Activity
From 12 Apr 2014 to 11 May 2014
09 May 2014
-
22:27 CoCoALib Design #551: Reconsider overloaded virtual mem fns
- Just a hint to find easily those bits of code which need to be reviewed: look for @using@ (without @std@) inside the ...
-
- The warning flag about @-Woverloaded-virtual@ suggests that it may be better to avoid overloading virtual fns, _i.e._...
-
- I've tested and checked in my revised impl.
Perhaps I was too paranoid after finding an ugly bug in @IsHomog@ (also ... -
- Wouldn't it be more natural to specify indexed symbols using this syntax: @RingElem(R, "x", [2,3])@?
-
- OK, I think I have figured out what the code is supposed to do. I've added some documentation (and improved variable...
-
- What is going on in the impl of builtin fn @RingElem@; source is in @BuiltinFunctions.C:949-977@???
Something *BAD*! -
- We could avoid the problem by *using different names for the various mem fns*; this idea should be considered (& then...
-
- Here is an argument against blindly applying the @using@ "trick".
The class @RingBase@ defines a mem fn @myNew(con... -
- We should employ the @using@ "trick" only if it the correct design! I believe it is a bad idea to use it just to mak...
-
11:53 CoCoA-5 Bug #545: Compiler g++ 4.2.1 (clang-503.0.40) on MacOSX 10.9: warnings and errors
- problem (2): done for Sugar too.
Now to be done for all concrete rings: myNew, myAssign, myIndets.
I think that sho... -
- Anna Maria Bigatti wrote:
> So it is probably useful to keep the compilation flag and to add those "using".
> I expec... -
- John Abbott wrote:
> Similar @overload-virtual@ warnings are produced even when compiling CoCoALib. I believe I unde... -
- added small test.
Closing.
-
- This is the best we can get.
Moreover new installation is easier (does not require set names or folder positioning)
... -
-
- A similar problem is storing HilbertFunction and resolution in the ideal/module
-
- There is a naive implementation (compute MinGens, then syz, then minimalize).
Here we are talking of the optimized i... -
- It works, but should be cleaned up.
Mainly we should store the resolutions in the ideal (together with Hilbert, Jane... -
- done and documented (I can't remember when).
Closing.
08 May 2014
-
- works on mac and linux.
On linux also deals automatically with 32/64 bit binaries
-
- I've been cleaning up my scripts and insterted them in CVS.
... as usual it is difficult to make things easy!
I hav... -
- I have checked/updated essentially all fn signatures in @CoCoAHelp.xml@.
I have done nothing about @TAGGED@.
Some... -
- I now think that it is appropriate for @FloatStr@, @DecimalStr@ and @ScientificStr@ to accept @RINGELEM@ and auto con...
-
12:30 CoCoALib Feature #218: CoCoALib normaliz interface
- I implemented additional return functions (step 1 in my update on issue #204).
The myComputation functions should be...
07 May 2014
-
- Similar @overload-virtual@ warnings are produced even when compiling CoCoALib. I believe I understand the problem, an...
-
- I note that @FloatStr@ used to be implemented in @approx.cpkg5@ and would accept a @RINGELEM@ argument (which was pre...
-
- I'm not sure how widespread this should be.
Perhaps the best approach would be to implement separately for a few fun... -
- It could be helpful sometimes to allow "automatic" conversion from @RINGELEM@ to @INT@ or @RAT@. I discovered this w...
-
- I've been cleaning up my scripts and insterted them in CVS.
... as usual it is difficult to make things easy!
I have ... -
- Satisfactory; may have to be reconsidered if/when the printing of rings is reimplemented (see #548)
-
-
- Works OK; the only remaining question is whether the default should be to capitalize or not. Currently (and for 5.1 ...
-
- Implemented, documented, tested.
Closing..
-
- Oscar reports that it works for him.
Closing.
-
- Anna is happy too; look OK in LaTeX/PDF.
Closing.
-
- Added more tests (on shorter random lists, since the code handles long and short lists separately).
Regarding the ...
06 May 2014
-
- It seems OK now (Anna's happy too), so closing.
-
- I have implemented option *(1)* _i.e._ eliminated all code to do with reading multiline string literals.
Documenta... -
- John Abbott wrote:
> Here is an idea: simple to implement, but perhaps not so nice to use?
> * a ring prints out si... -
- Here is an idea: simple to implement, but perhaps not so nice to use?
* a ring prints out simply as @ring(1)@, @ring... -
- Should this topic be in CoCoALib rather than CoCoA-5?
From the point of a user it would be very handy if the C5 ex... -
- And in CoCoA-5 would look like this:...
-
- I tried to see how it looks (just in RingFp.C).
This would be the output for test-RingHom1 .... not bad, I think
<p... -
- Each ring in CoCoA has an ID number (they start at 1 and are assigned incrementally). It'd be better to print those ...
-
- for some rings there is a unique implementation: *RingQQ()*, *RingZZ()*, *RingQQt(n)*.
For all the others the constr... -
- Ahhh! OK I have fixed @mpz_log@ now; sorry, I had overlooked a @M_LN2@.
-
- simpler example: both NewFractionField and NewRingTwinFloat seems necessary to trigger the error...
05 May 2014
-
- Made the changes requested in previous comment.
No idea why you are not seeing the updated @mpz_log.c@...?
-
- One addition that I forgot to mention: I had to include...
-
- I tried the same input with lower precisions and got @InsuffPrec@ errors; but when I quit CoCoA-5 there were quite a ...
-
- The following (C5 input) provokes a @SERIOUS@ error...
-
- It is inconvenient (perhaps even embarassing) that you cannot do this:...
-
- I suggest moving @Functions@ (currently built-in!?!) to the package @obsolete.cpkg5@
*20140508* OK done this (after ... -
- I have recently added a new C5 test (@exbugs.cocoa5@) and was surprised to find that it produces output. What happen...
-
- Could we condense the syntax and description so that the pages occupies less vertical space?
-
- The man page is much improved, and easier to find. Good enough for now; maybe experience will help improve it in the...
02 May 2014
-
- Current solution is satisfactory, so closing!
-
- Current approach is satisfactory (& even fairly backward compatible if you don't nest @/*...*/@ style comments)
Cl... -
- The current doc is satisfactory (& certainly much improved); maybe future experience with students will help make it ...
-
- I have added a new C5 test file @exbugs.cocoa5@ which contains a simple verification that @sorted@ works on some rand...
-
- After the mindboggling refresher course in regexps, I'm closing this because it all (magically?) works.
-
- It works fine both for me and for Anna -- after >2 weeks' testing. Closing!
-
- Adding note about original requester: @andrei_zarojanu@ (at @yahoo.com@)
I'll email him about our partial solution -... -
- A proper solution will be very costly and highly invasive (and Giovanni's not keen to help). Furthermore, it is prob...
-
- @TimeFrom@ is a satisfactory KISS solution. Still happy after 3 weeks, so closing.
-
- I like the current doc. Changing to feedback.
-
- I have modified th fn signature to teh following:...
-
- OK I have modified @mpz_log.c@; should be fine now.
Also I have CVSed everything.
@Christof: could you confirm th... -
- Thanks, that should make the compilation a lot easier! I can test it next week, is it CVSed?
Note that there is also ...
30 Apr 2014
-
- I have removed most uses of @size_t@ in CoCoALib.
The template fn @len@ now does an automatic cast to @long@ from @s... -
- I believe I have fixed the overloading problem, and replaced @M_PI@ with @4*atan(1.0)@.
Right now I don't have a g... -
- Problem (2) -- continued
compilation with @-Woverloaded-virtual@... -
- Anna Maria Bigatti wrote:
> Next to investigate
> Problem (2)
> [...]
@-Woverloaded-virtual@ is accepted by my compil... -
- problem (4)...
-
- problem (3)...
-
- Next to investigate
Problem (2)... -
- Problem (1) [solved]...
-
- Bruno Simoes found some warnings and a error in a cocoa-5 test.
*problem (1) struct/class* fixed
*problem (2) ove...
29 Apr 2014
-
- For the new normaliz version I did again a compilation on Windows. Here are my findings for CoCoALib 0.99533:
The ...
28 Apr 2014
-
- The problem was that after this step...
-
- Easier to debug (working on it)
[6, 28, 39, 1, 40, 41, 46, 7, 2, 20, 21, 3, 0, 5, 22, 23, 24, 38, 44, 37, 4, 30, 36,... -
- As I suspected the problem seems to depend only on the relative sizes of
the list elements; here is the same example... -
- The source code is in package @list.cpkg5@.
I found no examples of short lists; the source code uses @InsertionSort@... -
- ...
24 Apr 2014
-
- The mange page is easy to find, and easy to read/comprehend now.
Closing.
-
- The new pages are fine now. Closing.
-
- Closing because it's all been working fine for ages; no point in leaving it in feedback.
-
- Old fn is @FloatApprox@ obsolete -- there was no sense in making it decimal (not now that @FloatStr@ exists).
New ... -
- Completed doc & testing. Anna's happy too. Closing.
-
- I have now removed @float.cpkg5@.
Everything seems to work just fine.
Checked/updated doc too.
-
- This does not work...
-
- While doing this massive change removing all now useless *types* I'm also doing some cleaning in the documentation.
... -
- Now that I started this work I see that it *has to be finished* urgently.
As it was, the documentation wasn't just o... -
- Closing after interactive testing with Anna. Made some minor variations on the suggestions in comment-2.
Test cas...
23 Apr 2014
-
- More updates done in "commands" *syntax*.
Now I think that each *chapter* should be essentially empty and just give s... -
- Now in CoCoAHelp.xml we should tag the types as *type* or *rtn* in the syntax like this:...
-
- The old manual had a part about data-types.
Recover it and improve it. -
- I have produced a new starting script for CoCoA-5 which autodetermines the directory containing the script (and thus ...
-
- Make it easier for a Linux user to get the right version of CoCoA-5.
One solution would be to unify the 32-bit and 6... -
- Oscar, could you tell me if this works for you (and tell me what @uname -a@ gives?)
http://cocoa.dima.unige.it/downlo...
22 Apr 2014
-
- I have implemented @ProgressReporter@.
Checked in with doc and example!
State -> feedback!
-
- It would be handy to have utility that prints out occasional "progress reports" when conducting long iterative comput...
-
- Here is the very simple test case I used:...
-
- I have just tried the @NotWorks@ example from comment-5; it works correctly with the current Emacs UI code. Changing...
-
- Nobody thought anything for three days :-(
Anyway, I have revised the doc so it now has the structure proposed in ...
19 Apr 2014
-
- I'm undecided how best to organize the manual page(s) for strings.
Currently they're in a "funny" section of the manu... -
- I've sent the files -- they ought to be translated into english...
17 Apr 2014
-
- Works OK up to stack size of about 5400; at 5500 I get a SEGV. Why????
Also get SEGV without @--stacksize@ option; ... -
- I've implemented a "quick fix":
* ctor @RuntimeEnvironment@ now has extra arg for specifying @MaxStackSize@
* ctor fo... -
- *Anna* will check the manual!
-
- *The solution chosen is (c)*
-
-
- Want to discuss precise specifications of the fn.
-
- We will discuss the design/names @IntersectList@ @intersection@ etc??
-
-
-
- > JAA noticed that ex-PolyRing1.C is too long & complicated; split it into two (one about ctors, the other about doin...
-
- Renamed source files (&Makefile) & doc; changed references in doc.
-
- first prelease on web-page
-
- *make ValgrindTests* DONE (only the "throw-16b-leaks")
-
-
-
- A factorization is arbitrary.
Choose and implement functions to clean it up (e.g. grouping all equal factors in one ... -
- (for now) we have decided that factorization of denom is arbitrary.
There will be functions to clean up a factorizat... -
-
-
- I checked and confirm that the tests listed (those "leaking" *16 bytes*) are those which test errors (at least once)....
-
- tested, added test-matrix4 (which indeed gave IMMINENT DISASTER before recompiling ;-)
16 Apr 2014
-
- *Do the "quick fix"!* Not satisfactory, but a proper soln is going to be costly :-(
-
- JAA must give Anna Giulia's code!
-
- JAA suggests:
* one page about string literals, concat and indexing
* one page about "escape sequences" inside string... -
- JAA noticed that @ex-PolyRing1.C@ is too long & complicated; split it into two (one about ctors, the other about doin...
-
- I'm not an expert in the field, and in my ignorance find the "compact" form more satisfactory. I could even envisage...
-
- I guess you could use @raw(zero(myR))@ instead of @raw(useless)@; though perhaps the latter is clearer?
-
- JAA suggested to try ...
-
- I confirm the previous leaks (probably due to test for *throwing/catching errors*)
This "16-byte leak" is independent... -
- ...
-
- I tried @cocoa-5.0.9-text-mac10.6.tgz@ (28 Mar 2014) on my machine; it did not start...
-
- Some explanation on NmzIntegrate:
The possible input is basically the same as for Normaliz. This means the cone re... -
- Which example do you mean? For me all the newer ones (from update 10 on) are working. The older should work too if yo...
-
- The example in the description does not work.
Postponing this issue to Seoul version (with full manual to be publish... -
- Implementation is in @ToString.H/C@
@DecimalStr@ now defaults to 3 digits after decimal point.
-
- I tested the new *DecimalStr*: I'd prefer 3 digits after "." by default (instead of 4).
I find it more natural to gro... -
- Added copyright blurb to @TmpMorse..@ files.
-
- JAA has tried the proposed modification, and can report that it compiles and all tests pass.
*HOWEVER*, the relevant... -
- [reported by Roberto Pagaria <roberto-dot-pagaria-chiocciola-sns-dot-it>]
Ieri ho tentato di installare CoCoALib su ... -
- [I've already written this in #419]
I tested the new *DecimalStr*: I'd prefer 3 digits after "." by default (instead... -
- Just for the record: I did the experiment from within C-5.1, and the time taken (for just @FloatStr@) was 32.3s
-
- Using CoCoAlib-0.99533 @FloatStr@ uses about 1.5Gbytes extra memory when applied to
@N:=factorial(10^8);FloatStr(N);... -
- It seems to work OK for me too.
I do vaguely recall discovering what the problem was and then fixing it (but evident... -
- Feedback
Works fine for me, no problems in strings or comments. Has this been fixed already? -
-
- Final testing (thanks Oscar Fernández Ramos!)
There are problems when loading CUA-mode (for copying and pasting wi...
15 Apr 2014
-
- The final design is:...
-
-
-
-
-
- We should start by looking at the various current uses of @bool3@, and possibly any obvious future candidates.
-
- A new *@ClearDenom@* function has been added to @SparsePolyRing@; it follows very closely the code @ClearDenom2@ I wr...
-
- It works. Closing.
-
- Reopening because it was auto-closed when I closed #139 which this duplicates.
-
- Closing because it is duplicated
-
- Closing because it is duplicated
-
-
- This is info is in the file @doc/README@ -- it's been there for some time now!
-
- The current impl is simple, compact and works fine in the cases we have tried. Let's KISS until someone points out t...
-
-
- Now it should recognize all properly written keywords (*PriNT* is accepted but not properly written)
Now there is an ... -
-
-
-
- Now it takes 60s on my machine which is quite acceptable.
Closing.
-
-
-
-
- The doc is now much improved; no doubt it will be tweaked occasionally in the future, but this issue can now be close...
-
- No problems have arisen in 2 years. In ptic the fns *@IsPthPower2@* and *@PthRoot2@* have not made their absence fel...
-
- The mem fn @RingBase::myCharacteristic@ returns @void@ (supposedly for "efficiency", but who's going to calculate mil...
14 Apr 2014
-
- > JAA suggests putting it in *MatrixOperations*
(after talking about it on the phone) moved all *Is..* functions f... -
- John Abbott wrote:
> JAA suggests putting it in *@MatrixOperations@*
OK. Should we also move *IsZeroRow/Col*? -
- JAA suggests putting it in *@MatrixOperations@*
-
- Added to cocoa-5
I had to add *evalArgAsT1orT2orT3orT4orT5orT6orT7* in *Interpreter.H*.
-
- Implemented as described above.
Seems to work (added test @IsZero(ZeroMat(R, 12345,67890))@ and is fast).
But where... -
- JAA suggests as a general rule: try to avoid designing tests that produce much output (unless the test specifically c...
-
- I noticed that CoCoAInterpreter for 10.5 works on my 10.6.
... so John fixed it?
What about viceversa?
John: could... -
-
- Should this function work only for list of RINGELEM?
If so we can directly use *HasUniqueOwner*, otherwise is is ver... -
- I think that now the manual is more diligently updated than the *Translation table*.
Close this issue?
-
- Now the make target is done and works well. Moreover generating the TeX documentation is actually useful for checkin...
13 Apr 2014
-
- I tried implementing it in a clever way but got stuck.
(clever = ZeroImpl gives true, IdentityImpl gives false, ...)...
Also available in: Atom