On 21/01/2009 7:03 AM, Alex Mandel wrote: > Griggs, Donald wrote: >> Hi Alex >> >> I can't answer your specific questions, but I presume you've already >> studied the file format document (most recently mentioned by Roger >> Binn's recent post): >> >> http://www.sqlite.org/fileformat.html >> > Yes I did read through that, but I didn't see what I was looking for.
Have a look at section 3.3.2 (http://www.sqlite.org/fileformat.html#record_format) and when you get to mentions of "variable length integer" and start to wonder what that is, read section 3.3.1. > Is there a list of binary codes or other delimiters likely to be > encountered between records or rows, or is the magic of length of a > given record only stored in some coded journal section or index. > > So far things are looking good using the unix tr command and trying out > various filters to remove unneeded characters. I also vacuumed a copy of > the database and did a diff against the tr results from that to narrow > down where my missing data is in the file. > > Now I'm just trying to figure out how to slice it into records, which > may dependant on me knowing the data really well, the problem I'm > running into is that there's no consistent character between two integer > fields so it's hard to split them apart. Two levels of problem: (1) Finding where each record starts (2) Unpacking the record into columns You would certainly need to write code to do the record unpacking, and probably for the record finding as well. HTH, John _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
