Opened 11 years ago
Closed 10 years ago
#13315 closed enhancement (fixed)
Clean up MANIFEST.in, spkgdelauto and spkgdist
Reported by:  jdemeyer  Owned by:  tbd 

Priority:  minor  Milestone:  sage5.3 
Component:  distribution  Keywords:  
Cc:  Merged in:  sage5.3.beta1  
Authors:  Jeroen Demeyer  Reviewers:  John Palmieri 
Report Upstream:  N/A  Work issues:  
Branch:  Commit:  
Dependencies:  Stopgaps: 
Description (last modified by )
There are some obvious improvements and simplifications to be done to devel/sage/MANIFEST.in
and devel/sage/spkgdist
.
Apply 13315_manifest.patch
Attachments (1)
Change History (16)
comment:1 Changed 11 years ago by
Description:  modified (diff) 

comment:2 Changed 11 years ago by
Description:  modified (diff) 

Summary:  Clean up devel/sage/MANIFEST.in → Clean up devel/sage/MANIFEST.in and spkgdist 
comment:3 Changed 11 years ago by
Description:  modified (diff) 

comment:4 Changed 11 years ago by
Description:  modified (diff) 

Status:  new → needs_review 
comment:5 Changed 11 years ago by
Summary:  Clean up devel/sage/MANIFEST.in and spkgdist → Clean up MANIFEST.in, spkgdelauto and spkgdist 

comment:6 Changed 11 years ago by
comment:7 Changed 11 years ago by
At first glance this seems okay to me. But why not more clean up, for example

MANIFEST.in
diff git a/MANIFEST.in b/MANIFEST.in
a b 1 recursiveinclude * *.c *.cc *.cpp *.h *.hh *.hpp*.pyx *.pxd *.pxi *.txt *.tex2 recursiveinclude doc *.rst *. py *.inc1 globalinclude *.c *.cc *.cpp *.h *.hh *.hpp *.py *.pyx *.pxd *.pxi *.txt *.tex 2 recursiveinclude doc *.rst *.inc 3 3 graft .hg 4 4 include .hgignore .hgtags 5 5 include MANIFEST.in 6 6 include bundle export install pull sagepush 7 include module_list.py8 include fpickle_setup.py9 7 include spkgdelauto spkgdist spkginstall 10 8 include c_lib/SConstruct 11 9 include sage/ext/macapp/README 12 10 include sage/graphs/graph_decompositions/rankwidth_c/README 13 include sage/libs/gmp/__init__.py14 include sage/libs/polybori/__init__.py15 include sage/numerical/test.py16 include sage/rings/polynomial/padics/__init__.py17 11 include sage/schemes/hyperelliptic_curves/hypellfrob/README 18 include sage/server/notebook/templates/*19 include sage/server/notebook/templates/css/*20 include sage/server/notebook/templates/notebook/*21 include sage/server/notebook/templates/worksheet/*22 12 include sage/tests/french_book/README 13 graft sage/server/notebook/templates 14 graft sage/server/notebook/templates/css 15 graft sage/server/notebook/templates/notebook 16 graft sage/server/notebook/templates/worksheet 23 17 include doc/common/python.inv 24 18 include doc/common/updatepythoninv.sh 25 19 include doc/en/introspect/static/empty 26 20 include doc/en/introspect/templates/layout.html 27 include doc/en/numerical_sage/errors28 21 include doc/en/website/static/pdf.png 29 22 include doc/en/website/templates/index.html 30 23 include doc/en/a_tour_of_sage/eigen_plot.png
(Why do we include doc/en/numerical_sage/errors
? Why don't we include all *.py
files?)
comment:8 Changed 11 years ago by
Also, regarding line 22 of spkgdist:
cp pR .hg* doc sage c_lib tmp/
copying the entire doc directory can take a while, since we're copying all of the output. Would it be better to copy doc/common
and doc/[az][az]
?
comment:10 Changed 11 years ago by
Description:  modified (diff) 

Status:  needs_work → needs_review 
comment:11 Changed 11 years ago by
In line 1 of MANIFEST.in, should we replace recursiveinclude * ...
with globalinclude ...
? Then I think we can delete line 8 (include fpickle_setup.py module_list.py
). In my testing, the results are the same with these changes.
comment:12 Changed 11 years ago by
Status:  needs_review → needs_work 

Changed 11 years ago by
Attachment:  13315_manifest.patch added 

comment:13 Changed 11 years ago by
Status:  needs_work → needs_review 

comment:14 Changed 11 years ago by
Reviewers:  → John Palmieri 

Status:  needs_review → positive_review 
Okay, this looks good to me.
comment:15 Changed 10 years ago by
Merged in:  → sage5.3.beta1 

Resolution:  → fixed 
Status:  positive_review → closed 
I tested this patch by creating a new sdist (starting from sage5.2), and these are the only differences in files being added or not (this is the diff of
ls a1RF
output):sage5.2.
.sconsign.dblite