On 22.08.2013 14:26, Markus Schaber wrote: > Hi, > > Some days ago, I stumbled across a problem, and could analyze it a little bit > those days. > > Windows has a general limit on the path length of 260 characters on some > APIs, while others allow a length of about 32k characters (with some > restrictions). > > Most (if not all) of the SVN codebase seems to work fine with those long > pathes.
Yes, APR takes care to use the Unicode APIs and use \\?. > However, it seems that SQLite is not (yet) long path clean: Sigh... [...snip...] > I already mentioned this problem on IRC, where Bert meant that it is an issue > in SQLite itself which has its own windows abstracton layer. > > Now, some questions arise: > 1) Is this issue worth to file a bug report in the SVN database? (Up to now, > none of our customers has been bitten by that issue, but it is very likely > that one will show up sooner or later...) IMO it's a good idea to track this, but I propose you also file the report to SQLite (and put a link to that report in our report). > And for the folks more experienced with SQLite: > 2) Is the SQLite project open and responsive for bug reports? > 3) Is it possible to feed SQLite a "long" path using the \\?\ prefix? No idea, on either count. > 4) Is it possible to work around this issue by providing our own SQLite VFS > (which may delegate everything except file opening to the default one)? I haven't looked closely at SQLite's VFS API, but a quick look at the VFS object indicates that this would probably work -- as a stogap. It would be a lot better to have a fix in SQLite; it shouldn't be too hard to create a patch, never mind that the SQLite code is a mess. :) -- Brane -- Branko Čibej | Director of Subversion WANdisco // Non-Stop Data e. br...@wandisco.com
