- Package
bibtexback in Imports and no longer used conditionally inReadBib()since it's now back on CRAN. - Fixes for
ReadPDFs()(#95). useBytesis nowFALSEin allsub,grep, etc. calls in all package functions.- Fix for
GetPubMedByID()with multiple book results (#86). - Remove comma preceeding "et al" in
Cite,Citep, etc. calls (#84) - Add support for custom bibtypes with custom bibstyles (#83)
- Package
bibtexmoved to Suggests inDESCRIPTIONdue to it currently being orphaned on CRAN. Work is underway by the ROpenSci team to rectify this. The package can still be installed from GitHub. In the event thatbibtexis not installed, the functionsReadBib(),GetBibEntryWithDOI(),ReadCrossRef(), andReadZotero()throw an appropriate message asking the user to installbibtexand invisibly returnNULL. - The old CrossRef API can no longer be used. If
use.old.apiis set to \code{TRUE} in \code{ReadCrossRef()}, it will be ignored with a warning. GetDOIs()had to be removed due to changes to the CrossRef API. It will hopefully return in the next release.
- When working in single-byte locales, the
printmethod forBibEntryobjects is more robust against accented characters being converted to incorrect ones whenbib.style = "authoryear". Additionally, for this style, a period could be removed from the last initial in the first author's given name whenfirst.inits = TRUE. This has been corrected.
- The serial comma is now used when formatting name lists
- Fix issue that could lead to Unicode characters being converted to latin1 when printing (h/t joaochenriques #62)
- Fix issue with extracting DOI from CrossRef results
in
GetDOIs - Fixes for
ReadCrossRefwhenuse.old.apiisTRUE. Scores are sometimes not returned by the API call and when this occurs the entries will now be added to the outputBibEntryobject with a message indicating that no score was available. - A comma no longer appears before "et al." when the
max.namesoptions is set to1(h/t davidaknowles #56)
- Fix extraction of citation counts in
ReadGSthat was occasionally causing errors from somescholar.ids (h/t Miao Sun, #59) - Fix printing of thesis and report entries types in authoryear style when "type" field missing (h/t Hugo Grunson, #58)
- Fix for
PrintBibliographyfor the case ofBibEntryobjects with a single entry, an NA value could appear next to the year in the output (#60) - Fix issue that could lead to Unicode characters being converted to latin1 when printing (h/t joaochenriques #62)
GetPubMedByIDis better at extracting years and months from the results returned by NCBI Entrez (h/t Dale Steele #52)- The
as.data.framemethod forBibEntryobjects now correctly handles the case of a single entry with name list fields containing multiple names (h/t Damon Bayer #51)
+.BibEntryandmerge.BibEntrygain an argument ignore.case, which defaults toBibOptions()$ignore.case(TRUE) so that case is ignore when checking for duplicate fields (h/t Justin Calabrese #47)- Improved warning message when printing entries with unknown LaTeX macros (the entry key is now included). (h/t Justin Calabrese #49)
- The entry key is now included in warning messages when entries are
missing fields and
BibOptions()$check.entries == "warn"(h/t Justin Calabrese #48)
- Entries are now only checked once to ensure all required
fields are present in
ReadBib
PrintBibliographygains parameters "start" and "end" to allow for printing only a subset of all cited entries from a BibEntry object (h/t Joseph Casillas #45, #46)
- Use https for all links (h/t Katrin Leinweber)
- Use preferred DOI resovler (h/t Katrin Leinweber)
- Add support for latex macro ast for asterisks (h/t Melinda Higgins)
- Fix
GetDOIsto use https - Fix download of bibliographic info from DOI in
ReadPDFs
- Fix writing of BibEntry object to stdout in
WriteBib(h/t Stephane Plaisance) ReadBibwon't add an attribute "strings" if there are none present in read bib file (h/t Stephane Plaisance)
- Fix deletion of temporary file if user supplies a DOI to
ReadCrossRef(h/t Ben Raymond)
- Documentation example improvements
- Improve error handling for API query functions
- Package peer reviewed and accepted by rOpenSci (h/t Noam Ross, Carl Boettiger, and Amelia McNamara)
- Remove missing plot from Rhtml vignette
- URL field returned by
GetBibEntryWithDOIis now decoded properly - Fix hyperlinks from bibliography to citations in vignettes
- Remove a incorrect message occasionally output from the addition
operator for
BibEntryobjects
- Package now uses httr, xml2, jsonlite packages instead of RCurl, XML, RJSONIO for scaffolding
- No more R CMD check NOTE regarding foreign function call to bibtex (h/t Romain Francois)
- Fix printing when
BibOptions(style = 'yaml) - Remove invalid character in inst/Bib/RJC.bib
- Correct parsing of interval dates when creating unique labels for authoryear style citations
c.BibEntrythrows an error if not all objects arebibentryobjects- Fix typos in documentation
- Literal ampersands are now printed correctly (not as '&') (h/t Yue Hu)
- Ensure BibTeX month macros are processed properly by lubridate in non-English locales (h/t Sergio Oller)
- Unescape special characters in URL fields returned by CrossRef (h/t Michael Schubert)
- Remove square brackets from custom entry type names (h/t Hugh Parsonage)
- Feature involving
LaTeXmacros added in package version 0.12.0 can only be used for R 3.3.z and higher; this corrects cause of failed checks on R 3.2.z
- Updated calls to NCBI Entrez for functions
ReadPubMed,GetPubMedByID, etc. to use https as now required by NCBI (h/t Dale Steele and Anthony Crane) - Change reference to www.omegahat.org to www.omegahat.net (h/t Kurt Hornik)
- Documentation for
ReadPubMedis updated to reflect that the default number of entries returned (controlled by the argumentretmax) is 20 (h/t Dale Steele)
- Some
LaTeXmacros unknown to R are now defined as macros in the package, and will be parsed usingmacrosarg intools::parse_Rd(assuminggetRversion() >= "3.2.0"Note: corrected in 0.13.1 to begetRversion() >= "3.3.0")
- Parse
LaTeXmacro\textquotesinglein author names (h/t Bill Denney) - Avoid "Request-URI too large" errors in GetPubMedByID if requesting a large number of IDs (h/t Maurits Evers)
ReadCrossRefnow uses the newer CrossRef API and gains argumentsfilterandoffsetto use with the new API; an additional argumentuse.old.apiis added if the user wishes to use the old API (h/t Carl Boettiger)ReadCrossRefnow parses the results returned by CrossRef to create theBibEntryobject when using the new API; for the old API (and hence, older versions of the package) the query only returns DOIs andReadCrossRefwould then use the DOIs to request the corresponding BibTeX entries from CrossRef (i.e. less HTTP requests when using the new API)
- Fix generation of entry keys when the word used from the title for key generation contains a non-ascii character (h/t Mark Johnson)
- RefManageR will no longer hang due to a bug in
tools::latexToUtf8(PR#17138) that is occasionally encountered when that function processes an unknown macro (h/t Eric Bryant) - Entries with no title field can now be printed without error when
BibOptions()$check.entriesis not set to "error" (default is "error")
- Removed unnecessary use of local/parent.frame; fixes execution with bytecode compiler (h/t Tomas Kalibera)
- Fixed broken test involving
ReadPDFsdue to changed URL (h/t Kurt Hornik) as.data.frame.BibEntryworks for length one BibEntry with multiple authors (h/t Dale Steele)- Use
httr::GETto fixReadGS - Fixed broken tests in
test-authors.Rowing to changes topersonclass - Fixed
ReadCrossReftests and error message - Fixed printing for authoryear style (h/t Joseph Casillas)
- Name list fields (author, editor, etc.) provided to the function
BibEntryare now properly parsed when specified as they would be in BibTeX/BibLaTeX; e.g.author = "Smith, Bob A. and Doe, Jane".
- The 'key' field in
BibEntryobjects is now always enforced to be unique as.data.frame.BibEntryis faster and now works if duplicate keys are present; keys in (h/t Dale Steele)- Fix for
ReadCrossRefif downloaded BibTeX had leading whitespace (h/t Carl Boettiger) useBytes = TRUEused for all calls togrep,sub, etc. (h/t HI&RH Lord Ripley of England)- remove use of deprecated function
lubridate::new_interval - updated URL for the BibLaTeX manual
- Fix test in test-search.R that broke because of new year (h/t HI&RH Lord Ripley of England)
- add additional functions from utils and stats to NAMESPACE
- Use
bibtex >= 0.4.0.9000functiondo_read_bibto avoid.Externalcall andR checknote (request of HI&RH Lord Ripley of England)
- Improve parsing of dates in
ReadPDFs - Citations using
Citefamily of functions can now bepandocstyle, e.g.[@abibkey]by settingBibOptions(cite.style = "pandoc")(h/t Dale Steele) - Added note about locales when parsing string 'month' fields to
ReadBibhelp page (h/t Dieter Menne)
- Fixed merging
BibEntryobjects by multiple fields when no duplicates open.BibEntryfixed to not use partial matching of field names; e.g. an error would occur if the specified entry had a 'urldate' field, but no 'url' fieldopen.BibEntrywillmessageand not throw error if entry cannot be opened- Fixes for
ReadPDFswhen argumentuse.metadataisFALSE - Fix for
ReadPDFswhen when reading one file which is a JSTOR pdf - Fix for sorting by volume (
BibOptions(sorting = "anyvt")andBibOptions(sorting = "nyvt") - Fix for sorting by label (
BibOptions()sortingequal to "anyvt" or "anyt") GetBibEntryWithDOI.Rwill notstopif an error occurs downloading any of the DOIs (e.g., if one entry in thedoivector has a typo and the rest are valid)
GetPubMedByID: Now returns some additional fields including 'month' and 'issn' for articles; will print a warning if PubMed does not return the complete list of authors; will use the name of a collective if one is available and the individual authors are missing (h/t Dale Steele)
ReadBib: If a name list field in an entry cannot be parsed in the bib file, the entry will be ignored, but the rest of the file will still be processed and returned. In the past, this caused an error and no output would be returned.- 'Book' entries will now be parsed correctly by
GetPubMedByID(h/t Dale Steele) - Fix error/warning messages when entry is missing required fields (bug introduced in Version 0.8.45)
- Name lists containing a comma in braces will now be parsed correctly, e.g. "Buchalter, Louis and {Murder, Inc.} and Anastasia, Albert"
ReadCrossRefnow correctly handles the small number of cases where BibTeX information cannot be obtained for a particular DOI, which resulted in 'stack imbalance' warnings and no results being returned (h/t Norman L Guinasso Jr).ReadGSfixed to account for changes to Google "API" (h/t Norman L Guinasso Jr).- Improved parsing for BibTeX format names ending with a '}' (h/t Henrik Bengtsson).
- Printing references with
style = "html"would not always add an opening tag whenbib.style = "numeric"orbib.style = "alphabetic"(h/t Henrik Bengtsson). format.BibEntrywould ignore the.styleargument if called directly by the user. Note, this function should normally not need to be called directly. (h/t Henrik Bengtsson)
- Improved formatting of citation given to CrossRef for increased
chances of finding matches with
GetDOIsfunction (h/t Erich Studerus) - Additional parsing of 'month' field to accomodate days and ranges of
days and months. Example bib entries that will be parsed correctly
include
month = jun # "/" # jul,month = "20~" # jan,month = "20==25~" # dec,month = "10~" # jan # "/" # feb(request of Stephen Eglen) - Added argument 'group' to
ReadZoterofor specifying a groupID to query a group library instead of a user library (h/t Greg Blomquist).
- DOI's hyperlinks in Markdown format are now correct (h/t Stephen Eglen)
print.BibEntrywithBibOptions(style = "Biblatex")fixed (h/t Artem Klevtsov)unlist.BibEntryandRelistBibEntrynow retains@stringsandmheaderandmfooterattributes (see ?BibEntry) if they are present
- Added function
GetDOIswhich searches CrossRef for DOIs for the citations stored in aBibEntryobject
ReadCrossReffixed to account for change to CrossRef API endpoint. (h/t Carl Boettiger)- Abstracts returned by NCBI Entrez can be multiple parts. This is now handled correctly and the complete abstract will be returned in the 'abstract' field. (h/t Erich Studerus)
- DOI's were too naively extracted from NCBI Entrez results, resulting in some entries having 'doi' fields with length greater than one. Now fixed. (h/t Erich Studerus)
- Functions for interacting with NCBI Entrez return abstract of each article (request of Erich Studerus)
print.BibEntrywithBibOptions(style = "citation")now works properly- Examples calling web resources should no longer upset the check farm (h/t HI&RH Lord Ripley of England)
as.BibEntrywill create entry key if given abibentryobject with no key. Useful when citing packages withcitation.- PrintBibliography and Cite functions (Cite, Citet, etc.) accept
bibentryobjects in addition toBibEntryobjects. $<-.BibEntrywill now accept a single person object, so that a single author in a multi-author entry may be updated. An example may be found athelp("$<-.BibEntry"). (h/t Carl Boettiger)
- validated html
- changed example for
WriteBibthat occasionally failed check
- Cite functions work if a specified entry has no key. Note that keys should always be provided for all entries as they are required for all entries in a BibLaTeX bib file (h/t Carl Boettiger)
- Entries returned by Crossref that have entry type 'Data' which is not supported by default in BibLaTeX are converted to type 'Online' (h/t Carl Boettiger)
- Fix for
ReadGSwhen argumentcheck.entriesis FALSE or "warn" (h/t Francisco Rodriguez Sanchez) - Family names from Scholar in all caps are handled correctly in ReadGS
- Functions for interacting with PubMed return language of each article (h/t Dale Steele)
- Added CITATION file
- Updated License to explicitly include GPL-2 and GPL-3
- Fix for
names<-.BibEntry - Fix for
print.BibEntrywhen entry has urldate field but no url field - Corrections for some documentation typos
- Fix pmidrelated field when
batch.mode = FALSEinGetPubMedRelated - Fix for
LookupPubMedIDwhenindexargument specified open.BibEntrynow works properly- Fix for converting thesis entries in
toBibtex.BibEntry - Fix for
WriteBibwithbiblatexargument
- Added Vignettes including user manual and Rmd citation examples
- Added NEWS
- Added HTML output of Rmd and RHTML citation examples to doc/