rkward-devel — Discussion of RKWard development

Re: [rkward-devel] import/export CSV

[Thomas F. <tho...@ru...>]

Hi,

On Sunday 30 November 2014 23:19:48 meik michalke wrote:
> as promised, i completely re-wrote the "export tabular data" dialog. it's
> attached[*] as a pre-built plugin package so you can check it out (it also
> includes the rkwarddev script used to build it).

ok, some first comments, below. Minor stuff, only.
 
> it defaults to custom file format at the moment -- CSV or CSV2 could be
> better choices, but this deactivates some elements and could confuse some
> people, couldn't it?

See my last mail. I think defaulting to CSV (_not_ CVS2!) would be sensible.

Assorted comments:
- Why is the "append" option controlled by the predefined format?
- Uncommenting arguments, that do not apply, is a nifty idea, but does come 
with some drawbacks:
  - Bloats the generated code
  - Seems to suggest that these _could_ be customized for write.csv(2). 
However, the R help says: "append, col.names, sep, dec and qmethod cannot be 
altered." I'd suggest stripping these.
- Remove the "data.frame"-warning (see my last mail).
- For encoding specification, import_spss and import_stata share some code 
(currently in the form of a <snippet>. It might make sense to turn this into 
an embeddable plugin, and use it.

Regards
Thomas

Re: [rkward-devel] import/export CSV

[Thomas F. <tho...@ru...>]

Hi,

On Sunday 30 November 2014 23:19:48 meik michalke wrote:
> as promised, i completely re-wrote the "export tabular data" dialog. it's
> attached[*] as a pre-built plugin package so you can check it out (it also
> includes the rkwarddev script used to build it).

will give it a look, shortly.
 
> it defaults to custom file format at the moment -- CSV or CSV2 could be
> better choices, but this deactivates some elements and could confuse some
> people, couldn't it?

I wouldn't be too afraid of that, in particular, as the "File format" 
preselection, and the elements controlled by it are all on the same tab. State 
that on the .rkh-page to be on the safe side. Also, exporting to one of the 
standard formats is certainly the most common thing to do, and should be 
easiest.

> i have a question regarding the current plugin:
> 
> what plugin is referred to by "use the plugin 'write' for mere variables"?
> is it "Write vector / matrix"? generally, can i somehow refer to a plugin
> so that my dialog fetches its current name (dialog label) dynamically? i
> guess not -- would be cool to have it as a direct link, shown as the path
> from its root menu.

As to that "warning":
a) It seems to be pretty useless (and in fact rather misleading) in the first 
place. I suppose it might have had a real meaning some time in the past 
(before we were running plugin code inside local(), for instance.
b) If at all, it should go to the help page.

The plugin, referred to, would be "save_variables", inded. You can refer to 
that in the .rkh page (see above) using <link 
href="rkward://component/save_variables"/>, which will insert the correct 
label, automatically (and menu-path might be added, later).

> and one questions regarding the import CSV plugin: what does this logic
> statement do?
> 
>   <connect client="commentchar.enabled" governor="quickNone" />

Well, it disables specification of the commentchar (effectively disabling 
comments), if any of the pre-defined formats is selected. I believe the 
rationale for that is:
- For the predefined formats, code using read.csv(2) / read.delim(2) is 
generated, for easier copying of code.
- These default to comment.char="", instead of "#" for read.table().
- Cleanly covering that in the UI would have been rather cumbersome, and did 
not seem worth bothering (as comment-lines inside csv-files should be rather 
uncommon, in the first place).

Regards
Thomas

[rkward-devel] import/export CSV

[meik m. <mei...@un...>]

hi,

as promised, i completely re-wrote the "export tabular data" dialog. it's 
attached[*] as a pre-built plugin package so you can check it out (it also 
includes the rkwarddev script used to build it).

it defaults to custom file format at the moment -- CSV or CSV2 could be better 
choices, but this deactivates some elements and could confuse some people, 
couldn't it?

i have a question regarding the current plugin:

what plugin is referred to by "use the plugin 'write' for mere variables"? is 
it "Write vector / matrix"? generally, can i somehow refer to a plugin so that 
my dialog fetches its current name (dialog label) dynamically? i guess not -- 
would be cool to have it as a direct link, shown as the path from its root 
menu.

and one questions regarding the import CSV plugin: what does this logic 
statement do?

  <connect client="commentchar.enabled" governor="quickNone" />


viele grüße :: m.eik


[*] i am generally not a friend of attachments, but this one is so tiny that i 
hope you're ok with it ;-)
-- 
  dipl. psych. meik michalke
  institut f"ur experimentelle psychologie
  abt. f"ur diagnostik und differentielle psychologie
  heinrich-heine-universit"at d-40204 d"usseldorf

Re: [rkward-devel] RKWard 0.6.2 starts with an error

[Mohammad A. <m-...@nt...>]

Hi Thomas,
Thank you for your response. For the time being I will revert to an 
earlier version of RKWard until a stable version of 0.6.2 is released. I 
will be grateful if you would advice me on the time when the stable 
version will come out. Another issues  I did not report in my earlier 
e-mail is that some menue labels such as analysis and Data appear as no 
label ("no-label").

Regards,
Mohammad

On 30/11/2014 13:56, Thomas Friedrichsmeier wrote:
> Hi,
>
> On Saturday 29 November 2014 19:11:08 Mohammad Abbas wrote:
>> Hello,
>> Thanks for getting back to me.
> same, here. No need to withdraw your posts to the list. We use moderation for
> non-subscribers to weed out SPAM-posts, not to actually limit anything.
>
>> In response to your questions
>> 1) I installed RKWard using the (RKWard_O.6.2_KDE_4.10.2_R_3.1.1.exe)?
>> 2) RKWard Starts after the DOS window show the error message. However,
>> when accessing other R libraries such as Rcmdr and rattle they tend to
>> crash or "freeze up".
> Ok, I can see that part, too. This should now be fixed in our development
> version. As a workaround, until we can release a fixed version:
> - avoid using other graphical packages in combination with RKWard, or, if that
> is unavoidable:
> - before using any functionality of these packages, make sure, something is
> running inside RKWard's R Console. This could simply be:
>     Sys.sleep (30000)
> This will block most actions in RKWard, but you can simply unblock by
> interrupting.
>
>> I will run the code you given me and I will e-mail you.
> This will still be of interest, as it is an entirely unrelated problem.
>
> Regards
> Thomas

Re: [rkward-devel] RKWard 0.6.2 starts with an error

[Thomas F. <tho...@ru...>]

Hi,

On Saturday 29 November 2014 19:11:08 Mohammad Abbas wrote:
> Hello,
> Thanks for getting back to me.

same, here. No need to withdraw your posts to the list. We use moderation for 
non-subscribers to weed out SPAM-posts, not to actually limit anything.

> In response to your questions
> 1) I installed RKWard using the (RKWard_O.6.2_KDE_4.10.2_R_3.1.1.exe)?
> 2) RKWard Starts after the DOS window show the error message. However,
> when accessing other R libraries such as Rcmdr and rattle they tend to
> crash or "freeze up".

Ok, I can see that part, too. This should now be fixed in our development 
version. As a workaround, until we can release a fixed version:
- avoid using other graphical packages in combination with RKWard, or, if that 
is unavoidable:
- before using any functionality of these packages, make sure, something is 
running inside RKWard's R Console. This could simply be:
   Sys.sleep (30000)
This will block most actions in RKWard, but you can simply unblock by 
interrupting.

> I will run the code you given me and I will e-mail you.

This will still be of interest, as it is an entirely unrelated problem.

Regards
Thomas

Re: [rkward-devel] RKWard 0.6.2 starts with an error

[Mohammad A. <m-...@nt...>]

Hello,
Thanks for getting back to me.
In response to your questions
1) I installed RKWard using the (RKWard_O.6.2_KDE_4.10.2_R_3.1.1.exe)?
2) RKWard Starts after the DOS window show the error message. However, 
when accessing other R libraries such as Rcmdr and rattle they tend to 
crash or "freeze up".
3) I use windows 8.1  64bit

I will run the code you given me and I will e-mail you.

Kind Regards,
M.Abbas


On 29/11/2014 18:43, Thomas Friedrichsmeier wrote:
> Hi!
>
> On Saturday 29 November 2014 15:07:39 Mohammad Abbas wrote:
>> Installing the latest version of RKWard seems to start with an error. I
>> have a screen shot of the message and hope you can help to resolve it.
> I'll try. Some questions:
> 1) Did you install using the "custom installer" (install_rkward_0.6.2.exe) or
> the "installation bundle" (RKWard_O.6.2_KDE_4.10.2_R_3.1.1.exe)?
>
> 2) Does RKWard start in spite of these messages, or is this the last thing you
> see happening?
>
> 3) Which version of Windows?
>
> 4) Please open a command prompt, cd to your installation of rkward, and type
>    KDE\bin\rkward.exe --debug-level 4
> This should give a bit more output during startup, to help diagnose what
> exactly is going wrong.
>
> Thanks!
> Thomas

Re: [rkward-devel] RKWard 0.6.2 starts with an error

[Thomas F. <tho...@ru...>]

Hi!

On Saturday 29 November 2014 15:07:39 Mohammad Abbas wrote:
> Installing the latest version of RKWard seems to start with an error. I
> have a screen shot of the message and hope you can help to resolve it.

I'll try. Some questions:
1) Did you install using the "custom installer" (install_rkward_0.6.2.exe) or 
the "installation bundle" (RKWard_O.6.2_KDE_4.10.2_R_3.1.1.exe)?

2) Does RKWard start in spite of these messages, or is this the last thing you 
see happening?

3) Which version of Windows?

4) Please open a command prompt, cd to your installation of rkward, and type
  KDE\bin\rkward.exe --debug-level 4
This should give a bit more output during startup, to help diagnose what 
exactly is going wrong.

Thanks!
Thomas

[rkward-devel] RKWard 0.6.2 starts with an error

[Mohammad A. <m-...@nt...>]

Hi,
Installing the latest version of RKWard seems to start with an error. I 
have a screen shot of the message and hope you can help to resolve it.
Thank you
Dr. M. Abbas

Re: [rkward-devel] Question regarding menu ordering for plugins

[Thomas F. <tho...@ru...>]

Hi,

On Friday 28 November 2014 01:18:53 Stefan Rödiger wrote:
> My vote: alphabetical sorting only within each group

ok, I did that (and if we don't like it, it will be easy to remove the 
alphabetical sorting).

Now a lot depends on coming up with a sensible grouping. We'll have to see 
about that. Here are some details of the current implementation:

- You can define groups inside any menu like this:
  <group id="somegroup"/>
- If you want the group to be separated from other entries, use:
  <group id="somegroup" separated="true"/>
- Entries, menus, and groups can be appended to a specified group, using:
  <entry component="..." group="somegroup"/>
- In fact, it is also possible to define groups (without separator lines) 
implicitly:
  <entry component="first" group="a"/>
  <entry component="third"/>
  <entry component="second" group="a"/>
- Group names are specific to each menu. Group "a" in menu "Data" does not 
conflict with group "a" in menu "Analysis", for example.
- I believe, the most common use case is defining groups at the top, or at the 
bottom of a menu. For this, there are pre-defined groups "top" and "bottom" in 
each menu.
- Entries within each group are sorted, alphabetically. Groups appear in the 
order of declaration (unless appended to another group, of course).
- Menus and entries without group specification logically from a group (""), 
too.

Please experiment with this, if you care. As long as this is not released, it 
will be easily possible to change details. Afterwards that may be more 
difficult.

Regards
Thomas

[rkward-devel] News on plugin i18n

[Thomas F. <tho...@ru...>]

Hi!

Ok, I did make some more progress on plugin i18n (but still trying to figure 
out how to integrate best with KDE's translation infrastructure[1]). Here's 
some of the more important bits:

1. The message extraction script is finally complete (I think), although my 
statement from end of October that it would not get too much more complex was 
a bit premature...
On the upside it will also do some things that I had not originally thought of 
including. Basically, all you have to do now (for external plugins) is:
  scripts/update_plugin_messages.py my.pluginmap
On the first run this will create the .pot-file (and it will be in folder "po" 
next to your pluginmap). Place any translations (.po-files) next to it. Then 
run the script again to update extracted messages, merge with the 
translations, compile the translations and copy them to their destination dirs 
below "po".

2. Actually marking all relevant strings for i18n in the .js files is as 
cumbersome as I had feared, and then some. I've completed most of 
analysis.pluginmap, but there is a whole lot of work left to be done on the 
other plugins.

3. But as I was getting frustrated with all that header-pasting, I 
had an idea to make that particular part much easier, and cleaner.
You can now write (in .js):
  new Header (i18n ("Some analysis"))
      .add (i18n ("Some parameter"), i18n ("some value"))
      .print ();
Better yet, suppose "Some parameter" is controlled by a radio-element (with id 
"radio_id"), and essentially you just want to copy the setting from the UI for 
documenting it in the header, you can now do:
  new Header (i18n ("Some analysis")).addFromUI ("radio_id").print ();

Regards
Thomas

[1] http://lists.kde.org/?l=kde-i18n-doc&m=141700741608249&w=2

Re: [rkward-devel] file formats [was: rkward: erm, ok, here are the missing files to the last commit...]

[Thomas F. <tho...@ru...>]

Hi,

On Friday 28 November 2014 21:12:47 meik michalke wrote:
> Am Freitag, 28. November 2014, 08:04:57 schrieb Thomas Friedrichsmeier:
> > > btw. there's some other issues: RKWard doesn't recognise *.rda files
> > > yet;
> > 
> > Ok, will add.
> 
> i'd vote for RKWard being able to edit all text files it is using, i.e.,
> also adding *.xml and *.js to the list. plugin development would benefit
> from it, at least i'd use kate anyway.

true. Should now be handled correctly. (In fact it was always meant to open 
all text files, but did not recognize them, due to a bug...)

Regards
Thomas

[rkward-devel] file formats [was: rkward: erm, ok, here are the missing files to the last commit...]

[meik m. <mei...@un...>]

hi,

Am Freitag, 28. November 2014, 08:04:57 schrieb Thomas Friedrichsmeier:
> > btw. there's some other issues: RKWard doesn't recognise *.rda files yet;
> 
> Ok, will add.

i'd vote for RKWard being able to edit all text files it is using, i.e., also 
adding *.xml and *.js to the list. plugin development would benefit from it, 
at least i'd use kate anyway.


viele grüße :: m.eik

-- 
  dipl. psych. meik michalke
  institut f"ur experimentelle psychologie
  abt. f"ur diagnostik und differentielle psychologie
  heinrich-heine-universit"at d-40204 d"usseldorf

Re: [rkward-devel] [rkward-cvs] [rkward] rkward: erm, ok, here are the missing files to the last commit...

[Thomas F. <tho...@ru...>]

On Friday 28 November 2014 16:20:10 meik michalke wrote:
> good point -- i did that, and i also renamed the file to vnd.rkward.r.xml,
> which makes it a vendor-specific MIME type declaration. that makes it
> possible for us to define that MIME type without colliding with other
> packages doing the same.

Ah, yes, that sounds much better!

> i'll try and test if this and the protocol handler also works on OS X. can
> you check that on windows?

Will try to remember. I have an inkling it's going to require adding it to the 
registry, though...

> actually, that is something that has always annoyed me a bit: i start RKWard
> and want to *open* a previous workspace, but the first thing i get is a
> question whether to *save* the workspace. good thing if that was gone.

Yes. I'll try to come up with something.
 
> do we have to care for that single .Random.seed object when one wants to
> open another workspace anyway?

No, but how to tell it apart for other, more interesting objects, reliably. I 
think the solution will have to involve something like marking the workspace 
as "clean", at an appropriate point in the startup sequence, and then only 
asking to save, if it was modified after that.

> alternatively, RKWard could ask you how to deal with the input: replace
> current workspace or add all objects to the current one. do you know whether
> there's a difference between *.RData files containing saved workspaces and
> saved R objects? if there was, we could add a <magic> section to the MIME
> type, which would cause the first bits of a file to be read to get its
> type.

I don't think so. ?save.image says: "save.image() is just a short-cut for 
‘save my current workspace’, i.e., save(list = ls(all = TRUE), file = 
".RData")". Even if you could tell how it was saved, that still doesn't tell, 
reliably, how you'd like to use it. In general, there are three options:
  - Replace current workspace
  - Merge with current workspace (potentially overwriting objects)
  - Import into a sub-environment
And then, if a .workplace-file exists to go with that, for the first two, 
there's also the question, whether to load that, and if so, whether to close 
all other windows first...

So - a whole lot of options, really. I guess a plan could look like this: In 
the menu, provide two actions: One is the current behavior. The second 
(perhaps called "Import Workspace..."?) would offer all applicable choices. If 
there is a request to open a workspace file in a running session (from 
RKWard's own filebrowser, or externally), and the current workspace is not 
empty, use that second action, too.

Regards
Thomas

Re: [rkward-devel] [rkward-cvs] [rkward] rkward: erm, ok, here are the missing files to the last commit...

[meik m. <Mei...@un...>]

hi,

Am Freitag, 28. November 2014, 08:04:57 schrieb Thomas Friedrichsmeier:
> On Friday 28 November 2014 01:07:40 meik michalke wrote:
> > are you sure?
> 
> no. Except perhaps no the part about shipping both mime-types in one file.

good point -- i did that, and i also renamed the file to vnd.rkward.r.xml, 
which makes it a vendor-specific MIME type declaration. that makes it possible 
for us to define that MIME type without colliding with other packages doing 
the same. the original XML file is now being installed to 
/usr/share/mime/packages, but XDG does also automatically generate 
/usr/share/mime/text/r.xml and /usr/share/mime/application/rdata.xml from it. 
in effect it's the same result.

> I'm still somewhat worried about future clashes. _If_ R ever decides to
> start shipping these mime-types, RKWard is going to get the blame for
> conflicting names, I'm afraid.

i hope this is solved for good now. the hint to libreoffice.xml was useful and 
led me to the vnd.* files.

i'll try and test if this and the protocol handler also works on OS X. can you 
check that on windows?

> > and if you click on an *.RData file, you're asked *once* whether the
> > workspace should be saved -- after that, the current worspace is just
> > silently replaced with the content of any other RData file you click.
> 
> I changed that a bit, and yes, it might be buggy, but I think you should
> find, the behavior is now:
> - If a workspace has just been loaded, and not been touched, don't ask
> whether to save it before quitting / replacing it.
> - If you touch some object in between, you should be prompted, again.

right, i missed that part -- i was just baffled that rkward switched from one 
workspace to another without asking, and i was wondering what that would do to 
one's data if you accidentally clicked on a *.RData file... but that's a good 
solution now.

> - Now as to why you're prompted for the initial - probably empty -
> workspace: It's that .Random.seed that gets created some time in the middle
> of starting up. Did not think up a nice way to cope with that, yet.

actually, that is something that has always annoyed me a bit: i start RKWard 
and want to *open* a previous workspace, but the first thing i get is a 
question whether to *save* the workspace. good thing if that was gone.

do we have to care for that single .Random.seed object when one wants to open 
another workspace anyway?

> > single objects can't be added to the workspace this way either. there
> > should probably be two methods for handling *.RData files: complete
> > workspace or individual R objects,
> 
> True.

KDE knows this as service menus, i can try to write an according *.desktop 
file as well. then you'd have to actively open the file via right click. that 
would need another option for the rkward binary.

alternatively, RKWard could ask you how to deal with the input: replace 
current workspace or add all objects to the current one. do you know whether 
there's a difference between *.RData files containing saved workspaces and 
saved R objects? if there was, we could add a <magic> section to the MIME 
type, which would cause the first bits of a file to be read to get its type.

> > and you should always be asked before
> > your workspace is wiped.
> 
> Really?

no, its fine as it is (except maybe for the initial workspace).


viele grüße :: m.eik

-- 
  dipl. psych. meik michalke
  institut f"ur experimentelle psychologie
  abt. f"ur diagnostik und differentielle psychologie
  heinrich-heine-universit"at d-40204 d"usseldorf

Re: [rkward-devel] [rkward-cvs] [rkward] rkward: erm, ok, here are the missing files to the last commit...

[Thomas F. <tho...@ru...>]

Hi,

On Friday 28 November 2014 01:07:40 meik michalke wrote:
> Am Donnerstag, 27. November 2014, 19:43:58 schrieb Thomas Friedrichsmeier:
> > On Thursday 27 November 2014 15:44:45 m. eik michalke wrote:
> > > A  +12   -0    rkward/r.xml
> > > A  +13   -0    rkward/rdata.xml
> > 
> > these work nicely for me. However, I think they should be merged into one
> > mime file, and that should be called rkward.xml . So there's no
> > significant
> > risk that this could clash (now or later) with (R-) mime types declared
> > from other sides.
> 
> are you sure?

no. Except perhaps no the part about shipping both mime-types in one file.

> i mean, these file types are not really RKWard files, but
> ordinary R files, so i'd say their MIME type can only be R or R data,
> respectively. the connection between these MIME types and applications is
> defined in *.desktop files instead,

Ah, in fact I had missed that part. Sounds more sane, now that I understand 
that.

> where one MIME type can have multiple
> associations. i'd find it a bit confusing if there were such pseudo-MIME
> types like, say, firefox.xml, chrome.xml and rekonq.xml, all for *.html
> files.
> 
> of course it would be better to have these global MIME types rather defined
> by either R or KDE itself, so RKWard can just assume they're there. but
> from what i found on the net, this has not been accomplished for many many
> years. there still doesn't seem to be any package defining a MIME type for
> R files yet, at least in the realm of ubuntu and CRAN packages.

So probably the cleanest solution would be having them added to R... Oh well.

I'm still somewhat worried about future clashes. _If_ R ever decides to start 
shipping these mime-types, RKWard is going to get the blame for conflicting 
names, I'm afraid. That's why my first thought was naming it rkward.xml 
(similar to libreoffice.xml, which also declares .xls, for instance). But no, 
I'm not sure.

> btw. there's some other issues: RKWard doesn't recognise *.rda files yet;

Ok, will add.

> and if you click on an *.RData file, you're asked *once* whether the
> workspace should be saved -- after that, the current worspace is just
> silently replaced with the content of any other RData file you click.

I changed that a bit, and yes, it might be buggy, but I think you should find, 
the behavior is now:
- If a workspace has just been loaded, and not been touched, don't ask whether 
to save it before quitting / replacing it.
- If you touch some object in between, you should be prompted, again.
- Now as to why you're prompted for the initial - probably empty - workspace: 
It's that .Random.seed that gets created some time in the middle of starting 
up. Did not think up a nice way to cope with that, yet.

> single objects can't be added to the workspace this way either. there
> should probably be two methods for handling *.RData files: complete
> workspace or individual R objects,

True.

> and you should always be asked before
> your workspace is wiped.

Really?

Regards
Thomas

Re: [rkward-devel] Question regarding menu ordering for plugins

[Stefan R. <ste...@gm...>]

My vote: alphabetical sorting only within each group

Kind regards
Stefan

Re: [rkward-devel] [rkward-cvs] [rkward] rkward: erm, ok, here are the missing files to the last commit...

[meik m. <mei...@un...>]

hi,

Am Donnerstag, 27. November 2014, 19:43:58 schrieb Thomas Friedrichsmeier:
> On Thursday 27 November 2014 15:44:45 m. eik michalke wrote:
> > A  +12   -0    rkward/r.xml
> > A  +13   -0    rkward/rdata.xml
> 
> these work nicely for me. However, I think they should be merged into one
> mime file, and that should be called rkward.xml . So there's no significant
> risk that this could clash (now or later) with (R-) mime types declared
> from other sides.

are you sure? i mean, these file types are not really RKWard files, but 
ordinary R files, so i'd say their MIME type can only be R or R data, 
respectively. the connection between these MIME types and applications is 
defined in *.desktop files instead, where one MIME type can have multiple 
associations. i'd find it a bit confusing if there were such pseudo-MIME types 
like, say, firefox.xml, chrome.xml and rekonq.xml, all for *.html files.

of course it would be better to have these global MIME types rather defined by 
either R or KDE itself, so RKWard can just assume they're there. but from what 
i found on the net, this has not been accomplished for many many years. there 
still doesn't seem to be any package defining a MIME type for R files yet, at 
least in the realm of ubuntu and CRAN packages.

btw. there's some other issues: RKWard doesn't recognise *.rda files yet; and 
if you click on an *.RData file, you're asked *once* whether the workspace 
should be saved -- after that, the current worspace is just silently replaced 
with the content of any other RData file you click. single objects can't be 
added to the workspace this way either. there should probably be two methods 
for handling *.RData files: complete workspace or individual R objects, and 
you should always be asked before your workspace is wiped.


viele grüße :: m.eik

-- 
  dipl. psych. meik michalke
  institut f"ur experimentelle psychologie
  abt. f"ur diagnostik und differentielle psychologie
  heinrich-heine-universit"at d-40204 d"usseldorf

Re: [rkward-devel] [rkward-cvs] [rkward] rkward: erm, ok, here are the missing files to the last commit...

[Thomas F. <tho...@ru...>]

Hi,

On Thursday 27 November 2014 15:44:45 m. eik michalke wrote:
> Git commit 34ab37d96a82e0010dc805dd77e40d6b7aeb6c80 by m.eik michalke.
> Committed on 27/11/2014 at 15:43.
> Pushed by meikm into branch 'master'.
> 
> erm, ok, here are the missing files to the last commit...
> 
> A  +12   -0    rkward/r.xml
> A  +13   -0    rkward/rdata.xml

these work nicely for me. However, I think they should be merged into one mime 
file, and that should be called rkward.xml . So there's no significant risk 
that this could clash (now or later) with (R-) mime types declared from other 
sides.

Regards
Thomas

Re: [rkward-devel] run again link outside RKWard

[meik m. <Mei...@un...>]

hi,

Am Freitag, 21. November 2014, 17:25:16 schrieb Thomas Friedrichsmeier:
> On Wednesday 19 November 2014 11:06:21 meik michalke wrote:
> > i was curious if it was possible to get the run again links work also
> > outside of RKWard
[...]
> 2. Use
>   rkward --reuse myurl
> to open myurl in an existing RKWard instance (if any).

awesome, works just great! i have a working rkward.protocol for that and will 
add it to git once i found the correct way to also add it to the installation 
process.

where would you like the .protocol file best in the source tree? my guess 
would be the location of the .desktop file.

i'm now trying to get an entry in the context menu of KDE file managers, so 
you can open any R script or workspace image with a right click. this should 
also make it easy to enable RKWard to offer this in its file browser dialog.

proof of concept first, then we can refine it.


viele grüße :: m.eik

-- 
  dipl. psych. meik michalke
  institut f"ur experimentelle psychologie
  abt. f"ur diagnostik und differentielle psychologie
  heinrich-heine-universit"at d-40204 d"usseldorf

Re: [rkward-devel] feedback

[Aaron B. <ab...@sf...>]

>
> just to make sure: do you know that you can export plots in various formats
> directly from the graphics window? this was *not* possible in previeous
> versions of RKWard on mac, only after thomas added the RK() device.
>

1) Mind = blown

2) Last time I knew, this did not work, so I never touched it again!

3) Looks like I need to update that section of my manual as well...

[rkward-devel] Question regarding menu ordering for plugins

[Thomas F. <tho...@ru...>]

Hi!

I'm currently redoing some of the menu-building from .pluginmaps. In 
particular so we can keep information, on just where in the menu a particular 
plugin can be found. In this process, I plan on ditching the index="x"-
attribute that is currently controlling ordering of menu items. This never was 
terribly clever design. Instead, I consider simply sorting menu entries 
alphabetically by their label, automatically.

Does that sound like a good idea to you, or will it be better to give more 
control (and more responsibility) to the .pluginmap author? One alternative 
idea would be to replace the attribute "index" with "group". Entries would be 
placed in the menu in the order of declaration, but those belonging to the 
same "group" would stay next to each other. Would you like that better? Or a 
mixed approach, with alphabetical sorting only within each group?

Regards
Thomas

Re: [rkward-devel] Review Request 121245: Change rkward-devel macport to build from git repo

[Thomas F. <tho...@ru...>]

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/121245/
-----------------------------------------------------------

(Updated Nov. 25, 2014, 6:54 p.m.)


Review request for Rkward.


Repository: rkward


Description
-------

So we can shut down the svn repo on SF.net (actually not shutting it down, but marking it as out-of-date)


Diffs
-----

  macports/kde/rkward-devel/Portfile eb6f7a4 

Diff: https://git.reviewboard.kde.org/r/121245/diff/


Testing
-------

None. Which is the reason for this review request. (Well, another reason is that I wanted to test our reviewboard setup).


Thanks,

Thomas Friedrichsmeier

Re: [rkward-devel] Review Request 121245: Change rkward-devel macport to build from git repo

[Thomas F. <tho...@ru...>]

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/121245/
-----------------------------------------------------------

(Updated Nov. 25, 2014, 6:54 p.m.)


Review request for Rkward.


Repository: rkward


Description
-------

So we can shut down the svn repo on SF.net (actually not shutting it down, but marking it as out-of-date)


Diffs
-----

  macports/kde/rkward-devel/Portfile eb6f7a4 

Diff: https://git.reviewboard.kde.org/r/121245/diff/


Testing
-------

None. Which is the reason for this review request. (Well, another reason is that I wanted to test our reviewboard setup).


Thanks,

Thomas Friedrichsmeier

[rkward-devel] Review Request 121245: Change rkward-devel macport to build from git repo

[Thomas F. <tho...@ru...>]

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://git.reviewboard.kde.org/r/121245/
-----------------------------------------------------------

Review request for Rkward.


Repository: rkward


Description
-------

So we can shut down the svn repo on SF.net (actually not shutting it down, but marking it as out-of-date)


Diffs
-----

  macports/kde/rkward-devel/Portfile eb6f7a4 

Diff: https://git.reviewboard.kde.org/r/121245/diff/


Testing
-------

None. Which is the reason for this review request. (Well, another reason is that I wanted to test our reviewboard setup).


Thanks,

Thomas Friedrichsmeier

Re: [rkward-devel] feedback

[meik m. <mei...@un...>]

Am Sonntag, 23. November 2014, 21:14:16 schrieb Aaron Batty:
> K, I just tested. It turns out that it works okay in Windows, so I will now
> go yell at all those students who told me it didn't.
> 
> It does not, however, work on the Mac. Here is what happens when trying to
> drag something out of the output:
> 
> http://aaronbatty.net/media/RKWardDrag.png

just to make sure: do you know that you can export plots in various formats 
directly from the graphics window? this was *not* possible in previeous 
versions of RKWard on mac, only after thomas added the RK() device.

in most plot dialogs, you can use the preview checkbox to get the graphics 
device, and once it fits your needs, click "device" -> "export". you can then 
save your plot wherever you want and don't have to search hidden folders for 
that, and you're free to chose from various formats (PNG, PDF, SVG, EPS, even 
TikZ...).


viele grüßen :: m.eik

-- 
  dipl. psych. meik michalke
  institut f"ur experimentelle psychologie
  abt. f"ur diagnostik und differentielle psychologie
  heinrich-heine-universit"at d-40204 d"usseldorf