Repository navigation
det: a RETURN / XCTL COMMAREA passes LENGTH bytes (Part of #4817, #4695) - #5085
Merged
Merged
Conversation
…of a LENGTH past the DTO is by design (Part of #4817, #4695) CicsTask keeps min(LENGTH, bytes) of the bytes passed beside the DTO for RETURN TRANSID, RETURN IMMEDIATE and XCTL (a LINK's area is by address, not cut). #4695: returnedBytes / returned(Dto) documented and the refusal names the Object answer. Register X51. Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
4 tasks done
Contributor
Owner
Author
|
No description provided. |
squid-protocol
marked this pull request as ready for review
October 14, 2026 02:50
squid-protocol
enabled auto-merge (squash)
October 14, 2026 02:50
Owner
Author
|
No description provided. |
squid-protocol
added a commit
that referenced
this pull request
Oct 14, 2026
…EA-LENGTH (#4711) Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
squid-protocol
added a commit
that referenced
this pull request
Oct 14, 2026
…alibration per runtime rule (Part of #4711) (#5086) * WIP cobolrt rule catalogue (#4711): catalogue, checker, markers, comment-insensitive runtime compare Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * WIP cobolrt rule catalogue (#4711): two-way test, rendered page, #5073's new refusal classified Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * cobolrt rule catalogue: the generated-docs publish list pin names the new page (#4711) Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> * cobolrt rule catalogue: #5085's layoutBytes under RT-CICS-XCTL-COMMAREA-LENGTH (#4711) Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Part of #4817, #4695.
A RETURN TRANSID / RETURN IMMEDIATE / XCTL COMMAREA passes LENGTH bytes (the next task's EIBCALEN). The det port keeps the record's whole span beside its DTO, so
CicsTasknow cuts those bytes at LENGTH (areaOf), as the DTO's reading already was. A LINK's area is passed by address and is not cut.#4695 decision: the facade's
handleTransactionisObject-typed, so a RETURN with LENGTH past the DTO answers itsbyte[]by design; documented, andreturned(Dto.class)refuses it by name with a message pointing toObject. No emission change (det/cics.py/det/program.pyare #5081's).Register X51 (ASSUMED). Test:
tests/cobol_mainframe/test_commarea_length_bytes_4817.py(fails without the fix).Gates:
pr_gates --fast6/6,--ratchets7/7, nothing skipped. No golden or ledger movement.Cross-repo: none.
🤖 Generated with Claude Code