Archives de catégorie : code

To resolve conflits on binary files, you can checkout the remote version and your version with the `–theirs’ and `–ours’ checkout flags. This allows you to inspect both version, then write back the correct resulting file. $ git checkout –theirs … Continuer la lecture →

Here is a quick walkthrough on setting up a single RedMine code to serve multiple vhosts/clients with their own database. By default all uploaded files would be stored in the same `files’ subdir, so to have proper file upload isolation across the … Continuer la lecture →

Recently, I had to upgrade a Debian 4 (Etch) to Debian 5 (Lenny) at a client facility connected to Internet with a « slow » 2Mbps SDSL link. The problem is that downloading the packages while doing the upgrade on-site, during working … Continuer la lecture →

http://cd34.com/blog/web-security/hackers-bypass-htaccess-security-by-using-gets-rather-than-get/ After reading this article I immediately checked my application and found that it was subject to this issue. I started worrying and testing out how to mitigate this issue and protect my application. However, after calming down, I started looking for … Continuer la lecture →

util-linux(-ng) flock On Linux systems, there is a `flock’ command (from `util-linux’ or `util-linux-ng’) that can be used to run a command inside a file lock/unlock block. function run_lock_flock { LOCK=$1 shift touch « $LOCK » flock « $LOCK » « $@ » } portable way … Continuer la lecture →