9:52 a.m.
Hi,
there were a few accumulated changes because of my laziness, and I have
a couple of questions:
* "Wrote gfn file as\n" -- "wrote" as synonym for "saved"?
(Perhaps
using "saved" also uniformly in English then?)
* Does it exceed the length in context to translate "FEVD" as "FEVD
(Prognosefehlervarianzzerlegung)", which by the way is a nice example of
German composita words...?
* And similar for "IRF" -> "IRF (Impulsantwortfunktion)"
* line 16486: "Full" meaning complete, total, no space left...?
* In "eigenvalues of X'X, largest to smallest", there is an isolated
single quote. I remember once that I had introduced something like that
in the German translation ("Gauß'sche") which made German gretl crash?
Is this safe here now?
* "note: variance proportions columns sum to 1.0" - "note" as a noun
or
the imperative of the verb? (I guess both mean the same thing though...)
* "%s: invalid option argument": should it rather be "optional", or
what
does this mean here (line 6206)
* line 8673: "filter": noun or verb?
* line 17450: "condition index" -- context? I'm clueless.
At the risk of being redundant, let me remind my fellow translators of
this multilingual statistical glossary:
http://isi.cbs.nl/glossary/
thanks,
sven
10:17 a.m.
Am 24.11.2015 um 10:52 schrieb Sven Schreiber:
* In "eigenvalues of X'X, largest to smallest", there is an isolated
single quote. I remember once that I had introduced something like that
in the German translation ("Gauß'sche") which made German gretl crash?
Is this safe here now?
Well I tested it locally, and it seems to be ok.
One other thing: I noticed some SVAR-related translated strings, but in
the menu Model/Time series the entry "Structural VAR" is apparently
still English-only.
thanks,
sven
10:24 a.m.
Am 24.11.2015 um 11:17 schrieb Sven Schreiber:
Well I tested it locally, and it seems to be ok.
Sorry for populating this thread all by myself, but my first
git-commit-push ever is failing. I have read
http://ricardo.ecn.wfu.edu/~cottrell/gretl-git-basics.html#commit, but
git tells me "fatal: could not read from remote repository. / Make sure
you have access rights" etc.
I would guess that the step "export CVS_RSH=ssh" has to be replaced with
something git-related, or is this ssh key thing completely history now?
(What I did was: edit de.po, do "git commit po/de.po" apparently
successfully, build a test version of gretl locally, and then try to do
"git push".)
thanks,
sven
10:41 a.m.
Are you sure you cloned the rw repository?
On Tue, Nov 24, 2015 at 10:24 AM, Sven Schreiber <svetosch(a)gmx.net> wrote:
Am 24.11.2015 um 11:17 schrieb Sven Schreiber:
Well I tested it locally, and it seems to be ok.
>
>
Sorry for populating this thread all by myself, but my first
git-commit-push ever is failing. I have read
http://ricardo.ecn.wfu.edu/~cottrell/gretl-git-basics.html#commit, but
git tells me "fatal: could not read from remote repository. / Make sure you
have access rights" etc.
I would guess that the step "export CVS_RSH=ssh" has to be replaced with
something git-related, or is this ssh key thing completely history now?
Yes that environment var is not for Git. You should see "git config"
(and
there is some config file). But the typical username + password still
should be possible.
(What I did was: edit de.po, do "git commit po/de.po" apparently
successfully, build a test version of gretl locally, and then try to do
"git push".)
Before the commit you should have done a "git add -i" (-i for interactive).
12:02 p.m.
Am 24.11.2015 um 11:41 schrieb Hélio Guilherme:
Are you sure you cloned the rw repository?
No not sure -- it's been a while since I cloned it. "git pull" worked
though, but that is probably unrelated.
I would guess that the step "export CVS_RSH=ssh" has to be replaced
with something git-related, or is this ssh key thing completely
history now?
Yes that environment var is not for Git. You should see "git config"
(and there is some config file). But the typical username + password
still should be possible.
How exactly would the username + password work, at what stage, which
commands?
(What I did was: edit de.po, do "git commit po/de.po" apparently
successfully, build a test version of gretl locally, and then try to
do "git push".)
Before the commit you should have done a "git add -i" (-i for interactive).
Sure? The commit stage itself seems to have worked.
Thanks for helping,
sven
12:28 p.m.
On Tue, Nov 24, 2015 at 12:02 PM, Sven Schreiber <svetosch(a)gmx.net> wrote:
Am 24.11.2015 um 11:41 schrieb Hélio Guilherme:
> Are you sure you cloned the rw repository?
>
No not sure -- it's been a while since I cloned it. "git pull" worked
though, but that is probably unrelated.
Please note that I was using git for other projects before gretl
migrating. This
sourceforge+git is a bit different (we are pushing to our
own repository and then it is merged in master official repository).
You may change the repository in the config file to have the correct RW
URL. Or a safer way is to rename the your current local gretl-git directory
to free its name and do the cloning with the RW URL.
How exactly would the username + password work, at what stage, which
commands?
> This would happen when you want to push the changes. (Not really sure if
still
possible)
In normal git use, I do not have writing permissions on the master project,
so I must push to my forked repository (which I own), and then create a
Pull Request (PR) to the master project. They will reject or accept my PR,
and they do a merge if so.
This is the difference with sourceforge+git, because this PR and merge is
automatically (and transparent). Maybe Allin or Jack, have a better
visibility of this.
> (What I did was: edit de.po, do "git commit
po/de.po" apparently
> successfully, build a test version of gretl locally, and then try to
> do "git push".)
>
> Before the commit you should have done a "git add -i" (-i for
> interactive).
>
>
Sure? The commit stage itself seems to have worked.
No, it didn't. Looking at the po changes in
http://sourceforge.net/p/gretl/git/ci/master/tree/
I see my name and last commit 2 days ago.
(actually my HelioGuilherme66 is my name in Github, but the link goes to
the SourceForge user)
I don't remember if we must have a Github account, and if you must generate
a SSH key from the account Github profile page. I have done this before for
Linux (manually) and Windows is much easier (installing git tools + git
shell) it does the key generation for you.
12:33 p.m.
Am 24.11.2015 um 13:28 schrieb Hélio Guilherme:
On Tue, Nov 24, 2015 at 12:02 PM, Sven Schreiber <svetosch(a)gmx.net
<mailto:svetosch@gmx.net>> wrote:
Sure? The commit stage itself seems to have worked.
No, it didn't. Looking at the po changes in
http://sourceforge.net/p/gretl/git/ci/master/tree/
I see my name and last commit 2 days ago.
When I say the "commit stage" I mean the "git commit" which according
to
Allin's descriptions just commits to my local repo. The meaning of
"commit" thus would be different than in the old days (and the old
meaning would now be "git push"). I am aware that I have not managed to
upload anything on the server repo.
But perhaps I have some deeper misunderstanding than I thought...
thanks,
sven
1:11 p.m.
El 24/11/15 a las 13:33, Sven Schreiber escribió:
Am 24.11.2015 um 13:28 schrieb Hélio Guilherme:
>
>
> On Tue, Nov 24, 2015 at 12:02 PM, Sven Schreiber <svetosch(a)gmx.net
> <mailto:svetosch@gmx.net>> wrote:
>
> Sure? The commit stage itself seems to have worked.
>
>
> No, it didn't. Looking at the po changes in
> http://sourceforge.net/p/gretl/git/ci/master/tree/
> I see my name and last commit 2 days ago.
>
When I say the "commit stage" I mean the "git commit" which according
to Allin's descriptions just commits to my local repo. The meaning of
"commit" thus would be different than in the old days (and the old
meaning would now be "git push"). I am aware that I have not managed
to upload anything on the server repo.
But perhaps I have some deeper misunderstanding than I thought...
Read http://ricardo.ecn.wfu.edu/~cottrell/gretl-git-basics.html, again.
Depending on how you clone the gretl tree, the access is read-only or
read-write.
In Linux, if you execute
git clone ssh://USERNAME@git.code.sf.net/p/gretl/git gretl-git
(you will need to give your SF password) the gretl-git directory will be
defined as rw, including a directory '.git' with a 'config' file which
contains the line
url = ssh://USERNAME@git.code.sf.net/p/gretl/git
and then you can use a 'git push' command in this 'gretl-git' directory.
--
Ignacio Díaz-Emparanza
Departamento de Economía Aplicada III (Econometría y Estadística)
Universidad del País Vasco - Euskalherriko Unibertsitatea, UPV/EHU
Tfno: (+34) 94 601 3732
http://www.ehu.eus/ea3
3:05 p.m.
Am 24.11.2015 um 14:11 schrieb Ignacio Diaz-Emparanza:
El 24/11/15 a las 13:33, Sven Schreiber escribió:
Read http://ricardo.ecn.wfu.edu/~cottrell/gretl-git-basics.html, again.
Depending on how you clone the gretl tree, the access is read-only or
read-write.
Ah ok, that must be it. It feels kind of awkward that the decision to
possibly have write access later on must be made as the very first
thing, but well, so be it.
thanks (also to Allin),
sven
1:17 p.m.
On Tue, 24 Nov 2015, Sven Schreiber wrote:
But perhaps I have some deeper misunderstanding than I thought...
It really should be very straightforward, provided the repository is
"cloned" in the right way to allow read-write access. As per
http://ricardo.ecn.wfu.edu/~cottrell/gretl-git-basics.html .
If you're unsure of what you did at first, try a fresh cloning
with
git clone ssh://USERNAME@git.code.sf.net/p/gretl/git gretl-git
using your sourceforge USERNAME. Then ensure that your username and
email address are set correctly for the gretl context, using "git
config user.name" and "git config user.email".
Allin
2:20 p.m.
On Tue, 24 Nov 2015, Sven Schreiber wrote:
Sorry for populating this thread all by myself, but my first
git-commit-push
ever is failing. [...]
I would guess that the step "export CVS_RSH=ssh" has to be replaced with
something git-related, or is this ssh key thing completely history now?
Just to clarify this. The CVS_RSH environment variable is irrelevant
when using git, and no analogous environment variable is required.
However, the "ssh key thing" is not history; your git changes are
uploaded via ssh. That mechanism gets baked into your local copy when
you clone the repository the right way for write access, as in
git clone ssh://<whatever>
Allin
2:05 p.m.
On Tue, 24 Nov 2015, Sven Schreiber wrote:
Hi,
there were a few accumulated changes because of my laziness, and I have
a couple of questions:
* "Wrote gfn file as\n" -- "wrote" as synonym for "saved"?
(Perhaps
using "saved" also uniformly in English then?)
Yes, wrote == saved.
* Does it exceed the length in context to translate "FEVD"
as "FEVD
(Prognosefehlervarianzzerlegung)", which by the way is a nice example of
German composita words...?
* And similar for "IRF" -> "IRF (Impulsantwortfunktion)"
Those are SVAR strings, you'd need to see what context they arise in.
* line 16486: "Full" meaning complete, total, no space
left...?
Another SVAR string.
* In "eigenvalues of X'X, largest to smallest", there
is an isolated
single quote. I remember once that I had introduced something like that
in the German translation ("Gauß'sche") which made German gretl crash?
Is this safe here now?
Should be OK.
* "note: variance proportions columns sum to 1.0" -
"note" as a noun or
the imperative of the verb? (I guess both mean the same thing though...)
I'd interpret it as imperative but it could go either way.
* "%s: invalid option argument": should it rather be
"optional", or what
does this mean here (line 6206)
It refers to an argument that attaches to an option flag, as in
--output=display.
* line 8673: "filter": noun or verb?
Ah, verb. That's the label for the button you use to filter the
content of the function-package listing by JEL code.
* line 17450: "condition index" -- context? I'm
clueless.
Context is Belsley-Kuh-Welsch collinearity analysis.
As a general point, note that gretl.pot gives you (C) file and line
number for each translatable string. A glance at the relevant C file
will often (though admittedly not always) give you a good clue as to
the context.
Allin
3:13 p.m.
Am 24.11.2015 um 15:05 schrieb Allin Cottrell:
On Tue, 24 Nov 2015, Sven Schreiber wrote:
>
> * And similar for "IRF" -> "IRF (Impulsantwortfunktion)"
Those are SVAR strings, you'd need to see what context they arise in.
> * line 16486: "Full" meaning complete, total, no space left...?
Another SVAR string.
Right -- then I guess "Full" is from the bias correction choice.
> * "%s: invalid option argument": should it rather be "optional",
or
> what does this mean here (line 6206)
It refers to an argument that attaches to an option flag, as in
--output=display.
> * line 8673: "filter": noun or verb?
Ah, verb. That's the label for the button you use to filter the content
of the function-package listing by JEL code.
> * line 17450: "condition index" -- context? I'm clueless.
Context is Belsley-Kuh-Welsch collinearity analysis.
ok, thanks.
As a general point, note that gretl.pot gives you (C) file and line
number for each translatable string. A glance at the relevant C file
will often (though admittedly not always) give you a good clue as to the
context.
Yes, in principle I know that. However with the po tools that I have
tried so far, somehow I haven't managed to get an automatic lookup
working. And in the past when I looked manually it indeed seemed to be
the case that it didn't really help. (However, with the SVAR strings
above it would have been different, admittedly.)
thanks,
sven
3:30 p.m.
On Tue, 24 Nov 2015, Allin Cottrell wrote:
Yes, wrote == saved.
> * Does it exceed the length in context to translate "FEVD" as "FEVD
> (Prognosefehlervarianzzerlegung)", which by the way is a nice example of
> German composita words...?
>
> * And similar for "IRF" -> "IRF (Impulsantwortfunktion)"
Is there some excuse for using the word
"Donaudampfschiffahrtsgesellschaftskapitänskajütenschlüssel" in gretl?
Those are SVAR strings, you'd need to see what context they arise
in.
> * line 16486: "Full" meaning complete, total, no space left...?
Another SVAR string.
As in "Full bias correction"
As a general point, note that gretl.pot gives you (C) file and line
number
for each translatable string. A glance at the relevant C file will often
(though admittedly not always) give you a good clue as to the context.
And/or other po files; sometimes I take a look at es.po and fr.po for
inspiration.
-------------------------------------------------------
Riccardo (Jack) Lucchetti
Dipartimento di Scienze Economiche e Sociali (DiSES)
Università Politecnica delle Marche
(formerly known as Università di Ancona)
r.lucchetti(a)univpm.it
http://www2.econ.univpm.it/servizi/hpp/lucchetti
-------------------------------------------------------
3:43 p.m.
This is a very educative conversation:
Look at https://en.wikipedia.org/wiki/Donaudampfschiffahrtsgesellschaft
And from there:
See also[edit
<https://en.wikipedia.org/w/index.php?title=Donaudampfschiffahrtsgesellsch...
]
-
Donaudampfschifffahrtselektrizitätenhauptbetriebswerkbauunterbeamtengesellschaft
<https://en.wikipedia.org/wiki/Donaudampfschifffahrtselektrizit%C3%A4tenha...;,
a notoriously long word, of similar origins.
(You must take a deep breath before saying that word :D )
And/or other po files; sometimes I take a look at es.po and fr.po
for
inspiration.
And I look at it.po (is more similar to PT than ES ).
9:44 a.m.
Am 24.11.2015 um 16:43 schrieb Hélio Guilherme:
This is a very educative conversation:
Look at https://en.wikipedia.org/wiki/Donaudampfschiffahrtsgesellschaft
And from there:
See also[edit
<https://en.wikipedia.org/w/index.php?title=Donaudampfschiffahrtsgesellsch...>]
* Donaudampfschifffahrtselektrizitätenhauptbetriebswerkbauunterbeamtengesellschaft
<https://en.wikipedia.org/wiki/Donaudampfschifffahrtselektrizit%C3%A4tenha...;,
a notoriously long word, of similar origins.
(You must take a deep breath before saying that word :D )
Yeah, and BTW the spelling reform (20 years already !?) not only
introduced the triple-fff (or other triple-consonants in certain
circumstances), but also allowed more hyphens to split up these long
words, so that you can read them better...
well enough education already I guess,
sven
9:38 p.m.
Am 24.11.2015 um 15:05 schrieb Allin Cottrell:
On Tue, 24 Nov 2015, Sven Schreiber wrote:
> * And similar for "IRF" -> "IRF
(Impulsantwortfunktion)"
Those are SVAR strings, you'd need to see what context they arise in.
> * line 16486: "Full" meaning complete, total, no space left...?
Another SVAR string.
Ok somehow git and I aren't getting along too well. I have now cloned
the git repo again, this time with read-write access (I hope). Then I
have moved the updated de.po file in there, and I have done "git commit
de.po" on it. Indeed git later tells me at some occasion that my local
copy is 1 commit ahead of master, which sounds right.
But then I try to test for example the long translated strings in the
SVAR addon, so I configure (with --enable-build-addon) and build gretl
from that local copy. Then I do "sudo make install" as always, and start
gretl. But inside that newly built gretl, all the SVAR-related strings
(including the term "Structural VARs" itself) are still untranslated.
What do I have to do to get the translated strings of SVAR to appear?
What am I doing wrong? I believe (but not 100% sure) that there are some
other new strings that didn't exist (in the core gretl program, not an
addon) which appear translated alright.
thanks,
sven
1:31 a.m.
On Tue, 1 Dec 2015, Sven Schreiber wrote:
Am 24.11.2015 um 15:05 schrieb Allin Cottrell:
> On Tue, 24 Nov 2015, Sven Schreiber wrote:
>> * And similar for "IRF" -> "IRF (Impulsantwortfunktion)"
>
> Those are SVAR strings, you'd need to see what context they arise in.
>
>> * line 16486: "Full" meaning complete, total, no space left...?
>
> Another SVAR string.
Ok somehow git and I aren't getting along too well. I have now cloned the git
repo again, this time with read-write access (I hope). Then I have moved the
updated de.po file in there, and I have done "git commit de.po" on it. Indeed
git later tells me at some occasion that my local copy is 1 commit ahead of
master, which sounds right.
But then I try to test for example the long translated strings in the SVAR
addon, so I configure (with --enable-build-addon) and build gretl from that
local copy. Then I do "sudo make install" as always, and start gretl. But
inside that newly built gretl, all the SVAR-related strings (including the
term "Structural VARs" itself) are still untranslated.
What do I have to do to get the translated strings of SVAR to appear? What am
I doing wrong? I believe (but not 100% sure) that there are some other new
strings that didn't exist (in the core gretl program, not an addon) which
appear translated alright.
Could you cite an example or two of SVAR strings that are aren't
appearing in translation after your changes to de.po? The location
of the addons within the main source tree is new with git, and there
are a few complicating factors, but it seems to me this ought to
work.
Allin
7:48 a.m.
Am 02.12.2015 um 02:31 schrieb Allin Cottrell:
On Tue, 1 Dec 2015, Sven Schreiber wrote:
> Am 24.11.2015 um 15:05 schrieb Allin Cottrell:
>> On Tue, 24 Nov 2015, Sven Schreiber wrote:
>
>>> * And similar for "IRF" -> "IRF
(Impulsantwortfunktion)"
>>
>> Those are SVAR strings, you'd need to see what context they arise in.
>>
>>> * line 16486: "Full" meaning complete, total, no space left...?
>>
>> Another SVAR string.
>
> Ok somehow git and I aren't getting along too well. I have now cloned
> the git repo again, this time with read-write access (I hope). Then I
> have moved the updated de.po file in there, and I have done "git
> commit de.po" on it. Indeed git later tells me at some occasion that
> my local copy is 1 commit ahead of master, which sounds right.
>
Could you cite an example or two of SVAR strings that are aren't
appearing in translation after your changes to de.po? The location of
the addons within the main source tree is new with git, and there are a
few complicating factors, but it seems to me this ought to work.
Well for example the strings mentioned above. Or the menu entry
originally named "Structural VARs" in Model -> Time Series. (At least I
thought that the string "Structural VARs" was for the menu entry.)
thanks,
sven
3317
days inactive
3325
days old
18 comments
5 participants
participants (5)
-
Allin Cottrell -
Hélio Guilherme -
Ignacio Diaz-Emparanza -
Riccardo (Jack) Lucchetti -
Sven Schreiber