diffobj

v0.3.5

Options automatically fallback to factory defaults if they are unset (h/t @gadenbui).
#158: Calling diff* with do.call now works without warnings.
#117: Fix guide detection with very wide wrapped data.frames (h/t @bastician, @overvolting).

Add a print method for ses_dat return values that makes it easier to interpret the diff.
#152: Rewrite the fall-back “O(n)” algorithm that kicks in when there are max.diffs differences to be more robust (h/t @hadley, @DanChaltiel, @gadenbui).
Related to #152: max.diffs=0 used to mean the same as max.diffs=-1 (i.e. unlimited), but this was undocumented and an error. max.diffs=0 will now immediately fall back to the “O(n)” algorithm.

#143: Add ses_dat to provide a more computable version of ses (h/t @hadley).
#144: Re-encode strings to UTF-8 prior to comparison to avoid spurious encoding-only differences (h/t @hadley).
#142: Typos in standardGeneric in trim/guide generic definitions.
Drop attributes from inputs to diffChr (revealed as an issue by #142).
Banish ghosts of stringsAsFactors.

Tests explicitly set stringsAsFactors=TRUE so they don’t fail with the anticipated changed for R4.0.
#140: Bad link in ?ses.

This is a bugfix release.

#136: Documentation for ignore.white.space (h/t @flying-sheep) and max.diffs parameters listed incorrect defaults.
#135: Incorrect handling of potential meta data strings when unwrapping atomics would cause a “wrong sign in by argument” error (h/t @flying-sheep). We also fixed other bugs related to the handling of meta data in atomic vectors that were uncovered while debugging this issue.
#134: Forwarding ... to diff* functions no longer breaks substitution of arguments for diff banners (h/t @noamross)..
#133: diffFile considers files with equal content but different locations to be all.equal now (h/t @noamross).
#132: Duplicate pager slot for baseline Pager removed (h/t Bill Dunlap).

There are also several other small internal changes that in theory should not affect user facing behavior.

#129: Allow pager specification via lists rather than full Pager objects for easier changes to defaults. As part of this we changed StyleRaw objects to use default pager instead of PagerOff.
#126: Add embedding diffs in Shiny to vignette.
#119: ignore.whitespace now also ignores white space differences adjoining punctuation.
#118: New option to preserve temporary diff file output when using pager (see ?Pager).
#114: New options strip.sgr and sgr.supported allow finer control of what happens when input already contains ANSI CSI SGR and how ANSI CSI SGR is handled in string manipulations. Related to this, options(crayon.enabled=TRUE) is no longer set when capturing output prior to diff as it used to be. By default pre-existing ANSI CSI SGR is stripped with a warning prior to comparison.

#131: Fix missing slot in S4 class definition (discovered by Bill Dunlap).
#127: Width CSS conflicts with bootstrap (reported by @eckyu, debugged/fixed by @cpsievert).

Comply with CRAN directive to remove references to packages not in depends/imports/suggests in tests (these were run optionally before).
Fix bugs in corner case handling when we hit max.diffs.

#87: diffobj is now GPL (>=2) instead of GPL-3.
#81: Better handling of mixed UTF-8 / ASCII strings, reported by jennybc
#88: correctly handle trimming when empty lists are involved, reported by wch
#77: diffObj now favors dispatching to diffPrint unless diffPrint output is large
#82: diffChr and ses now treat NA as “NA” (needed with change in nchar(NA) in base R)
#85: Improved alignment of unwrapped atomic vector diffs
#83: Improve pager auto detection (note now ANSI output is only allowed by default if terminal supports ANSI colors and the system pager is less, see ?Pager for details)
#92, #80, #45: basic implementation of S4 guidelines and trimming (full resolution eventually with #33)
#84: simplify how to call diffChr for improved performance, including “optimization” of convert.hz.whitespace.
#64: fix line limit in corner case
More robust handling of external diff* methods and of how diffObj calls diffStr and diffPrint

#71: Buggy diffs b/w data frames when one has sequential row numbers and the other does not, loosely related to #38
#69: Improve performance on outputs with large print/show output, and other assorted minor optimizations
#72: Warn when style parameter overrides other user supplied parameters
#70: Improve word contrast in YB HTML mode
#63: Show all.equal output when objects are not all.equal but there are no visible differences
Add Mean Relative Indifference vignette and update vignette styling

#67: Fix CRAN Binaries
Clarified that C code is heavily modified and incompatible with original libmba implementation

First version on CRAN
#51: use RStudio viewport to display HTML diffs when running in RStudio (h/t Noam Ross)
#54: #55, scale HTML output to viewport width (see ?Style)
#53: default term colors computed on run instead of on package load
#56: disable wrap for HTML output
HTML output now captured with default width 80 since there is no explicit relationship between HTML viewport width and getOption("width")
The style parameter now accepts lists to use as instantiation arguments for Style objects (see ?Style)
Fix subtle rendering and formatting application flaws
Switch Travis shields to SVG per Gábor Csárdi
Improve in-hunk alignment of partially matching lines
Compile with -pedantic, fix related warnings Arun
Improved coverage and more robust testing
Several internal structure changes to accommodate improvements