Bug numbers refer to the BTS at http://pari.math.u-bordeaux.fr/Bugs/
Done for version 2.19.1 (released 14/11/2026):
  Fixed
BA   1- z=ffgen(23^3,'z);hyperellcharpoly(t^3+z*t+4) -> SEGV [#2703]      [F1]
     2- localbitprec(53); lfun(ellinit([1,2]), 100) -> FPE                [F2]
BA   3- rnfisabelian could return wrong result for complex fields [#2707] [F3]
     4- nfislocalpower: error when argument not an algebraic integer      [F4]
     5- x==0 incorrect for t_COMPLEX x (defined as |x| == 0)              [F5]
     6- logint(t_REAL, 2, &z); z was uninitialized                        [F6]
MJ   7- matdet(qfbredsl2(Qfb(2,-1,2^10000))[2])-> -1 instead of 1 [#2711] [F7]
MJ   8- e.tate inaccurate at p=2 [#2710]                                  [F8]
     9- polrootsreal(P, [a,b]) inaccurate for large (a-b) [#2714]         [F9]
BA  10- setrand(2); hyperellisisom(x^5+x+1, x^5+x+1) -> SEGV [#2718]      [F10]
BA  11- ellrank(E) could fail at low accuracy [#2719]                     [F11]
BA  12- b = ffgen(5^3,'b); hyperellcharpoly(x^5+4*x+b^2) ->SEGV [#2720]   [F12]
    13- ellpointtoz(E/Qp, P) accuracy issues                              [F13]
MJ  14- bnfinit:typo making Grenie-Molteni's residue bound larger [#2706] [F14]
SL  15- qfminimize([-5,0,-2;0,-1,0;-2,0,1]->div by 0 [#2721]              [F15]
BA  16- primecertisvalid had false positive [#2723]                       [F16]
    17- lerchphi domain errors [#2728]                                    [F17]

Changed
     1- normalization for 2-adic sqrt so that sqrt(1 + 4Z_2) = 1 + 4Z_2    [C1]
BA   2- use sigsetjmp instead of setjmp for bsd/darwin [#2715]             [C3]

Done for version 2.19.0 (released 28/09/2026):
  Fixed
     1- bestappr(-1/7, 6) -> 0 [instead of -1/6]
     2- abs(t_QUAD) would not return exact answer even if possible
     3- polredbest(x^3-2*x+1) -> SEGV [#2689]
     4- mflinear(MF in half integer weight, coefs in Q(chi)): wrong result
BA   5- ellisdivisible: potential oo-loop [#2695]
BA   6- bnrL1(,,0) over imaginary quadratic fields was wrong [#2693]
     7- matmodhnf: incorrect handling of empty matrix [#2699]
BA   8- ellisomat/nf could fail with CM curves
BA   9- polsubcyclofast(58,7) -> []
    10- factorpadic(1+O(3)*x^2,3,2) -> SEGV [#2701]
    11- bnfcertify and c_2 phase ("be honest") in bnfinit incorrectly skipped
        part of the checks in degree > 2.
BA  12- trace(Mod(x, Mod(1,2)*(x^2+x+1))) -> 0 [#2702]

  Added
     1- [libpari] QM_ishnf
     2- [libpari] fractor_lg, itor_lg, rdivii_lg, rtor_lg, stor_lg, utor_lg
     3- [libpari] ZX_realroots, ZX_nf_roots
     4- znorder: allow a znstar for the 'order' optional argument
     5- allow lfunqf(t_VEC) for lfunqf(matdiagonal(...))
     6- [libpari] vandermondeinverseinit_real
BA   7- New GP function galoisorder
BA   8- [libpari] group_istrivial, groupgen_export, groupelts_gen, group_iswss
BA   8- [libpari] group_istrivial, groupgen_export, groupelts_gen
BA   9- galoisidentify: support for groups of order <= 191
BA  10- [libpari] divisors_fact_degmax
    11- [libpari] prime_over_p_full

  Removed
     1- real_1_bit: use real_1
     2- group_export_GAP, group_export_MAGMA: use group_export
     3- [libpari] functions push_localprec, pop_localprec, get_localprec

  Changed
     1- bestappr(t_REAL) now choses B = 2^((realprec(x) + 1) \ 2) instead of
        2^realprec(x).
     2- [libpari] rename real_0_bit -> real_0_expo
     3- speed up numbpart
     4- '?' now prints a short introduction, move the list of help
        sections to '?*'
     5- allow arbitrary separable (rational) polynomials in the polred*
        family of functions. Don't assume irreducibility [#2689]
     6- readline completion: don't include yet undefined symbols
BA   7- group_ident, group_ident_trans, checkgroup: remove parameter S
BA   8- groupelts_to_group: if not WSS, return elts instead of NULL
     9- [libpari] The concept of a 'prec' (attached to the precision
        of a t_REAL) is now obsolete. Any positive integer is now a valid
        precision. The following utility functions have been deprecated
        bit_prec: use realprec
        lg2prec: use bit_accuracy
        nbits2extraprec: use nbits2prec
        ndec2nbits: use ndec2prec
        prec2ndec: use nbits2ndec
        prec2lg: use nbits2lg
