-
Notifications
You must be signed in to change notification settings - Fork 235
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Link to files, not topics #1109
Conversation
This comment has been minimized.
This comment has been minimized.
be377a4
to
b533014
Compare
I.e. if the name of the topic is the same as the name of the file, then we are ok.
Also implement proper warnings, with file names and line numbers, for re-exports as well.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
Actually, the approach in #1109 (comment) does not work, so we'll have to go with the more complicated solution. |
This is R CMD check's job.
But leave the rest of the links alone.
This comment has been minimized.
This comment has been minimized.
@hadley This is ready, fyi. |
26c74c8
to
b30da61
Compare
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
Co-authored-by: Hadley Wickham <h.wickham@gmail.com>
That we just broke...
@hadley Can you pls take another look? Thanks! |
…ccount for documentation handling in R-devel.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
A few last smaller things; only bother with them if we don't hear back on R-package-devel in a couple of days. (And then I'll do a release on Thursday or Friday)
R searches for topics in the linking package itself first, so these are not need to be qualified. Ignoring the 'mypackage::' part helps moving code between packages, e.g. for compat files.
@hadley: Thanks, this is ready to go. |
Thanks! |
…topic as indicated here r-lib/roxygen2#1109
….7.0 Grant McDermott (24): Add logitmfx tidiers Better regexp Add more examples Add probit tidiers (same as logit TBH) Updated NAMESPACE after adding new mfx tidiers Correct function name would be nice Just use logitmfx tidiers for probitmfx equivalents Catch cases where user doesn't specify atmean argument Add tests for mfxlogit Argh, correct function name again Better name Add probitmfx tests Squash into generic tidiers with various methods Missed on previous commit Right export names would help... Add contributor details Suggest packages Fix mfx rdnames and methods Retain SEs of input object mfx tidiers to NEWS.md Resolve Alex's PR commments Separate out betamfx tidiers Drop deprecated fix_data_frame function Update NAMESPACE Greg Macfarlane (6): Write tidy method for sarlm models Add glance and augment for sarlm objects; document Simplify logic of spatial regression tidying Write tidy method for sarlm models Add glance and augment for sarlm objects; document Simplify logic of spatial regression tidying Lukas Wallrich (3): Add option to return p.values to tidy.polr Update NEWS.md Update DESCRIPTION Max Kuhn (1): 0.7.0 revdeps Petr Hrobar (2): model added model added Simon Couch (61): restore .Rproj... oops🙈 add svd return value alexpghayes/modeltests#26, document() check when entries don't result in docs work through failing return_* entries (closes #856) small build fixes fix `codecov` and `pkgdown` actions use lavaan extractors in glance.lavaan (#835) explicitly coerce to tibble (closes #824) tidyverse -> tidymodels in links (closes #832) pass ddf arg to confint (#839) adjust covr install re: 4.0.0 error update gh actions R version install tidytemplate for pkgdown correct nobs and norig in glance.lavaan (closes #835) note deprecation of fix_data_frame() (#806, #862) restore n column re: chest revdepcheck (#862) refresh docs with devel modeltests version restore .resid column to augment.nls output restore sparse matrix tidiers (closes #470) small build fixes use_tidy_style() (closes #640) document after use_tidy_style() last few fixes to evalRd return_*() calls deprecate fix_data_frame() (closes #629, closes #806) add contributors🙂 glance.lm edge cases (closes #865) add helper for tidy methods small build fixes use modeltests for car:: unit tests unname columns generated from predict() (closes #866) for now, use devel modeltests for unit testing wait on modeltests release for car::Anova tests wait for modeltests release to check arguments document p.values arguments correct language about glance.aov changes (closes #864) warn on tidy_optim standard error (closes #529) deprecate summary.manova() address TODOs in NEWS.md spellcheck update to new tidyr::(un)nest() API (#831, closes #870) more vignette edits (closes #831) windows 3.6 build fix only manually install mnormt on windows windows 3.6 build failing re: new mnormt release add glance helper (#823) run document() refactor glance() methods (closes #823) a few more NEWS.md updates for 0.7.0 rename helpers for internal consistency ensure no docs for map_as_tidy_tibble small vignette style edits (closes #831) restore unit tests re: modeltests v0.1.2 (closes #869) update fixest tidiers with new internal helpers and documentation document tidy.map return value (closes #481) small build fixes remove duplicate entry workshop NEWS.md fix merge conflict fix url formatting replace `iris` with `trees`, fix badges formatting/spellcheck Simon P. Couch (54): resolve merge conflicts, use helpers in #847 tidiers recover original response name, clarify docs (#847) spellcheck/clarify automated docs add `spatialreg` contributions (#847) build fixes add unit testing for #847 build fixes more build fixes correct documentation skip sarlm examples skip redundant examples improve errors for scam objects (closes #186) use conf.int idiom (closes #868) address package lint increment version number documentation updates re: lint update cran comments clean up pkgdown yaml add myself as author per alex's request submit to CRAN! fixes from CRAN pretest failure - new `fixest` version update timeline submit round 2! link to man files rather than topics (r-lib/roxygen2#1109) cran submission 3 adjust docs for reexports specify base and stats namespaces explicitly export mfx aliased functions (#873) fix for internal .Rd linking more internal .Rd fixes link to quantreg file rather than topic note docs changes, update title clarify news re: new tidiers submission submit to CRAN update epi.2by2 tests submit to cran bump version write out iris in favor of modeldata::hpc_data update docs with reexports and iris -> hpc_data change bump version back soften deprecations adjust language about deprecations some check fixes re: deprecations tidy.dist fixes /dontrun examples with longest runtime update cran comments with new revdepchecks records restore finish_glance add deprecation notice to finish_glance more notes on reverse dependency checks submit to cran skip most unit tests on CRAN submit to cran accepted to CRAN🎉 karldw (10): Auto-remove trailing whitespace Start adding fixest Progress on fixest tidiers Progress on fixest tidiers Fixes to fixest tidiers Fix roxygen issues Address fixest PR comments Address additional fixest comments Bump required fixest version Remove deprecated `fix_data_frame` petrhrobar (2): Another model support added (line 31) Another model support added (line 31) topepo (1): revdeps on 05-26
External packages are simple, we just look up the file name and link to that.
Linking to the package being documented is harder, because at the time of the linking we don't know the file names yet. So we'll need to mark these parts in the docs, and post-process all files, ones the file names are known.These links do work with the topic, so we'll leave them alone.Internal links, to the package being documented.Not needed.Can we link to another package w/o specifying its name? Yes, probably shouldn't, though.Not relevant.Should we look up unqualified unknown link targets in dependent packages? Not just yet.Not releavant.One warning only for non-local unqualified links.No warning at all,R CMD check
will warn if it needs to.Resolve links to the base package, automatically?No need, we'll just leave unqualified links alone.Docs.Not really needed.@inheritedDotParams
intro.Closes #1108.