Docs: cross-link digits and substring function docs

[groeneai]

Changelog category (leave one):

• Documentation (changelog entry is not required)

Changelog entry (a user-readable short description of the changes that goes into CHANGELOG.md):

...

Description

Adds reciprocal "See also" cross-links between the in-source docs of digits and substring. digits(n, offset[, length]) is the numeric analogue of substring(s, offset[, length]), so each function's documentation now points to the other.

Requested by rschu1ze in #109012 (comment).

Docs-only, no behavior change. The links use the inline "See also" convention already used by scalar functions (for example initialQueryID -> other-functions#queryID, tokens -> string-functions#sparseGrams); FunctionDocumentation has no related_functions field.

Version info

• Merged into: 26.7.1.1302 (included in 26.7 and later)

[groeneai]

Pre-PR validation gate (click to expand)

#	Question	Answer
a	Deterministic repro?	N/A (docs-only, no failing behavior to reproduce)
b	Root cause explained?	N/A as a bug. Rationale: digits is the numeric analogue of substring; maintainer directive to cross-link their in-source docs
c	Fix matches root cause?	Yes. Reciprocal "See also" links are exactly the ask
d	Test intent preserved / new tests?	N/A. No tests changed. Doc-gating tests (02415/03921) validate doc presence, not prose; both functions stay documented
e	Demonstrated in both directions?	N/A (no behavior). Targeted compile of both changed TUs (digits.cpp.o, substring.cpp.o) => exit 0, proving the raw-string literals stay well-formed
f	General, not a narrow patch?	Reciprocal in BOTH directions (digits->substring and substring->digits) = full scope of the directive
g	Generalizes across inputs/wrappers?	N/A (no type-dependent code path). substringUTF8 intentionally not linked: the directive named substring only, so the link lives on the canonical byte-variant doc
h	Backward compatible?	Yes. Docs text only; no setting/format/wire change
i	Invariants/contracts preserved?	Yes. No code contract touched; added text contains no )" so raw-string literals remain valid

Session id: cron:clickhouse-author-slot-0:20260720-214300

[groeneai]

Internal second-model review (click to expand)

An independent second model reviewed this diff before it was opened.

• ❌ Blockers: none
• ⚠️ Majors: none
• 💡 Nits: none

Verdict: 0 findings. The change is a docs-only addition of two reciprocal
"See also" links inside existing FunctionDocumentation::Description string
literals; both changed translation units compile, and the guarded diff is
exactly the two intended files with no submodule drift.

[clickhouse-gh]

Workflow [PR], commit [31a71ec]

Summary: ✅

• Performance Comparison: Performance dashboard

---

AI Review

Summary

This PR adds reciprocal See also cross-links between the in-source docs for digits and substring. I found no review findings in the current head: the new links follow the same extensionless route style already used by other FunctionDocumentation entries, and the target pages/categories match the generated docs slugs (/sql-reference/functions/string-functions#substring and /sql-reference/functions/other-functions#digits). Residual risk is limited to docs rendering, which I did not rebuild locally.

Final Verdict

Status: ✅ Approve

[groeneai]

CI finish ledger - ae7c776

This is a docs-only change (2 files, +4 lines in FunctionDocumentation string literals in digits.cpp and substring.cpp). None of the failures below can be caused by the diff. Every failure has an owner.

Check / test	Reason	Owner / fixing PR
Integration tests (amd_asan_ubsan db-disk old-analyzer 3/6; amd_msan 7/8; amd_tsan 3/6) / test_replicated_database::test_replicated_table_structure_alter	chronic flaky (1212 hits / 539 PRs / 107 master in 30d)	#111029 (fixing PR, open)
Stateless (amd_llvm_coverage, ParallelReplicas, s3) / Logical error: Got read request from replica 1 for unknown stream ... (STID 5217-5b3d)	chronic trunk bug (33 hits / 31 PRs / 2 master in 30d)	a full-effort fix task owns it (root-caused: Default-mode announce-vs-request registration race in ParallelReplicasReadingCoordinator; blocked on the #109202 WONTFIX ruling, re-engaged with falsifying master evidence)
Stateless (amd_llvm_coverage, ParallelReplicas, s3) / 01271_show_privileges ERROR + Server died	crash victims of the STID 5217-5b3d abort above (same job/commit: "server is shutting down due to a fatal error (ABORTED)")	same as above
CH Inc sync	-	CH Inc sync (private, not actionable)

Session id: cron:our-pr-ci-monitor:20260721-093000

[clickhouse-gh]

LLVM Coverage Report

Metric	Baseline	Current	Δ
Lines	86.20%	86.20%	+0.00%
Functions	92.00%	92.00%	+0.00%
Branches	78.30%	78.30%	+0.00%

Changed lines: Changed C/C++ lines covered: 13/13 (100.00%) · Uncovered code

Full report · Diff report