On Wed, Feb 17, 2010 at 14:14, Ivan Zhakov <i...@visualsvn.com> wrote: > On Wed, Feb 17, 2010 at 10:12 PM, Greg Stein <gst...@gmail.com> wrote: >> On Wed, Feb 17, 2010 at 14:03, Ivan Zhakov <i...@visualsvn.com> wrote: >>> On Wed, Feb 17, 2010 at 9:57 PM, Greg Stein <gst...@gmail.com> wrote: >>>> Also: PRISTINE.size is the size on-disk, which may be different from >>>> the uncompressed size. >>>> >>>> The on-disk size is handier than the uncompressed size, I believe. >>>> >>> Just one note from wc-1 experience: It's good to have translated >>> pristine size to quickly answer that file modified without comparison >>> file contents at all. >> >> That value is stored in BASE_NODE and WORKING_NODE as the column >> 'translated_size'. The size needs to account for EOL and keyword >> translations. >> > Great, I didn't aware about this.
No problem. To further illuminate here... the *same* pristine could have variant translated_size values based on each node's set of properties (svn:keywords, svn:eol-style, etc). That is why we attached the translated_size to the BASE_NODE and WORKING_NODE tables. A/foo and A/bar could use the same pristine, but have different on-disk representations due to the properties attached to each. Hunh. Now that I think about it, if you change the properties on a node (ie. new props go into ACTUAL_NODE), then we do not have a modified translated_size. Darn it. I think we're going to need to add that... Cheers, -g
