Can you do a half screen scroll on the Edit Action Diagram screen o position the line to top of the screen ?
I don't believe so....or at least I've never seen a way to do it.
It used to be the 'recommendation' of 'consultants' that you try and work to a single page at a time. That would mean using emty case statements to provide nesting / commentary as needed, and expanding where you want to follow the storyline
so something like this
> USER: Validate fields
. ...--> F4 for company ?
. ...--> No company - go to select
. company number nam *FIELD
. Employment sequence nam *FIELD
. ** - - - - - - - - - - - -
. ...--> Start date check
. ...--> end date calcs
. ...--> todays date
. ...--> start date check
. ...--> ended date check to start date
. ...--> ended date check to today
Where the --> indicates (to me at least) that there is something worth expanding
if we z into the second of these we get
> --> No company - go to select
¦-DTL.Company number is blank
¦ 006 SEL Co & Pos by name - Company *
an 'empty' case statement gives you the chance to insert a bit of commentary, and not generate a subroutine, whilst encapsulating a lump of code, thus
. > --> example of a case statement
and you produce these by keying
ICF (enter) (f3) then
io (enter) then
fd (enter) and placing your text fragment. then the code.
the comment lines ** - - - - - are there purely to place markers inthe generated code delimiting a section of code - that 'date checking' one liner probably contains a huge amount of embedded/ generated code, which is nicely abstracted to that one line entry in the 2e code.
- not quite the answer to your direct question, but one method of engineering a way round it.
I been putting sequence inside my case's so I can zoom in on them. I seem sequence with hide/zoom is the best way I found.
Encapsulating /Abstracting code is good.
I used to insist on no empty case statements - 'Case, otherwise' being a confusing waste of space. (Sorry Arti) I've become converted though. Using a sequence generates as a subroutine, meaning a bit of CPU power used to store the address and branch/return, but more importantly, source which is highly fragmented should you ever need to look at it, or use debug on it.
Using a case to do the same job leaves you with the commentary line in the source code, and my foible for a --> (also stolen from a fellow contractor) enables a simple search of source code
scan for 'USER: ' to find user point
scan for '-->' to find code blocks of relevance.
and for especially exciting stuff, where a lot is happening, I add a rightmost <== to indicate the major drill down lines.
and for completeness, a compound case gets a header with ++> and a comment sequence is commented out (nothing generated) and given **>
so we get, one one page, a story, and SUBR's minimised. a scan of source will follow this page sequence of code. Using a different markef for 'top level code could be done, but hey, life is short. If you want to do ---> that's fine with me..
**> Notes in here
--> initialise values
--> Validate incoming parmeters
--> Take an action
--> Another action
--> major work is here <==
**> Change log