Opened 10 years ago
Closed 10 years ago
#6782 closed defect (fixed)
doctest failure in doc/en/constructions/calculus.rst due to upgrade to Maxima 5.19.1
Reported by: | drkirkby | Owned by: | AlexGhitza |
---|---|---|---|
Priority: | major | Milestone: | sage-4.1.2 |
Component: | calculus | Keywords: | maxima |
Cc: | Merged in: | ||
Authors: | Reviewers: | ||
Report Upstream: | Work issues: | ||
Branch: | Commit: | ||
Dependencies: | Stopgaps: |
Description
On Solaris (SPARC), the following test failed. Both ECL and Maxima were updated - ECL version 9.8.4, Maxima version 5.19.1
---------------------------------------------------------------------- | Sage Version 4.1.1, Release Date: 2009-08-14 | | Type notebook() for the GUI, and license() for information. | ---------------------------------------------------------------------- Thu Aug 20 20:02:37 BST 2009 dsage-trial tmp directory doesn't exist - creating ... This script will run the unit tests for DSage
<SNIP>
File "/export/home/drkirkby/sage/sage-4.1.1/devel/sage/doc/en/constructions/calculus.rst", line 117: sage: maxima(f).powerseries(x,0) Expected: ('sum((-1)^i2*2^(2*i2)*bern(2*i2)*x^(2*i2)/(i2*(2*i2)!),i2,1,inf))/2 Got: 'sum((-1)^i2*2^(2*i2-1)*bern(2*i2)*x^(2*i2)/(i2*(2*i2)!),i2,1,inf)
Attachments (1)
Change History (8)
comment:1 Changed 10 years ago by
- Summary changed from Doctest failure in sage-4.1.1/devel/sage/doc/en/constructions/calculus.rst to 1 doctest failure in sage-4.1.1/devel/sage/doc/en/constructions/calculus.rst
comment:2 Changed 10 years ago by
- Keywords maxima added
- Owner changed from burcin to AlexGhitza
- Status changed from new to assigned
- Summary changed from 1 doctest failure in sage-4.1.1/devel/sage/doc/en/constructions/calculus.rst to [with patch, needs review] doctest failure in doc/en/constructions/calculus.rst due to upgrade to Maxima 5.19.0
comment:3 Changed 10 years ago by
- Summary changed from [with patch, needs review] doctest failure in doc/en/constructions/calculus.rst due to upgrade to Maxima 5.19.0 to [with patch, needs review] doctest failure in doc/en/constructions/calculus.rst due to upgrade to Maxima 5.19.1
The log was generated with Maxima 5.19.1, not 5.19.0, so I'm changing the title slightly to reflect this.
comment:4 Changed 10 years ago by
Are these mathematically equivalent? It seems to me the factor in the numerator has changed from (2*i2) to (2*i2-1) and denominator from 2 to 1. That seems a different result to me.
I tried what I think the test is doing in Mathematica and got:
In[7]:= f=Log[Sin[x]/x] Sin[x] Out[7]= Log[------] x In[8]:= Series[f,{x,0,12}] 2 4 6 8 10 12 -x x x x x 691 x 13 Out[8]= --- - --- - ---- - ----- - ------ - ---------- + O[x] 6 180 2835 37800 467775 3831077250 In[9]:=
I've no idea what's right or not.
Dave
comment:5 Changed 10 years ago by
The difference is 2^(2*i2)/2
versus 2^(2*i2-1)
. Yes, they are the same.
comment:6 Changed 10 years ago by
- Summary changed from [with patch, needs review] doctest failure in doc/en/constructions/calculus.rst due to upgrade to Maxima 5.19.1 to doctest failure in doc/en/constructions/calculus.rst due to upgrade to Maxima 5.19.1
comment:7 Changed 10 years ago by
- Resolution set to fixed
- Status changed from assigned to closed
This is fixed by #6699.
This one is trivial; see attached patch.