Changes commited to PARI GIT during the last two weeks

91e0396466 | Karim Belabas | Thu, 11 Dec 2025 11:44:20 +0100

clarify forpart long help text, simplify short help

13	15	src/functions/programming/forpart

c0dfe82732 | Bill Allombert | Thu, 4 Dec 2025 13:17:20 +0100

forstep: use _gc_needed to avoid stack overflow

1	1	src/language/sumiter.c

8d174a268d | Bill Allombert | Thu, 4 Dec 2025 12:34:59 +0100

ZM_lll_norms: retune heuristic minimally for flatter

2	1	src/basemath/lll.c

8181146aa1 | Karim Belabas | Mon, 1 Dec 2025 21:13:15 +0100

mfeisenstein_i: optimize/simplify sum over divisors

12	17	src/basemath/mftrace.c

0d72d66ef5 | Karim Belabas | Sun, 30 Nov 2025 23:43:14 +0100

46- make sure factorlimit>= primelimit as documented

If not, then composites not detected in absZ_factor_limit_strict

Under gp -p 500000

install(absZ_factor_limit_strict,"GLD&")
absZ_factor_limit_strict(2181205156739883082072322466307767511435659730624751301062908246757922745017759415867401676157674404976691398629219689002172416000000000,10^6)

would include a composite at the end of the list

2	0	CHANGES
13	5	src/language/default.c

c6225f6e83 | Bill Allombert | Sun, 30 Nov 2025 15:24:57 +0100

Flxq_log_index_cubic: avoid non-working models

78	1	src/basemath/Flxq_log.c

7378c334ce | Karim Belabas | Sun, 30 Nov 2025 12:24:58 +0100

45- fix factoru_sign's batch trial division branch [#2648]

Two problems:
1) absZ_factor_limit_strict could include a composite in prime list
This created BIB (non-prime ideals and crash) in higher level calls

2) inconsistent behaviour wrt power test: cofactor not tested for
easy 2/3/5/7 powers in the batch trial division branch

- Fix bad logic in factoru_sign. No need for primality test: we have
enough information to decide.

- Fix inconsistent behaviour wrt power test: test cofactor for
  easy 2/3/5/7 powers in all branches. (Result would depend on
  tunings/primelimit in unpredicatble ways)

- For clarity, move the 2 snippets of code performing the power test out of
  the function

1	0	CHANGES
54	13	src/basemath/ifactor1.c
2	1	src/test/32/bnrclassfield
3	0	src/test/in/bnrclassfield

738e569de9 | Bill Allombert | Sun, 30 Nov 2025 00:22:16 +0100

smallirred_Flx: fix confusion long/ulong

3	3	src/basemath/Flxq_log.c

556bc56849 | Bill Allombert | Sat, 29 Nov 2025 14:26:20 +0100

test-character: update for pari_err_DOMAIN messages

4	4	src/test/32/character

682842a727 | Henri Cohen | Fri, 28 Nov 2025 22:37:28 +0100

corrected elleta when imag(w1/w2) < 0

35	38	src/basemath/elltrans.c
3	3	src/test/32/ell

2080b93f72 | Karim Belabas | Fri, 28 Nov 2025 04:16:30 +0100

ZpM_ZpC_gauss: missing GC

1	1	src/basemath/Zp.c

4f4869c0e7 | Bill Allombert | Thu, 27 Nov 2025 15:59:41 +0100

New libpari function ZpM_ZpC_gauss

1	0	CHANGES
4	0	doc/usersch5.tex
23	1	src/basemath/Zp.c
1	0	src/headers/paridecl.h

75290022c6 | Bill Allombert | Thu, 27 Nov 2025 14:55:47 +0100

znstar(disc): use pari_err_DOMAIN

1	1	src/basemath/char.c