BibArts is a package to administer bibliographical references in footnotes, and for creating a bibliography from these references simultaneously; it requires a program, for which source and Windows executable are provided. (A summary of contents is in English; the full documentation is in German.) === BibArts 2.6 is a LaTeX package to assist in making bibliographical features common in the arts and humanities (history, political science, etc.). bibarts.sty provides commands for quotation, register keywords, abbreviations, and especially for a formatted citation of literature, journals (periodicals), published documents, and unpublished archive documents. BibArts will also copy the arguments of all those commands into lists for an automatically generated appendix. These lists are optionally referring to page and footnote numbers in your text (index). BibArts has nothing to do with BibTeX, and it does not use any data bank except your own LaTeX text. The lists are created by bibsort. A file bibsort.exe is part of the package ( CTAN mirrors > BibArts > bibarts.zip ) and runs on newer Windows systems. Other users first have to create a binary file from bibsort.c (ANSI-C) with their own C-compiler. BibArts 2.6 is developed and tested on Windows 10 with the 2022-11-01 LaTeX 2e distribution, but it should be able to work even with LaTeX 2.09. BibArts 2.6 (9 files, 8 dated 2023-06-18): README.txt This file here: Version history bibarts.sty The LaTeX style file ba-short.pdf Short introduction (English) ba-short.tex Source of ba-short.pdf bibarts.pdf Full documentation (German) bibarts.tex Source of bibarts.pdf bibsort.exe bibarts-bin to create the lists bibsort.c Source of bibsort.exe COPYING License (dated 1993-11-28) === Changes from BibArts versions 1.x (1990s) to versions 2.x: Version 2.0 was a completely new package with massive extensions. Since, bibarts.sty helps to use slanted fonts (italics), and is able to set ibidem automatically in footnotes. Therefore, it is now possible to add volume and page numbers e.g. to the \vli command (\vli did also exist in 1.x for full references to literature), and the new \kli command (shortened references). Pre-defined text elements (captions) are provided as \bacaptionsenglish, \bacaptionsfrench, and \bacaptionsgerman (default). bibsort 2.x creates the index numbers (BibArts 2.x does not use MakeIndex). bibarts.sty 2.x starts an emulation for 1.3 texts, when you type \makebar, but you better keep copies of the package files of a BibArts 1.x, if you wrote texts with it. BibArts 2.x uses .aux files instead of a .bar file. Even if you set \makebar, any changes to commands \schrift, \barschrift, and \indschrift will be ignored. \verw and \punctuation do not exist any more; see examples for the new commands \frompagesep and \ntsep in bibarts.pdf. === Changes from BibArts version 2.0 (2015) to version 2.1 (2016): BibArts 2.0 set \footnotesep to 2ex, whereas 2.1 to 2.6 do *not* change the pre-set value. If you want to continue with the 2.0-distance between two footnotes, you will have to type \setlength{\footnotesep}{2ex} in your text. Some of the pre-defined text elements (captions) have been modernized. The \evkctitlename changed from {Short Titles} to {Shortened References}. And \gannouncektitname changed to "... im Folgenden". To restore its 2.0 def.: \renewcommand{\gannouncektitname}{ (\kern 0.015em im folgenden \baupcorr} bibarts.sty will even be loaded, if ~":;!?'`<> are active (catcode 13); and bibsort -g1, and bibsort -g2 are sorting "z (with an active ") as \ss now. You now may choose your own order of page and footnote numbers in the index (roman--arabic, arabic--roman, etc.). Type bibsort -s2 xxxx for page and ... -f2 xxxx for footnote numbers. xxxx are permutations of 4 letters out of nRrAas (a=alph, A=Alph, n=arabic, R=Roman, r=roman, s=fnsymbol); you always have to set n and s, and to choose R *or* A, and r *or* a. E.g. srnR means, that you can use \Roman in your text, but you do not have to. bibsort is able to evaluate the new fnsymbols (which expand to \TextOrMath). If bibsort should write into files with a different prefix as the .aux input file, you have to use -o now. And you may type bibsort , *or* bibsort -i (e.g. if the input file name begins with '-'). bibsort sorts the 'official' $Greek variables$ since version 2.0. To write single words in Old Greek, BibArts 2.1 also provides \Alpha [A], \Beta [B], \Epsilon [E], \Zeta [Z], \Eta [H=sort=>E], \Iota [I], \Kappa [K], \Mu [M], \Nu [N], \Rho [P==>R], \Tau [T], \Chi [X==>Ch], \Omicron [O], \omicron [o]. === Changes from BibArts version 2.1 (2016) to version 2.2 (2019): (1) bibsort, the BibArts Program to create the lists, has 3 new options: Use bibsort -h [...] to sort hyphens ( - \hy \fhy "= "~ ) as spaces. Use bibsort -b [...] if you want that your command \bago in an argument should define the position at which each comparison begins. E.g. ... \vli{J.}{Smith}{The \bago \ktit{Book}, London 2005} ... with -b will be sorted like without 'The'. (If you did not type \bago, the position is at the begin of the argument.) bibsort -n1 [...] will create blocks of footnote numbers only, if the entries are from footnotes on the same page, e.g.: 10^{3-6}, 11^{7-9} instead of 10^{3}--11^{9}. (2) If there are two entries of \kli{Smith}{Book} from two successive pages, but with *the same* footnote number, bibsort will print out e.g. 10^{3}, 11^{3} . bibsort 2.0 and 2.1 ignored 11^{3} in that case. (3) bibsort checks your input concerning the \ktit{Shortened Title}. You will find warnings on the screen and as comments in the created files. (4) \vli{J.}{Smith}{The \ktit{\onlyvoll{r}\onlykurz{R}ed Book}, 2006} has to be used (for index and ibidem), if the shortened title is not just a part of the full title ("red" and "Red"). In some cases, bibsort 2.1 printed out wrong only-arguments on the lists. This error has been fixed. And errors in the sorting of entries with only-commands were eliminated. (5) You may say now \renewcommand{\ntsep}{\bapoint\ } to print out a full stop between name and title; BibArts is avoiding e.g. "Smith, J..". (6) If you '\renewcommand' a pre-defined text element (caption), you sometimes will have to adapt the italic correction. BibArts provides the homogeneous command \bacorr instead of the four different commands \bakxxcorr, \baabkcorr, \balistcorr, and \bakntsepcorr (but those old commands are still valid). To print the lists, you may say now e.g.: \renewcommand{\frompagesep}{\bacorr ; } \printnumvli \printnumvkc (7) If you use BABEL, BibArts will not execute \originalTeX between list items (that's prepared by bibsort, if the catcode of " changes): It was useless with BABEL-German and caused errors with BABEL-French. With german.sty or ngerman.sty, the use of \originalTeX is still reproduced. [... but see BibArts 2.6.] (8) bibarts.sty provides new commands for the hyphenation in German: - \oldhyss to set a \ss, which is splitted s-s (to be used between vovels, e.g. au\oldhyss er; you have to use your sharp S else, e.g. da{\ss}); - \newhyss to set a \ss, which is splitted s-s only in Small Caps, and else is printing a sharp S, which is splitted as \language says (-\ss); - \hyss to execute \oldhyss, if \language is \l@german, and \newhyss else (e.g. if \language is \l@ngerman), so you can use \hyss in all languages; - commands to handle the "tripple consonant in front of vovel" in German: E.g. Sto\hyf figur, Scha\hyl leistung, Sta\hym mutter, Ke\hyn nummer, Ste\hyp pullover, Sta\hyr rahmen, and Schri\hyt tempo: If \language is \l@german, BibArts will PRINT AND SORT \hyf as f, and ELSE as ff, etc. + All commands are ready to be used in \MakeUppercase (also Dru\hyc ker). (9) If you \conferize, e.g. \kli{A}{B} will print a cross-reference to \vli{}{A}{\ktit{B}}. The INTERNAL KEYWORD is created from the letters, and some accents you use in {A} and {B}. The accents are REPRESENTED as \`=[ \'=] \"=* \ss=(ss \l=(l \L=(L \o=(o \O=(O \^=) \~=- \c=+ in the keyword. The active " and \= \. \b \d are not represented any more. === Changes from BibArts version 2.2 (2019) to version 2.3 (2021): There have been two new options for bibsort: -utf8 sorts UTF-8 characters as definded in utf8enc.dfu 2021/01/27 v1.2l, except \cyr...- and \CYR...-commands, which bibsort still ignores. bibsort is able to sort the accent \textcommabelow (which LaTeX uses from U+0218 to U+021B). The internal representation of accents did change (two instead of one ASCII 1 to 31). For numerous smaller changes see /* 2.3 comments */ in bibsort.c. Like in earlier versions, you can introduce e.g. a book in your text \xvli{John}{Smith} *{\vauthor{Peter}{Young}} {... [Title] ...} with Peter Young as co-author. (For further co-authors, use \midvauthor{FirstName}{LastName} in front of \vauthor.) Please always TYPE the FirstName first. -e exchanges the PRINT ORDER of the args of \vauthor (and \midvauthor) on the List of Literature. So, \printvli will PRINT: Smith, John / Young, Peter: ... [Title] ... If you start bibsort without -e, \printvli will still print: Smith, John / Peter Young: ... [Title] ... The SORT ORDER in *both* cases is (that did/does *not* change): Smith John Young Peter === Changes from BibArts version 2.3 (2021) to version 2.4 (2022) [2022a]: An error was fixed concerning "inner v-commands" on the List of Literature, and the List of Published Documents: In shortened "inner" references, BibArts 2.3 -- together with bibsort -e ... -- printed the first names of co-authors. Now, such shortened references only print the family names. bibsort now knows ten new UTF-8 characters (according to the changes in utf8enc.dfu from version 2021/01/27 v1.2l to 2021/06/21 v1.2n). === Changes from BibArts version 2.4 (2022[a]) to version 2.5 (2022b): (1) You may now put curly brackets around the \ktit{shortened title}, e.g.: \vli {K.} {Mayer} {The \textbf {\ktit{Book}}, Frankfurt/M.}. Older versions, despite \announcektit, did not print "(cited as ...)" in that case. Now, this error has been fixed, and \vli prints correctly: K. Mayer: The Book, Frankfurt/M. (cited as Mayer: Book [L]). (2) bibarts.sty now searches for a full stop at the end of the last argument of \vli commands. So, you may set \notannouncektit, and you will get: K. Mayer: The Book, Frankfurt/M. (BibArts 2.5 automatically prints "." instead of ".." at the end.) (3) BibArts now supports the name-year system. In this feature, the last argument of a shortened reference \kli consists of the year of publication (instead of a few words from the full title). In this new system, you also use \ktit in the full reference \vli, but in the optional *{argument} as: \footnote{\vli {J.} {Smith} *{\onlyvoll{\ktit{2022}}} {An Alternative}.} ... \footnote{\kli {Smith} {2022}.} %% <= automatically prints IBIDEM. The \ktit in the *{argument} of the full reference \vli has to be masked by \onlyvoll; only then, BibArts will print IBIDEM, if \kli{Smith}{2022} is in the following footnote. Instead of especially *{\onlyvoll{\ktit{2022}}}, you may use *{\starktit{2022}} as a shorter synonym; also see ba-short.pdf. After a start of the BibArts program bibsort, and a second start of LaTeX, \printvli prints the List of Literature from the given name-year example as: Smith, J. 2022: An Alternative. (4) Anyway, \ktit now is no more necessarily in the last argument of a \vli command. The new bibsort tells you whether you have always chosen the same argument (where "arg 3" means the optional *{argument}, and "arg 4" the last argument). To get the new features, bibsort, and bibarts.sty, both need to be updated to version 2.5. bibsort is sometimes referred to as bibarts-bin. (5) bibsort now knows further eleven UTF-8 characters (according to the changes in utf8enc.dfu from version 2021/06/21 v1.2n to 2022/06/07 v1.3c). === Changes from BibArts version 2.5 (2022b) to version 2.6 (now, 2023): Version 2.6 improves the use of BibArts with BABEL and UTF-8. For this purpose, some commands have been expanded: \baoriginalTeX and \bagermanTeX (for reproducing the catcode of " on BibArts' lists), \hy and \fhy (for printing a hyphen with a kerning), \abra and \fabra (for editorial additions in adjusted square brackets). BibArts 2.6 is checking whether LaTeX defines `\inputencodingname' as `utf8'; if not, [f]hy and [f]abra will not interpret UTF-8 multi-octets. Commands for compatibility with older versions of BibArts are added: \notnewhyko and \notnewabra are downgrading BibArts 2.6 to the older interpretation of [f]hy and [f]abra commands. (1) When you print a BibArts list (e.g., if you use \printvli), the list items are copies of the arguments of BibArts' commands in your LaTeX text file (e.g. \vli{FirstName}{LastName}{Title ...}). If such entries originate from environments with *different catcodes* for the double quote [U+0022], BibArts will reproduce such changes on its lists (from active " to passive " or vice versa). Now, additionally BABEL is supported for this purpose: (a) BibArts 2.6 tests whether you have loaded BABEL; if NOT, the new BibArts estimates (IF the catcode of " changes AND " is USED in a respective entry) that german.sty or ngerman.sty is loaded, *and* that the reason is that YOU have used \originalTeX, and \germanTeX or \ngermanTeX. So, on BibArts' lists, like in older versions, the automatically inserted \baoriginalTeX executes \originalTeX, and \bagermanTeX executes \germanTeX or \ngermanTeX. (b) BABEL's command \originalTeX seems never to change the " catcode, but the MEANING of the active double quote: {\originalTeX "`} is printing *two* characters. BibArts 2.6 is *not* able to reproduce BABEL's \originalTeX. Also BABEL's command \selectlanguage apparently does not change any catcode. If the catcode of " changes, older BibArts versions since BibArts 2.1 simply did nothing when BABEL was loaded. Now, BibArts 2.6 estimates that you have used BABEL's commands \shorthandoff{"} respectively \shorthandon{"} in your LaTeX file, and BibArts executes both commands on its lists - to reproduce the conditions of the place in your text where the entries come from. If you have loaded e.g. \usepackage[ngerman]{babel}, or - despite BibArts 2.6 provides no Dutch captions - \usepackage[dutch]{babel}, you are free to say \shorthandoff{"} ... \vli{}{M"uller}{Title} ... \shorthandon{"} ... \vli{}{M"uller}{Title} ...; and BibArts' sorting program bibsort will detect that: A start of bibsort will create two items (instead of one) on the lists (e.g. \printvli); and, if you start bibsort with option -g1 to sort [U+00FC] and "u as 'ue', then the \shorthandoff{"}-entry is sorted as Muller (BABEL prints a double quote character between 'M' an 'u'), whereas the \shorthandon{"}-entry is sorted as Mueller (BABEL prints "u as 'ü'): What you see is what is sorted. (c) Anyway, if it has a different meaning in your LaTeX text file when the catcode of " changes, you must still redefine the corresponding BibArts list commands. If you have loaded e.g. [n]german.sty, and YOU use \mdqoff and \mdqon (instead \originalTeX, and \germanTeX or \ngermanTeX), then define: \renewcommand{\baoriginalTeX}{\mdqoff} %No user commands! bibsort inserts \renewcommand{\bagermanTeX}{\mdqon} %\baoriginalTeX and \bagermanTeX. Such redefinitions, if ordered, will be executed ALWAYS: They OVERWRITE the case differentiation between (a) [n]german.sty and (b) BABEL. Together with BABEL, \mdqoff and \mdqon are known commands, *if* you have loaded BABEL with options [n]german or [n]austrian or [n]swissgerman. Then, \mdqoff and \mdqon simply execute \shorthandoff{"} and \shorthandon{"}; and thus all "X will be reproduced correctly - with or without both redefinitions. Commands which are (only) changing catcodes have one problem in common: If you place such a command INSIDE THE ARGUMENT of e.g. \footnote{...}, then it will not change what footnote text looks like ("u is PRINTED as without the command). The decisive catcode is the one which is valid when LaTeX BEGINS to read an argument. To achieve an impact, you have to say e.g.: \mdqoff ... .\footnote{\vli{}{Harry}{The "Big" Book}[2].} ... .\mdqon \footnote{\vli{}{M"uller}{Title}[3].} If you, despite, use such commands INSIDE footnote text, then older versions reproduced the changed VALUE of the catcode on BibArts' lists (e.g. M"uller was printed different on the list than in the footnote). Now, BibArts 2.6 makes the entry on a BibArts' list look exactly like the arguments of a BibArts' command which you use in a footnote. (But there is only a solution for active " in BibArts' commands used in footnotes and minipage footnotes.) (d) The following creates two items on \printvli now instead of three items: \shorthandon{"} \vli{}{"`M"'}{}. \shorthandoff{"} \vli{}{"`M"'}{}. \shorthandon{"} \vli{}{"`M"'}{}. (A use of „M“ = [U+201E]M[U+201C] instead of the German "`M"' creates one item; but the distance between M and “ is appropriate in T1, not OT1.) (2) \hy is printing a hyphen ('-') which, in difference to the normal minus character, allows to split a following word (word division at the end of a line). If \hy is followed by certain characters or letters, there will be a kerning additionally. The characters are STILL ` and ', \glq, ), ], \} and STILL some Latin letters: A, T, v, V, w, W, x, X, y, Y. That means that e.g. \hy T prints -T with a smaller distance between '-' and 'T'. Use \fhy instead \hy to prevent a line break between the '-' and following text. The kerning is even done when the letter has ONE accent: \", \., \=, \^, \', \`, \~, \b, \c, \d, \H, \k, \r, \u, \v, or the ACTIVE ". Now, \t and \textcommabelow are new among the accents as well as capital-accents like \capitalcedilla; and a kerning is also done in front of \AA, \lq, \grq, \rq, and (accent plus) \AE. If \hy or \fhy are in Small Caps, the new BibArts 2.6 does a kerning in front of \aa as well as (accent plus) \ae, a and t. \hy and \fhy now accept all equivalent UTF-8 multi-octets, as far as they are defined in utf8enc.dfu (2022/06/07 v1.3c); see the list in bibarts.pdf. Today, \hy Ärmel [U+00C4] or \hy Ångstrom [U+00C5] etc. are handled in the same way as \hy\"Armel or \hy\r{A}ngstrom. Instead of \hy ‘ [U+2018], you may also use \hy\textquoteleft or \hy\lq, and instead of \hy ’ [U+2019] you may also use \hy\textquoteright or \hy\rq now. Note that there is no kerning when \hy is followed by a comma (Ober\fhy, Mittel- und Unter\hy Bau) - in difference to \hy\glq Hallo\grq or \hy‚Hallo‘ ([U+201A], [U+2018]). If ` or ' are ACTIVE, \hy and \fhy, since BibArts 2.6, will insert the same kern as in front of the PASSIVE ` or '. With BABEL-activeacute-activegrave- catalan, you have to use e.g. \hy\`E or \hy È [U+00C8] instead \hy`E. It is *not necessary* for you to know whether characters are UTF-8 multi- octets: (i) If \hy or \fhy are immediately followed by a letter, then braces are never necessary; and if used, such braces simply will switch off any kerning. (ii) An accent may always be followed by braces {...}, e.g. \hy\"{A}. If you do not know whether you have an UTF-8 multi-octet, it will be better to use curly braces, because LaTeX itself WANTS braces in case of multi-octets as e.g \b{Ä} [U+00C4] - despite, as mentioned above, \hy and \fhy will not perform a kerning if letters have more than one accent. An error is fixed concerning the hyphenation of a word as e.g. \hy \"Armel (with [n]german.sty also \hy "Armel): In *T1*, it now can split at the end of a line. (The [n]german-BABEL does not split "Armel, while Ärmel would.) If you have used older versions of BibArts in one of your existing LaTeX text files, you can use BibArts 2.6 and type \notnewhyko. Then \hy and \fhy will not handle the new accents and characters; in particular, no kerning is done in front of UTF-8 multi-octets; and e.g. \hy\"Armel will not split Är- mel at the end of a line: Concerning \hy and \fhy, \notnewhyko starts a compatibility mode and any hy-kerning is done as with BibArts 2.5. \nothyko still switches off the kerning of \hy and \fhy. In *typewriter fonts*, where every character (including '-') should have the same width, the kerning is switched off automatically since BibArts 2.6. If you use \notnewhyko, you still have to set \nothyko in a typewriter environment yourself, or you may use the normal minus character there. Also, as in older versions, e.g. \hy{}T switches off the kerning between '-' and 'T'. (3) In quotations, you may use e.g. \abra{sic!} for editorial comments and for adding characters, which are missing in the text from which you are quoting from. On default, BibArts 2.6 CONTINUES to print this additions in square brackets, e.g. [sic!]. Use \fabra to prevent a line break between the closing bracket and following characters, e.g.: `misspel\fabra{l}ed' => `misspel[l]ed'. Certain arguments automatically are printed in subscript or superscript brackets: E.g. ``This is \fabra{`}strange'.'' (where the `inner' quotation may represent `...' as well as ``...'' in the original). The tiny square brackets cannot be shown here; but size (\tiny) and position of those square brackets are adjusted to the size and position of certain arguments, e.g.: \abra{.} \abra{,} \abra{-} \abra{...} \abra{\dots} The new \abra and \fabra also print the corresponding UTF-8 multi-octets in such tiny superscript or subscript brackets, e.g. \abra{…} [U+2026]. Alternatively, you can now use the corresponding LaTeX command, e.g. \abra{\textellipsis} - in addition to \abra{\dots}. Today, \abra{“} [U+201C] or \abra{\textquotedblleft} - and together with [n]german: \abra{"'} or \abra{\grqq} - can be used, each of which prints German right quotes in tiny brackets. For more examples see bibarts.pdf (version 2.6): "Argumente, bei denen \abra und \fabra angepasste Klammern drucken". The characters ` ' < > can be active or not; since BibArts 2.6, therefore, e.g. \abra{"<} and \abra{">} from [n]german would be accepted with active or passive < and > to print French quote characters. BABEL seems to accept \glqq, \grqq, \flqq and \frqq as well as \glq, \grq, \flq and \frq ALWAYS; so you can use e.g. \abra{\flq} as synonym for \abra{‹} [U+2039] - even if \usepackage[UKenglish]{babel} or \usepackage[french]{babel} is loaded. If you have used older versions of BibArts in your existing LaTeX text files before, you can use \notnewabra to avoid a profound change of your existing line break: With \notnewabra, only the arguments, which already older versions printed in adjusted brackets, will be printed in tiny brackets, whereas the `new' arguments will be printed as before (inside normal size brackets). But explicit errors will not be reproduced, e.g. concerning LaTeX' mathematical mode: In [n]german, BibArts 2.6 prints $\abra{"`}$ and $\abra{"'}$ inside normal-sized brackets, whereas BibArts 2.5 did produce an error message. In text mode, \abra{"`} and \abra{"'} are STILL printed in tiny brackets, but only if both arguments are commands to print double quote characters: If BABEL prints "` and "' as two characters, then normal-sized brackets [] are printed, regardless of whether you typed \notnewabra or not. And you now can use e.g. \babbrack{$_{1}$}, $ a\bambrack{\cdot}b=c $, or 30\batbrack{°} to print the tiny brackets at *b*ottom, *m*edium, or *t*op positions directly (but WITHOUT any automatic centring of the argument). For symmetry reasons, also \fbatbrack, \fbambrack and \fbabbrack exist. "french"-BABEL inserts space in front of active two-part punctuation marks. If you load \usepackage[french]{babel}, only :;!? are active characters, and BibArts will not do anything to reproduce changing catcodes on its lists. But BibArts 2.6 now also prints \abra{:}, \abra{;}, \abra{!} and \abra{?} with active :;!? as [:], [;], [!] and [?]; also on lists, no extra space is inserted (with \notnewabra, still [ :], [ ;], [ !], [ ?] is printed). But BibArts 2.6 still does not handle !` and ?`. "french"-BABEL itself has a solution, when you can not use \abra{¿} [U+00BF] and \abra{¡} [U+00A1]: E.g. {\NoAutoSpacing \abra{?`}}. This seems to work even INSIDE ARGUMENTS. In difference to e.g. D[a]mpfschifffahrt, LaTeX, at the end of a line, can split D\fabra{a}mpfschifffahrt (German word division: D[a]mpf-schiff-fahrt). In typewriter fonts you can use e.g. \banbrack{.} or \fbanbrack{.} to print brackets in *n*ormal size (tiny brackets would have an inappropriate width). The arguments of BibArts' ...abra and ...brack commands are always printed `upshape', and all, except the `bottom' versions, do an italics correction. E.g. \textit{(H.\bahasdot\abra{'})} switches the italics correction off. If you use BibArts 2.6, you will be able to say \renewcommand{\pabrao}{(} and \renewcommand{\pabrac}{)}. This will alter \abra, \fabra, \babbrack, \fbabbrack, \bambrack, \fbambrack, \batbrack, \fbatbrack, \banbrack and \fbanbrack. Then, e.g. \abra{sic!} will print (sic!) instead of [sic!]. (4) bibsort 2.5 with options -g1 and -g2 sorted "i, "I, "e and "E as i, I, e and E. In the same case, bibsort 2.6 sorts `active double quote in front of letter' as `letter with dieresis' now. Also see NUMGER in bibsort.c. (5) BibArts' support of german.sty 2.4a (1991/04/12, where e.g. ..."' expands to ...\active@dq \dq@prtct {'@} in the .toc-file) may end in the near future. So, if you have used BibArts together with such very old German style files: Keep back a copy of bibarts.sty 2.6. - In difference, the support of ngerman.sty v2.5e and german.sty v2.5e (1998/07/08, where ..."' expands to ...\grmn@active@dq@ ' in the .toc-file) probably will continue [[but the support of ...\active@dq \dq@prtct {'} from the compatibility mode which is included in v2.5e will probably also END]]. - Features relating to BABEL will probably be extended. === Published under the terms of the GNU General Public License. BibArts 2.6 (2023): (C) Timo Baumann, Versions 2.x (2015-2023)