Re: Multiple issues with CL Source editing

On 9/5/2017 11:24 AM, Duane Scott wrote:

First, I'm editing a CL source member in 9.5.1.2 on a V7R1M0. The source file type is PF38-SRC. (I recognize that could be part of the problem, but I hope to deal with that later.) The record length of the file is 92. When editing a member with a type of CLP, the RDi formats lines beyond the 80 character limit. The parser association for CLP is CL. Additionally, as expected, when the member is saved with lines over the 80 character limit, the member is saved with lines truncated at the 80th column. Yet when selecting (again) for editing the full script displays as if it were not truncated. Preferences show that the Line Length for "CL" is set to "*RCDLEN". When changed to "80", the editor properly reformats to the 80th column, but one would think that "*RCDLEN" would reformat to the proper 80th column.

I've had the CL formatter turned off for ages, but I turned it on to
test this out. Here's what I did:

Used an existing PF38-SRC file called QCLSRC, 92 recl
Used RDi 9.5.1.2 to add a new member BUCKCL
PGM
CPYTOIMPF.....150 columns long
ENDPGM
Ctrl-S to save
Receive message 'Line truncation occurred during save.'
With RDi still open, DSPPFM QCLSRC BUCKCL
See that indeed, the long CPYTOIMPF is truncated
F3 back to the command line
In RDi, close the member
In RDi, re-edit the member by clicking on the name in RSE

1) I do NOT see the chopped off end of the CPYTOIMPF line - it /is/
truncated.

2) The formatter seems to work as expected. I paste in the very long
line, move the cursor and the formatter makes it all fit into 80 columns.

Are you editing using IBM i Projects?

So, second, how does one clear the cache for the text on a member? I've "cleared cache" a few times, closed RDI, and every time I call up the source, all the data is there. A compile correctly lists errors as if the lines were truncated, but still the line shows as if the source were all there.

Terminology will kill us all yet. By 'text' I'm assuming you mean
'contents of the member'. There isn't a menu option to clear the cache
of actual source members, but you can do it yourself:

Close RDi
Using Windows File Explorer, drill down into the workspace:
RemoteSystemsTempFiles
IBM.i.machine.name
QSYS.LIB
YOURLIB.LIB
QCLSRC.FILE

The member will be in there.

Third is the problem of what to set as the record length for the CL Source file. Since the compiler will accept 100 column commands, and the settings on RDi are "*RCDLEN", "80" and "100", one would think that the new record length for CL Source Files should be 112. But record length for CL is not specifically discussed in more recent issues of manuals that I could find. RPG has quite a few discussions of longer settings being available. I'm not sure if we should stick to 80, move to 100, or go beyond. The fact that some of our programmers and operations staff still use SEU is of issue, but I'd like to hear some other viewpoints.

SEU handles 100 just fine.