CoCoA

27 Sep 2021

09:57 CoCoA-5 Bug #1610: IsInRadical: some more little bugs

Error for... Anna Maria Bigatti

09:55 CoCoA-5 Bug #1610: IsInRadical: some more little bugs

Error if *GradingDim* is 0 (lex, elim, ...) for call to *IsHomog*. Now fixed. Anna Maria Bigatti

09:52 CoCoA-5 Bug #1610 (Closed): IsInRadical: some more little bugs

I met some more bugs for extreme cases.
Usually, after isolating them, the solution is trivial.
List below. Anna Maria Bigatti

24 Sep 2021

22:37 CoCoA-5 Feature #1536 (Closed): Implement xel?

John Abbott

22:35 CoCoA-5 Bug #1215 (Closed): RationalSolve: gives "Error: must be non-zero"

Closing this issue because there is no clear indication of what to test.
I suspect the new design of @RationalSolve@... John Abbott

22:32 CoCoA-5 Bug #1537 (Closed): EmacUI: strange colours, sometimes

I have not seen the problem of unexpected colours for some time.
Reading through the notes suggests that Emacs's way... John Abbott

22:29 CoCoA-5 Bug #1574: ApproxSolveTF

Probably @ApproxSolveTF@ should have an accuracy-check-loop the same was as was done for @ApproxSolve@:
the heuristi... John Abbott

22:26 CoCoA-5 Bug #1573 (Closed): ApproxSolve: very imprecise

@ApproxSolveTF@ was surprisingly fast on the example given in the description (about 9s, vs 64s using @ApproxSolve@).... John Abbott

20:34 CoCoA-5 Bug #1573: ApproxSolve: very imprecise

I have just updated the manual entry for @ApproxSolv@.
I wonder what @ApproxSolveTF@ does... mmm??? John Abbott

22:21 CoCoA-5 Slug #1392 (Closed): ApproxSolve: another slow example

John Abbott

22:18 CoCoA-5 Bug #1539: Website: installation instructions for Microsoft

With luck the current instructions will be well tested over the next couple of weeks.
I hope we can close this issue... John Abbott

22:16 CoCoA-5 Bug #1531 (Closed): CRASH: arithmetic with HilbertSeries (tagged object)

There is already a test in *@exbugs.cocoa5@*
Closing after 11 months in feedback. John Abbott

21:58 CoCoA-5 Bug #713 (Closed): External libs: interrupting not easy

CoCoALib and CoCoA-5 do often monitor for ctrl-C. Inside CoCoA-5 the situation seems to be reasonably good.
The d... John Abbott

21:50 CoCoA-5 Bug #1216 (Closed): RationalSolve: gives wrong answer

This has been in feedback for 7 months.
I think we mention in the release notes that it is a backward INcompatible c... John Abbott

21:47 CoCoA-5 Feature #1293: Vim Plugin for CoCoA

Same comment as above (11 months ago). We should mention it to the participants at the minicourse. John Abbott

21:43 CoCoA-5 Design #1540 (Closed): Double power

John Abbott

22 Sep 2021

17:45 CoCoALib Bug #1585 (Closed): CRASH/ABORT: GMP overflow

As observed in comment 11, it would be very hard to try to prevent overflow in all cases.
It thus seems reasonable t... John Abbott

21 Sep 2021

11:31 CoCoALib Bug #1601: Compilation ambiguity

I think I know understand what happens: g++ v.11 has as default C++ version something newer than C++14.
My script @c... John Abbott

20 Sep 2021

21:33 CoCoALib Bug #1585: CRASH/ABORT: GMP overflow

I have worked around the @FloatStr@ problem (actually in @FloorLogBase@) mentioned in comment 4 above.
I just call t... John Abbott

20:52 CoCoALib Bug #1585: CRASH/ABORT: GMP overflow

I have put a new constant called *@OVERFLOW_BITS@* in *@config.H@*, and modified the impls so that they use this cons... John Abbott

16:49 CoCoALib Bug #1601: Compilation ambiguity

Florian succeeded in installing g++-10 on his computer; and he reports that CoCoALib compiled just fine (except a kno... John Abbott

16:17 CoCoALib Support #861 (Closed): Janet basis code: TmpJB files give some problems with C++11 (using CLANG/LLVM)

I have removed the commented-out code (and another function which existed only for that code).
It all compiles, and ... John Abbott

16:09 CoCoALib Support #1609 (New): Clean time.C

The file @time.C@ contains some hackery to work on Microsoft (or other non-unix-like) platforms.
Can we clean this... John Abbott

15:41 CoCoALib Design #1558 (Closed): CpuTimeLimit: more frequent clock checks

No problems have been reported in 7 months. So closing.
Maybe the imminent CoCoA school will find some "interesting... John Abbott

14:08 CoCoALib Design #1608 (Feedback): Domain of definition of NextPrime (and PrevPrime)

I have spoken to Anna who agrees to change the behaviour:
* both fns throw @BadArg@ (or similar) if given (strictly)... John Abbott

13:24 CoCoALib Design #1608 (In Progress): Domain of definition of NextPrime (and PrevPrime)

I am now inclined to allow 0 as arg to @NextPrime@; perhaps this is related to my preference to consider 0 as a "natu... John Abbott

16 Sep 2021

22:03 CoCoALib Design #1608 (Closed): Domain of definition of NextPrime (and PrevPrime)

Not so important: what is the domain of defn of @NextPrime@ and that of @PrevPrime@?
I tried to compute @NextPrime... John Abbott

13:51 CoCoALib Bug #1376 (Closed): GBasisTimeout: not working as expected

John Abbott

13:48 CoCoALib Bug #15 (Closed): Adjoint of a non-invertible matrix

John Abbott

13:47 CoCoALib Slug #897 (Closed): SimplestBigRatBetween: why is it so slow?

John Abbott

13:12 CoCoALib Slug #897: SimplestBigRatBetween: why is it so slow?

Closing after 11 months in feedback.
John Abbott

13:38 CoCoALib Design #786 (Closed): MemPool: review min and max loaf sizes

John Abbott

13:33 CoCoALib Bug #1484 (Closed): ker bug (quotientinghom for R/ideal())

Added test to @test-exbugs.C@.
Closing.
John Abbott

13:08 CoCoA-5 Bug #1573: ApproxSolve: very imprecise

tested on my Mac, now is gives (takes some time)... Anna Maria Bigatti

13:04 CoCoA-5 Bug #1573: ApproxSolve: very imprecise

My hack favours correctness over speed.
Is there a verbosity level which gives useful info? If so, document it (ri... John Abbott

12:54 CoCoALib Design #1538 (Closed): RingElem from string (ReadExpr)

John Abbott

12:49 CoCoALib Bug #1565 (Closed): IsInRadical: gives "weird" error

Deleted the comment out cruft. There is already a test in @ex-bugs.C@
Closing!
John Abbott

12:40 CoCoA-5 Slug #1270 (Closed): RationalSolve: use MinPolyQuot instead of elim

tested on Mac, ~0.1s Anna Maria Bigatti

12:38 CoCoA-5 Support #1287 (Closed): Better err mesg for easy typo in ring definition

Anna Maria Bigatti

12:36 CoCoA-5 Slug #1506 (Closed): Remove XMLFileName from Main.C

Anna Maria Bigatti

12:32 CoCoALib Bug #1570 (Closed): FloatStr sometimes produces NUL chars

John Abbott

12:29 CoCoALib Feature #1580: New fn prim for polys

Try to implement the idea in comment 4 (above).
John Abbott

12:27 CoCoALib Bug #1590 (Closed): CRASH: LinKerZZ

There seems to be documentation now.
I have just added a new test *@test-MatrixOps1.C@* which tests the input given ... John Abbott

12:15 CoCoALib Bug #1590: CRASH: LinKerZZ

works ok for me.
Anna Maria Bigatti

15 Sep 2021

11:56 CoCoALib Bug #1585: CRASH/ABORT: GMP overflow

I have discussed this with Anna; and had also written to the GMP people some time ago.
Overall conclusion: it is *... John Abbott

14 Sep 2021

12:07 CoCoALib Design #1606: Return type with const

Not yet ready: the awkward cases mentioned in the description are compiled... ooops :-(
We may need to change the ... John Abbott

12:01 CoCoALib Bug #1605 (Closed): binomial: domain of defn?

John Abbott

12:00 CoCoALib Feature #1598: RingHom: implement phi(X) as apply(phi, X) also for X vector and matrix

The manual entry for "Intro to RingHom" needs to be updated.
Must also add some good tests, e.g.... John Abbott

11:56 CoCoALib Bug #1473 (Closed): isystem not working as expected

John Abbott

11:53 CoCoA-5 Feature #1599 (Closed): ConcatStr

John Abbott

11:47 CoCoA-5 Slug #1597 (Closed): GetRow/GetRows are extraordinarily slow

John Abbott

11:46 CoCoA-5 Bug #1594 (Closed): Parser bug: missing close square bracket

John Abbott

11:45 CoCoA-5 Feature #1431 (Rejected): Juxtaposition of string literals

Given the example in comment 6 (above), and the fact that *@ConcatStrings@* is now quite fast. We have decided to re... John Abbott

13 Sep 2021

11:56 CoCoALib Design #1576: cmp for machine integers

Here is another approach which could be viewed as being "proper use of C++" (maybe):
* The only *@cmp@* function is ... John Abbott

08 Sep 2021

21:13 CoCoALib Design #1576: cmp for machine integers

It seems there may be some nice solutions in C++20, but it is to early [Sept 2021] to adopt features from C++20:
* t... John Abbott

20:49 CoCoALib Design #1576: cmp for machine integers

I think I may have an ugly solution. The idea is to use yet another template fn:... John Abbott

15:50 CoCoALib Design #1576: cmp for machine integers

Here is a simpler failing example:... John Abbott

01 Sep 2021

14:51 CoCoALib Design #1576: cmp for machine integers

Here is an example program which fails to compile because the call to @cmp@ in @main@ refers to the template rather t... John Abbott

14:28 CoCoALib Design #1576: cmp for machine integers

I think I have now understood what is causing the trouble:
creating a @ConstRefRingElem@ object from a @const RingEl... John Abbott