Activity
From 14 Jun 2016 to 13 Jul 2016
13 Jul 2016
- 13:14 Feature #902 (Resolved): New function: FrobeniusMat
- 11:19 Design #906 (New): External libraries: keep copy of compatible versions?
- We should keep copies of external libraries guaranteed compatible with cocoalib.
(also useful for compiling old ver...
07 Jul 2016
- 13:51 Bug #905: txt2tags: double quotes wrong in LaTeX
- I had a quick look at the txt2tags website. It seems likely that a "filter" can solve the problem, but I don't know ...
- 13:49 Bug #905 (Closed): txt2tags: double quotes wrong in LaTeX
- txt2tags does not correctly handle double quotes when generating LaTeX.
The end result is that the PDF generated h...
04 Jul 2016
- 13:45 Feature #904 (New): SmallFpDouble: which impl of InvMod to use?
- The source code for @SmallFpDoubleImpl@ includes its own impl of @InvMod@ which uses @double@ rather than an integral...
30 Jun 2016
- 13:59 Design #789: NumTheory: behaviour of InvMod when inverse does not exist
- I am still a bit perplexed by this issue.
Having @InvMod@ always throw seems to me to be the cleanest and safest s...
29 Jun 2016
- 17:54 Design #789: NumTheory: behaviour of InvMod when inverse does not exist
- Perhaps the best approach is not to worry about @noexcept@ at the moment since we have no experience of the feature. ...
- 17:42 Design #789: NumTheory: behaviour of InvMod when inverse does not exist
- If we do adopt the idea of different signatures for throwing and non-throwing versions, which approach is better:
* ... - 15:32 Design #789: NumTheory: behaviour of InvMod when inverse does not exist
- I have just thought of an argument in favour of Anna's proposed design (namely, different signatures).
In C++11 ther... - 15:21 Design #789: NumTheory: behaviour of InvMod when inverse does not exist
- John Abbott wrote:
> Currently the @enum ErrorActionIndicator@ is defined inside @NumTheory.H@;
May we call it @e... - 15:18 Design #789: NumTheory: behaviour of InvMod when inverse does not exist
- Currently the @enum ErrorActionIndicator@ is defined inside @NumTheory.H@; probably it should be moved to @utils.H@ s...
- 15:15 Design #789: NumTheory: behaviour of InvMod when inverse does not exist
- I have just checked in my impl. Perhaps Anna could have a look at it, and decide whether she thinks it is comprehens...
- 15:12 Design #789: NumTheory: behaviour of InvMod when inverse does not exist
- John Abbott wrote:
> Your approach requires me to add 5 new signatures (for a total of 10);
good point!
- 15:01 Design #789: NumTheory: behaviour of InvMod when inverse does not exist
- @Anna: your approach is quite possible but it is more awkward than what I have implemented.
My approach has an opt... - 14:45 Design #789: NumTheory: behaviour of InvMod when inverse does not exist
- I'd rather have (as I said in comment #10)...
- 12:42 Design #789: NumTheory: behaviour of InvMod when inverse does not exist
- I have implemented @enum ErrorActionIndicator { ThrowOnError, NoThrow };@
And the functions now have a 3rd arg @co...
28 Jun 2016
- 21:37 Design #789: NumTheory: behaviour of InvMod when inverse does not exist
- John Abbott wrote:
> Perhaps the @NoThrow@ arg as suggested in comment 9 is clearer?
clearer
- 19:41 Design #789: NumTheory: behaviour of InvMod when inverse does not exist
- Here is another proposal for the interface:
* @InvMod(r,m)@ throws if the inverse does not exist
* @InvMod(r,m, fla... - 07:26 Design #789: NumTheory: behaviour of InvMod when inverse does not exist
- John Abbott wrote:
> Another possible interface is an optional 3rd arg to say not to throw: _e.g._ @InvMod(res,mod,N...
27 Jun 2016
- 22:41 Design #789: NumTheory: behaviour of InvMod when inverse does not exist
- Another possible interface is an optional 3rd arg to say not to throw: _e.g._ @InvMod(res,mod,NoThrow)@. This avoids...
- 17:26 Design #789: NumTheory: behaviour of InvMod when inverse does not exist
- We can indeed supply two (or more?) fns.
I like @InvMod@ as it currently is: it can be used conveniently inside a fo... - 17:20 Design #789: NumTheory: behaviour of InvMod when inverse does not exist
- The documentation was already updated to say that @ERR:DivByZero@ is thrown if the inverse does not exist.
Note th... - 14:40 Design #789: NumTheory: behaviour of InvMod when inverse does not exist
- Thinking about it: I would prefer to have 0 if it is relatively costly to determine if it is divisible.
We also ha... - 14:05 Design #789: NumTheory: behaviour of InvMod when inverse does not exist
- The actual implementations in @NumTheory.C@ do throw rather than return 0 -- I do not know when the code was changed....
- 13:17 Feature #898: New function: cardinality of finite field?
- *@ExtensionDeg@* is certainly clearer than *@ExtnDeg@*. Right now I cannot think of any good reason to push for succ...
- 11:45 Feature #898: New function: cardinality of finite field?
- John Abbott wrote:
> I think it makes most sense to change @LogCardinality@ to @ExtnDeg@ (any other candidate names?... - 11:41 Feature #898 (In Progress): New function: cardinality of finite field?
- I think it makes most sense to change *@LogCardinality@* to *@ExtnDeg@* (any other candidate names?), and then to rej...
- 11:06 Feature #898: New function: cardinality of finite field?
- John Abbott wrote:
> I'm not enthusiastic about a function for computing the cardinality.
> * As Anna recalled, the... - 11:03 Feature #898: New function: cardinality of finite field?
- I'm not enthusiastic about a function for computing the cardinality.
* As Anna recalled, there is already *@LogCardi... - 08:41 Feature #898: New function: cardinality of finite field?
- JAA wrote in issue #107
> JAA wonders whether *@ExtnDeg@* may not be a better name for *@LogCardinality@*. The name ... - 08:37 Feature #898 (In Progress): New function: cardinality of finite field?
- We currently have the function @LogCardinality@ for a finite field.
Should we also have @cardinality@?
Should it wo... - 09:11 Feature #902: New function: FrobeniusMat
- done, cvs-ed.
Needs proper testing - 09:09 Feature #902 (Closed): New function: FrobeniusMat
- Convert code from CoCoA-5
- 09:08 Feature #900: New function: MinPoly of RingElem in quotient ring
- From existing functions in CoCoA-5
- 09:05 Feature #900 (Closed): New function: MinPoly of RingElem in quotient ring
- Convert code from CoCoA-5
- 09:06 Feature #901 (Closed): New function: PrimaryDecomposition0 in CoCoALib
- Convert code from CoCoA-5
- 08:55 Feature #899: IsMaximal, IsPrimary for IDEAL (in cocoalib)
- Implemented, need proper testing.
- 08:50 Feature #899 (Closed): IsMaximal, IsPrimary for IDEAL (in cocoalib)
- *2017-11*: actual setting of flags is discussed in #924
- 08:48 Feature #107: Recognizing finite fields
- Implementation of @IsMaximal@ for multivariate rings now allows working non-simple extensions.
25 Jun 2016
- 21:54 Design #789: NumTheory: behaviour of InvMod when inverse does not exist
- In this case I think that the "backward compatibility" aspect is of relatively little importance.
If someone has use... - 19:02 Bug #860 (Feedback): Check impl of RingTwinFloatImpl::myIsRational
- This has largely been completed; there remains only the mystery as to why @SimplestBigRatBetween@ is too slow -- this...
- 18:58 Slug #897 (Closed): SimplestBigRatBetween: why is it so slow?
- Investigations in issue #860 showed that @SimplestBigRatBetween@ was mysteriously slower than expected (and than an e...
- 16:40 Feature #896 (In Progress): myIsEqual, myCmp: direct comparisons between RingElem and MachineInt, BigInt and BigRat?
- I note there could be a "mild inconsistency" in the @RingTwinFloat@ implementation: let @N@ be a @BigInt@ and @x@ a t...
- 14:33 Feature #896: myIsEqual, myCmp: direct comparisons between RingElem and MachineInt, BigInt and BigRat?
- Here are the new mem fns which every ring would have to offer:
* @myIsEqual(RingElem, MachineInt)@
* @myIsEqual(Rin... - 14:28 Feature #896: myIsEqual, myCmp: direct comparisons between RingElem and MachineInt, BigInt and BigRat?
- Here are some pros and cons:
* *(+)* avoid making temporary ring-elem values (wasteful new-delete cycle);
* *(+)* s... - 14:09 Feature #896 (In Progress): myIsEqual, myCmp: direct comparisons between RingElem and MachineInt, BigInt and BigRat?
- Currently all comparisons between a @RingElem@ and a @MachineInt@ (or @BigInt@ or @BigRat@) entails first converting ...
- 14:11 Design #859 (Closed): Twin-float: comparisons and equality test
- I have updated the documentation for @RingTwinFloat@.
In practice the current impl of @myCmp@ deals first with the... - 12:21 Support #696 (Closed): test-OrderedRing: activate or eliminate?
- 12:19 Bug #853 (Closed): NearestInt can needlessly throw InsufficientPrecision
- This has been in feedback for 3 months without any reports of problems.
Apparently the (new-ish?) test @test-Ordered... - 12:08 Bug #858 (Closed): floor for TwinFloat can produce ERR::SERIOUS
- I have added a new test (@test-bug7.C@) for this issue.
Since there have been no problems for 3 months, I am closing... - 11:40 Bug #895: NumDigits: sometimes gives wrong answer
- I have just tried a quick experiment comparing speeds: let @N := 10^(10^9);@
* @NumDigits(N,10)@ takes less than 0....
24 Jun 2016
- 16:53 Bug #895: NumDigits: sometimes gives wrong answer
- Which variants of @NumDigits@ should exist and what should they be called?
* base 10 possible name @NumDigits10@
*... - 16:26 Bug #895 (In Progress): NumDigits: sometimes gives wrong answer
- The GMP function is guaranteed to be fast (very fast!).
The correct definition is @NumDigits(N,base) = 1 + FloorLo... - 16:22 Bug #895 (Closed): NumDigits: sometimes gives wrong answer
- @NumDigits@ is a direct call to the CoCoALib fn called @NumDigits@ which is a direct call to the GMP fn @mpz_sizeinba...
22 Jun 2016
- 16:15 Design #891: Replace auto_ptr in preparation for C++11?
- I have looked on the internet for "clever" ways to make our code compatible with both C++03 and C++11 (using @auto_pt...
- 13:26 Design #891: Replace auto_ptr in preparation for C++11?
- Just a quick note about doing a global replacement of @auto_ptr@ by @unique_ptr@. Open all files in emacs, then use ...
- 13:24 Design #891 (In Progress): Replace auto_ptr in preparation for C++11?
- I have just globally replaced @auto_ptr@ with @unique_ptr@.
Just 1 line in @TmpJBMill.C@ needed to be changed to get...
21 Jun 2016
- 14:47 Design #894 (In Progress): strict enum types: C++11 extension
- Apparently C++11 offers "strongly typed enum" which do not have automatic conversion to integer values.
This seems... - 14:37 Feature #873: bool3: add some logical operations?
- On the other hand, if we choose not to be "clever" then we can use normal (inline?) functions which always evaluate a...
- 14:17 Feature #873: bool3: add some logical operations?
- I would like that the definitions of @and3@ and @or3@ behave just like the built-in logical operators @operator&&@ an...
- 13:12 Design #891: Replace auto_ptr in preparation for C++11?
- Here are the files which use @std::auto_ptr@:
@GlobalManager.H@
@MemPool.H@
@SparsePolyRing.H@
@SugarDegree.H@
@...
20 Jun 2016
- 15:01 Design #849: Cleanup MinPoly code
- The function @MinPoly@ (and @MinPolyQuot@) should be straight calls to @MinPolyDef/MinPolyDefQuot@
- 14:54 Design #849: Cleanup MinPoly code
- The current functions should be renamed @MinPolyXXQuot(f, I, z)@
17 Jun 2016
- 22:24 Design #891: Replace auto_ptr in preparation for C++11?
- A quick check we should do is to string-replace all @auto_ptr@ with @unique_ptr@ and see if the code compiles cleanly...
- 22:23 Design #891: Replace auto_ptr in preparation for C++11?
- Obviously we are not the first to encounter this problem.
My understanding is that @std::unique_ptr@ is intended a... - 22:17 Design #891 (Closed): Replace auto_ptr in preparation for C++11?
- Compiling CoCoALib with @--std=c++11@ produces lots of warnings about @std::auto_ptr@ being deprecated
Reportedly ... - 15:51 Feature #357: Constructor for vectors? CoCoAVector
- We have these (for short lists)...
16 Jun 2016
- 17:18 Design #827: NewPositiveMat also for matrices over QQ? Also NewIntegerOrdMat. (now called MakeTermOrd)
- What is the status of this issue?
I'm hoping it can be closed quickly and easily for the next (imminent) release.
- 17:14 Bug #814: PPOrdering: matrix ordering, what rings are allowed.
- What is the answer to the original question?
Comment 1 proposed an answer, and asked for discussion. The other co... - 17:09 Feature #800: PPMonoidSparse: impl of sparse PPs
- As reported in issue #842 execution speed is sometimes very disappointing, so currently it may not be such a good ide...
- 17:03 Design #789: NumTheory: behaviour of InvMod when inverse does not exist
- We could also offer 2 functions: @InvMod@ which throws, and @InvModNoThrow@ which returns 0.
To maintain backward ...
15 Jun 2016
- 14:35 Bug #547: elim using RingTwinFloat triggers error "SERIOUS"
- I have confirmed that both examples still trigger @ERR::SERIOUS@ with 5.1.5 on CoCoALib 0.99543.
Also available in: Atom