The citation label is formed by these rules, easily applicable by a human, or by a computer program like this one:
- (1)
- Take the first author's last name, dropping apostrophes, Jr/Sr/generation numbers, protecting braces, and eliminating accents (e.g., J{\"a}nsch -> Jaensch, and Jind\v{r}ich -> Jindrich), using multi-letter transliterations if that is conventional. Preserve hyphenated names, like Baeza-Yates, in full.
- (2)
- Append a colon.
- (3)
- Append the four-digit year of publication.
- (4)
- Append another colon.
- (5)
- Pick the initial letters of at most three of the leading important words in the title that begin with a letter, excluding articles, prepositions, and TeX math mode, and append those letters.
For example, given the title {Euler}'s Constant to $1271$ Places, (from an article by Donald E. Knuth in Mathematics of Computation, 16(79) 275--281, July 1962) this recipe produces ECP.
- (6)
- If the resulting citation label is already in use, add a letter a, b, c, ... to make it unique. In those rare cases when there are more than 26 such collisions, add additional letters, producing suffixes written in a base-26 number system in ascending order: a..z, aa..az, ba..bz, ..., za..zz, aaa..aaz, ..., zza..zzz, aaaa..aaaz, ..., zzza..zzzz, ....
This will produce a label like Smith:1994:ABC.
The reason for including a four-digit year is that the worldwide Y2K problem at the millennium change amply demonstrated the foolishness of two-digit year abbreviations. Also, some bibliographies may be historical, with entries dating back hundreds of years. Using a four-digit year will keep sorts of otherwise identical keys in chronological order, and putting the year before the key derived from the title will facilitate sorting by year by, e.g., bibsort(1).
Because any change in citation labels must be accompanied by a change in citations in all documents that use the bibliography, it is not sufficient to just produce a new bibliography file with changed labels. Consequently, the output of biblabel is expected to be saved, and subsequently used with citesub(1) to actually carry out the substitutions efficiently. If no documents other than the bibliography file itself need to be changed, then a simple UNIX or IBM PC DOS pipeline of the form
biblabel <foo.bib | citesub -f - foo.bib >foo.new
will produce a new bibliography file with all of the citation labels changed to the new standardized form.
To avoid confusion between labels with common prefixes, such as Smith80 and Smith80a, citesub(1) will check for leading context of a left brace, quote, comma, whitespace, or beginning of line and trailing context of a right brace, comma, quote, percent, whitespace, or end of line so as to match these styles:
@Book{Smith:1980:ABC,
crossref = "Smith:1980:ABC",
crossref = {Smith:1980:ABC},
\cite{Smith:1980:ABC}
\cite{Smith:1980:ABC,Jones:1994:DEF}
\cite{%
Smith:1980:ABC,%
Jones:1994:DEF%
}
Created labels are guaranteed to be unique within the input files provided on the command line.
However, in a larger project, one may wish to exclude labels that are already in use in other bibliographies. To provide for this, the --used-file option can be specified to define the name of a file of labels that are already in use.
When a label is found in use, and the current file matches the in-use label filename, the label is considered to be unused; otherwise, repeated runs through this program would keep changing already-assigned labels.
All options are parsed before any input bibliography files are read, no matter what their order on the command line.
The leading hyphen that distinguishes an option from a filename may be doubled, for compatibility with GNU and POSIX conventions. Thus, -author and --author are equivalent.
Except for the options described below, all other command-line words are assumed to be input files. Should such a filename begin with a hyphen, it must be disguised by a leading absolute or relative directory path, e.g., /tmp/-foo.bib or ./-foo.bib.
This option may be abbreviated --c.
Multiple --corporate-file options may be specified.
See also the INITIALIZATION FILES section below.
To avoid disastrous overwriting of a bibliography file in the event of a command-line mistake, the output file must not yet exist.
Unlike the other dump options, which are processed at startup time, this option is processed only at the end of a successful execution.
To avoid disastrous overwriting of a bibliography file in the event of a command-line mistake, the output file must not yet exist.
To avoid disastrous overwriting of a bibliography file in the event of a command-line mistake, the output file must not yet exist.
Multiple --ignore-file options may be specified.
See also the INITIALIZATION FILES section below.
Without this option, a braced corporate author/editor string of {Free Software Foundation} is reduced to FSF. With this option, it becomes Free-Software-Foundation.
Single-word corporate names are never abbreviated to an initial: IBM remains that way, instead of being reduced to I.
Each line consists of a whitespace-separated pair of filename and citation label. Inclusion of the filename in which the label is already in use is required, both so that it can be used in diagnostic messages, and to avoid unnecessary changes to labels in the current file.
This option can be used in a multi-file bibliography collection to guarantee unique citation labels across the entire collection.
Multiple --used-file options may be specified.
See also the INITIALIZATION FILES section below.
These files may contain:
Each line consists of a whitespace-separated pair of filename and citation label.
To make it possible to override the built-in ignore list, if the special word @RESET@ appears in any ignore-list file, then it, and all entries in the internal list, are immediately forgotten.
For consistency, this special word is also recognized in labels-in-use files, but has limited utility since there is no built-in citation label table. It could nevertheless be useful if biblabel were wrapped inside another script, or a shell alias, which themselves provided command-line initialization files.
Here is the built-in ignore list, taken directly from the output file created by the --dump-ignore-file option. To conserve space here, the original one-word-per-line list has been reformatted into paragraphs of words with common initial letters.
%% Title: Dump of ignore list %% CreationDate: Fri Mar 9 10:16:31 MST 2001 %% Creator: biblabel version 0.04 [06-Mar-2001] %% For: Nelson H. F. Beebe <beebe@suncore.math.utah.edu> %% Directory: /u/sy/beebe/tex/biblabel/biblabel-0.04 a ab aber als also am an and any are as at auf aus aux away az be bei bin bir bist but by cum da dans das dat de dei dem den der des det di die dos down e een eene egy ei ein eine einen einer eines eit el en er es et ett eyn eyne for from fuer fur gehabt gl gli ha hab habe haben habt had haette hai has hast hat hatte have he heis hen hena henas het hin hinar hinir hinn hith ho hoi i il in into is ist its ka ke l la las le les lo los mia mit n na ne nicht nji not o oben oder of off ohne on onto or os others out over pas s seid sie sind so sur t ta that the these this those to uber um uma un una und une uno unter unto up via vom von with without y yr zu zum zur
biblabel also has a much shorter built-in list of words to be ignored when forming corporate name abbreviations. The list is small, because all words from the main ignore list are also excluded when forming such abbreviations. biblabel augments the built-in list with the contents of any biblabel.cig file in the current directory, plus any files specified with --corporate-file options.
Here is the built-in corporate ignore list, taken directly from the output file created by the --dump-corporate-file option. To conserve space here, the original one-word-per-line list has been reformatted into paragraphs of words with common initial letters.
%% Title: Dump of corporate ignore list %% CreationDate: Fri Mar 9 10:22:54 MST 2001 %% Creator: biblabel version 0.04 [06-Mar-2001] %% For: Nelson H. F. Beebe <beebe@suncore.math.utah.edu> %% Directory: /u/sy/beebe/tex/biblabel/biblabel-0.04 co company corp corporation gmbh group inc incorporated limited ltd staff team
There are no internal defaults for the list of citation labels already in use. That list is initialized from any biblabel.use file in the current directory, plus any files specified with --used-file options.
Here, in alphabetical order, are the messages that biblabel can produce, with brief explanations. Each is prefixed with an uppercase word denoting the severity level.
This practice may be changing in American English: the widely-followed Chicago Manual of Style, 14th ed., University of Chicago Press (Chicago and London), 1993, ISBN 0-226-10389-7, notes in section 8.55 on p. 307:
Traditionally, Jr. and Sr. have been set off with commas, whereas I, II, III, IV, and so on have not. This tradition is still widely followed, and the University of Chicago Press recognizes and accepts it; but the Press now also accepts, and in fact recommends, that the commas be omitted in both cases.
On the other hand, the MLA Handbook for Writers of Research Papers, 4th ed., Modern Language Association of America (New York), 1996, ISBN 0-87352-565-5, in section 4.6.1 on p. 110, wants a comma before all such suffixes.
Strunk and White's The Elements of Style 3rd ed., Macmillan (New York), 1979, ISBNs 0-02-418230-3 and 0-02-418220-6, on p. 3 says to omit the comma before Jr., but to include it before Ph.D. and S.J. suffixes.
Words into Type, 3rd ed., Prentice Hall (Englewood Cliffs, NJ), 1974, ISBN 0-13-964262-5, recommends the comma before Jr. and Sr., but notes that newspapers frequently omit it.
The ACS Style Guide, 2nd ed., American Chemical Society (Washington, DC), 1997, ISBNs 0-8412-3461-2 and 0-8412-3462-0, requires a comma before Jr. and Sr., but says to treat roman numeral suffixes according to the person's preference.
In other words, style manuals don't agree! biblabel's diagnostic is thus only informational.
This message cannot be raised unless the underlying awk(1) implementation supports 8-bit characters in regular-expression patterns. gawk(1) and mawk(1) do, but others may not.
Nelson H. F. Beebe
University of Utah
Department of Mathematics, 110 LCB
155 S 1400 E RM 233
Salt Lake City, UT 84112-0090
USA
Email: beebe@math.utah.edu, beebe@acm.org,
beebe@computer.org (Internet)
WWW URL: http://www.math.utah.edu/~beebe
Telephone: +1 801 581 5254
FAX: +1 801 581 4148
ftp://ftp.math.utah.edu/pub/tex/bib/ http://www.math.utah.edu/pub/tex/bib/index-table-b.html#biblabel
in the files
where x.yy is the current version. Each of the popular archive format unpacks into an identical distribution tree in a subdirectory named biblabel-x.yy. [Caution: older software distributions may omit the leading subdirectory prefix in some archive formats.]biblabel-x.yy.jar biblabel-x.yy.shar.gz biblabel-x.yy.tar.gz biblabel-x.yy.zip biblabel-x.yy.zoo
That site is mirrored to several other Internet archives, so you may also be able to find it elsewhere on the Internet; try searching for the string biblabel at one or more of the popular Web search sites, such as
http://search.microsoft.com/ http://www.altavista.com/ http://www.dejanews.com/ http://www.dogpile.com/ http://www.euroseek.net/ http://www.excite.com/ http://www.go2net.com/ http://www.google.com/ http://www.hotbot.com/ http://www.infoseek.com/ http://www.inktomi.com/ http://www.lycos.com/ http://www.northernlight.com/ http://www.snap.com/ http://www.stpt.com/ http://www.websmostlinked.com/ http://www.yahoo.com/
######################################################################## ######################################################################## ######################################################################## ### ### ### biblabel: generate standardized BibTeX citation labels ### ### ### ### Copyright (C) 1994, 1996, 1997, 2001, 2006 Nelson H. F. Beebe ### ### ### ### This program is covered by the GNU General Public License (GPL), ### ### version 2 or later, available as the file COPYING in the program ### ### source distribution, and on the Internet at ### ### ### ### ftp://ftp.gnu.org/gnu/GPL ### ### ### ### http://www.gnu.org/copyleft/gpl.html ### ### ### ### This program is free software; you can redistribute it and/or ### ### modify it under the terms of the GNU General Public License as ### ### published by the Free Software Foundation; either version 2 of ### ### the License, or (at your option) any later version. ### ### ### ### This program is distributed in the hope that it will be useful, ### ### but WITHOUT ANY WARRANTY; without even the implied warranty of ### ### MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ### ### GNU General Public License for more details. ### ### ### ### You should have received a copy of the GNU General Public ### ### License along with this program; if not, write to the Free ### ### Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, ### ### MA 02111-1307 USA ### ######################################################################## ######################################################################## ########################################################################