Activity
From 07 Mar 2020 to 05 Apr 2020
26 Mar 2020
-
11:40 CoCoA-5 Design #990: CoCoA-5 distribution: tidying
- It's a bit late to make a change for the current version, but we could do it for the next release...
I suggest postp... -
- Here is a quick summary of what I think the installation "script" for CoCoALib does:
* create new dir with name like... -
- I have a prototype impl, much like the impl for *@HomogCompt@*.
It is disappointingly slow for larger polynomials:...
24 Mar 2020
-
22:37 CoCoA-5 Support #1222 (Feedback): Release CoCoA-5.3.0
- release for windows done (maybe)
-
- This is a little bit like *@HomogCompt(f,0)@*. See issue #1439.
20 Mar 2020
-
- John Abbott wrote:
> *NOTE*: the linux version unpacks into *@cocoa-5.3.0/@* whereas the Mac version unpacks into *@... -
- *NOTE*: the linux version unpacks into *@cocoa-5.3.0/@* whereas the Mac version unpacks into *@CoCoA-5.3@*. *Intenti...
-
- The section in the web page with "Notes" looks a bit strange: _e.g._ "New and renamed files" seems to have two empty ...
-
- I suggest putting the SHA-256 checksum in the "Notes" section on the web page for source releases.
Probably the best... -
- *IMPORTANT* see the discussion in issue #1443 (about "illegal instruction") before doing much here!
-
-
- I have just checked in a (very) minor update to the man page for @RandomLinearForm@.
I have added a see-also to @Hom... -
- Robbiano requested the new function @LinearForm@ (see #1439), and clearly had in mind that it should be linear wrt th...
-
- I suggest making the manual entry for *@RandomLinearForm@* more explicit, now that we have @HomogCompt@ which respect...
-
- Anna says she has now managed to get successful compilation with N-3.8.4.
She should register (to this issue) the ti... -
- I have postponed this issue to the next release. This release we used the workaround to compile on a different machi...
-
- I think this issue has run its course, so I'm closing it.
The important points are:
* comment 24 (all of it)
* c...
14 Mar 2020
-
- OK, I do not know what is going on. This is *really annoying!*
I have compiled CoCoA-5 with the *@generic@* opti... -
- The illegal instruction seems to be coming from the statically linked GMP library -- that is *@libgmp.a@* (and/or @l...
-
- There's a new version of GMP!! Released almost 2 months ago, but I didn't notice.... odd!
12 Mar 2020
-
- John checked
-
- I have sent several stripped executables to Matteo; some have worked, some have not.
Anyway, I am now convinced that... -
- Last night I made a release version (stripped, with statically linked @libgmpxx.a@, @libgmp.a@, @libreadline.a@, @lib...
-
- It could also be that the "illegal instruction" does not come from CoCoA at all, but is in one of the already compile...
-
- There is information on the web page:...
-
- It seems that *@--march=i686@* is *not correct*. It says that it does not support x86_64 instruction set :-(
Inves... -
- We need to do this piecemeal as many different classes must be considered. Besided modifying the class, we will prob...
-
- C++14 (maybe even C++11?) offer an appealing notion of "move ctor" which avoid some deep copies.
We should adapt o... -
- What should we do with this issue? The correct "solution" is for us to learn the foibles of C++.
It does highlig... -
- What is the status this issue? Is it now fully resolved?
Anna succeeded (2020-03-11) in compiling on her Mac, pres... -
- Find a way to automate what has been discovered in issue #755.
Presumably it would be nice if I could just "downlo...
11 Mar 2020
-
- No one seems to care about this issue; I'm pretty sure it has lost all relevance, so I'm closing it.
-
- Regarding comment 7: I spotted the code which checked for 64-bit platform, and complained if not.
Now I'm undecide... -
- Added doc for CoCoALib.
Still no test :-/
Close anyway?
-
- I have implemented @HomogCompt@. Also accessible from CoCoA-5 (with man page).
No tests, no CoCoALib doc (oops... I... -
- What should the fn name be? *@HomogForm@* or *@HomogCompt@*?
-
- John Abbott wrote:
> I now have a prototype of *@HomogForm(f,d)@* in C++.
> It seems to be about 30 times faster th... -
- I now have a prototype of *@HomogForm(f,d)@* in C++.
It seems to be about 30 times faster than @LinearForm@ in CoCoA... -
- adjusted the page and the release-xxx.sh file to produce the correct files and instructions for publishing.
-
- I think I might be slightly confused. What I wrote above in comments 11 and 12 actually refers to the "binary" relea...
-
- Actually this issue was about helping "install" the CoCoA executable on *LINUX*.
Probably we should handle two cas... -
- Yes, that is clearer, thanks! As you say, it is also a good idea to test it with some "volunteers" ;-)
It is easi... -
- John Abbott wrote:
> The problem was that he had unpacked the zip file in @c:\@, which created a directory with a na... -
- I'm still waiting for feedback from Matteo.
-
- John Abbott wrote:
> Florian suggests that the name *@LinKerZZ@* is superfluous, and that *@LinKer@* should automati...
10 Mar 2020
-
- I have just found on StackOverflow that one can use @gcc@ to find where libraries are:...
-
- It seems that I must specify both @libtermcap@ and @libtinfo@ statically (and in that order!). On my Ubuntu box they...
-
- @libreadline@ on my Ubuntu machine depends on the *@tinfo@* library; I have a static version at *@/usr/lib/x86_64-lin...
-
- Florian has already put a prototype impl (CoCoA-5) in issue #1304. This should be ported to C++.
NTL has HNF whic... -
- It would be nice to have HNF in CoCoALib.
I can see two cases:
* *(special)* over @ZZ@
* *(general)* over a more... -
- Florian suggests that the name *@LinKerZZ@* is superfluous, and that *@LinKer@* should automatically call @LinKerZZ@ ...
-
-
- I have changed the @release-linux.sh@ script so that it strips @CoCoAInterpreter@ before making the TGZ file.
We c... -
- So, should we put a stripped version of the executable in the official release, or leave the symbol table in there?
... -
- After running @strip@ on the executable the size dropped to 15Mbytes. Wow!
Now, why is the Mac version so big? ;... -
- Why is the last binary release for Linux about 81Mbyte when that for Mac is only 27Mbyte?
The Linux one is surely ... -
- -Maybe the correct flag is *@-march=i686@*.- *NO! This did not work* (see comment 5)
Presumably this has to be ... -
- I do not know what the problem can be at the moment.
I would not expect stripping a binary to cause such a problem... -
- I sent a compressed stripped binary to Matteo via email, but he reports a problem with *@Illegal instruction@*.
In... -
- A student here had trouble installing CoCoA-5.2.5 on a Microsoft Windows 10 computer.
The problem was that he had ...
09 Mar 2020
-
- John Abbott wrote:
> Do we have a function which computes a list of homog compts?
no, we explicitely decided not ... -
- Do we have a function which computes a list of homog compts?
-
- I have also added a quick check that there are no writable files in the subtree to be released.
I have modified @r... -
- So far I have made a small change to @release-linux.sh@.
I have added a line after the @MakeTGZ@ line:... -
- I'll check once we have a source TGZ release. When that is clean, I'll close this issue (hopefully today or tomorrow...
-
- I have just checked the contents of @cocoa-5.3-mac.tgz@.
There are three files whose names begin with dot-underscore... -
- Anna Maria Bigatti wrote:
> John Abbott wrote:
> > Maybe the commands in the script @FIX_PERMS.sh@ can be put into ... -
- John Abbott wrote:
> Maybe the commands in the script @FIX_PERMS.sh@ can be put into a shell function in @release-co... -
-
- This is almost resolved: we add some extra time when computing the RGB in the main loop.
This works well, but the qu... -
- Closing
-
- Anticipated because it is an example of the main new feature of this release: @GetLine@
(the reason why it is 5.3.0)
-
- Many useful GFan-based functions are implemented in CoCoA-5.
Port them into CoCoALib.
-
- updated starting page: https://cocoa.dima.unige.it/redmine/
-
- updated starting page: https://cocoa.dima.unige.it/redmine/
-
- vandal ;-)
08 Mar 2020
-
- I found this behaviour after deliberately feeding random input to CoCoA-5. It survived :-)
-
- CoCoA correctly gives a parse error given the input *@8.a;@*
The error message is:...
07 Mar 2020
-
- updated RelNotes
Checked in everything
-
- documentation done
(but checkin later -- flat battery) -
- Done (in NotBuiltin.cpkg5)
Funny cases:
if f = 0 returns 0
if GradingDim<>1 returns error
use weights: es:
<pr... -
- Add function which returns the sum of the summends of degree 1.
-
- Notes for me. In general: M matrix ordering.
If first row of M all positive
m m m 1
M 0
otherwise
1 1 1...
Also available in: Atom