CTAN Comprehensive TeX Archive Network

CTAN up­date: xint

Date: Jan­uary 8, 2017 10:19:17 AM CET
Jean-François Burnol sub­mit­ted an up­date to the xint pack­age. Ver­sion num­ber: 1.2k 2017-01-06 Li­cense type: lp­pl1.3 Sum­mary de­scrip­tion: Ex­pand­able op­er­a­tions on long num­bers An­nounce­ment text:
- The xint­frac.sty float­ing point macros since 1.2f round their ar­gu­ments to the tar­get pre­ci­sion P be­fore fur­ther pro­cess­ing. This round­ing is now ex­act (aka cor­rect) in all cases, even with frac­tions hav­ing long nu­mer­a­tors and de­nom­i­na­tors. This change has lit­tle in­flu­ence on float ex­pres­sions, as the \xint­floa­t­expr parser han­dles there the / sym­bol as an op­er­a­tor hence it does not (ex­cept for spe­cial con­structs) get to see frac­tions as such. - Half-in­te­ger pow­ers A^x (only avail­able in float ex­pres­sions, not via macros) pro­ceed by an in­te­ger power and then a square-root ex­trac­tion: the 1.2f im­ple­men­ta­tion did the lat­ter on an al­ready rounded value, 1.2k keeps some of the guard dig­its to make the com­puted value Z closer to the ex­act one: a dif­fer­ence of less than 0.52 ulp(Z) is guar­an­teed in all cases. - Macro \xint­new­dummy is made a pub­lic one, it serves to de­clare ad­di­tional let­ters as dummy vari­ables in ex­pres­sions. This is for Uni­code en­gines, mainly, as all Latin let­ters are al­ready pre­de­fined to act as such. See CHANGES.pdf or CHANGES.html for the de­tails and more.
This pack­age is lo­cated at http://mir­ror.ctan.org/macros/generic/xint More in­for­ma­tion is at http://www.ctan.org/pkg/xint
Thanks for the up­load. For the CTAN Team Man­fred Lotz We are sup­ported by the TeX user groups. Please join a users group; see http://www.tug.org/user­groups.html .

xint – Ex­pand­able op­er­a­tions on long num­bers

The xint bun­dle main mod­ules are:

xint­tools
util­i­ties of in­de­pen­dent in­ter­est such as ex­pand­able and non-ex­pand­able loops,
xint­core
ex­pand­able macros im­ple­ment­ing ad­di­tion, sub­trac­tion, mul­ti­pli­ca­tion, di­vi­sion, and pow­ers for ar­bi­trar­ily long in­te­gers,
xint
ex­ten­sion of xint­core,
xint­frac
ex­tends the scope of xint to dec­i­mal num­bers, to num­bers us­ing sci­en­tific no­ta­tion and also to (ex­act) frac­tions,
xin­t­expr
pro­vides ex­pand­able parsers of nu­meric ex­pres­sions us­ing the stan­dard in­fix no­ta­tions, paren­the­ses, built-in func­tions, user de­fin­able func­tions and vari­ables (and more ...) which do ei­ther ex­act eval­u­a­tions (also with frac­tions) or float­ing point eval­u­a­tions un­der a user cho­sen pre­ci­sion.

Fur­ther mod­ules of the bun­dle are: xin­tk­er­nel (sup­port macros for all the bun­dle con­stituents), xint­bin­hex (con­ver­sion to and from hex­adec­i­mal and bi­nary bases), xint­gcd (pro­vides gcd() and lcm() func­tions to xin­t­expr), xintseries (eval­u­ates nu­mer­i­cally par­tial sums of se­ries and power se­ries with frac­tional co­ef­fi­cients), and xintcfrac (ded­i­cated to the com­pu­ta­tion and dis­play of con­tin­ued frac­tions).

All com­pu­ta­tions are com­pat­i­ble with ex­pan­sion-only con­text.

The pack­ages may be used with Plain , , or (a pri­ori) any other macro for­mat built upon .

Pack­agexint
Ver­sion1.3b 2018-05-18
Copy­right2013–2018 Jean-François Burnol
Main­tainerJean-François Burnol

An­nounce­ments

more

Guest Book Sitemap Contact Contact Author