Changes commited to PARI GIT during the last two weeks

bnfinit: avoid catastrophic loss of accuracy in log. units embeddings

2	1	src/basemath/buch2.c
2	1	src/test/32/bnf
3	0	src/test/in/bnf

hnflll: avoid desperation GC

1	7	src/basemath/hnf_snf.c

init_red_mod_units: take possible catastrophic loss of accuracy into account

1	1	src/basemath/buch2.c

improve doc

3	4	src/functions/number_theoretical/znchartoprimitive

remove bnrconductormod flag

22	22	src/basemath/buch3.c
1	1	src/basemath/kummer.c
1	1	src/headers/paridecl.h

91- [libpari] bnrconductor_raw, bnrconductor_factored

52- bnrconductor_i is now obsolete, use bnrconductor_raw or bnrconductormod

2	0	CHANGES
1	0	COMPAT
13	3	doc/usersch6.tex
1	1	src/basemath/base3.c
91	40	src/basemath/buch3.c
24	28	src/basemath/kummer.c
11	3	src/basemath/lfunutils.c
2	0	src/headers/paridecl.h
3	3	src/modules/stark.c

further list_Hecke cleanups + update bench

53	70	src/basemath/kummer.c
6	6	src/test/32/bnrclassfield

kill warnings

3	5	src/basemath/kummer.c

cleanup rnfkummer struct + logall

26	29	src/basemath/kummer.c

rnfkummer cleanup

3	6	src/basemath/kummer.c

cleanup build_list_Hecke, factor out matlogall, remove primlist struct

57	68	src/basemath/kummer.c

subgroup -> H

10	13	src/basemath/kummer.c

remove _rnfkummer_step18

8	19	src/basemath/kummer.c

rnfkummer: remove xell

2	3	src/basemath/kummer.c

rnfkummer: remove enumeration; do everything with linear algebra

26	56	src/basemath/kummer.c

remove useless ok_congruence

11	50	src/basemath/kummer.c

rnfkummer: remove useless vecMsup

11	26	src/basemath/kummer.c

printing of "t_MAT of t_VECSMALL" broken

fix SEGV and uniformize : always add parentheses, no matter whether
strictly necessary or not. Nobody will want to setup matrices of
t_CLOSURE and look at the result anyway

17	8	src/language/es.c

80- ldata_vecan(an, N, prec): make sure the function returns N coefficients

1	0	CHANGES
2	2	doc/usersch8.tex
3	0	src/basemath/lfunutils.c

get_prlist: don't add last prime if Ldone already

1	1	src/basemath/kummer.c

idealsqr, idealpow: don't compute mod zkmultable_capZ(m)

can be very costly, hardly ever improves significantly the modulus

4	4	src/basemath/base4.c

mpqs: find_nonsingular_sub: fix off-by-one error

3	4	src/basemath/F2v.c

compiler: countmatrixelts did not count noarg correctly

4	3	src/language/compile.c

New libpari function stack_calloc_align

3	0	doc/usersch5.tex
1	0	src/headers/paridecl.h
9	0	src/kernel/none/level1.h
1	1	src/modules/mpqs.c

rewrite get_prlist to avoid invimsubgroup

simpler and avoid a bnr with generators

141	146	src/basemath/kummer.c
3	3	src/test/32/bnrclassfield

stack_malloc_align: fix typo (e instead of k)

1	1	src/kernel/none/level1.h

Buchray_i: one more use of cyc0 instead of cyc

3	4	src/basemath/buch3.c
1	0	src/test/32/bnr
3	0	src/test/in/bnr

minor cleanup invimsubgroup

2	2	src/basemath/kummer.c

update bench

11	3	src/test/32/rnfkummer
1	3	src/test/in/rnfkummer

bnrinit(,,,MOD): wrong relations

can't reduce every occurence of cyc modulo MOD

4	4	src/basemath/buch3.c
2	1	src/test/32/bnr
2	2	src/test/32/rnfkummer
4	0	src/test/in/bnr

rnfkummer: missing p in prototype

1	1	src/functions/number_fields/rnfkummer

FindModulus: cleanup + document better algorithm

30	31	src/modules/stark.c

FindModulus: cleanup

13	12	src/modules/stark.c

Buchray -> Buchraymod

1	1	src/basemath/mftrace.c

cleanup nfgrunwaldwang (variable scope + gcopy -> shallowcopy)

5	5	src/modules/algebras.c

nfgrunwaldwang: don't modify argument

1	0	CHANGES
1	0	src/modules/algebras.c
3	1	src/test/32/algebras
6	0	src/test/in/algebras

update bench

3	11	src/test/32/rnfkummer
3	1	src/test/in/rnfkummer

bignf -> bnf

5	5	src/headers/paridecl.h

update bench

13	21	src/test/32/rnfkummer
1	1	src/test/in/rnfkummer

update bench

2	2	src/test/32/bnrclassfield

kill warnings

2	2	src/basemath/buch3.c
2	2	src/basemath/lfunutils.c
1	2	src/modules/stark.c

rnfkummer now obsolete. Change docs.

1	1	CHANGES
8	5	src/functions/number_fields/bnrinit
3	14	src/functions/number_fields/rnfkummer

fix error messages

4	3	src/basemath/buch3.c

bnr_char_sanitize + move varn(T) test out of bnr_subgroup_sanitize

17	7	src/basemath/buch3.c
11	2	src/basemath/kummer.c
14	19	src/basemath/lfunutils.c
1	0	src/headers/paridecl.h
1	3	src/modules/stark.c

use bnr_subgroup_sanitize

3	3	src/modules/stark.c

export bnrclassfield_sanitize -> bnr_subgroup_sanitize

1	1	CHANGES
1	1	doc/usersch6.tex
37	7	src/basemath/buch3.c
2	31	src/basemath/kummer.c
2	1	src/headers/paridecl.h
5	6	src/modules/stark.c

bnrconductor_i -> bnrconductormod

3	3	src/basemath/lfunutils.c
1	1	src/modules/stark.c

minor cleanup FindModulus

6	6	src/modules/stark.c

bnrconductor_i -> bnrconductormod

5	4	src/basemath/buch3.c

51- rnfconductor: flag to compute Cl_f/[K:Q] instead of Cl_f

2	1	CHANGES
7	6	src/basemath/buch3.c
1	1	src/basemath/lfunutils.c
26	10	src/functions/number_fields/rnfconductor
1	0	src/headers/paridecl.h
1	1	src/test/in/bnrL1

bnrclassfield_sanitize: more stringent type checks

4	1	src/basemath/kummer.c
2	2	src/test/32/bnrclassfield

bnrclassfield: bnrconductor -> bnrconductormod

1	1	src/basemath/kummer.c

bnrclassfield_sanitize: use bnrconductormod

16	6	src/basemath/kummer.c
2	2	src/test/32/bnrclassfield

fix log_prk + compute bnrisprincipal modulo exponent

- restore ordfa format for DL in residue field (the fa part had
  disappeared)
- N needs not be coprime to mod (only to its p-primary part and that's
  enough)
- force 'mod' in bnrisprincipal when generator not needed:
  avoid computing expensive DL in residue fields if units kill off a
  large part of (Z_K/f)^*
- simplify sprk_get_L2

56	24	src/basemath/base3.c
1	0	src/basemath/buch3.c

remove 'all' argument to rnfkummer

1	1	src/basemath/mftrace.c
1	1	src/modules/algebras.c

lfunabelianrelinit: remove useless Buchray call

2	2	src/basemath/lfunutils.c

cleanups bnrisprincipalmod + improve doc

8	15	src/basemath/buch3.c
1	1	src/functions/number_fields/bnrclassfield

doc typo

1	1	src/functions/number_fields/bnrisprincipal

50- rnfkummer: removed "all" flag, use bnrclassfield !

1	0	CHANGES
1	0	COMPAT
5	2	src/basemath/buch3.c
70	285	src/basemath/kummer.c
4	10	src/functions/number_fields/rnfkummer
2	1	src/headers/paridecl.h
29	42	src/test/32/rnfkummer
8	26	src/test/in/rnfkummer

bnr_log_gen_pr: remove nf argument

8	9	src/basemath/buch3.c

add test

8	1	src/test/32/bnr
10	0	src/test/in/bnr

bnrisprincipalmod: reduce mod cycray in all cases

2	6	src/basemath/buch3.c

49- optional argument to bnrinit to compute Cl_f/n, avoids tough discrete logs

1	0	CHANGES
10	6	src/basemath/base3.c
24	9	src/basemath/buch3.c
1	1	src/basemath/kummer.c
15	1	src/functions/number_fields/bnrclassfield
16	4	src/functions/number_fields/bnrinit
18	4	src/functions/number_fields/bnrisprincipal
4	0	src/headers/paridecl.h
2	1	src/test/32/bnrclassfield
16	13	src/test/32/compat
3	0	src/test/in/bnrclassfield

cleanup after merging aurel-bnrisprincipalmod

- buch4.c: use sprk instead of bid
- base3.c: implement log_gen_pr and log_prk1 for sprk (instead of bid)

82	58	src/basemath/base3.c
12	15	src/basemath/buch4.c
2	0	src/headers/paripriv.h

New libpari functions bnrisprincipalmod and ideallogmod

1	0	CHANGES
79	36	src/basemath/base3.c
20	3	src/basemath/buch3.c
1	1	src/basemath/char.c
12	10	src/basemath/kummer.c
4	0	src/functions/number_fields/bnrisprincipal
6	2	src/functions/number_fields/ideallog
2	0	src/headers/paridecl.h
1	0	src/headers/parinf.h
1	1	src/headers/paripriv.h
2	2	src/test/32/algebras
6	1	src/test/32/bnr
11	8	src/test/32/bnrclassfield
1	0	src/test/in/algebras
5	0	src/test/in/bnr
10	0	src/test/in/bnrclassfield

avoid overfull hbox

1	1	src/functions/modular_symbols/mscosets

48- gammamellininv*(A,...): allow A to be an L-function

1	0	CHANGES
7	2	doc/usersch8.tex
1	2	src/basemath/lfun.c
6	2	src/basemath/lfunutils.c
43	30	src/basemath/mellininv.c
3	3	src/functions/transcendental/gammamellininvasymp
3	1	src/functions/transcendental/gammamellininvinit
1	0	src/headers/paridecl.h
5	2	src/test/32/gammamellininv
4	0	src/test/in/gammamellininv

Makefile.SH: fix dft when static=true

3	3	config/Makefile.SH

gammamellininvinit: convert Vga to floating point to avoid coef explosion

3	4	src/basemath/mellininv.c

ellisomat: fix error message and documentation

1	1	src/basemath/ellisog.c
11	10	src/functions/elliptic_curves/ellisomat
4	1	src/test/32/ellisomat
3	0	src/test/in/ellisomat

test-eval: disable syntax error test that depended on bison version

Newer bison displays "end of file" instead of $end

0	2	src/test/32/eval
1	1	src/test/in/eval

p2squarenf: avoid discrete logs in the residue field and extra Idealstar

+ cleanups

64	39	src/basemath/buch4.c

fix display of t_MAT(t_CLOSURE)

5	1	src/language/es.c
3	1	src/test/32/print
6	0	src/test/in/print

isfactor: return 0 for t_CLOSURE

3	0	src/language/es.c

fix comment

1	2	src/basemath/base3.c

polcoeff -> polcoef

1	1	doc/usersch2.tex

change some polcoeff (deprecated) to polcoef

2	2	src/functions/modular_forms/mfperiodpol
1	1	src/functions/polynomials/diffop
2	2	src/functions/sums/laurentseries

apply_U: don't modify input

3	4	src/basemath/base3.c

fix test-compat

3	2	src/test/32/compat

lemma7nf: minor cleanup

5	5	src/basemath/buch4.c

add sqrtint(,&r) example

1	1	examples/contfrac.gp

doc typo

1	1	src/functions/number_theoretical/sqrtint

47- add optional argument to sqrtint [set remainder]  [#2229]

1	0	CHANGES
13	0	src/basemath/arith1.c
13	6	src/functions/number_theoretical/sqrtint
1	0	src/headers/paridecl.h
7	1	src/test/32/sqrtn
7	0	src/test/in/sqrtn