Changes commited to PARI GIT during the last two weeks

11- sinc(t_SER) more accurate

1	0	CHANGES
22	4	src/basemath/trans1.c
10	1	src/test/32/trans2
2	0	src/test/in/trans2

vecpowuu: special case B <= 2

14	1	src/basemath/lfun.c

improve doc

3	3	src/functions/number_fields/nfhnf

19- genus2red(3*(4*x^6+6*x^5+3*x^4+8*x^3+9*x^2-3),3) -> bug in litredp [#2053]

2	0	CHANGES
18	6	src/modules/genus2red.c
7	1	src/test/32/genus2red
5	0	src/test/in/genus2red

make FpX_quad_root static

1	1	src/basemath/FpX_factor.c
0	1	src/headers/paripriv.h
1	1	src/modules/stark.c

add ellisotree

1	0	doc/refcard-ell.tex

ZV_to_ZpV & friends: reuse l = lg(z)

2	2	src/basemath/polarit1.c

lfunmf: test for non-integral weight [#2058]

1	0	src/basemath/mftrace.c

polval -> ZX_pval

5	16	src/modules/genus2red.c

cleanups polymini [comments + RgX_to_06 / RgX_to_03 ]

33	19	src/modules/genus2red.c

doc updates

4	2	doc/refcard-nf.tex
3	2	src/functions/number_fields/galoisgetname

explicitly disallow "q-adics" when q is not prime

19	13	doc/usersch2.tex

Revert "19- a = 7/6+O(6^5); lift(a*2) -> invalid 14/6 [#2057]"

This reverts commit f6e1d1844c596b68d31f31f1a08d770df44c89d7.

This makes padic_to_Q 5 times slower for tiny inputs, worse for larger
inputs, all for a BIB. Not worth it.

0	2	CHANGES
6	5	src/basemath/gen3.c
1	2	src/test/32/padic
0	2	src/test/in/padic

19- a = 7/6+O(6^5); lift(a*2) -> invalid 14/6 [#2057]

2	0	CHANGES
5	6	src/basemath/gen3.c
2	1	src/test/32/padic
2	0	src/test/in/padic

18- bnrinit(,m,1) could return generators not coprime to m (when 2 || Nm)

1	0	CHANGES
1	1	src/basemath/base3.c
5	0	src/test/in/bnr

simplify rnfconductor: use bounds instead of ZpX_primedec

16	16	src/basemath/buch3.c

divrem: poldivrem -> RgX_divrem

1	1	src/basemath/gen3.c

fix C++ compilation

1	1	src/basemath/base2.c

revert previous patch, document eta_product_ZXn as shallow

2	1	doc/usersch5.tex
1	1	src/basemath/lfunutils.c

eta_product_ZXn: add GC at end

1	1	src/basemath/lfunutils.c

typo in eta_product_ZXn

2	2	src/basemath/lfunutils.c

17- factor(N, lim) was using primes <= lim instead of < lim as documented

1	0	CHANGES
1	1	src/basemath/ifactor1.c
4	1	src/test/32/factor
1	0	src/test/in/factor

cleanup rnfconductor

26	23	src/basemath/buch3.c

rnfconductor: put back generators

test cheaply (ZpX_primedec) whether primes are ramified instead

0	2	CHANGES
24	2	src/basemath/buch3.c
1	3	src/functions/number_fields/rnfconductor
5	5	src/test/32/rnf

modify round4 to return prime decomposition type

Can return primedecomposition itself in ZpX_primedec: when maxord_i
returns NULL, S.phi certifies S.chi of S.nu-Eisenstein type mod p.

Later.

27	1	src/basemath/base2.c
1	0	src/headers/paridecl.h

11- rnfconductor(bnf,T) no longer returns a bnr with explicit generators

[very costly in tough cases]

2	0	CHANGES
1	1	src/basemath/buch3.c
3	1	src/functions/number_fields/rnfconductor
5	5	src/test/32/rnf

cleanup cxEk and use rigorous bounds

3	1	doc/usersch5.tex
8	7	src/basemath/trans3.c

13- [libpari] cxEk

1	1	CHANGES
4	0	doc/usersch5.tex
2	36	src/basemath/elliptic.c
27	7	src/basemath/trans3.c
1	0	src/headers/paridecl.h
0	1	src/headers/paripriv.h

12- [libpari] cxtoreal

1	0	CHANGES
8	0	doc/usersch5.tex
4	8	src/basemath/elliptic.c
1	3	src/basemath/trans2.c
1	1	src/basemath/trans3.c
2	1	src/headers/paridecl.h
4	0	src/headers/pariinl.h
3	5	src/language/intnum.c

emscriten_draw: output plot as base64 data 

25	1	src/systems/emscripten/emscripten.c

New GP function galoisgetname. Require an updated galpol

1	0	CHANGES
20	0	src/functions/number_fields/galoisgetname
1	0	src/headers/paridecl.h
23	1	src/modules/galpol.c

New GP function galoisgetgroup. Require an updated galpol

19	0	src/functions/number_fields/galoisgetgroup
1	0	src/headers/paridecl.h
24	0	src/modules/galpol.c

lfunzetakinit_Galois: factor as Artin L functions

27	34	src/basemath/lfunutils.c
2	2	src/test/32/lfuntype

eta_product_ZXn: add random GC

6	0	src/basemath/lfunutils.c

11- [libpari] eta_ZXn, eta_product_ZXn

unify eta product code, return ZXn instead of t_SER, rename functions

1	0	CHANGES
6	0	doc/usersch5.tex
22	14	src/basemath/lfunutils.c
5	5	src/basemath/mftrace.c
6	6	src/basemath/trans3.c
2	1	src/headers/paridecl.h

doc typo

1	1	src/functions/modular_forms/mffrometaquo

10- improved eta(x^k)

1	0	CHANGES
7	5	src/basemath/lfunutils.c
11	8	src/basemath/trans3.c

fix idealfactor_limit(nf, t_INT or t_FRAC, lim)

was including cofactor> lim (as in Z)

15	4	src/basemath/base4.c
9	1	src/test/32/ideal
6	1	src/test/in/ideal

fix rnfconductor implementation when 'lim' is present

4	1	src/basemath/buch3.c

galoisidentify: add support for deg 132, 135, 136, 140

6	1	src/modules/groupid.c

16- galoischartable: wrong results

1	0	CHANGES
66	51	src/basemath/galconj.c
4	2	src/test/in/galoischartable

kill spurious warning

1	0	src/basemath/buch2.c

improve mfwt1newdimall (Pollack's criterion)

12	4	src/basemath/mftrace.c

gp2c: _badtype : fix check for gal

1	1	src/functions/gp2c_internal/_badtype

fix and cleanup compute_R

21	31	src/basemath/buch2.c
1	1	src/headers/paripriv.h

compute_R: yet another place where loss of accuracy is fatal [#2054]

1	1	src/basemath/buch2.c

getfu: avoid SEGV when low accuracy => lg(A) < RU

1	0	src/basemath/buch2.c

15- elleisnum([1,I/2^100],4) -> exponent overflow

1	0	CHANGES
6	0	src/basemath/elliptic.c
6	0	src/basemath/trans3.c
3	1	src/test/32/ell
2	0	src/test/in/ell

set_gamma: don't increase accuracy when \gamma = Id

4	3	src/basemath/elliptic.c

add difficult be_honest example from [#2054]

improves code coverage [but adds 20s to bench timings]

2	1	src/test/32/nf
6	0	src/test/in/nf

revert regression introduced in 8e990e70 [#2055]

3	3	src/basemath/hnf_snf.c
24	0	src/test/32/mathnf
2	0	src/test/in/mathnf

is_pm1 -> equali1

16	16	src/basemath/gen1.c

off-by-1 error in bestappr_real [#2054]

3	2	src/basemath/arith1.c

(inefficient) missing reduction in be_honest [#2054]

3	3	src/basemath/buch2.c

Revert 1bb20211: galoisfrobeniuslift: skip lift of degree < mindep

This patch is not correct

3	6	src/basemath/galconj.c

cleanup Fl_powers_pre

2	2	src/basemath/arith1.c

cleanup install-doc targets

7	10	config/Makefile.SH