Re: Squiggly yellow lines

I understand that technically the parser is working properly; I just
offered a simple example of Terry's situation.

Anyway,

I can't find an option to turn the squigglies off other than to cripple the
software and revert back to the on-demand outline. Lacking an option to
"not display the squigglies", I wonder if the COLOR of the squigglies is
hard-coded as yellow? I can't seem to find a preference for coloring
whatever you call them other than squigglies (what are they called,
anyway?). If there were a preference, I could set the color to the same as
the background and I wouldn't have to see them anymore, but still would
have the yellow triangles and right-side blobs.

Stu






On Thu, Nov 7, 2013 at 9:51 AM, Edmund Reinhardt <
edmund.reinhardt@xxxxxxxxxx> wrote:

Thanks for the example Stuart,
I have worked through it and RDi is behaving properly in each instance.
Without a semicolon, the next line is interpreted as being a continuation
of the current incomplete one and therefore is not parsed correctly. In
the example below, the next line is a FOR statement, so its ENDFOR
statement is orphaned etc.
One improvement we have delivered in the meantime is to automatically
provide an ENDSL; as soon as you hit enter on the SELECT;

But the yellow squigglies are giving you feedback that your syntax is
broken and my tooling is not going to provide an accurate model until it is
fixed. If you were to run a compile you would likely get many more
messages. This is of course why you done try to compile on every
keystroke, but getting instant feedback, does tell you if you should even
attempt a compile because the code is incomplete.

This is how the most modern IDEs provide feedback on the syntactical
correctness of your code as you enter it. It may give you some incentive
to make sure that your source is roughly syntactically correct, in which
case our tooling can provide you better support.

I understand that this may be distracting, but as you get used to it, I
think you will find it useful immediate feedback. If you know you are only
part way through coding a statement, you can safely ignore them. But if
you think you are working with valid code and you see them, it is a very
useful flag, that you missed something.
As to the fact that the errors are further down in the code, I understand
that this would be confusing as it may not be obvious which syntax error is
causing the complaint at a later point in the source. A good strategy is to
fix the first line with an annotation first, because that fix may cause the
rest of the source to be understandable and all other annotations to
disappear. You can tell which line is the first offender by looking at the
top yellow indicator on the RIGHT-HAND side of the editor. The right-hand
side has annotations for errors on every line in the entire source. If you
click on them, it will take you to the offending line. The left-hand side
only has annotations for the lines that are currently visible.

In many ways this is not unlike fixing compile errors, except that this
feedback is immediate and only highlights MAJOR syntactic problems.

Happy coding,



Edmund (E.H.) Reinhardt
COBOL IDE on AIX, DDS, WebFacing, System i Application Development,
Rational Developer for Power



2D barcode Phone: 1-905-413-3125 | Phone: 1-905-854-6195
IBM
- encoded E-mail: edmund.reinhardt@xxxxxxxxxx
with AIX COBOL:
8200 Warden Ave
contact www.youtube.com/user/IBMRational#g/c/62DF24D5BCD43501
Markham, ON L6G 1C7
information "In every thing give thanks: for this is the will of
Canada
God in Christ Jesus concerning you." - 1
Thessalonians 5:18






From: Stuart Rowe <rowestu@xxxxxxxxx>
To: "Rational Developer for IBM i / Websphere Development Studio
Client for System i & iSeries" <wdsci-l@xxxxxxxxxxxx>,
Date: 07/11/2013 09:56 AM
Subject: Re: [WDSCI-L] Squiggly yellow lines
Sent by: wdsci-l-bounces@xxxxxxxxxxxx



Edmund,

I have witnessed the squigglies almost always extending well past the scope
of the unrecongnizable grammar. It is never limited to the "line" or
"statement" I am keying. Usually it wil stop at something concretely
recognizable, like the beginning of a procedure. I've gotten used to
ignoring them because usually the hover message on the yellow triangle (on
the left) is generally meaningless.

Example (nonsensical of course):

d x s 5i 0
d someData s 1a dim( 10 )

/free

for x = 1 to 10;

if someData( x ) = *blanks;
someData( x ) = 'X';
endif;

endfor;

return;

/end-free

Insert a line before the "for", and key just an "s" on it. See it light up
with yellow squiggles? Now put "elect" without a semi on the end. Lights
up differently. Now add the semi. Different again.
Now, add a line and put a "when" on it, no ending semi. Now add the
semicolon.
Now add a line and put "end" on it (not endsl) and see it light up well
down source as well.

Imagine a hundred lines of code between the error and the next recognizable
syntax -- it gets distracting as Terry pointed out.

Stu




On Wed, Nov 6, 2013 at 4:21 PM, Edmund Reinhardt <
edmund.reinhardt@xxxxxxxxxx> wrote:

Hi Terry,
As of the live RPG support in 8.5, RDi is now building a model of your

RPG

with every key stroke and maintaining the outline view among other

things.

If we are unable to build this model due to a severe RPG syntax error
(severe enough that our grammar gets confused and is unable to recognize
the rest of the file) then this is highlighted with an annotation

(which

does show up as the triangle, the exact syntax that is confusing the
grammar is highlighted with the squiggly underline. This is similar to
the behaviour you get with the best editing tools in Eclipse.
You could avoid this behaviour by going to the static outline, but I
sincerely recommend learning to take advantage of the behaviour. This

live

model is going to be the basis of a lot of great tooling we want to

provide

you.
If it is complaining about valid RPG, then this is something I want to

know

about right away, because this will mean that all of the tooling on top

of

the live model will not work beyond that point in the file.


Regards,



Edmund (E.H.) Reinhardt
COBOL IDE on AIX, DDS, WebFacing, System i Application Development,
Rational Developer for Power



2D barcode Phone: 1-905-413-3125 | Phone: 1-905-854-6195
IBM
- encoded E-mail: edmund.reinhardt@xxxxxxxxxx
with AIX COBOL:
8200 Warden Ave
contact www.youtube.com/user/IBMRational#g/c/62DF24D5BCD43501
Markham, ON L6G 1C7
information "In every thing give thanks: for this is the will of
Canada
God in Christ Jesus concerning you." - 1
Thessalonians 5:18






From: Terry Hertel <T.Hertel@xxxxxxxxxxxxxxxxx>
To: "'wdsci-l@xxxxxxxxxxxx'" <wdsci-l@xxxxxxxxxxxx>,
Date: 06/11/2013 04:34 PM
Subject: [WDSCI-L] Squiggly yellow lines
Sent by: wdsci-l-bounces@xxxxxxxxxxxx



I am using RDi 9.0.0.1 and just made my first coding error (tongue in

cheek

of course). Seriously, while I was entering a new select group, the

editor

added yellow squiggly lines to the uncompleted lines as well as adding a
yellow triangle to the left of the line number on some of the lines. I
assumed it was because I hadn't entered the ENDSL line yet. I wasn't sure
how I felt about this at first but then decided I kind of liked it until

I

found that the yellow squiggly lines and yellow triangles are sometimes
placed on lines well beyond where the error occurred; lines that had no
issues. First, is this the expected behavior in RDi? Second, how does RDi
decide where to start adding the yellow squigglies and where to quit?
Third, is there a preference to turn them off if the lines that aren't in
error become distracting?
--
This is the Rational Developer for IBM i / Websphere Development Studio
Client for System i & iSeries (WDSCI-L) mailing list
To post a message email: WDSCI-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/wdsci-l
or email: WDSCI-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/wdsci-l.


--
This is the Rational Developer for IBM i / Websphere Development Studio
Client for System i & iSeries (WDSCI-L) mailing list
To post a message email: WDSCI-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/wdsci-l
or email: WDSCI-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/wdsci-l.

--
This is the Rational Developer for IBM i / Websphere Development Studio
Client for System i & iSeries (WDSCI-L) mailing list
To post a message email: WDSCI-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/wdsci-l
or email: WDSCI-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/wdsci-l.


--
This is the Rational Developer for IBM i / Websphere Development Studio
Client for System i & iSeries (WDSCI-L) mailing list
To post a message email: WDSCI-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/wdsci-l
or email: WDSCI-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/wdsci-l.