+1
also "Update Javadoc" if javadoc is present and the method signature doesn't
matches it


"Igor Karpov" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> I think all ideas seem faily reasonable and worth implementing. Different
> people may like different approaches.
> My opinion is based on my experience, so that a user looks at the current
> method (remembering the context and what exactly what this method does),
and
> think "Well, this had better to be documented"..
> The cursor is positioned somewhere inside of the method, and it's annoying
to
> carefully position it (mouse, also).. The intention is "well, I want to
javadoc
> it" and the reasonable way of doing this is "alt+insert javadoc".
> I like the "todo" approach also, but it's differerent - more like a
reminder,
> and the cursor must be positioned to the javadoc body anyway.
>
> -----Original Message-----
> From: "Jacques Morel" <[EMAIL PROTECTED]>
> To: [EMAIL PROTECTED]
> Date: Thu, 18 Jul 2002 00:22:53 -0500
> Subject: Re: javadoc comments generating
>
> >
> > Paul,
> > Why being so aggressive? Can't we have a calm and respectful debate
here? Do
> > you want to turn into another Hani? I am not sure I can bear more than
one
> > ;-)
> >
> > My initial post was made because like others, I am concerned that
features I
> > do not think add any value, could be implemented.
> > However I think you are making a reasonable case here to make javadoc
> > writing easier. I am still not entirely sure how to do it best but your
> > suggestion do make it easier.
> >
> > However I would rather fix the positioning pre-requisite of javadoc
> > generation.
> > Why not having, like Igor Karpov proposed, an action to generate javadoc
of
> > the current member. It could be invoked from within the body through a
> > shortcut (not Alt-Insert though since it provide access to only class
level
> > actions) or from the structure view by selecting the to-be-documented
> > member(s) and invoking "Generate member javadoc" from the context menu.
> > Multi-selection would allow generation of the whole file.
> >
> > I like your idea to have TODO markers inserted BTW when you generate
more
> > than one member at once.
> >
> > Jacques
> >
> >
> > "Paul Bradshaw" <[EMAIL PROTECTED]> wrote in message
> > ah47v6$5f2$[EMAIL PROTECTED]">news:ah47v6$5f2$[EMAIL PROTECTED]...
> > > Did you read my message at all?  It's not just "/**", it's "position
> > cursor,
> > > [enter][up-arrow]/**[Enter], type stuff, then find next method.  All
I'm
> > > saying is that a global "add javadoc template with a todo" woudl make
it
> > so
> > > much easier -- click blue bar, add javadoc, click blue bar, add
javadoc...
> > I
> > > don't understand the hatred of something so simple and obvious (and
> > > something you don't have to use anyway, and which wouldn't add any
huge
> > > burden to IDEA if you didn't use it).  I'd also love it if it did
things
> > > like examine existing javadoc, and if a parameter was missing, add it
> > (with
> > > a TODO), and if a parameter was obsolete, delete it.  I'm just looking
for
> > > ways of making the managing of javadoc simpler by automating some of
the
> > > drugery.  Isn't that what IDEA is supposed to be about?
> > >
> > >
> > > "Thomas Singer" <[EMAIL PROTECTED]> wrote in message
> > > [EMAIL PROTECTED]">news:[EMAIL PROTECTED]...
> > > > Completely agree with Scott's words.
> > > >
> > > > Paul, since you need to take a look at the method itself to write
the
> > > > JavaDoc's content, what's the problem with /**?
> > > > Or do you just write useless space-wasters like:
> > > >
> > > >   /**
> > > >    * @param forAllUsers New value of property forAllUsers.
> > > >    */
> > > >   public void setForAllUsers(boolean forAllUsers) {
> > > >       this.forAllUsers = forAllUsers;
> > > >   }
> > > >
> > > >   or the killer-one:
> > > >
> > > >   /** Setter for property forAllUsers.
> > > >    * @param forAllUsers New value of property forAllUsers.
> > > >    */
> > > >   public void setForAllUsers(boolean forAllUsers) {
> > > >       this.forAllUsers = forAllUsers;
> > > >   }
> > > >
> > > > Tom
> > > >
> > > >
> > > > On Tue, 16 Jul 2002 13:19:00 -0500, "Scott Sirovy"
> > > > <[EMAIL PROTECTED]> wrote:
> > > >
> > > > > I add javadoc when it helps, not just to add size to my .java
files.
> > > > >
> > > > > The functionality you mention could be a plugin (that you can
install
> > > and I
> > > > > don't have to) or an external tool (that you configure and I do
not).
> > > > >
> > > > > Take a look at http://www.doclet.com and search for either DocWiz
or
> > > > > CommentMaster.  Feel free to configure an external tool for them.
> > This
> > > > > isn't something JetBrains should waste their time duplicating.
> > > > >
> > > > > -sms
> > > > >
> > > > > "Paul Bradshaw" <[EMAIL PROTECTED]> wrote in message
> > > > > ah1kd7$mh4$[EMAIL PROTECTED]">news:ah1kd7$mh4$[EMAIL PROTECTED]...
> > > > > > Well, it's more than just three characters.  I have to go to the
> > > method
> > > > > > name, press enter to "open up" a blank line in which to type the
> > > > > /**[ENTER],
> > > > > > and then up arrow back up to the blank line.  A grand total of
six
> > > > > > keystrokes NOT including the positioning.
> > > > > >
> > > > > > I'd LOVE to see an "auto generate javadoc stubs" function that
would
> > > plug
> > > > > in
> > > > > > a TODO comment at the top of each one.  Then I could generate
all
> > the
> > > > > > javadoc stubs for a class at once, and then immediately navigate
to
> > > each
> > > > > one
> > > > > > by clicking on the blue to-do bars.  Talk about a time and
energy
> > > saver...
> > > > > >
> > > > > > "Jacques Morel" <[EMAIL PROTECTED]> wrote in message
> > > > > > agv710$i87$[EMAIL PROTECTED]">news:agv710$i87$[EMAIL PROTECTED]...
> > > > > > > I humbly disagree. What would you gain by generating text that
> > would
> > > > > have
> > > > > > > being automatically inferred by the javadoc generator? The
batch
> > > > > > generation
> > > > > > > won't change how your html javadoc looks like, do we agree? So
why
> > > don't
> > > > > > you
> > > > > > > defer writing the javadoc shell of a member until you want to
> > write
> > > some
> > > > > > > real javadoc. At that point typing /** is not too much I
think. It
> > > is
> > > > > just
> > > > > > 3
> > > > > > > characters. In addition you will get the following benefits:
> > > > > > > 1) a guaranteed up-to-date javadoc of your member
> > > > > > > 2) less comment junk that makes your code more readable
> > > > > > > I think the previous benefits easily offset the burden of
typing 3
> > > > > > > characters for each members don't you think?
> > > > > > >
> > > > > > > Or do I miss something?
> > > > > > >
> > > > > > > Jacques
> > > > >
> > > > >
> > > >
> > >
> > >
> >
> >
> > _______________________________________________
> > Eap-features mailing list
> > [EMAIL PROTECTED]
> > http://lists.jetbrains.com/mailman/listinfo/eap-features
> >
>
>
> ------------------------------------
> Mail.Ru - ������, ��������, �������!
> ------------------------------------


_______________________________________________
Eap-features mailing list
[EMAIL PROTECTED]
http://lists.jetbrains.com/mailman/listinfo/eap-features

Reply via email to