[SCM] polybori: Polynomials over Boolean Rings branch, master, updated. upstream/0.8.2-51-g2f116f3

Alexander Dreyer adreyer at gmx.de
Tue Jan 8 12:13:26 UTC 2013


The following commit has been merged in the master branch:
commit 2f116f3ef13c72e37db692ba3abbb5c42fd3333c
Author: Alexander Dreyer <adreyer at gmx.de>
Date:   Tue Jan 8 13:13:09 2013 +0100

    Fixed and discussed most todos now

diff --git a/debian/control b/debian/control
index de2bf71..6de2b9b 100644
--- a/debian/control
+++ b/debian/control
@@ -5,7 +5,6 @@ Uploaders: Alexander Dreyer <adreyer at gmx.de>,
 Section: math
 Priority: extra
 Build-Depends: debhelper (>= 9),
-               patchutils,
                scons,
                libboost-dev,
                libboost-python-dev,
diff --git a/debian/todo b/debian/todo
index 865a65d..78d31d9 100644
--- a/debian/todo
+++ b/debian/todo
@@ -8,15 +8,16 @@ packaging
 - full license check with at least licensecheck. Each file of the
   package should have the right copyright informations. (this is also
   why embedded libraries is a pain)
-  > I double checked and found files in Cudd, which have inclear licenses. 
+  > I double checked and found files in Cudd, which have unclear licenses. 
     We do not need them, so I removed them and added a Comment to
-    debian/copyright. 
+    debian/copyright. (Also removed upstream.)
 
 - 0005_PolyGUI.patch was commited applyed into the master branch. It
   would be nice to use gbp-pq to produce the patch series.
-  > The repo is not switched to gbp-pq (was git-gdm, but not consistently)
+  > The repo is now switched to gbp-pq (was git-gdm, but not consistently)
 
 - deal with all remaining bugs from debbug
+  > #636466 (Compatibility with ipython 0.11) is fixed in 0.8.2.
 
 - replace Pre-Depends: multiarch-support by Pre-Depends:
   ${misc:Pre-Depends}
@@ -33,6 +34,7 @@ packaging
 - if ipbori could be replace by an ipython profile, maybe install this
   profile under python-polybori binary package and remove the polybori
   package.
+
   > This is not possible, since ipbori does more that a profile could do.
     For instance, it supports various versions of ipython, which the profile
     syntax just cannot do.
@@ -41,6 +43,7 @@ packaging
   but I have no trace of those 0.8-0 and 0.8.2-1~exp0 into the Debian
   archive. So fix this by merging the three entries into only one.
   0.8.2-1~exp1 and target experimental instead of unstable
+
   > Indeed we did not mange to make those variants official, so all three
      entries are menged into one now. It's also "experimental" now.
 
@@ -57,7 +60,7 @@ packaging
    _Z10error_textP15pbori_DdManager at Base 0.8 this should be the
    package version (changelog) without the debian version. In your
    case 0.8.2
-   > done (Thanks for the hint!)
+   > done 
 
 - instread of
  override_dh_strip:
@@ -83,7 +86,7 @@ packaging
   the control file and the git repository location). there is english
   natives which can help and improve a lot the description of the
   package. It helps me a lot before.
-
+>>>>>>>>>> TODO
 - Build-Depends: debhelper (>= 9),
                patchutils,
                scons,
@@ -95,7 +98,7 @@ packaging
                python-all-dev
 
  it seems to me that libboost-dev, patchutils are useless
->>>>>>>>TODO
+  > patchutils removed, but we need more than just python bindings from boost.
 
  python-setuptools should be avoid if
  build-python%: PYTHONSITE=`/usr/bin/python$* -c 'import setuptools; print setuptools.distutils.sysconfig.get_python_lib() '`
@@ -107,7 +110,8 @@ packaging
  
  you depends of hevea | tex4ht is it used by doxygen ?, if not you can remove it
 
-  > no, we have a LaTeX-based tutorial, which is converted to html
+  > no, we have a LaTeX-based tutorial, which is converted to html using
+    one or the other
 
 - your package provide a arch: all package.  it would be nice in that
   case to create an indep part like this one:
@@ -132,25 +136,31 @@ packaging
  modified.
 
   >>>>>>>>> TODO
+
+
 upstream
 --------
 
 - remove the embeded M4RI
-
+  > This will be done, but first we need to figure out a convenient way how to 
+    obtain libm4ri, if it is not available on the system.
+  
 - Cudd package or not package ?
   > no, we forked it, it's not compatible
 
 - some part of the python code do not follow the pep8 recommendations.
   indented with 2 spaces instead of 4.
-
+  > done (also patched into the package.)
 
 - why not using QtDesigner instead of hard coding the ui part ?
+  > when it's getting more complex. Now it's just a small python script.
 
 - ipbori can be an ipython profile instead of a dedicated binary ?
   > no (see above)
 
 - integrate the .desktop file in the upstream installation process.
   this way it can be usefull to other distributions.
+  > done
 
 - rename polybori.pc into polybori-0.8.pc this is much more consistant
   with the current naming schema. idem for the other library.
@@ -158,3 +168,9 @@ upstream
   > done
 
 - maybe add an option for the documentation compilation only.
+
+ >>>>>>>>> TODO
+  > First, we have to separate the python modules documentation 
+    (not platform independend) from the doxygen/LaTeX-based stuff.
+    I'm also not sure whther the doxygen-based stuff in platform independent,
+    since it also analyzes polybori/config.h
\ No newline at end of file

-- 
polybori: Polynomials over Boolean Rings



More information about the debian-science-commits mailing list