Sync guide (DSPLY and DCL-PI rules) and regenerate IBM i Compatibility
IBM i conformance results, run 35928181340
IBM i conformance results, run 35925174093
IBM i conformance results, run 35924791705
IBM i conformance results, run 35921577004
IBM i conformance results, run 35910236812
IBM i conformance results, run 35900619685
IBM i conformance results, run 35892882059
IBM i conformance results, run 35892263897
IBM i conformance results, run 35891517321
IBM i conformance results, run 35889870067
Regenerate: 61% agreement, plus KEYED and TAG rules Agreement 58% -> 61% (135 of 221) after the KEYED, TAG and SQL decimal-point fixes. Portability notes gain two rules that cost real debugging to find: KEYED is a free-form DCL-F keyword and belongs in position 34 as 'K' in fixed format, and TAG declares its label in Factor 1 while GOTO names its target in Factor 2 — mirror images, easy to get backwards.
Record formats: state the display-file exception Database access names the file, display access names the format. The previous wording implied OpenRPG has no record formats at all, which is wrong — EXFMT names one, and a display file genuinely is a set of named formats.
Note record formats as a platform difference Record-level access on externally-described files names the record format on IBM i and the file in OpenRPG, because record formats exist only where DB2 tables are native file objects. Readers writing portable source need to know that up front.
Regenerate from scripts/conformance-summary.py The page is now produced whole by the generator — statistics and portability guidance alike — so it cannot drift from the data it describes. Edit the script, not this page.
Report compatibility as an agreement rate Was a rejection count, which read as a defect count. Now leads with the share of sources where OpenRPG and IBM's compiler reach the same verdict - counting a shared rejection as the success it is, since it means OpenRPG refuses the same invalid RPG IBM does. Adds the full matrix and a breakdown of the disagreements, separating test-environment causes (missing IBM i objects, non-standalone fragments, SQL locale) from genuine language differences.
Drop the 'Running the check yourself' section The page is for readers who want to know how compatible OpenRPG is and what to avoid for portable source, not for running the verification harness — that is a maintainer workflow and lives in the repo.
Add IBM i Compatibility page Records how OpenRPG's test corpus fares under IBM's own ILE RPG compiler on a real IBM i 7.5, and — more usefully for readers — the portability rules that verification surfaced: fixed-format column adjustment, definition-type entries, /FREE indentation, **FREE in copy members, built-ins in traditional C-spec factors, and the DSPLY and object-name limits. Numbers are generated from ibmi-conformance-baseline.json rather than typed, and the framing makes clear that a rejection is not automatically a defect: 43 of the 121 are negative tests being correctly rejected.
Sync from docs/GUIDE.md: EXTPROC case-sensitivity fix
Sync from docs/GUIDE.md: Xcode CLT and ncurses dependency notes
Sync from docs/GUIDE.md: installers-first Installation chapter
Publish the OpenRPG User's Guide as one page per chapter Generated from docs/GUIDE.md by scripts/sync-wiki.py in the main repo.
Initial Home page