I just did an HTDump on the database that is not 'cooperating'. The dump of the documents file seems fine. However when I did an HTDump on the word database I rec'd absolutely nothing, guess that would explain why my searches aren't finging anything! The results of the htdump -d only gives the header line...nothing else.
My db.words.db file is 1.5G though, so I'm wondering why nothing was dumped? Any ideas on this one? I didn't compile with large file support, but since the largest database is 1.5G I shouldn't have to worry about this? I did run htdig on a new database I set up that had only 6000 files, and everything seems fine, I also did an htdump on this database, and the word and document dump files seemed to be fine. Right now I'm working off of two years worth of data. This year we will be adding another, so the databases will grow by 1/3 this year. I'd like to keep the database searches quick, and keep the databases on the smaller side...as the words database will definitely be larger than 2G by the end of this year. I'm considering splitting the database up into two smaller databases for different filings, rather than keeping one large database. Any pointers/suggestions for working with large databases such as this one would truly be appreciated! Thanks! Abbie -----Original Message----- From: Jim Cole [mailto:[EMAIL PROTECTED] Sent: Tuesday, March 04, 2003 8:58 PM To: Abbie Greene Subject: Re: [htdig] RE: No matches 3.2.0b4 On Tuesday, March 4, 2003, at 05:42 AM, Abbie Greene wrote: > Nope, I'm receiving no error message when I do a search. I double > checked to be sure that it was pointing to the write database > directory, and correct conf file...all seem fine. It's strange when I > reran rundig as root on a very small subsection of files (about 100), > searches came out You might want to try the htdump utility to dump the databases to text files so that you can see what is there. If the words/URLs you are interested in show up in the dump, then in theory htsearch should be able to find them. If the command fails, or you end up with a bunch of garbage in the files, then there is a good chance that the databases were somehow corrupted. > fine. The database size is 2.3 GB total. I did not compile for large > file support, should this have been done and if so, how? Also I > haven't found anything in regards to running htsearch from the command > line... Compiling for large file support requires that you specify the need for such support when you initially run configure. I believe the option is --enable-bigfile; if you do ./configure --help, you should find the option described in the output. This support is required for cases where single database files exceed the 2 GB limit. Your operating system (or more specifically file systems and libraries) must also support large files. If your largest single database file is less than 2 GB, ht://Dig shouldn't require large file support. > Would I just run ./htsearch [search word] like that? If you just go to the directory and type ./htsearch and hit return, you will be prompted for search terms. Invoking htsearch in this way causes it to use the default configuration file. Jim ------------------------------------------------------- This SF.net email is sponsored by: Etnus, makers of TotalView, The debugger for complex code. Debugging C/C++ programs can leave you feeling lost and disoriented. TotalView can help you find your way. Available on major UNIX and Linux platforms. Try it free. www.etnus.com _______________________________________________ htdig-general mailing list <[EMAIL PROTECTED]> To unsubscribe, send a message to <[EMAIL PROTECTED]> with a subject of unsubscribe FAQ: http://htdig.sourceforge.net/FAQ.html
