Real-time collaboration for Jupyter Notebooks, Linux Terminals, LaTeX, VS Code, R IDE, and more,
all in one place.
Real-time collaboration for Jupyter Notebooks, Linux Terminals, LaTeX, VS Code, R IDE, and more,
all in one place.
| Download
GAP 4.8.9 installation with standard packages -- copy to your CoCalc project to get it
Project: cocalc-sagemath-dev-slelievre
Views: 4183461[1X3 [33X[0;0YChanges between [5XGAP[105X[101X[1X 4.6 and [5XGAP[105X[101X[1X 4.7[133X[101X23[33X[0;0YThis chapter contains an overview of most important changes introduced in4[5XGAP[105X 4.7.2 release (the first public release of [5XGAP[105X 4.7). It also contains5information about subsequent update releases for [5XGAP[105X 4.7.[133X678[1X3.1 [33X[0;0Y[5XGAP[105X[101X[1X 4.7.2 (December 2013)[133X[101X91011[1X3.1-1 [33X[0;0YChanges in the core [5XGAP[105X[101X[1X system introduced in [5XGAP[105X[101X[1X 4.7[133X[101X1213[33X[0;0YImproved and extended functionality:[133X1415[30X [33X[0;6YThe methods for computing conjugacy classes of permutation groups have16been rewritten from scratch to enable potential use for groups in17different representations. As a byproduct the resulting code is18(sometimes notably) faster. It also now is possible to calculate19canonical conjugacy class representatives in permutation groups, which20can be beneficial when calculating character tables.[133X2122[30X [33X[0;6YThe methods for determining (conjugacy classes of) subgroups in23non-solvable groups have been substantially improved in speed and24scope for groups with multiple nonabelian composition factors.[133X2526[30X [33X[0;6YThere is a new method for calculating the maximal subgroups of a27permutation group (with chief factors of width less or equal 5)28without calculating the whole subgroup lattice.[133X2930[30X [33X[0;6YIf available, information from the table of marks library is used to31speed up subgroup calculations in almost simple factor groups.[133X3233[30X [33X[0;6YThe broader availability of maximal subgroups is used to improve the34calculation of double cosets.[133X3536[30X [33X[0;6YTo illustrate the improvements listed above, one could try, for37example[133X3839[4X [32X Example [32X[104X40[4X[28Xg:=WreathProduct(MathieuGroup(11),Group((1,2)));[128X[104X41[4X[28XLength(ConjugacyClassesSubgroups(g));[128X[104X42[4X[32X[104X4344[33X[0;6Yand[133X4546[4X [32X Example [32X[104X47[4X[28Xg:=SemidirectProduct(GL(3,5),GF(5)^3);[128X[104X48[4X[28Xg:=Image(IsomorphismPermGroup(g));[128X[104X49[4X[28XMaximalSubgroupClassReps(g);[128X[104X50[4X[32X[104X5152[30X [33X[0;6YComputing the exponent of a finite group [22XG[122X could be extremely slow.53This was due to a slow default method being used, which computed all54conjugacy classes of elements in order to compute the exponent. We now55instead compute Sylow subgroups [22XP_1[122X, ..., [22XP_k[122X of [22XG[122X and use the easily56verified equality [22Xexp(G) = exp(P_1) x ... x exp(P_k)[122X. This is usually57at least as fast and in many cases orders of magnitude faster.[133X5859[4X [32X Example [32X[104X60[4X[25Xgap>[125X [27XG:=SmallGroup(2^7*9,33);;[127X[104X61[4X[25Xgap>[125X [27XH:=DirectProduct(G, ElementaryAbelianGroup(2^10));;[127X[104X62[4X[25Xgap>[125X [27XExponent(H); # should take at most a few milliseconds[127X[104X63[4X[28X72[128X[104X64[4X[25Xgap>[125X [27XK := PerfectGroup(2688,3);;[127X[104X65[4X[25Xgap>[125X [27XExponent(K); # should take at most a few seconds[127X[104X66[4X[28X168[128X[104X67[4X[32X[104X6869[30X [33X[0;6YThe functionality in [5XGAP[105X for transformations and transformation70semigroups has been rewritten and extended. Partial permutations and71inverse semigroups have been newly implemented. The documentation for72transformations and transformation semigroups has been improved.73Transformations and partial permutations are implemented in the [5XGAP[105X74kernel. Methods for calculating attributes of transformations and75partial permutations, and taking products, and so are also implemented76in the kernel. The new implementations are largely backwards77compatible; some exceptions are given below.[133X7879[33X[0;6YThe degree of a transformation [10Xf[110X is usually defined as the largest80positive integer where [10Xf[110X is defined. In previous versions of [5XGAP[105X,81transformations were only defined on positive integers less than their82degree, it was only possible to multiply transformations of equal83degree, and a transformation did not act on any point exceeding its84degree. Starting with [5XGAP[105X 4.7, transformations behave more like85permutations, in that they fix unspecified points and it is possible86to multiply arbitrary transformations.[133X8788[30X [33X[0;12Yin the display of a transformation, the trailing fixed points89are no longer printed. More precisely, in the display of a90transformation [10Xf[110X if [10Xn[110X is the largest value such that [10Xn^f<>n[110X or91[10Xi^f=n[110X for some [10Xi<>n[110X, then the values exceeding [10Xn[110X are not92printed.[133X9394[30X [33X[0;12Ythe display for semigroups of transformations now includes more95information, for example [10X<transformation semigroup on 10 pts96with 10 generators>[110X and [10X<inverse partial perm semigroup on 1097pts with 10 generators>[110X.[133X9899[30X [33X[0;12Ytransformations which define a permutation can be inverted, and100groups of transformations can be created.[133X101102[33X[0;6YFurther information regarding transformations and partial103permutations, can be found in the relevant chapters of the reference104manual.[133X105106[33X[0;6YThe code for Rees matrix semigroups has been completely rewritten to107fix the numerous bugs in the previous versions. The display of a Rees108matrix semigroup has also been improved to include the numbers of rows109and columns, and the underlying semigroup. Again the new110implementations should be backwards compatible with the exception that111the display is different.[133X112113[33X[0;6YThe code for magmas with a zero adjoined has been improved so that it114is possible to access more information about the original magma. The115display has also been changed to indicate that the created magma is a116magma with zero adjoined (incorporating the display of the underlying117magma). Elements of a magma with zero are also printed so that it is118clear that they belong to a magma with zero.[133X119120[33X[0;6YIf a semigroup is created by generators in the category121IsMultiplicativeElementWithOneCollection and CanEasilyCompareElements,122then it is now checked if the One of the generators is given as a123generator. In this case, the semigroup is created as a monoid.[133X124125[30X [33X[0;6YAdded a new operation [2XGrowthFunctionOfGroup[102X ([14XReference:126GrowthFunctionOfGroup[114X) that gives sizes of distance spheres in the127Cayley graph of a group.[133X128129[30X [33X[0;6YA new group constructor [2XFreeAbelianGroup[102X ([14XReference: FreeAbelianGroup[114X)130for free abelian groups has been added. By default, it creates131suitable fp groups. Though free abelian groups groups do not offer132much functionality right now, in the future other implementations may133be provided, e.g. by the [5XPolycyclic[105X package.[133X134135[30X [33X[0;6YThe message about halving the pool size at startup is only shown when136[10X-D[110X command line option is used (see [14X'Reference: Command Line137Options'[114X). [Suggested by Volker Braun][133X138139[30X [33X[0;6YAn info class called [2XInfoObsolete[102X ([14XReference: InfoObsolete[114X) with the140default level 0 is introduced. Setting it to 1 will trigger warnings141at runtime if an obsolete variable declared with142[10XDeclareObsoleteSynonym[110X is used. This is recommended for testing [5XGAP[105X143distribution and packages.[133X144145[30X [33X[0;6YThe [5XGAP[105X help system now recognises some common different spelling146patterns (for example, -ise/-ize, -isation/-ization, solvable/soluble)147and searches for all possible spelling options even when the synonyms148are not declared.[133X149150[30X [33X[0;6YAdded new function [2XCite[102X ([14XReference: Cite[114X) which produces citation151samples for [5XGAP[105X and packages.[133X152153[30X [33X[0;6YIt is now possible to compile [5XGAP[105X with user-supplied [10XCFLAGS[110X which now154will not be overwritten by [5XGAP[105X default settings. [Suggested by Jeroen155Demeyer][133X156157[33X[0;0YFixed bugs:[133X158159[30X [33X[0;6Y[2XUnion[102X ([14XReference: Union[114X) had [22XO(n^3)[122X behaviour when given many ranges160(e.g. it could take 10 seconds to find a union of 1000 1-element161sets). The new implementation reduces that to [22XO(n log n)[122X (and 4ms for162the 10 second example), at the cost of not merging ranges as well as163before in some rare cases.[133X164165[30X [33X[0;6Y[10XIsLatticeOrderBinaryRelation[110X only checked the existence of upper166bounds but not the uniqueness of the least upper bound (and dually for167lower bounds), so in some cases it could return the wrong answer.168[Reported by Attila Egri-Nagy][133X169170[30X [33X[0;6Y[2XLowIndexSubgroupsFpGroup[102X ([14XReference: LowIndexSubgroupsFpGroup[114X)171triggered a break loop if the list of generators of the 2nd argument172contained the identity element of the group. [Reported by Ignat173Soroko][133X174175[30X [33X[0;6YFixed regression in heuristics used by176[2XNaturalHomomorphismByNormalSubgroup[102X ([14XReference:177NaturalHomomorphismByNormalSubgroup[114X) that could produce a permutation178representation of an unreasonably large degree. [Reported by Izumi179Miyamoto][133X180181[30X [33X[0;6YFixed inconsistent behaviour of [10XQuotientMod( Integers, r, s, m )[110X in182the case where [3Xs[103X and [3Xm[103X are not coprime. This fix also corrects the183division behaviour of [10XZmodnZ[110X objects, see [2XQuotientMod[102X ([14XReference:184QuotientMod[114X) and [2XZmodnZ[102X ([14XReference: ZmodnZ[114X). [Reported by Mark185Dickinson][133X186187[30X [33X[0;6YFixed an oversight in the loading process causing [2XOnQuit[102X ([14XReference:188OnQuit[114X) not resetting the options stack after exiting the break loop.[133X189190[30X [33X[0;6YEmpty strings were treated slightly differently than other strings in191the [5XGAP[105X kernel, for historical reasons. This resulted in various192inconsistencies. For example, [10XIsStringRep("")[110X returned true, but a193method installed for arguments of type [10XIsStringRep[110X would NOT be194invoked when called with an empty string.[133X195196[33X[0;6YWe remove this special case in the [5XGAP[105X kernel (which dates back the197very early days of [5XGAP[105X 4 in 1996). This uncovered one issue in the198kernel function [10XPOSITION_SUBSTRING[110X (when calling it with an empty199string as second argument), which was also fixed.[133X200201[30X [33X[0;6YThe parser for floating point numbers contained a bug that could cause202[5XGAP[105X to crash or to get into a state where the only action left to the203user was to exit [5XGAP[105X via Ctrl-D. For example, entering four dots with204spaces between them on the [5XGAP[105X prompt and then pressing the return key205caused [5XGAP[105X to exit.[133X206207[33X[0;6YThe reason was (ironically) an error check in the innards of the float208parser code which invoked the [5XGAP[105X [10XError()[110X function at a point where it209should not have.[133X210211[30X [33X[0;6YRemoving the last character in a string was supposed to overwrite the212old removed character in memory with a zero byte, but failed to do so213due to an off-by-one error. For most [5XGAP[105X operations, this has no214visible effect, except for those which directly operate on the215underlying memory representation of strings. For example, when trying216to use such a string to reference a record entry, a (strange) error217could be triggered.[133X218219[30X [33X[0;6Y[2XViewString[102X ([14XReference: ViewString[114X) and [2XDisplayString[102X ([14XReference:220DisplayString[114X) are now handling strings, characters and immediate FFEs221in a consistent manner.[133X222223[30X [33X[0;6YMultiple fixes to the build process for less common Debian platforms224(arm, ia64, mips, sparc, GNU/Hurd). [Suggested by Bill Allombert][133X225226[30X [33X[0;6YFixes for several regressions in the [11Xgac[111X script. [Suggested by Bill227Allombert][133X228229[33X[0;0YChanged functionality:[133X230231[30X [33X[0;6YIt is not possible now to call [2XWreathProduct[102X ([14XReference:232WreathProduct[114X) with 2nd argument [3XH[103X not being a permutation group,233without using the 3rd argument specifying the permutation234representation. This is an incompatible change but it will produce an235error instead of a wrong result. The former behaviour of [2XWreathProduct[102X236([14XReference: WreathProduct[114X) may now be achieved by using237[2XStandardWreathProduct[102X ([14XReference: StandardWreathProduct[114X) which returns238the wreath product for the (right regular) permutation action of [3XH[103X on239its elements.[133X240241[30X [33X[0;6YThe function [10XViewLength[110X to specify the maximal number of lines that242are printed in [2XViewObj[102X ([14XReference: ViewObj[114X) became obsolete, since243there was already a user preference [10XViewLength[110X to specify this. The244value of this preference is also accessible in [10XGAPInfo.ViewLength[110X.[133X245246247[1X3.1-2 [33X[0;0YNew and updated packages since [5XGAP[105X[101X[1X 4.6.5[133X[101X248249[33X[0;0YAt the time of the release of [5XGAP[105X 4.6.5 there were 107 packages250redistributed with [5XGAP[105X. The first public release of [5XGAP[105X 4.7 contains 114251packages.[133X252253[33X[0;0YOne of essential changes is that the [5XCitrus[105X package by J.Mitchell has been254renamed to [5XSemigroups[105X. The package has been completely overhauled, the255performance has been improved, and the code has been generalized so that in256the future the same code can be used to compute with other types of257semigroups.[133X258259[33X[0;0YFurthermore, new packages that have been added to the redistribution since260the release of [5XGAP[105X 4.6.5 are:[133X261262[30X [33X[0;6Y[5X4ti2interface[105X package by Sebastian Gutsche, providing an interface to263[5X4ti2[105X, a software package for algebraic, geometric and combinatorial264problems on linear spaces ([7Xhttp://www.4ti2.de[107X).[133X265266[30X [33X[0;6Y[5XCoReLG[105X by Heiko Dietrich, Paolo Faccin and Willem de Graaf for267calculations in real semisimple Lie algebras.[133X268269[30X [33X[0;6Y[5XIntPic[105X package by Manuel Delgado, aimed at providing a simple way of270getting a pictorial view of sets of integers. The main goal of the271package is producing [5XTikz[105X code for arrays of integers. The code272produced is to be included in a LaTeX file, which can then be273processed. Some of the integers are emphasized by using different274colors for the cells containing them.[133X275276[30X [33X[0;6Y[5XLieRing[105X by Serena Cicalo and Willem de Graaf for constructing277finitely-presented Lie rings and calculating the Lazard278correspondence. The package also provides a database of small [22Xn[122X-Engel279Lie rings.[133X280281[30X [33X[0;6Y[5XLiePRing[105X package by Michael Vaughan-Lee and Bettina Eick, introducing282a new datastructure for nilpotent Lie rings of prime-power order. This283allows to define such Lie rings for specific primes as well as for284symbolic primes and other symbolic parameters. The package also285includes a database of nilpotent Lie rings of order at most [22Xp^7[122X for286all primes [22Xp > 3[122X.[133X287288[30X [33X[0;6Y[5XModIsom[105X by Bettina Eick, which contains various methods for computing289with nilpotent associative algebras. In particular, it contains a290method to determine the automorphism group and to test isomorphisms of291such algebras over finite fields and of modular group algebras of292finite [22Xp[122X-groups. Further, it contains a nilpotent quotient algorithm293for finitely presented associative algebras and a method to determine294Kurosh algebras.[133X295296[30X [33X[0;6Y[5XSLA[105X by Willem de Graaf for computations with simple Lie algebras. The297main topics of the package are nilpotent orbits, theta-groups and298semisimple subalgebras.[133X299300[33X[0;0YFurthermore, some packages have been upgraded substantially since the301[5XGAP[105X 4.6.5 release:[133X302303[30X [33X[0;6Y[5XANUPQ[105X package by Greg Gamble, Werner Nickel and Eamonn O'Brien has304been updated after Max Horn joined it as a maintainer. As a result, it305is now much easier to install and use it with the current [5XGAP[105X release.[133X306307[30X [33X[0;6Y[5XWedderga[105X package by Osnel Broche Cristo, Allen Herman, Alexander308Konovalov, Aurora Olivieri, Gabriela Olteanu, Ángel del Río and Inneke309Van Gelder has been extended to include functions for calculating310local and global Schur indices of ordinary irreducible characters of311finite groups, cyclotomic algebras over abelian number fields, and312rational quaternion algebras (contribution by Allen Herman).[133X313314315[1X3.2 [33X[0;0Y[5XGAP[105X[101X[1X 4.7.3 (February 2014)[133X[101X316317[33X[0;0YFixed bugs which could lead to incorrect results:[133X318319[30X [33X[0;6YIncorrect result returned by [10XAutomorphismGroup(PSp(4,2^n))[110X. [Reported320by Anvita][133X321322[30X [33X[0;6YThe [2XOrder[102X ([14XReference: Order[114X) method for group homomorphisms newly323introduced in [5XGAP[105X 4.7 had a bug that caused it to sometimes return324incorrect results. [Reported by Benjamin Sambale][133X325326[33X[0;0YFixed bugs that could lead to break loops:[133X327328[30X [33X[0;6YSeveral bugs were fixed and missing methods were introduced in the new329code for transformations, partial permutations and semigroups that was330first included in [5XGAP[105X 4.7. Some minor corrections were made in the331documentation for transformations.[133X332333[30X [33X[0;6YBreak loop in [10XIsomorphismFpMonoid[110X when prefixes in generators names334were longer than one letter. [Reported by Dmytro Savchuk and Yevgen335Muntyan][133X336337[30X [33X[0;6YBreak loop while displaying the result of338[2XMagmaWithInversesByMultiplicationTable[102X ([14XReference:339MagmaWithInversesByMultiplicationTable[114X). [Reported by Grahame Erskine][133X340341[33X[0;0YImproved functionality:[133X342343[30X [33X[0;6YBetter detection of UTF-8 terminal encoding on some systems.344[Suggested by Andries Brouwer][133X345346347[1X3.3 [33X[0;0Y[5XGAP[105X[101X[1X 4.7.4 (February 2014)[133X[101X348349[33X[0;0YThis release was prepared immediately after [5XGAP[105X 4.7.3 to revert the fix of350the error handling for the single quote at the end of an input line,351contained in [5XGAP[105X 4.7.3. It happened that (only on Windows) the fix caused352error messages in one of the packages.[133X353354355[1X3.4 [33X[0;0Y[5XGAP[105X[101X[1X 4.7.5 (May 2014)[133X[101X356357[33X[0;0YFixed bugs which could lead to incorrect results:[133X358359[30X [33X[0;6Y[2XInstallValue[102X ([14XReference: InstallValue[114X) cannot handle immediate values,360characters or booleans for technical reasons. A check for such values361was introduced to trigger an error message and prevent incorrect362results caused by this. [Reported by Sebastian Gutsche][133X363364[30X [33X[0;6Y[2XKnowsDictionary[102X ([14XReference: KnowsDictionary[114X) and [2XLookupDictionary[102X365([14XReference: LookupDictionary[114X) methods for [10XIsListLookupDictionary[110X were366using [2XPositionFirstComponent[102X ([14XReference: PositionFirstComponent[114X); the367latter is only valid on sorted lists, but in [10XIsListLookupDictionary[110X368the underlying list is NOT sorted in general, leading to bogus369results.[133X370371[33X[0;0YOther fixed bugs:[133X372373[30X [33X[0;6YA bug in [10XDirectProductElementsFamily[110X which used374[2XCanEasilyCompareElements[102X ([14XReference: CanEasilyCompareElements[114X) instead375of [2XCanEasilySortElements[102X ([14XReference: CanEasilySortElements[114X).[133X376377[30X [33X[0;6YFixed wrong [10XInfolevel[110X message that caused a break loop for some378automorphism group computations.[133X379380[30X [33X[0;6YFixed an error that sometimes caused a break loop in [2XHallSubgroup[102X381([14XReference: HallSubgroup[114X). [Reported by Benjamin Sambale][133X382383[30X [33X[0;6YFixed a rare error in computation of conjugacy classes of a finite384group by homomorphic images, providing fallback to a default385algorithm.[133X386387[30X [33X[0;6YFixed an error in the calculation of Frattini subgroup in the case of388the trivial radical.[133X389390[30X [33X[0;6YSeveral minor bugs were fixed in the documentation, kernel, and391library code for transformations.[133X392393[30X [33X[0;6YFixed errors in [2XNumberPerfectGroups[102X ([14XReference: NumberPerfectGroups[114X)394and [2XNumberPerfectLibraryGroups[102X ([14XReference: NumberPerfectLibraryGroups[114X)395not being aware that there are no perfect groups of odd order.[133X396397[30X [33X[0;6YRestored the ability to build [5XGAP[105X on OS X 10.4 and 10.5 which was398accidentally broken in the previous [5XGAP[105X release by using the build399option not supported by these versions.[133X400401[30X [33X[0;6YFixed some problems for ia64 and sparc architectures. [Reported by402Bill Allombert and Volker Braun][133X403404[33X[0;0YNew package added for the redistribution with [5XGAP[105X:[133X405406[30X [33X[0;6Y[5Xpermut[105X package by A.Ballester-Bolinches, E.Cosme-Llópez, and407R.Esteban-Romero to deal with permutability in finite groups.[133X408409410[1X3.5 [33X[0;0Y[5XGAP[105X[101X[1X 4.7.6 (November 2014)[133X[101X411412[33X[0;0YFixed bugs which could lead to incorrect results:[133X413414[30X [33X[0;6YA bug that may cause [2XShortestVectors[102X ([14XReference: ShortestVectors[114X) to415return an incomplete list. [Reported by Florian Beye][133X416417[30X [33X[0;6YA bug that may lead to incorrect results and infinite loops when [5XGAP[105X418is compiled without GMP support using gcc 4.9.[133X419420[30X [33X[0;6YA bug that may cause [2XOrthogonalEmbeddings[102X ([14XReference:421OrthogonalEmbeddings[114X) to return an incomplete result. [Reported by422Benjamin Sambale][133X423424[33X[0;0YFixed bugs that could lead to break loops:[133X425426[30X [33X[0;6Y[2XClosureGroup[102X ([14XReference: ClosureGroup[114X) should be used instead of427[2XClosureSubgroup[102X ([14XReference: ClosureSubgroup[114X) in case there is no428parent group, otherwise some calculations such as e.g. [2XNormalSubgroups[102X429([14XReference: NormalSubgroups[114X) may fail. [Reported by Dmitrii Pasechnik][133X430431[30X [33X[0;6YFixed a line in the code that used a hard-coded identity permutation,432not a generic identity element of a group. [Reported by Toshio Sumi][133X433434[30X [33X[0;6YFixed a problem in the new code for calculating maximal subgroups that435caused a break loop for some groups from the transitive groups436library. [Reported by Petr Savicky][133X437438[30X [33X[0;6YFixed a problem in [2XClosureSubgroup[102X ([14XReference: ClosureSubgroup[114X) not439accepting some groups without [2XParent[102X ([14XReference: Parent[114X). [Reported by440Inneke van Gelder][133X441442[33X[0;0YOther fixed bugs:[133X443444[30X [33X[0;6YEliminated a number of compiler warnings detected with some newer445versions of [5XC[105X compilers.[133X446447[30X [33X[0;6YSome minor bugs in the transformation and partial permutation code and448documentation were resolved.[133X449450451[1X3.6 [33X[0;0Y[5XGAP[105X[101X[1X 4.7.7 (February 2015)[133X[101X452453[33X[0;0YNew features:[133X454455[30X [33X[0;6YIntroduced some arithmetic operations for infinity and negative456infinity, see [14XReference: infinity[114X.[133X457458[30X [33X[0;6YIntroduced new property [2XIsGeneratorsOfSemigroup[102X ([14XReference:459IsGeneratorsOfSemigroup[114X) which reflects wheter the list or collection460generates a semigroup.[133X461462[33X[0;0YFixed bugs which could lead to incorrect results:[133X463464[30X [33X[0;6YFixed a bug in [2XUnion[102X ([14XReference: Union[114X) (actually, in the internal465library function [10XJoinRanges[110X) caused by downward running ranges.466[Reported by Matt Fayers][133X467468[30X [33X[0;6YFixed a bug where recursive records might be printed with the wrong469component name, coming from component names being ordered differently470in two different pieces of code. [Reported by Thomas Breuer][133X471472[30X [33X[0;6YThe usage of [10Xabs[110X in [11Xsrc/gmpints.c[111X was replaced by [10XAbsInt[110X. The former473is defined to operate on 32-bit integers even if [5XGAP[105X is compiled in47464-bit mode. That lead to truncating [5XGAP[105X integers and caused a crash475in [2XRemInt[102X ([14XReference: RemInt[114X), reported by Willem De Graaf and Heiko476Dietrich. Using [10XAbsInt[110X fixes the crash, and ensures the correct477behaviour on 32-bit and 64-bit builds.[133X478479[33X[0;0YFixed bugs that could lead to break loops:[133X480481[30X [33X[0;6YA problem with [2XProbabilityShapes[102X ([14XReference: ProbabilityShapes[114X) not482setting frequencies list for small degrees. [Reported by Daniel483Błażewicz and independently by Mathieu Gagne][133X484485[30X [33X[0;6YAn error when generating a free monoid of rank infinity. [Reported by486Nick Loughlin][133X487488[30X [33X[0;6YSeveral bugs with the code for Rees matrix semigroups not handling489trivial cases properly.[133X490491[30X [33X[0;6YA bug in [2XIsomorphismTypeInfoFiniteSimpleGroup[102X ([14XReference:492IsomorphismTypeInfoFiniteSimpleGroup[114X) affecting one particular group493due to a misformatting in a routine that translates between the494Chevalley type and the name used in the table (in this case, [10X"T"[110X was495used instead of [10X["T"][110X). [Reported by Petr Savicky][133X496497[33X[0;0YOther fixed bugs:[133X498499[30X [33X[0;6YThe [2XBasis[102X ([14XReference: Basis[114X) method for full homomorphism spaces of500linear mappings did not set basis vectors which could be obtained by501[2XGeneratorsOfLeftModule[102X ([14XReference: GeneratorsOfLeftModule[114X).[133X502503[30X [33X[0;6YA problem with [2XGaloisType[102X ([14XReference: GaloisType[114X) entering an infinite504loop in the routine for approximating a root. [Reported by Daniel505Błażewicz][133X506507[30X [33X[0;6YFixed the crash when [5XGAP[105X is called when the environment variables [10XHOME[110X508or [10XPATH[110X are unset. [Reported by Bill Allombert][133X509510[33X[0;0YFurthermore, new packages that have been added to the redistribution since511the release of [5XGAP[105X 4.7.6 are:[133X512513[30X [33X[0;6Y[5Xjson[105X package by Christopher Jefferson, providing a mapping between the514[5XJSON[105X markup language and [5XGAP[105X[133X515516[30X [33X[0;6Y[5XSglPPow[105X package by Bettina Eick and Michael Vaughan-Lee, providing the517database of [22Xp[122X-groups of order [22Xp^7[122X for [22Xp > 11[122X, and of order [22X3^8[122X.[133X518519520[1X3.7 [33X[0;0Y[5XGAP[105X[101X[1X 4.7.8 (June 2015)[133X[101X521522[33X[0;0YFixed bugs which could lead to incorrect results:[133X523524[30X [33X[0;6YAdded two groups of degree 1575 which were missing in the library of525first primitive groups. [Reported by Gordon Royle][133X526527[30X [33X[0;6YFixed the error in the code for algebra module elements in packed528representation caused by the use of [2XObjectify[102X ([14XReference: Objectify[114X)529with the type of the given object instead of [2XObjByExtRep[102X ([14XReference:530ObjByExtRep[114X) as recommended in [14X'Reference: Further Improvements in531Implementing Residue Class Rings'[114X. The problem was that after532calculating [10Xu+v[110X where one of the summands was known to be zero, this533knowledge was wrongly passed to the sum via the type. [Reported by534Istvan Szollosi][133X535536[30X [33X[0;6YFixed a bug in [2XPowerMod[102X ([14XReference: PowerMod[114X) causing wrong results537for univariate Laurent polynomials when the two polynomial arguments538are stored with the same non-zero shift. [Reported by Max Horn][133X539540[33X[0;0YFurthermore, new packages that have been added to the redistribution since541the release of [5XGAP[105X 4.7.7 are:[133X542543[30X [33X[0;6Y[5XPatternClass[105X by Michael Albert, Ruth Hoffmann and Steve Linton,544allowing to explore the permutation pattern classes build by token545passing networks. Amongst other things, it can compute the basis of a546permutation pattern class, create automata from token passing networks547and check if the deterministic automaton is a possible representative548of a token passing network.[133X549550[30X [33X[0;6Y[5XQPA[105X by Edward Green and Øyvind Solberg, providing data structures and551algorithms for computations with finite dimensional quotients of path552algebras, and with finitely generated modules over such algebras. It553implements data structures for quivers, quotients of path algebras,554and modules, homomorphisms and complexes of modules over quotients of555path algebras.[133X556557558559