>
>
> On 9/12/17, Denis Kudriashov  wrote:
> > There is also PDFTalk from Christian Haider
> > https://wiki.pdftalk.de/doku.php.
> > But it needs to be ported into Pharo.
> >
> > 2017-09-11 21:03 GMT+02:00 Alejandro Infante
> > :
> >
> >> Hi!
> >> I would like to explore having support for Unicode characters. I want to
> >> help for it.
> >>
> >> Who is the right person to talk about it?
> >>
> >> Cheers!
> >> Alejandro
> >>
> >
>
>
>

Re: [Pharo-users] Accessor Creation in Nautilus Browser

[Alistair Grant]

Hi Rob,

On 12 September 2017 at 03:32, Rob Rothwell  wrote:
> Hello,
>
> Is there a way to create accessors for more than one instance variable at a
> time with the Nautilus browser?
>
> "Class -> Refactoring -> Inst Var Refactoring -> Accessors" shows available
> instance variables, but only one can be selected...
>
> Thank you,
>
> Rob

Class -> Refactoring -> Class Refactoring -> Generate Accessors

Cheers,
Alistair

Re: [Pharo-users] Pharo 6.1 debugger - suggestions break once, does funny things?

[Nicolai Hess]

2017-09-11 1:26 GMT+02:00 Tim Mackinnon :

> While I’m on a debugger trip… anyone else also notice that the
> Suggestions, Break once debug option acts weirdly? I seem to get a debugger
> but it halts several lines after where I have said, and if I start stepping
> the debugger seems to jump to funny places and show values that don’t line
> up with what I know is happening? If I use Suggestions, Break it seems to
> work correctly (apart from the Step Through I mentioned in another thread).
>
> Tim
>


Thats because the "Break once" creates a MetaLink that when executing this
Metalink it removes itself from the methode prior to the break-command. If
now  the debugger opens the method, it executes a context to a method that
already had changed. (the actual method bytecode only exists in the current
context).

You can see the "real" bytecode of the contexts method if you select GTs
bytecode view.

Re: [Pharo-users] Who is maintaining of Artefact?

[Christian Haider]

Von: Pharo-users [mailto:pharo-users-boun...@lists.pharo.org] Im Auftrag von 
Denis Kudriashov
Gesendet: Dienstag, 12. September 2017 11:43
An: Any question about pharo is welcome 
Betreff: Re: [Pharo-users] Who is maintaining of Artefact?

 

Hi Christian.

 

2017-09-12 11:38 GMT+02:00 Christian Haider 
mailto:christian.hai...@smalltalked-visuals.com> >:

Moin,

 

the port is almost done. Then I will add documentation and publish the code 
under MIT licence.

 

Cool. And code repo is not public yet?

 

No, still in private store. I will publish it to Cincom’s public store when I 
am ready.

If you want to have a look, I could tell you privately where my store is… 

 

Including the code used to port the library from VW to Gemstone using a novel 
approach.

 

The heard of the “PDF engine” is the type system which allows the assignment of 
Smalltalk classes to raw PDF objects.

In the new version, the PDF types are decoupled from the Smalltalk classes, so 
that classes can be renamed freely (i.e. adding prefixes).

 

I think that porting the new version to Pharo (and VA and …) has gotten much 
easier. 

I hope that someone would take the challenge :).

 

Definitely

 

 

Cheers,

Christian

 

 

Von: Pharo-users [mailto:pharo-users-boun...@lists.pharo.org 
 ] Im Auftrag von Denis Kudriashov
Gesendet: Dienstag, 12. September 2017 11:12
An: Any question about pharo is welcome mailto:pharo-users@lists.pharo.org> >
Betreff: Re: [Pharo-users] Who is maintaining of Artefact?

 

Look at slides PDFtalk for Gemstone 
  from last ESUG about 
porting it to Gemstone (which is done).

Maybe it is similar as porting GLORP into Pharo. Somebody can comment on this 
process.

 

2017-09-12 11:04 GMT+02:00 H. Hirzel mailto:hannes.hir...@gmail.com> >:

Looks like a worthwhile exercise

PDFTalk https://wiki.pdftalk.de/doku.php

"The syntax is covered comprehensively, which means that any PDF file
can be read and any PDF object can be written in a conforming way.

Typical PDF objects are dictionaries of which many are specialized as
Smalltalk objects.

The library code is annotated with the original descriptions and
definitions from the specification, so that learning about PDF itself
is well supported. "

I wonder what the main issues would be  when porting this to Pharo.

--Hannes


On 9/12/17, Denis Kudriashov mailto:dionisi...@gmail.com> > wrote:
> There is also PDFTalk from Christian Haider
> https://wiki.pdftalk.de/doku.php.
> But it needs to be ported into Pharo.
>
> 2017-09-11 21:03 GMT+02:00 Alejandro Infante
> mailto:alejandroinfant...@gmail.com> >:
>
>> Hi!
>> I would like to explore having support for Unicode characters. I want to
>> help for it.
>>
>> Who is the right person to talk about it?
>>
>> Cheers!
>> Alejandro
>>
>

 

 

Re: [Pharo-users] Pharo 6.1 debugger - suggestions break once, does funny things?

[Tim Mackinnon]

That makes sense - but in a practical usability sense its broken - it doesn’t 
give me a useful breakpoint to step through code where I expect. Should I raise 
a Fogbugz for this?

Or should we just remove this feature if we can’t do it right?

Tim

> On 12 Sep 2017, at 11:09, Nicolai Hess  wrote:
> 
> 
> 
> 2017-09-11 1:26 GMT+02:00 Tim Mackinnon  >:
> While I’m on a debugger trip… anyone else also notice that the Suggestions, 
> Break once debug option acts weirdly? I seem to get a debugger but it halts 
> several lines after where I have said, and if I start stepping the debugger 
> seems to jump to funny places and show values that don’t line up with what I 
> know is happening? If I use Suggestions, Break it seems to work correctly 
> (apart from the Step Through I mentioned in another thread).
> 
> Tim
> 
> 
> Thats because the "Break once" creates a MetaLink that when executing this 
> Metalink it removes itself from the methode prior to the break-command. If 
> now  the debugger opens the method, it executes a context to a method that 
> already had changed. (the actual method bytecode only exists in the current 
> context).
> 
> You can see the "real" bytecode of the contexts method if you select GTs 
> bytecode view.
> 

Re: [Pharo-users] Pharo 6.1 debugger - suggestions break once, does funny things?

[Marcus Denker]

> On 12 Sep 2017, at 12:44, Tim Mackinnon  wrote:
> 
> That makes sense - but in a practical usability sense its broken - it doesn’t 
> give me a useful breakpoint to step through code where I expect. Should I 
> raise a Fogbugz for this?
> 
Yes, please add an issue. I wanted to see if I can improve the debugger  for 
this… 

Marcus

Re: [Pharo-users] Accessor Creation in Nautilus Browser

[Rob Rothwell]

On Tue, Sep 12, 2017 at 5:45 AM, Alistair Grant 
wrote:

> Hi Rob,
>
> On 12 September 2017 at 03:32, Rob Rothwell 
> wrote:
> > Hello,
> >
> > Is there a way to create accessors for more than one instance variable
> at a
> > time with the Nautilus browser?
> >
> > "Class -> Refactoring -> Inst Var Refactoring -> Accessors" shows
> available
> > instance variables, but only one can be selected...
> >
> > Thank you,
> >
> > Rob
>
> Class -> Refactoring -> Class Refactoring -> Generate Accessors
>
>
  Ha!  Thank you...my brain kept seeing "Class" and saying, "But I'm not ON
the Class side!"

Thanks again,

Cheers,
> Alistair
>
>

Re: [Pharo-users] help needed regarding conversion of hex to binary

[Ben Coman]

On Tue, Sep 12, 2017 at 2:55 AM, Casimiro de Almeida Barreto <
casimiro.barr...@gmail.com> wrote:

> Is there a (easy and ready) way to convert hexa to binary. I mean, convert
> data that was transformed
>
> x := myData asByteArray hex.
>
> back to binary?
>
>
>
I thought I'd try what usually works...  World > Tools > Finder > Examples
First to test what we know...
  #(9 10 11 12) asByteArray  .  '090a0b0c'
results in  #hex

but the reverse...
  '090a0b0c' .  #(9 10 11 12) asByteArray  .
doesn't show any result,
so perhaps there is no easy way.

But this might give you a starting point...
   hexString := #(9 10 11 12) asByteArray hex.
   (NumberParser on: '16r', hexString) nextInteger asByteArray " ==>  #[9
10 11 12]"

cheers -ben

Re: [Pharo-users] help needed regarding conversion of hex to binary

[Sven Van Caekenberghe]

ByteArray readHexFrom: #[ 0 1 2 3 4 5 6 7 8 9 10 ] hex.

> On 12 Sep 2017, at 14:47, Ben Coman  wrote:
> 
> 
> 
> On Tue, Sep 12, 2017 at 2:55 AM, Casimiro de Almeida Barreto 
>  wrote:
> Is there a (easy and ready) way to convert hexa to binary. I mean, convert 
> data that was transformed
> 
> x := myData asByteArray hex.
> 
> back to binary?
> 
> 
> 
> I thought I'd try what usually works...  World > Tools > Finder > Examples
> First to test what we know...
>   #(9 10 11 12) asByteArray  .  '090a0b0c' 
> results in  #hex
> 
> but the reverse...
>   '090a0b0c' .  #(9 10 11 12) asByteArray  .  
> doesn't show any result, 
> so perhaps there is no easy way.
> 
> But this might give you a starting point...
>hexString := #(9 10 11 12) asByteArray hex.
>(NumberParser on: '16r', hexString) nextInteger asByteArray " ==>  #[9 10 
> 11 12]"
> 
> cheers -ben

Re: [Pharo-users] Who is maintaining of Artefact?

[Stephane Ducasse]

super!

Stef

On Tue, Sep 12, 2017 at 11:38 AM, Christian Haider
 wrote:
> Moin,
>
>
>
> the port is almost done. Then I will add documentation and publish the code
> under MIT licence.
>
>
>
> Including the code used to port the library from VW to Gemstone using a
> novel approach.
>
>
>
> The heard of the “PDF engine” is the type system which allows the assignment
> of Smalltalk classes to raw PDF objects.
>
> In the new version, the PDF types are decoupled from the Smalltalk classes,
> so that classes can be renamed freely (i.e. adding prefixes).
>
>
>
> I think that porting the new version to Pharo (and VA and …) has gotten much
> easier.
>
> I hope that someone would take the challenge J.
>
>
>
> Cheers,
>
> Christian
>
>
>
>
>
> Von: Pharo-users [mailto:pharo-users-boun...@lists.pharo.org] Im Auftrag von
> Denis Kudriashov
> Gesendet: Dienstag, 12. September 2017 11:12
> An: Any question about pharo is welcome 
> Betreff: Re: [Pharo-users] Who is maintaining of Artefact?
>
>
>
> Look at slides PDFtalk for Gemstone from last ESUG about porting it to
> Gemstone (which is done).
>
> Maybe it is similar as porting GLORP into Pharo. Somebody can comment on
> this process.
>
>
>
> 2017-09-12 11:04 GMT+02:00 H. Hirzel :
>
> Looks like a worthwhile exercise
>
> PDFTalk https://wiki.pdftalk.de/doku.php
>
> "The syntax is covered comprehensively, which means that any PDF file
> can be read and any PDF object can be written in a conforming way.
>
> Typical PDF objects are dictionaries of which many are specialized as
> Smalltalk objects.
>
> The library code is annotated with the original descriptions and
> definitions from the specification, so that learning about PDF itself
> is well supported. "
>
> I wonder what the main issues would be  when porting this to Pharo.
>
> --Hannes
>
>
> On 9/12/17, Denis Kudriashov  wrote:
>> There is also PDFTalk from Christian Haider
>> https://wiki.pdftalk.de/doku.php.
>> But it needs to be ported into Pharo.
>>
>> 2017-09-11 21:03 GMT+02:00 Alejandro Infante
>> :
>>
>>> Hi!
>>> I would like to explore having support for Unicode characters. I want to
>>> help for it.
>>>
>>> Who is the right person to talk about it?
>>>
>>> Cheers!
>>> Alejandro
>>>
>>
>
>

Re: [Pharo-users] About neoCSV

[Stephane Ducasse]

Yes.
I read all the doc and is it there?
Because yes this is exactly what I was looking for.

On Tue, Sep 12, 2017 at 9:14 AM, Sven Van Caekenberghe  wrote:
> I am not sure I understand 100% what you want, but consider this example:
>
> (NeoCSVReader on: 'foo bar baz
> 1 2 3
> 11 22 33' readStream)
> separator: $ ;
> namedColumnsConfiguration;
> upToEnd.
>
> => an Array(a Dictionary(#bar->'2' #baz->'3' #foo->'1' ) a 
> Dictionary(#bar->'22' #baz->'33' #foo->'11' ))
>
> Since you did not use $, but space I had to set that using #separator:
>
> #namedColumnsConfiguration uses the header as keys and reads dictionaries 
> instead of arrays.
>
> Instead of #upToEnd you can also iterate with #do:
>
> Is this what you are looking for ?
>
>> On 12 Sep 2017, at 08:23, Stephane Ducasse  wrote:
>>
>> Hi sven
>>
>> I have a question with NeoCVS
>> Do you offer the possibility to manage the data of one line based on
>> the description given by the first line
>>
>> Ex
>>
>> foo bar baz
>> 1 2 3
>> 11 22 33
>>
>> =>
>>
>> res doWith: [:tag :data |
>> Transcript show: tag ; space ; show: data printString ;cr
>> ]
>>
>> foo 1 bar 2 baz 3
>> foo 11 bar 22 baz 33
>>
>> Stef
>>
>
>

Re: [Pharo-users] Who is maintaining of Artefact?

[Ben Coman]

On Tue, Sep 12, 2017 at 5:38 PM, Christian Haider <
christian.hai...@smalltalked-visuals.com> wrote:

> Moin,
>
>
>
> the port is almost done. Then I will add documentation and publish the
> code under MIT licence.
>
>
>
> Including the code used to port the library from VW to Gemstone using a
> novel approach.
>
>
>
> The heard of the “PDF engine” is the type system which allows the
> assignment of Smalltalk classes to raw PDF objects.
>
> In the new version, the PDF types are decoupled from the Smalltalk
> classes, so that classes can be renamed freely (i.e. adding prefixes).
>

Thats great news Christian.  Pharo's lack of namespaces and your desire to
avoid prefixing classes made me wonder how to keep VW and Pharo ports
synchronised - which added friction to my minor attempt to port it to
Pharo.

>
>
> I think that porting the new version to Pharo (and VA and …) has gotten
> much easier.
>
> I hope that someone would take the challenge J.
>

I'm busy with other things for a while, but I'll add it back on my list of
things to look at.

cheers -ben


>
>
> Cheers,
>
> Christian
>
>
>
>
>
> *Von:* Pharo-users [mailto:pharo-users-boun...@lists.pharo.org] *Im
> Auftrag von *Denis Kudriashov
> *Gesendet:* Dienstag, 12. September 2017 11:12
> *An:* Any question about pharo is welcome 
> *Betreff:* Re: [Pharo-users] Who is maintaining of Artefact?
>
>
>
> Look at slides PDFtalk for Gemstone
>  from last ESUG
> about porting it to Gemstone (which is done).
>
> Maybe it is similar as porting GLORP into Pharo. Somebody can comment on
> this process.
>
>
>
> 2017-09-12 11:04 GMT+02:00 H. Hirzel :
>
> Looks like a worthwhile exercise
>
> PDFTalk https://wiki.pdftalk.de/doku.php
>
> "The syntax is covered comprehensively, which means that any PDF file
> can be read and any PDF object can be written in a conforming way.
>
> Typical PDF objects are dictionaries of which many are specialized as
> Smalltalk objects.
>
> The library code is annotated with the original descriptions and
> definitions from the specification, so that learning about PDF itself
> is well supported. "
>
> I wonder what the main issues would be  when porting this to Pharo.
>
> --Hannes
>
>
> On 9/12/17, Denis Kudriashov  wrote:
> > There is also PDFTalk from Christian Haider
> > https://wiki.pdftalk.de/doku.php.
> > But it needs to be ported into Pharo.
> >
> > 2017-09-11 21:03 GMT+02:00 Alejandro Infante
> > :
> >
> >> Hi!
> >> I would like to explore having support for Unicode characters. I want to
> >> help for it.
> >>
> >> Who is the right person to talk about it?
> >>
> >> Cheers!
> >> Alejandro
> >>
> >
>
>
>

Re: [Pharo-users] About neoCSV

[Sven Van Caekenberghe]

I guess #namedColumnsConfiguration is relatively recent [oct 2015] (although it 
is just a convenience method, the underlying technology was already there), so 
it is probably not part of the external docs.
 
> 
> On 12 Sep 2017, at 15:26, Stephane Ducasse  wrote:
> 
> Yes.
> I read all the doc and is it there?
> Because yes this is exactly what I was looking for.
> 
> On Tue, Sep 12, 2017 at 9:14 AM, Sven Van Caekenberghe  wrote:
>> I am not sure I understand 100% what you want, but consider this example:
>> 
>> (NeoCSVReader on: 'foo bar baz
>> 1 2 3
>> 11 22 33' readStream)
>>separator: $ ;
>>namedColumnsConfiguration;
>>upToEnd.
>> 
>> => an Array(a Dictionary(#bar->'2' #baz->'3' #foo->'1' ) a 
>> Dictionary(#bar->'22' #baz->'33' #foo->'11' ))
>> 
>> Since you did not use $, but space I had to set that using #separator:
>> 
>> #namedColumnsConfiguration uses the header as keys and reads dictionaries 
>> instead of arrays.
>> 
>> Instead of #upToEnd you can also iterate with #do:
>> 
>> Is this what you are looking for ?
>> 
>>> On 12 Sep 2017, at 08:23, Stephane Ducasse  wrote:
>>> 
>>> Hi sven
>>> 
>>> I have a question with NeoCVS
>>> Do you offer the possibility to manage the data of one line based on
>>> the description given by the first line
>>> 
>>> Ex
>>> 
>>> foo bar baz
>>> 1 2 3
>>> 11 22 33
>>> 
>>> =>
>>> 
>>> res doWith: [:tag :data |
>>>Transcript show: tag ; space ; show: data printString ;cr
>>> ]
>>> 
>>> foo 1 bar 2 baz 3
>>> foo 11 bar 22 baz 33
>>> 
>>> Stef
>>> 
>> 
>> 
> 

Re: [Pharo-users] DictionaryValueHolder question

[Peter Uhnák]

Hi Rob,

this is by omission. Originally CollectionValueHolder didn't have it either
until I've introduced ValueAdded/ValueRemoved. So feel free to incorporate
it in a meaningful fashion to a DictionaryHolder.

Peter

On Tue, Sep 12, 2017 at 3:46 AM, Rob Rothwell 
wrote:

> Hello,
>
> Unlike the CollectionValueHolder, the DictionaryValueHolder only notifies
> you that something has changed, rather than tell you what item has been
> added or removed.  Does anyone know if this is by design, or just
> incomplete?
>
> If incomplete, what do you think about using the 
> ValueAdded/ValueChanged/ValueRemoved
> announcements and passing the association in newValue/oldValue?
>
> Thank you,
>
> Rob
>
>
>

Re: [Pharo-users] DictionaryValueHolder question

[Rob Rothwell]

Hi Peter,

Thank you very much for the clarification; I'll see what I can come up with!

Take care,

Rob

On Tue, Sep 12, 2017 at 11:18 AM, Peter Uhnák  wrote:

> Hi Rob,
>
> this is by omission. Originally CollectionValueHolder didn't have it
> either until I've introduced ValueAdded/ValueRemoved. So feel free to
> incorporate it in a meaningful fashion to a DictionaryHolder.
>
> Peter
>
> On Tue, Sep 12, 2017 at 3:46 AM, Rob Rothwell 
> wrote:
>
>> Hello,
>>
>> Unlike the CollectionValueHolder, the DictionaryValueHolder only notifies
>> you that something has changed, rather than tell you what item has been
>> added or removed.  Does anyone know if this is by design, or just
>> incomplete?
>>
>> If incomplete, what do you think about using the
>> ValueAdded/ValueChanged/ValueRemoved announcements and passing the
>> association in newValue/oldValue?
>>
>> Thank you,
>>
>> Rob
>>
>>
>>
>

Re: [Pharo-users] Gofer loads wrong version?

[Herby Vojčík]

Bump.

Can you pls help me with how to load proper version / finding what is 
wrong here?


As shown in replies, it loads two versions (eg. it seems it loads 
everything it finds in the repo).


Herby

Herby Vojčík wrote:

Hello!

As I need to load specific version of Glorp with my fix, and I did not
find out how to force-override it in my baseline, I tried to load it
post-the-baseline via Gofer:

(Gofer new smalltalkhubUser: 'DBXTalk' project: 'Glorp')
package: 'Glorp';
version: 'Glorp-HerbyVojcik.127';
load.

But it actually looks like this: [see attached bogusgofer.png].

OTOH, after the image loads the tests pass and monticello browser looks
like this: [see attached mbloaded127.png].

Am I doing something wrong? I am a bit afraid what is actually loaded /
if it won't break b/c of inconsistent state.

Herby

Re: [Pharo-users] Who is maintaining of Artefact?

[Stephan Eggermont]

On 12-09-17 11:11, Denis Kudriashov wrote:
> Maybe it is similar as porting GLORP into Pharo. Somebody can comment 
on this process.


The important point is that the original code is in another dialect, and 
the idea is not to fork, as the original maintainer develops it further. 
That means that it is not a one-time translation, but is a set of 
transformation rules and platform specific packages that is applied 
every time something changes on either side.


Stephan

Re: [Pharo-users] Who is maintaining of Artefact?

[Christian Haider]

> von Stephan Eggermont
> 
> On 12-09-17 11:11, Denis Kudriashov wrote:
>  > Maybe it is similar as porting GLORP into Pharo. Somebody can comment
> on this process.
> 
> The important point is that the original code is in another dialect, and the 
> idea
> is not to fork, as the original maintainer develops it further.
> That means that it is not a one-time translation, but is a set of 
> transformation
> rules and platform specific packages that is applied every time something
> changes on either side.

Stefan, you are good :-)!

Exactly! The full code transformation from VW to GS is described as a value, 
which gets interpreted by a generic code fileout machine.
A new version of the GS fileout can be generated at any time without human 
intervention.
A SystemChange (the name of the value class) could be interpreted in different 
ways.

Currently, the transform is one way from VW to GS. 

I am very interested in changes people may do to the library, but for now, I 
will keep the master in VW and integrate contributions by hand.
I am sure that, if the need arises, there will be a way to deal with a two-way 
synchronization. But let's have a contribution first :-)

Christian