Parsing geonames
in [Ruby]
|
From: Josh Cheek on 2 Jun 2010 03:05 [Note: parts of this message were removed to make it a legal post.] On Wed, Jun 2, 2010 at 1:30 AM, 12 34 <rubyforum(a)web.knobby.ws> wrote: > Michael Fellinger wrote: > > On Wed, Jun 2, 2010 at 2:57 PM, 12 34 <rubyforum(a)web.knobby.ws> wrote: > >> 41.505928,-81.594582 > > > > iota ~ % irb > >>> require 'geonames' > > => true > >>> places = Geonames::WebService.find_nearby_place_name(41.505928, > -81.594582) > > => [#<Geonames::Toponym:0x000000027dd158 @name="Little Italy", > > @alternate_names=nil, @latitude=41.5089406, @longitude=-81.5981817, > > @geoname_id="5161005", @country_code="US", @country_name="United > > States", @feature_class="P", @feature_code="PPLX", > > @feature_class_name=nil, @feature_code_name=nil, @population=nil, > > @elevation=nil, @distance=0.4495>] > >>> places.first.country_name > > => "United States" > > Thanks. I had tried places.country_name among others. Why the "first"? > > Because places is an array (think a list of geonames). You can tell it is an Array because it is surrounded by brackets. You have to tell it which element in the Array you are interested in, in this case the one at index zero (the first index). http://ruby-doc.org/core/classes/Array.html > places_nearby[0].country_code works too. > > The docs didn't suggest how to parse the result. > > Thank you for the answers, but I'm not sure why it's working. But I can > proceed. > > -- > Posted via http://www.ruby-forum.com/. > >
From: 12 34 on 2 Jun 2010 19:44 I'm confused. It's an array of one element, but that element has some kind of parts. The "parts" and how you access them are not clear to me. I can follow the model. It seems a hash would make more sense or multiple elements in the array (except one couldn't access them by name. And why an array of one element? Why bother with the array? Confused, but thanks for the help. I do what I need to do, but understanding it would help in the future. At the moment I getting timeouts errors, so can't test much, but I did earlier. -- Posted via http://www.ruby-forum.com/.
From: Michael Fellinger on 3 Jun 2010 03:47 On Thu, Jun 3, 2010 at 8:44 AM, 12 34 <rubyforum(a)web.knobby.ws> wrote: > I'm confused. It's an array of one element, but that element has some > kind of parts. The "parts" and how you access them are not clear to me. Toponym is just trying to make the access easier, you are right that the structure is very Hash-alike, and I would argue that either a Hash or a Struct would have been more sensible. The author of the geonames library doesn't use much idomatic ruby anywhere, so I will argue that he is just not very familiar with Ruby and this library might even be his first project to be out in the wild. Now to Toponym itself, for the purpose of discussion I put the source for this class online: http://pastie.org/990140 It's the kind of code that would make a seasoned Ruby developer run screaming with waving hands. Not only is he using 4 spaces for indentation ;) but also the seldom-used attr in combination with attr_writer for the exact same names, which would be much better handled with a single attr_accessor statement. module Geonames class Toponym attr_accessor :geoname_id, :name, :alternate_names, :country_code, :country_name, :population, :elevation, :feature_class, :feature_class_name, :feature_code,:feature_code_name, :latitude, :longitude, :distance end end Same functionality, less boilerplate. But to instantiate this class, Adam is using following code (which warranted another pastie http://pastie.org/990146 ) As you can see, it's pretty verbose and unlike anything you would be comfortable with after spending more time with Ruby. Actually, after going through this code, I'm really itching to rewrite it, we'll see how that goes :) > I can follow the model. It seems a hash would make more sense or > multiple elements in the array (except one couldn't access them by name. > And why an array of one element? Why bother with the array? Because it may return more than one toponym or none at all, having an Array makes the result easier to handle. I couldn't find any examples of coordinates that would return more than one toponym. Checking the source showed that it parses some XML, which may have more than one, so although I cannot confirm that they exists they are certainly allowed by the API. > Confused, but thanks for the help. I do what I need to do, but > understanding it would help in the future. > > At the moment I getting timeouts errors, so can't test much, but I did > earlier. -- Michael Fellinger CTO, The Rubyists, LLC
From: 12 34 on 3 Jun 2010 13:15 Michael Thank you so much for taking the time to explain this. I had looked a couple of my Ruby books and was finding nothing. I was ready to really dig in this morning, but obviously would not have found the answer. If you want a beta tester I'm in. I'm not a programmer of any kind and can only write basic Ruby with lots of help. If you decide to go ahead you might look at gps2photo and ExifTool. Another thing to consider is what source is being used and offering alternatives. I think ExifTool does this I think. We do need more mature gems to match what's available in perl. Thanks again. -- Posted via http://www.ruby-forum.com/.
From: botp on 3 Jun 2010 23:26
On Thu, Jun 3, 2010 at 7:44 AM, 12 34 <rubyforum(a)web.knobby.ws> wrote: > I'm confused. It's an array of one element, but that element has some > kind of parts. The "parts" and how you access them are not clear to me. read about geonames > I can follow the model. It seems a hash would make more sense or > multiple elements in the array (except one couldn't access them by name. > And why an array of one element? Why bother with the array? because you can get possibly multiple places. if you want to get a lot of places, specify a radius, possibly a big one just to test .. kind regards -botp |