Conversation
* Don't rely on transitive include of `"fmt/core.h"` We got it through `cpp11/protect.hpp`, but we should not rely on that * Correctly use `fmt::runtime()` on runtime strings passed to `fmt::format()` In fmt, there is a `FMT_CONSTEVAL` macro that resolves to `consteval` on "new enough" C++ (otherwise it doesn't do anything). For R 4.6+, the default C++ used is finally "new enough" (`__cplusplus > 201703L`). This causes all `fmt::format()` calls to require a constant expression for `const char*` and `std::string&` input, which we are not currently doing via `fmt_arg`. We have a runtime provided string, which must now be wrapped in `fmt::runtime()`, which is what we should have been doing all along. * Add `r_ns_env()` and use in `get_namespace()` Throwing an informative (and tested!) error when we can't find the package namespace * Define `RCPP_NO_R_HEADERS_CHECK` before all `#include <Rcpp.h>` usage `#include <cpp11/R.hpp>` sets everything up the right way, and otherwise we get a warning from Rcpp which doesn't seem to end up being relevant for this use case * Use `r_env_has()` + `r_env_get()` To avoid triggering a NOTE about usage of `Rf_findVarInFrame3()` on R 4.5, where technically we had the tools to avoid that
* Teach `cpp_source()` how to source multiple files * NEWS bullet * Link to PR
…violations (r-lib#493) * Add failing test * Use tag types to generate attribute specific internals * NEWS bullet * Formatting
* Update `cran-comments.md` * Increment version number to 0.5.5
Bring the fork up to date with r-lib/cpp11, keeping the fork's fork-local changes and taking upstream's advances. Conflicts resolved: * R/vendor.R: kept the fork's `date` and `overwrite` arguments to `cpp_vendor()`, re-applied on top of upstream's Air-formatted shape. * R/source.R: took upstream's version wholesale. Upstream's multi-file `cpp_source()` (r-lib#492) supersedes the fork's vectorization of `file`, and already carries the fork's "report the first missing file" fix. * inst/include/cpp11/protect.hpp: took upstream's version wholesale. Upstream's tag-type ODR fix (r-lib#493) addresses the root cause of the `cpp11::warning()` unwind crash (r-lib#295) that the fork worked around by dropping the `const char*` overloads and copying the format string. Also reformatted the fork's `[[deprecated]]` conversion operators in inst/include/cpp11/sexp.hpp with clang-format; they were over the 90 column limit. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01K8MneV8KqHYUuC8fWV3X5Q
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
2 participants
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.
Merges
r-lib/cpp11@7d0e4ac(upstreammain) into the fork'smain. The merge base wasadfb240(usethis::use_claude_code()(r-lib#485)); upstream has since added the Air formatter,cpp_source()multi-file support (r-lib#492), the tag-type ODR fix forcpp11::stop()/cpp11::warning()(r-lib#493), the removal of non-APIR_NamespaceRegistryusage (r-lib#488), and two releases. The guiding principle was to keep the fork's intent while taking upstream's advances.Three files conflicted.
R/vendor.R— the fork addsdateandoverwritearguments tocpp_vendor(); upstream reformatted the function with Air. Resolved by re-applying the fork'soverwritebranch andas.Date(date)on top of upstream's Air-formatted shape, rather than reverting the reformatting.R/source.R— resolved by taking upstream's version wholesale. Upstream's multi-filecpp_source()(r-lib#492) is a strict superset of the fork's "vectorizefile" change: it usesvcapply(file, generate_cpp_name)for the per-file names and a freshly generated randomgenerate_package_name()for the shared library, which avoids the name collisions the fork'stools::file_path_sans_ext(name[[1]])approach could hit. Upstream also already carries the fork's fix to report the first missing file in the "Can't findfile" error. The mergedR/source.Ris byte-identical to upstream's.inst/include/cpp11/protect.hpp— resolved by taking upstream's version wholesale. See "Dropped as superseded" below.Two other fork changes merged cleanly but are worth noting:
R/register.R'ssub()→gsub()fix for package names with multiple dots survived upstream's large refactor of that file untouched, and the fork-only headers (declarations.hpp,external_pointer.hpp,r_vector.hpp,sexp.hpp) pluscpp11test/src/test-external_pointer.cppwere not touched upstream at all.Dropped as superseded
The fork's
protect.hppchange — "Copy strings before callingRf_warningcall()to avoid weird unwind behavior", which removed thewarning(const char*, ...)overloads and changedwarning(const std::string&, ...)to take the format string by value — has been dropped. That was a workaround for r-lib#295, wherecpp11::warning(const char*)crashes butcpp11::warning(const std::string&)does not. Upstream's r-lib#493 fixes the root cause:Rf_errorcall()andRf_warningcall()have the same function type, socpp11::stop()andcpp11::warning()were instantiating one and the samedetail::closure/detail::apply()template, and under ODR the[[noreturn]]-flavoured instantiation could win and make the return path inwarning()unreachable. Upstream addsdetail::return_tag/detail::no_return_tagto the template parameters so the two get distinct instantiations, and ships a regression test (cpp11test/src/template-1-stop.cpp,template-2-warn.cpp,cpp11test/tests/testthat/test-template.R) that reproduces the original crash. Keeping the fork's workaround on top would have needlessly removed theconst char*overloads from the public API and pessimised the remaining ones, so upstream's version is taken unchanged and that test now passes here.Other changes
inst/include/cpp11/sexp.hppwas reformatted withclang-format. The fork's three[[deprecated]]conversion operators were written on single lines well past the project's 90 column limit and already failedclang-format --dry-run -Werroronmainbefore this merge; only whitespace changed, the attributes and bodies are untouched.Verification
air format .leaves the tree unchanged.clang-format --dry-run -Werroris clean on all five.hpp/.cppfiles that differ fromupstream/main. The R suite passes (testthat::test_dir("tests/testthat"), 0 failures; also re-run withNOT_CRAN=trueso the sixskip_on_cran()tests actually execute — still 0 failures). The C++ suite passes afterdevtools::install()+devtools::clean_dll("./cpp11test"), 0 failures, including both the fork'sexternal_pointer-C++context and upstream's newtemplateregression test.🤖 Generated with Claude Code
https://claude.ai/code/session_01K8MneV8KqHYUuC8fWV3X5Q
Generated by Claude Code