I wrote this in about a day and for a couple of reasons.
Google Code is shutting down - going into an archive only mode. But I feel like it's still useful to have your own offline copy for research and/or offline archival purposes (or archive to another service).
FlossMole [1] did write some tools to scrape Google Code and other services - but the instructions were lacking on exactly what you needed to do to run it. I wanted a no-non sense tool that pretty much anyone could use.
One important point about this library is that it does scrape wiki and issues - but for downloads and repo locations it merely presents the URLs to you and you need to decide what you want to do wit that (not really hard - either curl the releases or svn co/hg clone/git clone the repo). I did this to minimize the crawling load.
Yes there is also an export to github - but again I am off the mindset that it is important to archive data rather than just copy it to another service (I think the archive team agrees with that as well [2]).
If you still aren't convinced this could be useful - I think it would be pretty awesome if someone grabbed all the code repos and created a public OpenGork [3] instance to be able to do an advanced search through all the code (it even support regex searching). I think that would be pretty amazing to offer as an online and offline tool.
I expanded it to support srchub - which I run. Feel free to use it against it (all you have to do is replace the URL to the production instance). If there is interest I would be open to expanding it to support gitlab, github, and bitbucket.
This tool isn't perfect as I made it quickly but I think it does the job. I am open to issues or problems with it. And if you think it sucks - as the subversion authors once said "patches welcome".
1 comment
[ 2.8 ms ] story [ 14.8 ms ] threadGoogle Code is shutting down - going into an archive only mode. But I feel like it's still useful to have your own offline copy for research and/or offline archival purposes (or archive to another service).
FlossMole [1] did write some tools to scrape Google Code and other services - but the instructions were lacking on exactly what you needed to do to run it. I wanted a no-non sense tool that pretty much anyone could use.
One important point about this library is that it does scrape wiki and issues - but for downloads and repo locations it merely presents the URLs to you and you need to decide what you want to do wit that (not really hard - either curl the releases or svn co/hg clone/git clone the repo). I did this to minimize the crawling load.
Yes there is also an export to github - but again I am off the mindset that it is important to archive data rather than just copy it to another service (I think the archive team agrees with that as well [2]).
If you still aren't convinced this could be useful - I think it would be pretty awesome if someone grabbed all the code repos and created a public OpenGork [3] instance to be able to do an advanced search through all the code (it even support regex searching). I think that would be pretty amazing to offer as an online and offline tool.
I expanded it to support srchub - which I run. Feel free to use it against it (all you have to do is replace the URL to the production instance). If there is interest I would be open to expanding it to support gitlab, github, and bitbucket.
This tool isn't perfect as I made it quickly but I think it does the job. I am open to issues or problems with it. And if you think it sucks - as the subversion authors once said "patches welcome".
[1] https://code.google.com/p/flossmole/source/browse/#svn%2FFLO...
[2] http://www.archiveteam.org/index.php?title=Google_Code
[3] http://opengrok.github.io/OpenGrok/