Parses LaTeX Documents for Errors

Checks LaTeX documents and .bib files for typing errors, such as spelling errors, incorrect quotation marks. Also provides useful functions for parsing and linting bibliography files.


Development page for TeXCheckR. See also https://github.com/hughparsonage/grattanReporter

News

TeXCheckR 0.6.0

Enhancements:

  • validate_bibliography now errors if url fields contain escaped symbols.
  • check_dashes now does not error if an emdash occurs in a protasis within an list, or if requested.
  • check_spelling :
    • now obeys a directive % ignore_spelling_in_file: <file.tex> for files where the spelling may be ignored (such as tables with symbols, jargon, or esoteric nomenclature).
    • gains an argument known.correct.fixed for words (rather than patterns) to add.

Bug fixes:

  • check_unclosed_parentheses no longer errors or panics over unclosed parentheses occurring within optional arguments to \begin{enumerate} such as label constructions.
  • labels around the chapter checks now contemplate multiple labels on the same line, and distinguish between \label{} and \labelenumi.

Other

  • Test involving readr did not properly check for it being installed

TeXCheckR 0.5.1

  • Allow fread_bib to optionally retain braces in value.

  • New functions:

    • split_report to split a report into chapters

    • minimal_bib to produce a minimal bibliography from a spun document

    • veto_sic to locate and remove words preceding [sic] (especially during spell checks)

    • strip_comments is now exported

    • read_tex_document to take the root file of a LaTeX document and read the entire document (including the contents \input and \include)

    • (Unexported) check_unclosed_parentheses to check (some) instances of unclosed or unopened parentheses

  • stringi and readr have been moved to Suggests to minimize compile times (as on Travis-CI)

  • check_spelling now skip words 'preceding' an editorial [sic].

  • check_spelling tries to fill skipped commands with whitespace equal to its contents to replace (to provide more accurate positions if there is an error).

  • halt in report_error now provides an option to not halt if a check fails.

  • Experimental utility functions fill_nth_LaTeX_argument and locate_nth_LaTeX_argument.

  • extract_valid_abbrevations (and thus check_spelling) considers the word on optional when backtracking.

  • check_xrefs defaults now permits lowercase forms. By default, consistency is enforced (whether upper or lower).

  • separate_sentences gains a hanging_footnotes argument, indenting footnotes if set to TRUE.

  • Add some proper-nouns to the dictionary.

  • Bug fixes:

    • check labels now test for spaces rather than check dashes
    • lint_bib does not add commas after @string fields.
    • inputs_of does not emit arcane warning if multiple \end{document}s exist.
    • check_escapes do not check content within tikzpicture environment or within \url commands.
    • check_dashes does not check display equation lines.
    • check_labels does not check that caption commands in the preamble are associated with a label. Similarly, it does not check the labels associated with \addchap and \chapter unless they have an opening brace. (For example, within \renewcommand).
    • check_quote_marks does not check quote marks beyond \end{document}
    • check_sentence_ending_periods does not check the contents of \hl{ (which may be used to markup content during editing).
    • check_spelling honours non-default settings of dict_lang in files through \input or \include, does not check the contents of tikz environements, does not look at the optional arguments of \printbibliography
    • check_consecutive_words no longer requires grattan.cls to run. Thanks to @jonocarroll for reporting.
    • check_footnote_typography no longer falsely claims a footnote lacks a terminal full stop when it ends in a list like enumerate or itemize (and the terminal item ends with a full stop).
  • Internal:

    • readr, stringi are now in Suggests to reduce compile time (especially on Travis-CI)

TeXCheckR 0.4.3

  • Use hunspell's ignore option when a dictionary addition is not respected
  • Fix spurious unit test broken under new hunspell.
  • Fix issue with check_footnote_typography where the RStudio pops to the end of the first footnote, not the first wrong footntoe. Thanks to CC for reporting.

TeXCheckR 0.4.2

  • Fixed inputs_of to select \include and \inputs only
  • Export inputs_of

TeXCheckR 0.4.1

  • Fixed an issue whereby pdflatex may not run in certain environments (in particular CRAN where it caused an egregious hanging of the package queue). Tests now skip in such cases.

TeXCheckR 0.4.0

  • Major internal functions parse_tex, extract_mandatory_LaTeX_argument, and extract_optional_LaTeX_argument.

TeXCheckR 0.3.3

  • check_footnote_typography can now provide the precise location of incorrect punctuation, instead of just reporting the head of footnote.
  • Use extract_LaTeX_argument now can handles multiline arguments.
  • Added a NEWS.md file to track changes to the package.

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.

install.packages("TeXCheckR")

0.6.0 by Hugh Parsonage, 8 months ago


https://github.com/HughParsonage/TeXCheckR


Report a bug at https://github.com/HughParsonage/TeXCheckR/issues


Browse source code at https://github.com/cran/TeXCheckR


Authors: Hugh Parsonage [aut, cre]


Documentation:   PDF Manual  


GPL-2 license


Imports clisymbols, crayon, data.table, fastmatch, hunspell, hutils, magrittr, rstudioapi, stats, tools, zoo

Suggests devtools, readr, stringi, testthat, tinytex


Suggested by hutilscpp.


See at CRAN