Changes commited to PARI GIT during the last two weeks

forfactored doc typo

1	1	src/functions/programming/forfactored

forfactored doc typos

2	2	src/functions/programming/forfactored

update doc

33	24	src/functions/programming/forfactored

improve rec_order

- avoid useless powering when e = 1
- gc_GEN -> gc_INT

11	10	src/basemath/bb_group.c

doc primelimit <= 2^32

4	4	src/functions/default/primelimit

ZXQXQ_minpoly: fix typo

1	1	src/basemath/polarit3.c

gen_pow_init: make gc clean

1	1	src/basemath/bb_group.c

ZXQXQ_minpoly: finally handle denominator as documented.

18	13	src/basemath/polarit3.c
1	1	src/functions/linear_algebra/minpoly
1	1	src/headers/paripriv.h

ZXQ_minpoly: finally handle denominator as documented.

15	13	src/basemath/polarit3.c
1	1	src/functions/linear_algebra/minpoly
1	1	src/headers/paripriv.h

gen_order: missing GC

1	1	src/basemath/bb_group.c

implement fingerprint for t_CLOSURE

0	6	doc/usersch5.tex
7	1	src/language/hash.c
4	2	src/test/32/fingerprint
3	0	src/test/in/fingerprint

qfisom.c:fingerprint  avoid name clash with new GP function

2	2	src/basemath/qfisom.c

hash_zv: off-by-1 error (x[1] hashed twice)

2	2	src/language/hash.c

fingerprint: handle small negative integers properly

remi2n(x, b) has the sign of x; make sur x>= 0

19	16	src/language/hash.c

convert hash32 -> fingerprint

allow b-bit fingerprints instead of hardcoding 32 bits.

1	2	CHANGES
15	0	src/functions/programming/fingerprint
0	13	src/functions/programming/hash32
1	1	src/headers/paridecl.h
42	55	src/language/hash.c
113	0	src/test/32/fingerprint
0	107	src/test/32/hash
112	0	src/test/in/fingerprint
0	108	src/test/in/hash

hash32 documentation: we return 0 on t_CLOSURE

1	1	doc/usersch5.tex

fix has_GEN32(t_STR)

6	2	src/language/hash.c
7	6	src/test/32/hash
1	0	src/test/in/hash

fix hash bench

make t_FFELT a special case
implement t_CLOSURE and t_ERROR trivially for now

11	0	src/language/hash.c
11	7	src/test/32/hash
6	0	src/test/in/hash

add hash bench result

102	0	src/test/32/hash

62- new GP function hash32

2	0	CHANGES
10	1	doc/usersch5.tex
13	0	src/functions/programming/hash32
1	0	src/headers/paridecl.h
77	0	src/language/hash.c
101	0	src/test/in/hash

hash_GEN: make result independent of multiprecision kernel

still depends on 32/64 bit word size

13	7	src/language/hash.c

ZXQXQ_minpoly: Fix support for denominator + add doc

7	0	doc/usersch5.tex
1	1	src/basemath/polarit3.c

ZXQ_minpoly: Fix support for denominator

1	1	doc/usersch5.tex
2	2	src/basemath/polarit3.c

New libpari function ZXQXQ_minpoly

1	0	CHANGES
64	0	src/basemath/polarit3.c
6	0	src/functions/linear_algebra/minpoly
1	0	src/headers/paridecl.h
1	0	src/headers/paripriv.h

ZXQ_minpoly: handle bad primes correctly

3	2	src/basemath/polarit3.c

improve doc

2	2	src/functions/conversions/denominator
2	2	src/functions/conversions/numerator
2	1	src/functions/number_theoretical/content

RgXY_factor: use simplify on prospective factors

4	2	src/basemath/polarit2.c
6	3	src/test/32/factor
1	0	src/test/in/factor

reword comment

5	4	src/basemath/gen1.c

gmul(t_SER, t_POL): cleanup

11	11	src/basemath/gen1.c

localprec: typo in documentation

localprec no longer returns the prec value

1	2	src/functions/programming/localprec

refactor the init_ser / fill_ser interface -> pol2ser

34	31	src/basemath/gen1.c

RgX_to_ser_*: comments + rename variable

12	9	src/basemath/Ser.c

fix issues with valuations of t_SER over non-domains

? a=Mod(3,9) + O(x); b=Mod(3,9) + O(x); a*b
%1 = Mod(0, 9)*x^-1 + O(x^0)
? a=Mod(3,9)*x^0; b=Mod(3,9) + O(x^2); a*b
%2 = Mod(0, 9)*x^-1 + O(x^0)
? a=Mod(3,9)*x^0; b=Mod(3,9) + O(x); a*b
%3 = Mod(0, 9)*x^-1 + O(x^0)

? a=Mod(3,9)+O(x^2); b=Mod(3,9) + O(x^2); print(a*b)
Mod(0, 9) + O(x^2) \\ not properly normalized

As well as returning the "wrong" 0:

? a=Mod(0,9)*x^0; b=Mod(1,2) + O(x); a*b
%4 = Mod(0, 9) \\ expected Mod(0,1)

42	10	src/basemath/gen1.c
5	1	src/test/32/ser
4	0	src/test/in/ser