* Modernise messaging infrastructure with cli support
Rewrites message_(), warning_(), stop_() to use cli::cli_inform(),
cli::cli_warn(), and cli::cli_abort() when the cli package is available,
with a fully functional plain-text fallback for environments without cli.
Key changes:
- New cli_to_plain() helper converts cli inline markup ({.fun}, {.arg},
{.val}, {.field}, {.cls}, {.pkg}, {.href}, {.url}, etc.) to readable
plain-text equivalents for the non-cli fallback path
- word_wrap() simplified: drops add_fn, ANSI re-index algorithm, RStudio
link injection, and operator spacing hack; returns pasted input unchanged
when cli is available
- stop_() no longer references AMR_env$cli_abort; uses pkg_is_available()
directly; passes sys.call() objects to cli::cli_abort() call= argument
- Removed add_fn parameter from message_(), warning_(), and word_wrap()
- All call sites across R/ updated: add_fn arguments removed, some paste0-
based string construction converted to cli glue syntax ({.fun as.mo},
{.arg col_mo}, {n} results, etc.)
- cli already listed in Suggests; no DESCRIPTION dependency changes needed
https://claude.ai/code/session_01XHWLohiSTdZvCutwD7ag2b
* Replace {.fun} with {.help} for all exported functions in messaging
All function names referenced via {.fun …} in cli-style messages are
exported in NAMESPACE, so {.help …} is the appropriate markup — it
renders as a clickable help link rather than plain function styling.
https://claude.ai/code/session_01XHWLohiSTdZvCutwD7ag2b
* Qualify all {.help} tags with AMR:: and convert backtick ?func references
- Add AMR:: namespace prefix and trailing () to all {.help} cli markup
so they render as clickable help links (e.g. {.help AMR::as.sir}())
- Convert `?funcname` backtick-quoted help references to {.help AMR::funcname}()
in aa_helper_functions.R, custom_eucast_rules.R, interpretive_rules.R,
key_antimicrobials.R, mo.R, plotting.R, resistance_predict.R, and sir.R
- Skipped `?proportion` in sir_calc.R as 'proportion' is not exported
https://claude.ai/code/session_01XHWLohiSTdZvCutwD7ag2b
* Require cli >= 3.0.0 for cli_inform/cli_warn/cli_abort availability checks
cli_inform, cli_warn, and cli_abort were introduced in cli 3.0.0.
Add min_version = "3.0.0" (as character) to all four pkg_is_available("cli")
checks so older cli versions fall back to base R messaging.
https://claude.ai/code/session_01XHWLohiSTdZvCutwD7ag2b
* Implement cli::code_highlight() for R code examples in messages (issue #191)
Add highlight_code() helper that wraps cli::code_highlight() when cli >= 3.0.0
is available, falling back to plain code otherwise. Apply it to all inline
R code examples embedded in message/warning/stop strings across the package.
Also convert remaining backtick-quoted function and argument references in
messaging calls to proper cli markup: {.help AMR::fn}(), {.arg arg},
{.code expr}, and {.pkg pkg} throughout ab.R, ab_from_text.R, av_from_text.R,
amr_selectors.R, count.R, custom_antimicrobials.R, custom_microorganisms.R,
interpretive_rules.R, mo.R, mo_property.R, sir.R, sir_calc.R.
Fixes#191https://claude.ai/code/session_01XHWLohiSTdZvCutwD7ag2b
* Fix {.help} markup to use correct cli link format [{.fun fn}](AMR::fn)
Replace all instances of {.help AMR::fn}() (incorrect format with manual
parentheses outside the link) with {.help [{.fun fn}](AMR::fn)} which is
the correct cli hyperlink syntax: the display text [{.fun fn}] renders the
function name with parentheses automatically, and (AMR::fn) is the link target.
Also update the plain-text fallback handler in aa_helper_functions.R to
extract the display text from the [text](topic) markdown link format,
so that non-cli environments show just the function name (e.g. `fn()`),
not the raw link markup.
Dynamic cases in amr_selectors.R and mo_property.R also updated.
https://claude.ai/code/session_01XHWLohiSTdZvCutwD7ag2b
* Add {.topic} markup for non-function help page references
Replace {.code ?AMR-options} and backtick-style ?AMR-options / ?AMR-deprecated
references with proper {.topic AMR-options} / {.topic AMR-deprecated} cli markup
in count.R, interpretive_rules.R, proportion.R, and zz_deprecated.R.
Add {.topic} fallback handler to format_message() in aa_helper_functions.R:
plain-text environments render {.topic foo} as ?foo, and the [text](topic)
link form extracts just the display text (same pattern as {.help}).
Also convert remaining backtick function/arg references in proportion.R to
{.help [{.fun ...}](AMR::...)}, {.arg}, and {.code} markup for consistency.
Note: zzz.R intentionally keeps the backtick form since its startup message
goes through packageStartupMessage() which bypasses our cli infrastructure.
https://claude.ai/code/session_01XHWLohiSTdZvCutwD7ag2b
* Fix {.topic} to use required pkg::topic format with display text
{.topic} in cli requires a package-qualified topic reference to generate
a valid x-r-help:pkg::topic URI. Bare {.topic AMR-options} produced a
malformed x-r-help:AMR-options URI (no package prefix).
Use the [display_text](pkg::topic) form throughout:
{.topic [AMR-options](AMR::AMR-options)}
{.topic [AMR-deprecated](AMR::AMR-deprecated)}
The hyphen in the topic name is fine as a URI string even though
AMR::AMR-options is not a valid R symbol expression.
The fallback handler in format_message() already handles the [text](uri)
form by extracting the display text, so plain-text output is unchanged.
https://claude.ai/code/session_01XHWLohiSTdZvCutwD7ag2b
* Fix regexec() calls: remove perl=TRUE unsupported in older R
regexec() only gained the perl argument in R 4.1.0. The CI matrix
covers oldrel-1 through oldrel-4 (R 3.x/4.0.x), so perl=TRUE caused
an 'unused argument' error on every message_() call in those
environments.
All four affected regexec() calls use POSIX-extended compatible
patterns, so dropping perl=TRUE is safe.
https://claude.ai/code/session_01XHWLohiSTdZvCutwD7ag2b
* Slim CI matrix for PRs to ubuntu-latest / r-release only
For pull requests, check-recent now runs a single job (ubuntu-latest,
r-release) via a setup job that emits the matrix as JSON. On push and
schedule the full matrix is unchanged (devel + release on all OSes,
oldrel-1 through oldrel-4).
Also removed the pull_request trigger from check-recent-dev-pkgs; the
dev-packages check only needs to run on push/schedule.
https://claude.ai/code/session_01XHWLohiSTdZvCutwD7ag2b
* Restrict dev-versions and old-tinytest CI to main branch only
Both workflows were triggering on every push to every branch.
Narrowed push trigger to [main] so they only run after merging,
not on every feature/PR branch push.
https://claude.ai/code/session_01XHWLohiSTdZvCutwD7ag2b
* Update NEWS.md to continuous log + add concise style rules to CLAUDE.md
NEWS.md is now a single continuous log under one heading per dev series,
not a new section per version bump. CLAUDE.md documents: only replace
line 1 (heading), append new entries, keep them extremely concise with
no trailing full stop.
Merged 9035 and 9036 entries into one section; condensed verbose 9036
bullets; added CI workflow change entry.
https://claude.ai/code/session_01XHWLohiSTdZvCutwD7ag2b
* Replace single-quoted literals in messaging calls with cli markup
Converted bare 'value' strings inside stop_(), warning_(), message_()
to appropriate cli markup:
- {.val}: option values ('drug', 'dose', 'administration', 'SDD', 'logbook')
- {.cls}: class names ('sir', 'mo')
- {.field}: column names ('mo' in mo_source)
- {.code}: object/dataset names ('clinical_breakpoints')
Files changed: ab_from_text.R, av_from_text.R, sir.R, sir_calc.R, mo_source.R
https://claude.ai/code/session_01XHWLohiSTdZvCutwD7ag2b
* Apply {.topic}, {.cls}, and {.field} markup in sir.R messaging
- 'clinical_breakpoints' (dataset): {.code} -> {.topic [clinical_breakpoints](AMR::clinical_breakpoints)}
- "is of class" context: extract bad_col/bad_cls/exp_cls vars and use {.cls} + {.field} in glue syntax
- Column references in as.sir() messages: font_bold(col) with surrounding quotes -> {.field {col}}
https://claude.ai/code/session_01XHWLohiSTdZvCutwD7ag2b
* Replace glue-style dynamic markup with paste0() construction
{.field {variable}} and {.cls {variable}} patterns rely on glue
evaluation which is not safe in a zero-dependency package. Replace
all four occurrences with paste0("{.field ", var, "}") so the value
is baked into the markup string before reaching message_()/stop_().
https://claude.ai/code/session_01XHWLohiSTdZvCutwD7ag2b
* Limit push trigger to main in check-recent workflow
push: branches: '**' caused both the push event (9-worker matrix) and
the pull_request event (1-worker matrix) to fire simultaneously on every
PR commit. Restricting push to [main] means PR pushes only trigger the
pull_request path (1 worker), while direct pushes to main still get the
full 9-worker matrix.
https://claude.ai/code/session_01XHWLohiSTdZvCutwD7ag2b
* Limit push trigger to main in code-coverage workflow
Same fix as check-recent: push: branches: '**' caused the workflow to
run twice per PR commit (once for push, once for pull_request). Restricting
push to [main] ensures coverage runs only once per PR update.
https://claude.ai/code/session_01XHWLohiSTdZvCutwD7ag2b
* Replace bare backticks with cli inline markup across all messaging calls
- {.arg} for argument names in stop_/warning_/message_ calls
- {.cls} after "of class" text in format_class() and elsewhere
- {.fun} for function names (replaces `fn()` pattern)
- {.pkg} for tidyverse package names (dplyr, ggplot2)
- {.code} for code literals (TRUE, FALSE, expressions)
- Rewrite print.ab: use cli named-vector with * bullets and code
highlighting when cli >= 3.0.0; keep plain-text fallback otherwise
- Fix typo in as.sir(): "of must be" -> "or must be"
- switch sir.R verbose notes from message() to message_()
https://claude.ai/code/session_01XHWLohiSTdZvCutwD7ag2b
* Pre-evaluate inline expressions, add format_inline_(), fix print.ab
- All bare {variable}/{expression} in message_()/warning_()/stop_() calls
are now pre-evaluated via paste0(), so users without cli/glue never see
raw template syntax (mo_source.R, first_isolate.R, join_microorganisms.R,
antibiogram.R, atc_online.R)
- Add format_inline_() helper: formats a cli-markup string and returns it
(not emits it), using cli::format_inline() when available and cli_to_plain()
otherwise
- Rewrite .onAttach to use format_inline_() for all packageStartupMessage
calls; also adds {.topic} link and {.code} markup for option names
- print.ab: pre-evaluate function_name via paste0 (no .envir needed),
apply highlight_code() to each example bullet for R syntax highlighting
- join_microorganisms: pre-evaluate {type} and {nrow(...)} expressions
https://claude.ai/code/session_01XHWLohiSTdZvCutwD7ag2b
* fixes
* Replace all "in \`funcname()\`:" with {.help [{.fun funcname}](AMR::funcname)}
Converts all "in `funcname()`:" prefixes in warning_()/message_()/stop_()
calls to the full {.help} link format for clickable help in supported
terminals. Also fixes adjacent backtick argument names to {.arg}.
Files changed: ab.R, ab_property.R, av.R, av_property.R, antibiogram.R,
key_antimicrobials.R, mdro.R, mic.R, mo.R, plotting.R
https://claude.ai/code/session_01XHWLohiSTdZvCutwD7ag2b
* fixes
* definitive
* version fix
---------
Co-authored-by: Claude <noreply@anthropic.com>
* mdro(): infer base drug resistance from drug+inhibitor combination columns (#209)
When a base beta-lactam column (e.g., piperacillin/PIP) is absent but a
corresponding drug+inhibitor combination (e.g., piperacillin/tazobactam/TZP)
is present and resistant, resistance in the base drug is now correctly
inferred. This is clinically sound: resistance in a combination implies the
inhibitor provided no benefit, so the base drug is also resistant.
Susceptibility in a combination is NOT propagated to the base drug (the
inhibitor may be responsible for susceptibility), so only R values are
inferred; missing base drugs remain NA otherwise.
Implementation details:
- Uses AB_BETALACTAMS_WITH_INHIBITOR to identify all beta-lactam+inhibitor
combinations present in the user's data
- Derives base drug AB codes by stripping the "/inhibitor" part from names
- Creates synthetic proxy columns (.sir_proxy_<AB>) in x, set to "R" when
any matching combination is R, otherwise NA
- Proxy columns are added to cols_ab before drug variable assignment,
so all existing guideline logic benefits without any changes
- Multiple combos for the same base drug are OR-ed (any R → R)
- Adds internal ab_without_inhibitor() helper for the name->base mapping
- Verbose mode reports which combinations are used for inference
Bumps version: 3.0.1.9028 -> 3.0.1.9029
https://claude.ai/code/session_01Cp154UtssHg84bw38xiiTG
* Add sir.R/mic.R fixes and mdro() unit tests; bump to 3.0.1.9030
R/sir.R (line 571):
Guard purely numeric strings (e.g. "1", "8") from the Unicode letter
filter. Values matching the broad SIR regex but consisting only of digits
must not be stripped; add `x %unlike% "^[0-9+]$"` predicate.
R/mic.R (lines 220-222):
Preserve the letter 'e' during Unicode-letter removal so that MIC values
in scientific notation (e.g. "1e-3", "2.5e-2") survive the cleaning step.
- Line 220: [\\p{L}] → [^e\\P{L}] (remove all letters except 'e')
- Line 222: [^0-9.><= -]+ → [^0-9e.><= -]+ (allow 'e' in whitelist)
tests/testthat/test-mdro.R:
New tests for the drug+inhibitor inference added in the previous commit
(issue #209):
- TZP=R with no PIP column → PIP inferred R → MDRO class elevated
- TZP=S with no PIP column → proxy col is NA (not S) → class lower
- verbose mode emits "Inferring resistance" message
- AMC=R with no AMX column runs without error (Enterococcus faecium)
https://claude.ai/code/session_01Cp154UtssHg84bw38xiiTG
* Fix version to single bump (9029) and update CLAUDE.md versioning rules
CLAUDE.md: Rewrite the "Version and date bump" subsection to document that:
- Exactly ONE version bump is allowed per PR (PRs are squash-merged into one
commit on the default branch, so one commit = one version increment)
- The correct version is computed from git history:
currentversion="${currenttag}.$((commits_since_tag + 9001 + 1))"
with the +1 accounting for the PR's own squash commit not yet on the
default branch
- Fall back to incrementing DESCRIPTION's version by 1 if git describe fails
- The Date: field tracks the date of the *last* PR commit (updated each time)
DESCRIPTION / NEWS.md: Correct the version from 3.0.1.9030 back to 3.0.1.9029.
Two version bumps were made across two commits in this PR; since it will be
squash-merged as one commit only one bump is correct. Also update Date to
today (2026-03-07).
https://claude.ai/code/session_01Cp154UtssHg84bw38xiiTG
* Fix stats::setNames, test accessor bug, and version script verification
R/mdro.R:
Qualify setNames() as stats::setNames() in the drug+inhibitor inference
block to satisfy R CMD CHECK's global-function checks.
tests/testthat/test-mdro.R:
mdro() with verbose=FALSE returns an atomic ordered factor, not a
data.frame. Fix three test errors introduced in the previous commit:
- Line 320: result_no_pip$MDRO -> result_no_pip (factor, no $ accessor)
- Line 328: result_tzp_s$MDRO / result_no_pip$MDRO -> direct factor refs
- Line 347: expect_inherits(..., "data.frame") -> c("factor","ordered")
Also fix the comment on line 347 to match the actual return type.
Version: confirmed at 3.0.1.9029 (no further bump; one bump already made
this PR). git describe failed (no tags in dev environment) — fallback
applies. The +1 in CLAUDE.md's formula is correct for tagged repos:
currentcommit + 9001 + 1 = 27 + 9001 + 1 = 9029 ✓
https://claude.ai/code/session_01Cp154UtssHg84bw38xiiTG
* Fix unit tests: use mrgn guideline and expect_message() for proxy tests
Three failures corrected:
1. Classification tests (lines 321, 329): The EUCAST guideline for
P. aeruginosa already has OR logic (PIP OR TZP), so TZP=R alone
satisfies it regardless of whether the PIP proxy exists. Switch to
guideline="mrgn": the MRGN 4MRGN criterion for P. aeruginosa
requires PIP=R explicitly (lines 1488-1496 of mdro.R), with no TZP
fallback. Without the proxy: PIP missing -> not 4MRGN -> level 1.
With the proxy (TZP=R infers PIP=R): 4MRGN reached -> level 3.
The TZP=S case leaves proxy=NA, so PIP is still absent effectively
-> level 1, which is < level 3 as expected.
2. Verbose/message test (line 335): message_() routes through message()
to stderr, not cat() to stdout. expect_output() only captures stdout
so it always saw nothing. Fix: use expect_message() instead, and
remove the inner suppressMessages() that was swallowing the message
before expect_message() could capture it.
Also trim two stale lines left over from the old expect_output block.
https://claude.ai/code/session_01Cp154UtssHg84bw38xiiTG
---------
Co-authored-by: Claude <noreply@anthropic.com>
