Saving and restoring with YAML

ruby-talk
1

Hi all,

I have a quick question about YAML. From the documentation it seems as
though it was designed for, or is well suited to saving/restoring an
object’s state, but from looking at it I can’t quite tell what the ideal
way of doing this is. Say I have a class that looks like:

class Foo
attr_accessor :bar, :arr, :buzz

def initialize
@bar = true
@arr = []
@buzz = "Here is some string"
end
end

What would the “save” and “load” methods of that class look like if I
wanted to use YAML?

Ben

2

Ben Giddings wrote:

Hi all,

I have a quick question about YAML. From the documentation it seems as
though it was designed for, or is well suited to saving/restoring an
object’s state, but from looking at it I can’t quite tell what the ideal
way of doing this is. Say I have a class that looks like:

class Foo
attr_accessor :bar, :arr, :buzz

def initialize
@bar = true
@arr =
@buzz = “Here is some string”
end
end

What would the “save” and “load” methods of that class look like if I
wanted to use YAML?

If you require ‘yaml’, your class already has #to_yaml, which is a save
method, and you can use YAML.load to get it back:

require ‘yaml’
foo = Foo.new
YAML.load(foo.to_yaml)

Maybe I’m misunderstanding your question. Do you mean you want instances
of Foo to always be serialized in YAML format, even when using Marshal?

3

Ok, silly question.

I am writing a script to determine my router’s WAN ip address and then to
email me once an hour in case it changes. Currently I am running a web
server at work that returns a page with the client’s ip address. I need to
parse out the info on the page so I can extract the ip address of my router
when my script/program connects.

I am using the html-parser, sgml-parser and formatter ruby libraries
provided from raa and I have made the changes to the regexp regarding image
width and height. So I’m good there.

In my test.rb file I say:

···

h = Net::HTTP.new(‘www.zachstestip.com’ , 80 )
resp,data = h.get(’/index.php’ , nil )

w = DumbWriter.new
f = AbstractFormatter.new(w)
p = HTMLParser.new(f)
p.feed(data)
p.close

Here comes the silly part. The function “feed” is inherited by sgml-parser
to html-parser. It passes “data” along to the sgml-parser function
"goahead". It prints everything to stdout or stderr( i dont know, but it
makes it to my screen =), but there is no print, put, etc… etc… call to
send it there!!! I cant for the life of me determine where in the feed or
goahead functions are outputting my parsed results from data! This is damn
silly of me to ask I know, but how is it getting to my CLI?

In the “goahead” function there is a giant while loop. If i place a print or
puts statement at the right before the loop and right after the loop, then
nothing is outputted( except for my explicit print/puts statements).

Am I losing it?

Zach

4

Joel VanderWerf wrote:

If you require ‘yaml’, your class already has #to_yaml, which is a save
method, and you can use YAML.load to get it back:

Ah, ok. I didn’t realize that kind of magic happened. Cool. But to_yaml
isn’t a “save” method, it’s a “to-string” type method, right? So should my
save method be something like

def save
File.open(“foo_save.yaml”, “w”) {|file| file.puts(self.to_yaml) }
end

and load like:

def Foo.load
File.open(“foo_save.yaml”, “r”) {
>file>
YAML.load(file)
}
end

If this is the way to do it, that’s great. It would be nice, however, if
a) it were more symmetrical so you could have a YAML.save(file) as well as
a YAML.load(file)
b) YAML.load (and YAML.save) could do the File opening, writing, and
closing as well.

Ben

5

Zach Dennis wrote:

Ok, silly question.

I am writing a script to determine my router’s WAN ip address and then to
email me once an hour in case it changes. Currently I am running a web
server at work that returns a page with the client’s ip address. I need to
parse out the info on the page so I can extract the ip address of my router
when my script/program connects.

I am using the html-parser, sgml-parser and formatter ruby libraries
provided from raa and I have made the changes to the regexp regarding image
width and height. So I’m good there.

In my test.rb file I say:

h = Net::HTTP.new(‘www.zachstestip.com’ , 80 )
resp,data = h.get(‘/index.php’ , nil )

w = DumbWriter.new
f = AbstractFormatter.new(w)
p = HTMLParser.new(f)
p.feed(data)
p.close

Here comes the silly part. The function “feed” is inherited by sgml-parser
to html-parser. It passes “data” along to the sgml-parser function
“goahead”. It prints everything to stdout or stderr( i dont know, but it
makes it to my screen =), but there is no print, put, etc… etc… call to
send it there!!! I cant for the life of me determine where in the feed or
goahead functions are outputting my parsed results from data! This is damn
silly of me to ask I know, but how is it getting to my CLI?

In the “goahead” function there is a giant while loop. If i place a print or
puts statement at the right before the loop and right after the loop, then
nothing is outputted( except for my explicit print/puts statements).

Am I losing it?

Why not just qualify your IP address with something like >>>>IP<<<< and
then you can regex for it like this:

 match = />>>>(.+)<<<</.match(HTML)

 match[1] => your IP address

Sean O'Dell
6

This doesn’t answer your questions about Ruby, but most of what you want
exists already.

Look at http://www.dyndns.org. I’ve been using them for a year or so.
Every 5 minutes, a Perl daemon (ddclient) on my system wakes up, grabs
the WAN address from my Linksys box, and if it’s changed, updates
dyndns. I can ssh into my system at home using the name
’tidal.dyndns.org’, even though the address actually belongs to my ISP.
It works great, and it’s free.

Steve

7

Ok, silly question.

I am writing a script to determine my router’s WAN ip address and then to
email me once an hour in case it changes. Currently I am running a web
server at work that returns a page with the client’s ip address. I need to
parse out the info on the page so I can extract the ip address of my router
when my script/program connects.

check out dyndns.org - they have scripts for just about every router that does
this.

I am using the html-parser, sgml-parser and formatter ruby libraries
provided from raa and I have made the changes to the regexp regarding image
width and height. So I’m good there.

In my test.rb file I say:

h = Net::HTTP.new(‘www.zachstestip.com’ , 80 )
resp,data = h.get(‘/index.php’ , nil )

w = DumbWriter.new
f = AbstractFormatter.new(w)
p = HTMLParser.new(f)
p.feed(data)
p.close

one thing i might point out here - i myself have spent hours trying to figure
out weird bugs after naming a variable ‘p’. worth a check…

Here comes the silly part. The function “feed” is inherited by sgml-parser
to html-parser. It passes “data” along to the sgml-parser function
“goahead”. It prints everything to stdout or stderr( i dont know, but it
makes it to my screen =), but there is no print, put, etc… etc… call to
send it there!!! I cant for the life of me determine where in the feed or
goahead functions are outputting my parsed results from data! This is damn
silly of me to ask I know, but how is it getting to my CLI?

In the “goahead” function there is a giant while loop. If i place a print or
puts statement at the right before the loop and right after the loop, then
nothing is outputted( except for my explicit print/puts statements).

you could also try something like this to track the problem:

alias __p p
alias __print print
alias __puts puts

def p(*args);STDERR.p(caller.join(“\n”)); __p(*args);end
def print(*args);STDERR.print(caller.join(“\n”)); __print(*args);end
def puts(*args);STDERR.puts(caller.join(“\n”)); __puts(*args);end

i’m note sure you’d need all three but… you get the picture.

-a

···

On Wed, 1 Oct 2003, Zach Dennis wrote:

Ara Howard
NOAA Forecast Systems Laboratory
Information and Technology Services
Data Systems Group
R/FST 325 Broadway
Boulder, CO 80305-3328
Email: ara.t.howard@noaa.gov
Phone: 303-497-7238
Fax: 303-497-7259
The difference between art and science is that science is what we understand
well enough to explain to a computer. Art is everything else.
– Donald Knuth, “Discover”
~ > /bin/sh -c ‘for lang in ruby perl; do $lang -e “print "\x3a\x2d\x29\x0a"”; done’
====================================

8

Zach Dennis wrote:

I am using the html-parser, sgml-parser and formatter ruby libraries
provided from raa and I have made the changes to the regexp regarding image
width and height. So I’m good there.

I think the HTML parser might be abandoned (RAA says the last update was
2001-07-10 13:35:40 GMT).

You might have better luck using (my) htmltokenizer. It has a really
simple interface, and it might be more what you need:

http://raa.ruby-lang.org/list.rhtml?name=htmltokenizer

If you really want to use the html-parser, sorry, I can’t help you. I
never managed to understand how to work it, which is why I ported the
htmltokenizer.

Ben

9

h = Net::HTTP.new(‘www.zachstestip.com’ , 80 )
resp,data = h.get(‘/index.php’ , nil )

w = DumbWriter.new
f = AbstractFormatter.new(w)
p = HTMLParser.new(f)
p.feed(data)
p.close

Here comes the silly part. The function “feed” is inherited by
sgml-parser to html-parser. It passes “data” along to the sgml-parser
function “goahead”. It prints everything to stdout or stderr( i dont
know, but it makes it to my screen =), but there is no print, put,
etc… etc… call to send it there!!! I cant for the life of me
determine where in the feed or goahead functions are outputting my
parsed results from data! This is damn silly of me to ask I know, but
how is it getting to my CLI?

Through the DumbWriter. Check its implementation in
…\Ruby\lib\ruby\site_ruby\formatter.rb
that’s where the “write” statements live.

Often times when you want to parse HTML, it is simpler to use
the (misleadingly named) SGMLParser. Anyway these libraries are
direct ports of python modules, and can only be understood by
checking the documentation of the originals.
See eg: http://www.python.org/doc/1.5.2/lib/module-sgmllib.html
And usage examples (in python :wink:
Strip tags and Javascript from HTML page, leaving only safe tags « Python recipes « ActiveState Code
http://www.oreilly.com/catalog/pythonsl/chapter/ch05.html#t4

Cheers,

Bernard.

10

Thank you all for your replies, but I do not want to work around the
solution I would like to work the solution. When attempting to learn a new
language I try to think of small projects that I can do to use different
parts of the language. This is just one of them. I am going to attempt to
figure out what Ara had mentioned. Sometimes I have to reinvent little
wheels so I know I am using the language right when it comes to building
module pieces of big 99^nth polygon shapes with 147 sides.

-Zach

11

I’ll check it out! When you say simple, can I extract data from forms in
html page by chance?

Thanks,

-Zach

···

-----Original Message-----
From: Ben Giddings [mailto:bg-rubytalk@infofiend.com]
Sent: Wednesday, October 01, 2003 12:43 PM
To: ruby-talk ML
Subject: Re: HTML-Parser / SGML-Parser

Zach Dennis wrote:

I am using the html-parser, sgml-parser and formatter ruby libraries
provided from raa and I have made the changes to the regexp regarding
image
width and height. So I’m good there.

I think the HTML parser might be abandoned (RAA says the last update was
2001-07-10 13:35:40 GMT).

You might have better luck using (my) htmltokenizer. It has a really
simple interface, and it might be more what you need:

http://raa.ruby-lang.org/list.rhtml?name=htmltokenizer

If you really want to use the html-parser, sorry, I can’t help you. I
never managed to understand how to work it, which is why I ported the
htmltokenizer.

Ben

12

Zach Dennis wrote:

I’ll check it out! When you say simple, can I extract data from forms in
html page by chance?

You should be able to.

If you say:

while token = tokenizer.getTag(‘input’)
next unless ‘ip_addr’ == token.attr_hash[‘name’]

puts token.attr_hash[‘value’]
end

I think that will do what you want. I’m not sure if the syntax is perfect
since I’m doing this from memory, but it should be close enough to get you
started.

Ben