CoCoA » CoCoALib

15 Jan 2021

21:15 Feature #206: Matrix equation solving: LinKer

The source of *@LinKerBasis@* looks a bit dodgy.
Given a matrix it does what is expected; but it also works for a _l... John Abbott

21:07 Feature #206: Matrix equation solving: LinKer

This was originally in issue #122; moving it to here.
What is the difference between @LinKer@ and @LinKerBasis@?
... John Abbott

21:09 Feature #122 (Closed): Porting "LinKer"

This seems to be a *duplicate* of #206; since there is more info under that index, I'm closing this one. John Abbott

09:18 Feature #122: Porting "LinKer"

What is the difference between *@LinKer@* and *@LinKerBasis@*?
Is @LinKerBasis@ essentially the same as @GetCols(Lin... John Abbott

18:01 Support #1265 (In Progress): Unregistered TXT files in doc

Progress chart:
* *@SmallPrime.txt@* removed because very out of date
* _fill in as progress is made_ John Abbott

17:55 Design #1563 (Feedback): BigRat: ctor from machine int

I have now changed all calls @BigRat(...,1)@ into just @BigRat(...)@ in examples, tests, and source code.
Checking i... John Abbott

17:08 Design #1563 (Resolved): BigRat: ctor from machine int

I have added 2 new ctors (one from @MachineInt@ and one from @BigInt@).
They are @explicit@ (as is the ctor for @Big... John Abbott

17:05 Design #1563: BigRat: ctor from machine int

I have just discussed with Anna. Bruns's argument is persuasive: our original decision was to protect the user again... John Abbott

16:18 Design #1563 (In Progress): BigRat: ctor from machine int

There were 2 reasons for prohibiting ctor from a machine int: ambiguity of BigInt(0) since 0 could also be interprete... John Abbott

15:39 Design #1563 (Closed): BigRat: ctor from machine int

Bruns asks the following "why?"... John Abbott

17:13 Design #1180: BigRat(0) unexpectedly compiles! (calls ctor with mpq_t arg)

*IMPORTANT UPDATE*(2021-01-15) this decision has now been reversed in issue #1563 John Abbott

10:02 Feature #1559: Default ctor for rings

The class *@ring@* does have a default ctor (which assigns @RingZZ()@ as initial value).
In contrast, @SparsePolyRin... John Abbott

14 Jan 2021

14:31 Feature #206: Matrix equation solving: LinKer

The fn *@LinKerZZ@* is available in my copy of CoCoA-5, but there is no doc.
The doc for CoCoALib seems explicit.
... John Abbott

14:27 Feature #122: Porting "LinKer"

Has this already been done? There are two fns in CoCoA-5: @LinKer@ and @LinKerBasis@.
What about @LinKerZZ@?
John Abbott

12 Jan 2021

14:32 Design #1558: CpuTimeLimit: more frequent clock checks

I have modified the impl so that it triggers only when the CPU time limit has been exceeded, but it does try to call ... John Abbott

11 Jan 2021

11:09 Design #1562 (New): matrix.H and MatrixView.H

To be able to use the type @ConstMatrixView@ I need to include @matrix.H@ rather than @MatrixView.H@.
Is this reas... John Abbott

10:19 Support #1561 (New): Doc for module term orderings

The documentation for module term orderings is unhelpful (to put it politely).
I wanted to find out whether a @Fre... John Abbott

08:55 Design #1558: CpuTimeLimit: more frequent clock checks

John Abbott wrote:
> Now I am wondering whether it may not make more sense to base @CpuTimeLimit@ on the "steady clo... Anna Maria Bigatti

09 Jan 2021

18:20 Feature #1559 (In Progress): Default ctor for rings

Winfried Bruns sent the following by email:
"I wanted to define a class that contains two sparesepoly rings as mem... John Abbott

08 Jan 2021

18:46 Bug #1376 (Resolved): GBasisTimeout: not working as expected

Using just the one example "french students" example, the current (2021-01-08) version of @CpuTimeLimit@ works accept... John Abbott

09:48 Bug #1376: GBasisTimeout: not working as expected

Now I think it should be possible to make @CpuTimeLimit@ check the time more frequently without much overhead (by che... John Abbott

18:11 Design #1558 (Resolved): CpuTimeLimit: more frequent clock checks

I now have a version of @CpuTimeLimit@, which actually uses only "steady clock" rather than @CpuTime@.
It seems to... John Abbott

16:11 Design #1558: CpuTimeLimit: more frequent clock checks

Now I am wondering whether it may not make more sense to base @CpuTimeLimit@ on the "steady clock" (but then the nam... John Abbott

10:07 Design #1558 (In Progress): CpuTimeLimit: more frequent clock checks

Here is the code I used as a speed check:... John Abbott

09:38 Design #1558 (Closed): CpuTimeLimit: more frequent clock checks

The impl of @CpuTimeLimit@ is a bit complicated because the @CpuTime@ function is more expensive than I had expected.... John Abbott

13:03 Design #1391 (Closed): RingElems: syntax with [ and ] ?

I have added some tests (to @test-RingElem5.C@).
All seems OK. Closing. John Abbott

11:58 Bug #1484: ker bug (quotientinghom for R/ideal())

Anna! check ring for empty lists! Anna Maria Bigatti

07 Jan 2021

20:54 Bug #1376: GBasisTimeout: not working as expected

The relevant source is probably in *@SparsePolyOps-reduce.C@* around line 133.
The timer object is constructed in... John Abbott

20:42 Bug #1376: GBasisTimeout: not working as expected

Setting the timeout to 20s worked very well. Setting it to 25s triggered an interruption after 95s.
Setting it to 9... John Abbott

20:19 Bug #1376: GBasisTimeout: not working as expected

Why are there no test cases for this issue? Without them how can i know if it has been satisfactorily solved?
Usi... John Abbott

20:26 Support #1555 (In Progress): Software licence

Here is a suggestion:
* find out how many different copyright messages we need (headers, source, tests, examples, do... John Abbott

20:04 Design #855: Copyright and authorship?

I have added Bigatti to all files which had just Abbott as copyright holder.
Anna should now add Abbott to all files... John Abbott

20:02 Slug #1557 (Resolved): Reading list of rationals is too slow

There was a bug in my first version of the faster code: @in.get(ch)@ does not put EOF into @ch@ when EOF is reached -... John Abbott

06 Jan 2021

11:40 Slug #1557: Reading list of rationals is too slow

I have found a potentially useful entry on *StackOverflow*: link @https://stackoverflow.com/questions/9272276/can-you... John Abbott

11:32 Slug #1557 (In Progress): Reading list of rationals is too slow

I'm still puzzled: @ScanUnsignedIntegerLiteral@ continue to be slower than whatever the CoCoA-5 interpreter does; cou... John Abbott

05 Jan 2021

21:55 Slug #1557: Reading list of rationals is too slow

The profiler indicates that major costs are @hgcd@ (from GMP) and @ScanUnsignedIntegerLiteral@ (from CoCoALib).
JA... John Abbott

21:43 Slug #1557 (Closed): Reading list of rationals is too slow

Winfried Bruns sent me a file with about 100000 rationals (total size about 130Mbytes).
If I read the file in using ... John Abbott

28 Dec 2020

17:56 Support #1555: Software licence

Here are what we need to do:
* ensure that all copyright lines are correct (Anna and me)
* remove the GPL blurb fro... John Abbott

22 Dec 2020

16:16 Bug #1376: GBasisTimeout: not working as expected

What is the status of this issue? John Abbott

16:10 Bug #1542: gfan does not "deinitialize"

A non-solution is to modify the documentation to say that @GMPAllocator@ does not work well with GFan (or else the u... John Abbott

21 Dec 2020

12:33 Support #1555 (Closed): Software licence

Gereon Kremer asks whether we are willing to release some or all of the source for CoCoA/CoCoALib under a different l... John Abbott