<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN"
"http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
<head><meta http-equiv="content-type" content="text/html; charset=utf-8" /><style type="text/css"><!--
#msg dl { border: 1px #006 solid; background: #369; padding: 6px; color: #fff; }
#msg dt { float: left; width: 6em; font-weight: bold; }
#msg dt:after { content:':';}
#msg dl, #msg dt, #msg ul, #msg li, #header, #footer { font-family: verdana,arial,helvetica,sans-serif; font-size: 10pt; }
#msg dl a { font-weight: bold}
#msg dl a:link { color:#fc3; }
#msg dl a:active { color:#ff0; }
#msg dl a:visited { color:#cc6; }
h3 { font-family: verdana,arial,helvetica,sans-serif; font-size: 10pt; font-weight: bold; }
#msg pre { overflow: auto; background: #ffc; border: 1px #fc0 solid; padding: 6px; }
#msg ul, pre { overflow: auto; }
#header, #footer { color: #fff; background: #636; border: 1px #300 solid; padding: 6px; }
#patch { width: 100%; }
#patch h4 {font-family: verdana,arial,helvetica,sans-serif;font-size:10pt;padding:8px;background:#369;color:#fff;margin:0;}
#patch .propset h4, #patch .binary h4 {margin:0;}
#patch pre {padding:0;line-height:1.2em;margin:0;}
#patch .diff {width:100%;background:#eee;padding: 0 0 10px 0;overflow:auto;}
#patch .propset .diff, #patch .binary .diff {padding:10px 0;}
#patch span {display:block;padding:0 10px;}
#patch .modfile, #patch .addfile, #patch .delfile, #patch .propset, #patch .binary, #patch .copfile {border:1px solid #ccc;margin:10px 0;}
#patch ins {background:#dfd;text-decoration:none;display:block;padding:0 10px;}
#patch del {background:#fdd;text-decoration:none;display:block;padding:0 10px;}
#patch .lines, .info {color:#888;background:#fff;}
--></style>
<title>[599] trunk/alexandria: provider webster.it, now working well enough</title>
</head>
<body>
<div id="msg">
<dl>
<dt>Revision</dt> <dd>599</dd>
<dt>Author</dt> <dd>laurusnobilis</dd>
<dt>Date</dt> <dd>2007-02-15 16:28:19 -0500 (Thu, 15 Feb 2007)</dd>
</dl>
<h3>Log Message</h3>
<pre>provider webster.it, now working well enough</pre>
<h3>Modified Paths</h3>
<ul>
<li><a href="#trunkalexandrialibalexandriabook_providerswebster_itrb">trunk/alexandria/lib/alexandria/book_providers/webster_it.rb</a></li>
<li><a href="#trunkalexandrialibalexandriabook_providersrb">trunk/alexandria/lib/alexandria/book_providers.rb</a></li>
<li><a href="#trunkalexandriateststest_providersrb">trunk/alexandria/tests/test_providers.rb</a></li>
</ul>
</div>
<div id="patch">
<h3>Diff</h3>
<a id="trunkalexandrialibalexandriabook_providerswebster_itrb"></a>
<div class="modfile"><h4>Modified: trunk/alexandria/lib/alexandria/book_providers/webster_it.rb (598 => 599)</h4>
<pre class="diff"><span>
<span class="info">--- trunk/alexandria/lib/alexandria/book_providers/webster_it.rb 2007-02-15 18:06:57 UTC (rev 598)
+++ trunk/alexandria/lib/alexandria/book_providers/webster_it.rb 2007-02-15 21:28:19 UTC (rev 599)
</span><span class="lines">@@ -27,6 +27,7 @@
</span><span class="cx"> BASE_URI = "http://www.libreriauniversitaria.it" # also "http://www.webster.it"
</span><span class="cx"> CACHE_DIR = File.join(Alexandria::Library::DIR, '.webster_it_cache')
</span><span class="cx"> REFERER = BASE_URI
</span><ins>+ LOCALE = "BIT" # possible locales are: with isbn: "BIT", "BUS", "BUK", with ean: "BDE", "MIT"
</ins><span class="cx"> def initialize
</span><span class="cx"> super("Webster_it", "Webster Italia")
</span><span class="cx"> FileUtils.mkdir_p(CACHE_DIR) unless File.exists?(CACHE_DIR)
</span><span class="lines">@@ -38,16 +39,16 @@
</span><span class="cx"> req = BASE_URI + "/"
</span><span class="cx"> req += case type
</span><span class="cx"> when SEARCH_BY_ISBN
</span><del>- "BIT/"
</del><ins>+ "#{LOCALE}/"
</ins><span class="cx">
</span><span class="cx"> when SEARCH_BY_TITLE
</span><del>- "c_search.php?noinput=1&shelf=BIT&title_query="
</del><ins>+ "c_search.php?noinput=1&shelf=#{LOCALE}&title_query="
</ins><span class="cx">
</span><span class="cx"> when SEARCH_BY_AUTHORS
</span><del>- "c_search.php?noinput=1&shelf=BIT&author_query="
</del><ins>+ "c_search.php?noinput=1&shelf=#{LOCALE}&author_query="
</ins><span class="cx">
</span><span class="cx"> when SEARCH_BY_KEYWORD
</span><del>- "c_search.php?noinput=1&shelf=BIT&subject_query="
</del><ins>+ "c_search.php?noinput=1&shelf=#{LOCALE}&subject_query="
</ins><span class="cx">
</span><span class="cx"> else
</span><span class="cx"> raise InvalidSearchTypeError
</span><span class="lines">@@ -62,12 +63,12 @@
</span><span class="cx"> p req if $DEBUG
</span><span class="cx"> data = transport.get(URI.parse(req))
</span><span class="cx"> if type == SEARCH_BY_ISBN
</span><del>- to_book(data) rescue raise NoResultsError
</del><ins>+ to_book(data) #rescue raise NoResultsError
</ins><span class="cx"> else
</span><span class="cx"> begin
</span><span class="cx"> results = []
</span><span class="cx"> each_book_page(data) do |code, title|
</span><del>- results << to_book(transport.get(URI.parse(BASE_URI + "/BIT/" + code)))
</del><ins>+ results << to_book(transport.get(URI.parse(BASE_URI + "/#{LOCALE}/" + code)))
</ins><span class="cx"> end
</span><span class="cx"> return results
</span><span class="cx"> rescue
</span><span class="lines">@@ -78,7 +79,7 @@
</span><span class="cx">
</span><span class="cx"> def url(book)
</span><span class="cx"> return nil unless book.isbn
</span><del>- BASE_URI + "/BIT/" + Library.canonicalise_isbn(book.isbn)
</del><ins>+ BASE_URI + "/#{LOCALE}/" + Library.canonicalise_isbn(book.isbn)
</ins><span class="cx"> end
</span><span class="cx">
</span><span class="cx"> #######
</span><span class="lines">@@ -86,17 +87,19 @@
</span><span class="cx"> #######
</span><span class="cx">
</span><span class="cx"> def to_book(data)
</span><ins>+ data = data.convert("UTF-8", "iso-8859-15")
+
</ins><span class="cx"> raise unless md = /<li><span class="product_label">Titolo:<\/span><span class="product_text"> ([^<]+)/.match(data)
</span><span class="cx"> title = CGI.unescape(md[1].strip)
</span><span class="cx"> if md = /<span class="product_heading_volume">([^<]+)/.match(data)
</span><span class="cx"> title += " " + CGI.unescape(md[1].strip)
</span><span class="cx"> end
</span><ins>+
</ins><span class="cx"> authors = []
</span><del>-
- if md = /<li><span class="product_label">Autor([ei]):<\/span> <span class="product_text"><a href="([^>]+)>([^<]+)/.match(data)
</del><ins>+ if md = /<li><span class="product_label">Autor([ei]):<\/span> <span class="product_text"><a href="([^>]+)>([^<]+)/.match(data)
</ins><span class="cx"> authors = [CGI.unescape(md[3].strip)]
</span><del>-# md[1].split(', ').each { |a| authors << CGI.unescape(a.strip) }
- end
</del><ins>+# md[1].strip.split(', ').each { |a| authors << CGI.unescape(a.strip) }
+ end
</ins><span class="cx">
</span><span class="cx"> raise unless md = /<li><span class="product_label">ISBN:<\/span> <span class="product_text">([^<]+)/.match(data)
</span><span class="cx"> isbn = "978" + md[1].strip[0..8]
</span><span class="lines">@@ -106,20 +109,20 @@
</span><span class="cx"> publisher = CGI.unescape(md[2].strip)
</span><span class="cx">
</span><span class="cx"> if md = /<li><span class="product_label">Pagine:<\/span> <span class="product_text">([^<]+)/.match(data)
</span><del>- edition = "p. " + CGI.unescape(md[1].strip)
</del><ins>+ edition = CGI.unescape(md[1].strip) + "p. "
</ins><span class="cx"> else
</span><span class="cx"> edition = nil
</span><span class="cx"> end
</span><span class="cx">
</span><span class="cx"> publish_year = nil
</span><del>- if md = /<li><span class="product_label">Data di Pubblicazione:<\/span> <span class="product_text">([^"]+)/.match(data)
- publish_year = CGI.unescape(md[1].strip).to_i
</del><ins>+ if md = /<li><span class="product_label">Data di Pubblicazione:<\/span> <span class="product_text">([^<]+)/.match(data)
+ publish_year = CGI.unescape(md[1].strip)[-4 .. -1].to_i
</ins><span class="cx"> publish_year = nil if publish_year == 0
</span><span class="cx"> end
</span><span class="cx">
</span><span class="cx"> if data =~ /javascript:popImage/
</span><span class="cx">
</span><del>- cover_url = BASE_URI + "/data/images/BIT/" + isbn[9 .. 11] + "/" + isbn + "p.jpg" # use "g" instead of "p" for bigger image
</del><ins>+ cover_url = BASE_URI + "/data/images/#{LOCALE}/" + isbn[9 .. 11] + "/" + isbn + "p.jpg" # use "g" instead of "p" for bigger image
</ins><span class="cx"> cover_filename = isbn + ".tmp"
</span><span class="cx"> Dir.chdir(CACHE_DIR) do
</span><span class="cx"> File.open(cover_filename, "w") do |file|
</span><span class="lines">@@ -139,7 +142,7 @@
</span><span class="cx"> end
</span><span class="cx">
</span><span class="cx"> def each_book_page(data)
</span><del>- raise if data.scan(/<tr ><td width="10%" align="center"">&nbsp;<a href="BIT\/([^\/]+)/) { |a| yield a}.empty?
</del><ins>+ raise if data.scan(/<tr ><td width="10%" align="center"">&nbsp;<a href="#{LOCALE}\/([^\/]+)/) { |a| yield a}.empty?
</ins><span class="cx"> end
</span><span class="cx">
</span><span class="cx"> def clean_cache
</span><span class="lines">@@ -154,4 +157,3 @@
</span><span class="cx"> end
</span><span class="cx"> end
</span><span class="cx"> end
</span><del>-
</del></span></pre></div>
<a id="trunkalexandrialibalexandriabook_providersrb"></a>
<div class="modfile"><h4>Modified: trunk/alexandria/lib/alexandria/book_providers.rb (598 => 599)</h4>
<pre class="diff"><span>
<span class="info">--- trunk/alexandria/lib/alexandria/book_providers.rb 2007-02-15 18:06:57 UTC (rev 598)
+++ trunk/alexandria/lib/alexandria/book_providers.rb 2007-02-15 21:28:19 UTC (rev 599)
</span><span class="lines">@@ -230,7 +230,7 @@
</span><span class="cx"> end
</span><span class="cx">
</span><span class="cx"> require 'alexandria/book_providers/amazon'
</span><del>- #require 'alexandria/book_providers/bn'
</del><ins>+ require 'alexandria/book_providers/bn'
</ins><span class="cx"> require 'alexandria/book_providers/proxis'
</span><span class="cx"> #require 'alexandria/book_providers/mcu'
</span><span class="cx"> require 'alexandria/book_providers/amadeus'
</span><span class="lines">@@ -239,6 +239,7 @@
</span><span class="cx"> require 'alexandria/book_providers/adlibris'
</span><span class="cx"> require 'alexandria/book_providers/ls'
</span><span class="cx"> require 'alexandria/book_providers/bol_it'
</span><ins>+ require 'alexandria/book_providers/webster_it'
</ins><span class="cx">
</span><span class="cx">
</span><span class="cx"> # mechanize is optional
</span></span></pre></div>
<a id="trunkalexandriateststest_providersrb"></a>
<div class="modfile"><h4>Modified: trunk/alexandria/tests/test_providers.rb (598 => 599)</h4>
<pre class="diff"><span>
<span class="info">--- trunk/alexandria/tests/test_providers.rb 2007-02-15 18:06:57 UTC (rev 598)
+++ trunk/alexandria/tests/test_providers.rb 2007-02-15 21:28:19 UTC (rev 599)
</span><span class="lines">@@ -83,6 +83,11 @@
</span><span class="cx"> '9788817012980')
</span><span class="cx"> end
</span><span class="cx">
</span><ins>+ def test_webster
+ __test_provider(Alexandria::BookProviders::Webster_itProvider,
+ '9788817012980')
+ end
+
</ins><span class="cx"> def test_renaud
</span><span class="cx"> __test_provider(Alexandria::BookProviders::RENAUDProvider,
</span><span class="cx"> '9782894723388')
</span></span></pre>
</div>
</div>
</body>
</html>