How to use ruby read/write Berkeley DB file?

Ruby
1

Hi, all!

I create a Berkeley DB file for Vsftpd, I want user Ruby add new Vsftpd
virtual users via edit the Berkeley DB file.

I create BDB file as follow:
bash> $ cat > test.txt << EOF

bar
Hello,
foo
World!
EOF

bash> $ db48_load -T -t hash -f ./test.txt ./test.db

bash> $ file test.db
test.db: Berkeley DB (Hash, version 9, native byte-order)

bash> $ db48_dump -p test.db
VERSION=3
format=print
type=hash
h_nelem=2
db_pagesize=4096
HEADER=END
foo
World!
bar
Hello,
DATA=END

bash> $ irb
2.0.0-p353 :001 > Dir.entries(Dir.pwd())
=> [".", “…”, “test.txt”, “test.db”]
2.0.0-p353 :002 > require ‘dbm’
=> true
2.0.0-p353 :003 > DBM.open("./test.db") do |db|
2.0.0-p353 :004 > db.each_pair {|key, value| print “#{key}:
#{value}\n”}
2.0.0-p353 :005?> end
=> #DBM:0x00000002185590
2.0.0-p353 :006 > Dir.entries(Dir.pwd())
=> [".", “…”, “test.txt”, “test.db”, “test.db.pag”, “test.db.dir”]
2.0.0-p353 :007 >

bash> $ file test.db.dir
test.db.dir: GNU dbm 2.x database

I do not understand, why DBM cannot read Berkeley DB file, but can
create a
GNU dmb file?

Thanks & Regards

2

On 01/02/2014 01:21 AM, Terry Zheng wrote:

World!
type=hash
2.0.0-p353 :001 > Dir.entries(Dir.pwd())

bash> $ file test.db.dir
test.db.dir: GNU dbm 2.x database

I do not understand, why DBM cannot read Berkeley DB file, but can
create a GNU dmb file?

Thanks & Regards

Did you try not providing the file extension? In other words:

DBM.open(“test”) do |db|
db.each_pair {|key, value| print “#{key}: #{value}\n”}
end

According to the documentation, it will figure out the extension
automatically: http://rdoc.info/stdlib/dbm/DBM:initialize

As you can see, it thought you wanted to create a new DB called
“test.db” and created the “test.db.pag” and “test.db.dir” files itself.

-Justin