Changes commited to PARI GIT during the last two weeks

galoisconj4_main: simplify error message

1	1	src/basemath/galconj.c

New libpari function uisnilpotent

1	0	CHANGES
3	0	doc/usersch5.tex
18	0	src/basemath/ispower.c
1	0	src/headers/paridecl.h

test-factormod: lower parisize to 272M

2	2	src/test/32/factormod
1	1	src/test/in/factormod

mtpthread_queue_submit: remove nested BLOCK_SIGINT_START

9	13	src/mt/pthread.c

New libpari function ellheegner_z (from Deping Huang)

1	0	CHANGES
35	0	src/basemath/ellanal.c
3	0	src/functions/elliptic_curves/ellheegner
1	0	src/headers/paridecl.h

rename gred_rfrac2 -> gred_rfrac and transtype -> base_change

+ add comments

20	17	src/basemath/gen1.c

genus2_type5ns_2: handle multiplicative reduction

21	10	src/basemath/hyperell.c
7	1	src/test/32/genus2red
3	0	src/test/in/genus2red

genus2charpoly: use FpXQV_ellcharpoly, remove nfellcharpoly

18	34	src/basemath/hyperell.c

New libpari function FpXQV_ellcharpoly

1	0	CHANGES
6	0	doc/usersch7.tex
96	0	src/basemath/FpE.c
1	0	src/headers/paridecl.h

New private functions ell_to_a4a6_{FpXQ, Flxq,F2xq}

33	0	src/basemath/F2xqE.c
6	89	src/basemath/FF.c
36	0	src/basemath/FlxqE.c
14	0	src/basemath/FpE.c
9	0	src/headers/paripriv.h

gred_rfrac2: handle case when base change of num or den yields 0 [#2683]

When users input rational functions over incoherent but compatible
base rings (e.g., Z/NZ and Z) we try to recover by performing base
changes (e.g., to Z/NZ). If either num or den then becomes 0, the old
code was incorrect in various ways.

10	3	src/basemath/gen1.c
5	0	src/test/32/pol
4	0	src/test/in/pol

usersch7: F_2^n: Add missing \ before *

2	2	doc/usersch7.tex

doc: elllocalred: add linebreak to separate kod from v

3	1	src/functions/elliptic_curves/elllocalred

genus2charpoly: implement type5 non-split for p=2

Now, genus2charpoly(W,2) should give the correct result except that when the
correct result is of degree 1 or 2, it might return 1 instead.

131	11	src/basemath/hyperell.c

test-genus2red: add tests for p=2

1075	467	src/test/32/genus2red
8	1	src/test/in/genus2red

genus2red: remove component pic for p=2 (not correct)

1	0	src/functions/elliptic_curves/genus2red
4	1	src/modules/genus2red.c
3	4	src/test/32/genus2red

get_Flx_algebra: e->v was not set

1	1	src/basemath/Flx.c

genus2charpoly: do not compute P when p=2.

3	1	src/basemath/hyperell.c

New libpari functions ZXX_pval, ZXV_pval

1	1	CHANGES
6	0	doc/usersch5.tex
16	0	src/basemath/gen2.c
2	0	src/headers/paridecl.h

genus2_eulerfact2: remove spurious parameter F

(next version of genus2_type5 will need both P and Q)

3	2	src/basemath/hyperell.c
1	1	src/basemath/lfunutils.c
1	1	src/headers/paripriv.h

New libpari function ZXX_shifti

1	0	CHANGES
3	0	doc/usersch5.tex
13	0	src/basemath/ZX.c
1	0	src/headers/paridecl.h

doc: get_F*_algebra section: resort

5	4	doc/usersch5.tex

New libpari function get_Flx_algebra

1	1	CHANGES
3	0	doc/usersch5.tex
25	0	src/basemath/Flx.c
1	0	src/headers/paridecl.h

New libpari function get_Flxq_algebra

1	0	CHANGES
3	0	doc/usersch5.tex
20	7	src/basemath/Flx.c
1	0	src/headers/paridecl.h

gen_parapply_percent, parvector, parmatrix: don't leave object on stack

The one in gen_parapply_percent could corrupt the stack. The other two
would waste a small number of stack words.

3	3	src/basemath/alglin3.c
13	7	src/language/eval.c

test-variable: also test the GP function 'variables'

8	0	src/test/32/variable
7	0	src/test/in/variable