Re: [Gpsbabel-code] ANN: GPSBabel moved from cvs/Sourceforget to svn/Code.google.com
Brought to you by:
robertl
From: tsteven4 <tst...@qw...> - 2011-07-26 12:38:21
|
Thanks. Happy to have to have the code in subversion. Build seems broken this morning (./configure && make) > make: *** No rule to make target `garmin_fit.o', needed by > `gpsbabel'. Stop. > > ~/work/gpsbabel-read-only/gpsbabel% svn info > Path: . > URL: http://gpsbabel.googlecode.com/svn/trunk/gpsbabel > Repository Root: http://gpsbabel.googlecode.com/svn > Repository UUID: f51c46e8-681c-474f-0cfe-069cfd0219fb > Revision: 4079 > Node Kind: directory > Schedule: normal > Last Changed Author: rob...@gm... > Last Changed Rev: 4079 > Last Changed Date: 2011-07-25 23:59:31 -0600 (Mon, 25 Jul 2011) On 7/25/2011 10:36 PM, Robert Lipe wrote: > CVS access to GPSBabel was recently down for five days. This is a > recurring problem with SourceForge. The decision to move to a more > reliable code host - and a new SCM - was thus somewhat forced, but > relatively easy and it's something I've wanted to do for a while. > Sourceforge served us well in the early days, but it's time for us to > move on. > > Effective immediately, the main repository for GPSBabel is > http://code.google.com/p/gpsbabel/ > Relying on third parties always has risks, but I'm pretty confident > that Google shares my take that recurring multiple day outages are not > good. > > Why SVN? Because tools like Mercurial and Git add complexity and > solve a lot of problems we don't have. It's very familiar to CVS > users and adds things > like atomic submits and the ability to rename files. (Ooooh. Aaaah.) > It's also free, commonly available, and is cross-platform. > > I've done a large conversion of the project, minus the out of date > web pages which were effectively machine generated from the xmldoc > source, out > of date, and expensive to convert. All the revision history and > branches and tags are available in SVN. > > While I was wreaking havoc, I reformatted the entire project with > astyle, as described a few months ago, and the entire project now > compiles (and successfully runs the test suite) in both C and C++. > This means the code currently has the worst of both languages. > > I'll also be restructuring the source into directories (formats/, > filters/, core/ seem obvious) discontinuing the autoconf/configure and > eternally out of date MSVC projects in favor of QMake, and generally > moving to a stronger C++ model where we can use Qt even in the command > line version. We can finally have reasonable support for times > before 1970 and sub-second times without jumping through hoops, > portable threads, sensible containers, better translation support and > string operations that don't crush your soul. Things are likely to be > bumpy for a while, especially for any outstanding work that spans > those big commits. If you're a C++ guy that's cringed at the > pseudo-OO model we used, or can just offer a hand cleaning warnings, > I'd welcome a hand. I don't plan to go nuts replacing proven code > (that it's not clear anyone ever really uses anyway) just for fun, but > I will be focusing on some of the areas where we've long hurt like > timestamps, string manipulations, and better file access. (Something > as "simple" as traversing a directory across OSes in C was just a chore.) > > I'll leave the CVS trees open and readable for a few weeks in case > there are any outstanding patches against them, but I plan to cvs rm * > and do a commit that points people to the current repo. > > The source and history on the web is accessible at > > http://code.google.com/p/gpsbabel/source/browse/#svn%2Ftrunk%2Fgpsbabel > > > To check out the trunk into a local directory named 'gpsbabel': > > svn checkout http://gpsbabel.googlecode.com/svn/trunk/gpsbabel gpsbabel > > svn log and svn update all work like their SVN counterparts. Diffs > accept a range like 'svn diff -r999:100' That's 99% of what we do in > this project. > > > ------------------------------------------------------------------------------ > Magic Quadrant for Content-Aware Data Loss Prevention > Research study explores the data loss prevention market. Includes in-depth > analysis on the changes within the DLP market, and the criteria used to > evaluate the strengths and weaknesses of these DLP solutions. > http://www.accelacomm.com/jaw/sfnl/114/51385063/ > > > _______________________________________________ > Gpsbabel-code mailing list http://www.gpsbabel.org > Gps...@li... > https://lists.sourceforge.net/lists/listinfo/gpsbabel-code |