2009-01-20 merge
by Vasil KolevToday was one of the days, in which I kept thinking “why the hell am I still sober at work”…
One of the sites we have is being worked on again, so I had to check on it. Turned out that there was a tremendous difference between it ant what we had in the version control, and the mess was really high-class. The merging took 3-4 hours – looking at what has to be cleaned up, what to be ignored, what to go under versioning, if there’s a file full replaced just because of change from unix to dos newlines and other similar crap. I also found out what heights the programmers can reach without version control – in a directory I found the following variations of one file:
index_111.php index111.php index_11.php index123.php index12.php index_1_.php index_1.php index2.php indexcopy.php index_current.php index_last.php indexnow2.php index_old1.php index_old2.php index_old.php index_OLD.php index_oold.php _index.php index____.php index___.php index__.php index_.php indexshot.php indexxx.php
Long live the ingenuity and lack of rational thought.
And to top this, not all ignores are in place yet (so svn status says that some files from some caches are weird, but at least they have never been committed). In the end the whole thing is about 700MB, 5800 files and svn status takes 11 seconds on the second run …
(then some people wonder why do I say that programmers aren’t human)