CoCoA » CoCoALib

28 May 2012

17:10 Feature #153: Shorter simpler names for some bool3 values and functions

Proposal for shorter names: *@IsTrue3@* and *@IsFalse3@* and *@IsUncertain3@*
Propose replacing @DefinitelyTrue@, ... John Abbott

16:25 Feature #157: Separate ThreadsafeCounter from symbol.C

JAA is not sure how to test the threadsafe counter since we don't know how to compile and run multithreaded programs.... John Abbott

16:20 Bug #22: Rename PPMonoidEvZZ?

A "very small exp" version may be slightly faster because more PPs fit in the cache, but I doubt the slight gain woul... John Abbott

16:13 Feature #156 (Closed): Brand new symbol(s)

I've cleaned the code, and updated the documentation.
Note: Task #157 will make a further change to this one.
A... John Abbott

15:54 Feature #40: Squarefree factorization - Alessio d'Ali`

Alessio has written a prototype test program; he will improve it.
He has added some detail to the report.
John Abbott

14:11 Support #160 (Closed): Cleanup doc for rings

Anna Maria Bigatti

11:31 Feature #167: Integer operations

I've moved (into *@IntOperations@*) and reorganized all the integer functions. There were lots of consequential chan... John Abbott

25 May 2012

14:42 Feature #167: Integer operations

Problem originally arose in CoCoA-5, but the solution lies in changes to CoCoALib.
John Abbott

14:41 Feature #167 (In Progress): Integer operations

The hard part will be all the consequential changes. John Abbott

24 May 2012

16:19 Feature #167 (Closed): Integer operations

While adding *@IsDivisible@* for integers, it was not clear where to put the declarations and definitions. So far th... John Abbott

22 May 2012

12:06 Feature #165: FractionField -- only of TrueGCDDomain?

I've just looked at the code for @FractionFieldImpl@. I must make a separate impl for arithmetic if GCD is not avai... John Abbott

21 May 2012

17:11 Feature #165 (In Progress): FractionField -- only of TrueGCDDomain?

After speaking to Anna and Renzo, we concluded that it is better to keep @FractionField@ general.
Anna proposed cr... John Abbott

15:09 Feature #165: FractionField -- only of TrueGCDDomain?

John Abbott wrote:
> JAA proposes: the base ring of a *@FractionField@* must be a @TrueGCDDomain@ (and this must be ... Anna Maria Bigatti

14:36 Feature #165: FractionField -- only of TrueGCDDomain?

If my proposal is accepted, someone will have to look through the existing code and remove any checks which will have... John Abbott

12:51 Feature #165 (Closed): FractionField -- only of TrueGCDDomain?

I've started implementing @IsTrueGCDDomain@.
I noticed that the ctor for a *@FractionField@* checks that the base ... John Abbott

15:59 Feature #40: Squarefree factorization - Alessio d'Ali`

Alessio has a first solid version of his report.
We spoke of how to generate tables of timings automatically; Alessi... John Abbott

11:50 Feature #41: Squarefree factorization - overhead

Oh dear, I keep forgetting to log time here. I'll just make a guess.
John Abbott

16 May 2012

10:16 Support #160: Cleanup doc for rings

A richer template for the documentation is offered in @empty.txt@ for homogeneity in class descriptions and facilitat... Anna Maria Bigatti

07:46 Support #160 (Closed): Cleanup doc for rings

Some documentation has already been rewritten.
Check that all files have appropriate links to operations in the righ... Anna Maria Bigatti

07:57 Support #161 (New): Convert old "task table" into redmine

Redmine seems a lot easier to keep up-to-date than our old task table page.
We should then remove the task table and... Anna Maria Bigatti

11 May 2012

19:01 Bug #155 (Closed): gcd: multivariate over non-prime finite field

Anna Maria Bigatti

12:36 Support #75: Documentation review by Caleo

Alessandra reports that the doc for *@PPMonoid@* did not explain clearly what a PPMonoid is, nor why CoCoALib offers ... John Abbott

10 May 2012

18:34 Bug #135 (In Progress): Revise interface to SmallFpImpl & friends

John Abbott

18:34 Bug #135: Revise interface to SmallFpImpl & friends

Added fns equivalent to many of the procedures for arithmetic.
Needed to add several other fns. Development proce... John Abbott

18:16 Feature #157: Separate ThreadsafeCounter from symbol.C

Consider making a BOOST-free (and possibly non-threadsafe) version so that CoCoALib can be compiled without requiring... John Abbott

16:40 Feature #157 (Closed): Separate ThreadsafeCounter from symbol.C

The *@ThreadsafeCounter@* recently added to @symbol.C@ will be useful elsewhere (_e.g._ in rings for the ring counter... John Abbott

17:16 Feature #156: Brand new symbol(s)

It is simpler to move the "special" ctor into the private zone, and remove the *@AnonymousMarker@*. This should take... John Abbott

16:48 Feature #156: Brand new symbol(s)

Check through existing code, and use anonymous symbols where appropriate. Then check everything!
John Abbott

16:38 Feature #156: Brand new symbol(s)

I have opted not to implement *@IsAnonymous@* because I cannot think of when it might actually be useful.
The curr... John Abbott

15:53 Feature #156 (In Progress): Brand new symbol(s)

JAA has completed a first impl. The new public fns (pseudo-ctors) are:
* *@NewSymbol()@* which creates a single new... John Abbott

15:44 Bug #155 (Resolved): gcd: multivariate over non-prime finite field

A first impl of nameless symbols together with a change to line 1480 in @TmpGReductor.C@ has solved the problem.
A... John Abbott

09 May 2012

17:18 Feature #156: Brand new symbol(s)

The current suggestion for imlpementation is the following:
* internally an anonymous symbol has empty head
* the p... John Abbott

08 May 2012

14:32 Feature #142: Improve threadsafety

Anna Maria Bigatti wrote:
> *@Elim@* (and a few other functions) create a temporary ring.
Yes. Temporary rings are ... John Abbott

12:48 Feature #142: Improve threadsafety

John Abbott wrote:
> No doubt the problem is due to the temporary rings (inside @Elim@) not self destroying when the... Anna Maria Bigatti

09:49 Feature #156: Brand new symbol(s)

JAA suggests that internally the head of an "anonymous" symbol be the empty string; the accessor function then checks... John Abbott

07 May 2012

18:43 Feature #156: Brand new symbol(s)

John Abbott wrote:
> Here are some more proposals for the head: *@??@* or *@#@* or *@_@*
>
> (...)
>
> Right now ... Anna Maria Bigatti

17:15 Feature #156: Brand new symbol(s)

Here are some more proposals for the head: *@??@* or *@#@* or *@_@*
Printed value would look like: *@3*??[17]^... John Abbott

17:01 Feature #156: Brand new symbol(s)

This is hardly important, but might be beneficial in a multithreaded environment. Possibly it would be "cleaner" too... John Abbott

16:58 Feature #156: Brand new symbol(s)

Here are 3 proposals for new PPMonoid pseudo-ctor signatures:... John Abbott

16:54 Feature #156: Brand new symbol(s)

JAA's first suggestion is to create "gensyms" containing a single index (in the hope that the index will not overflow... John Abbott

16:46 Feature #156 (Closed): Brand new symbol(s)

The pseudo-ctors for PPMonoids all require that names be specified for the indets.
It can occur that we want to cr... John Abbott

17:06 Bug #155: gcd: multivariate over non-prime finite field

The guilty line is 1480 in *@TmpGReductor.C@*
My example used an unindexed *@x@* in the field of coeffs -- this tri... John Abbott

16:08 Bug #155 (Closed): gcd: multivariate over non-prime finite field

GCD produces an unexpected error. Here is an example:... John Abbott

15:46 Feature #40: Squarefree factorization - Alessio d'Ali`

Alessio has started writing his report on Bernardin's article.
There are some typos in the article. JAA has given... John Abbott

14:30 Bug #154 (In Progress): GCD normalization (e.g. monic)

GCDs in a poly ring over a field are defined only upto a constant factor.
If the field is a FractionField then there... John Abbott

12:47 Feature #50: Polynomial content

The function *@ContentWRT@* gives result with strange scale factors sometimes.... John Abbott

11:59 Feature #69 (Closed): p-th root

This issue has effectively been replaced by #107.
John Abbott

11:57 Feature #107: Recognizing finite fields

Should there also be a fn called *@PthPower@*?
I note that *@IsPthPower2@* and *@PthRoot2@* have become forgotten;... John Abbott

04 May 2012

21:38 Feature #153: Shorter simpler names for some bool3 values and functions

Spent about 45 mins discussing with Anna. Differences of opinion remain.
John Abbott

21:36 Feature #153 (Closed): Shorter simpler names for some bool3 values and functions

We need shorter names for *@IsDefinitelyTrue@* and *@IsDefinitelyFalse@*.
Also need a shorter way of expressing th... John Abbott

21:30 Feature #138: Automatic conversion from bool3 to bool?

I have modified *@operator<<@* for OpenMath streams so that they handle explicitly machine integers (rather than *@Ma... John Abbott

12:36 Support #75: Documentation review by Caleo

Alessandra reports:
* the examples 2,3,4 of *@ex-RingWeyl@* are almost identical -- this cannot be what is intended!
... John Abbott

12:13 Feature #152 (Closed): Replace bool3 by tribool

In some quiet moment switch to using BOOST's *@tribool@* (or C++11 equivalent if it exists).
It seems pointless try... John Abbott

03 May 2012

15:32 Feature #151: Iterated CanonicalHom

There is already some code called *@TmpChainCanonicalHom@* in @CanonicalHom.[HC]@: it looks whether there is a "chain... Anna Maria Bigatti

15:25 Feature #151 (New): Iterated CanonicalHom

Implement *@NaturalMap@* and iterated *@CanonicalHom@*
John Abbott

15:26 Feature #150: RingOf rather than owner? (and than AmbientRing, ...)

*@RingOf@* can be applied to: ideal, RingElem, matrix, HomogList.
John Abbott

15:03 Feature #150 (In Progress): RingOf rather than owner? (and than AmbientRing, ...)

In CoCoA-5 to get the ring to which a RingElem belongs we use the fn *@RingOf@*.
Currently in CoCoALib the analogous... John Abbott

02 May 2012

19:37 Bug #149 (New): Makefile dependencies

There are problems with dependencies for optional external libraries.
How to make these dependencies visible only ... John Abbott

01 May 2012

11:44 Feature #147 (In Progress): Buchberger-Moeller: impl via modular reduction

Impl the idea presented in AKR "Zero-dim Schemes" which uses modular reduction to get the form of the answer quickly ... John Abbott

11:39 Feature #146 (New): Buchberger-Moeller: input conversions

Variants of BM which accept more convenient input (e.g. points or 0-dim ideals).
Simply convert inputs to sets of ... John Abbott

11:37 Feature #145 (New): Buchberger-Moeller: fast modular impl

Impl BM over a small prime finite field. Aim for speed.
Input ideals are represented as multiplication matrices (... John Abbott

11:35 Feature #144 (Closed): Buchberger-Moeller: generic impl

Implement generic BM algorithm over any field.
Aim is to make impl clean rather than fast.
Each input ideal is re... John Abbott

10:17 Feature #143 (In Progress): Buchberger-Moeller (parent task)

Robbiano wants to have Buchberger-Moeller in C5. It makes most sense to implement it in CoCoALib and then make it vi... John Abbott

30 Apr 2012

17:58 Feature #142: Improve threadsafety

If ref counting is removed (seems to be highly desirable to avoid it in multithreaded programs) then CoCoALib needs t... John Abbott

17:51 Feature #142 (In Progress): Improve threadsafety

Currently CoCoALib can be made (largely?) threadsafe by compiling with the flag *@-DCoCoA_THREADSAFE_HACK@*. As its ... John Abbott