Path: csiph.com!usenet.pasdenom.info!gegeweb.org!de-l.enfer-du-nord.net!feeder1.enfer-du-nord.net!feeds.phibee-telecom.net!newsfeed.xs4all.nl!newsfeed6.news.xs4all.nl!xs4all!post.news.xs4all.nl!not-for-mail
Newsgroups: comp.lang.python
Date: Mon, 9 Jul 2012 06:33:31 -0700 (PDT)
In-Reply-To: <mailman.1945.1341838446.4697.python-list@python.org>
Complaints-To: groups-abuse@google.com
Injection-Info: glegroupsg2000goo.googlegroups.com; posting-host=81.83.162.247; posting-account=9-In_woAAACDnDFyCeZyZytl1ttJ-5Gp
References: <c28a17b7-7d20-4ba1-9adc-cf5e479c3764@googlegroups.com> <mailman.1941.1341832027.4697.python-list@python.org> <2081b59a-40af-4fc8-a9c2-6d76b2d61e96@googlegroups.com> <mailman.1945.1341838446.4697.python-list@python.org>
User-Agent: G2/1.0
MIME-Version: 1.0
Subject: Re: Deditor 0.4.0
From: Kruptein <darragh.ssa@gmail.com>
To: comp.lang.python@googlegroups.com
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Cc: Kruptein <darragh.ssa@gmail.com>, python-list@python.org
Precedence: list
Message-ID: <mailman.1947.1341840815.4697.python-list@python.org>
Lines: 146
NNTP-Posting-Host: 2001:888:2000:d::a6
Xref: csiph.com comp.lang.python:25083

Op maandag 9 juli 2012 14:54:03 UTC+2 schreef Jean-Michel Pichavant het vol=
gende:
> Kruptein wrote:
> > Op maandag 9 juli 2012 13:05:58 UTC+2 schreef Jean-Michel Pichavant het=
 volgende:
> >  =20
> >> Kruptein wrote:
> >>    =20
> >>> Hey I released a new version of my python-focused text-editor.
> >>> you can download it at http://launchpad.net/deditor
> >>>
> >>> What is it?
> >>> Deditor is aimed to be a text-editor which can be used as a basic tex=
t-editor as gedit or with the right plugins become a full-feature ide.
> >>> I focus on making it a python specific editor but you can use it to e=
dit other files as well.
> >>> It is python specific because the plugin system that is used (DPlug) =
is written in python and thus all plugins are written in python.  Also some=
 of the plugins there is one plugin that comes bundled with deditor that ad=
ds some python specific features like code-analyzing, code-inspection, auto=
-completion etc.
> >>>
> >>> So plugins are the main core of the program you can disable and enabl=
e plugins at your will.  The plugins that come bundled are made to increase=
 the joy of programming.
> >>>
> >>> NOTICE that this I don't regard this as a stable product yet.  There =
is no high risk in data loss and there are definitely no privacy concerns b=
ut you should still be aware that there are bugs/malfunctions and that you =
will encounter them.
> >>>
> >>> I hope some of you like it and if you don't try to give some construc=
tive criticism :)
> >>>
> >>>
> >>> Most changes in regard to previous release are under the hood.  so ex=
isting users might not notice a lot.
> >>> Also the customize plugin's second configuration tab: syntax highligh=
ting is bugged try to avoid it :p
> >>>  =20
> >>>      =20
> >> Good job.
> >>
> >> I have 2 question:
> >> Aren't you reinventing the wheel ?
> >> No way to use tabs for tabulation ??? (I'm CEO of the 'space haters cl=
ub')
> >>
> >> Otherwise, I found this bug :
> >> Start the editor, click twice on 'Analyse code', and you'll get a trac=
eback.
> >> I looked into the related code:
> >>
> >> def dUpdate(self, trig, value=3DNone):
> >>     if trig =3D=3D "tabclosed":
> >>         return self.tabClosed(value)
> >>
> >> def tabClosed(self, tab):
> >>     """Checks if the tab that is closing is a project"""
> >>         if tab.parentTab =3D=3D None:
> >>             pass
> >>
> >>
> >> Your method interfaces are quite inconsistent, I sense this is quite=
=20
> >> global to your code:
> >>
> >> 1/ dUpdate has None as default value for 'value'
> >> 2/ tab.parentTab will fail if tab is None
> >> 3/ on the traceback you get from the bug above, a bool has been given,=
=20
> >> bool has no parentTab.
> >>
> >>
> >> Here are 2 advices:
> >>
> >> 1/ Default values are for the fool/weak/noob/..., your pick. I tend to=
=20
> >> use them only for backward compatibilty or for interface stability.=20
> >> Otherwise, it's better not to use them. Some people still use them=20
> >> appropriately but in your I think it may help.
> >>
> >> 2/ In your docstrings, note down the expected type of your parameters=
=20
> >> (only 1 type ! (except for the None alternative)) and stick to it.
> >>
> >> Regards,
> >>
> >> JM
> >>    =20
> >
> > To begin with not everything that I want to see is yet implemented.  A =
lot still has to be done for example indentation although I really hate tab=
s ;D I will give the option in a future release maybe even the next one.
> >
> > I'll check the bug as it probably is indeed a bug.
> >
> > Regarding the default value dUpdate is a method from the plugin system =
and is called when a trigger is updated.  some triggers give an additional =
value for example a trigger that registers pressed buttons could potentiall=
y give which exact key as value.  However some triggers don't need this add=
itional value.  If a plugin needs both types of triggers the method will ne=
ed this default value to work.  Also the dev version already changed that t=
here always will be sent a value so no need for the default value anymore.
> >
> >  =20
> If you can, use the **kwargs syntax.
> def foo(tabs, **kwargs):
>     # optional arg
>     if kwargs.get('saveFile', False):
>        self.saveFile(tab.file)
>     # required arg
>     if kwargs['saveFile']:
>         self.saveFile()
>   =20
>=20
> The type of kwargs will be consistent, it will be always a dictionnary,=
=20
> possibly an empty one, but still a dictionnary, and best part of it, if=
=20
> you change the number of argument in the call, you don't break your=20
> interface.
>=20
> > Regarding the docstrings: as deditor still is under heavy development n=
ot all docstrings are correct or even there.
> >
> >  =20
> Yep, but writing the expected type of your parameters always help, in=20
> particular during the dev phase. you can save the verbose part of the=20
> documentation for later.
>=20
> Cheers,
>=20
> JM

Yeah I started to use kwargs in some methods but it's only usefull in some =
cases imo.


I'm a pretty decent python dev imo but I know that there is still a lot tha=
t I've to learn so often if I get to know something knew I tend to rewrite =
a large portion of my program(s) and thus I often am too lazy to write prop=
er docstrings as they would change anyway (<- very bad coding practice thou=
gh ^^)

Thanks for the feedback anyways :)