Get books: Fix Gutenberg store and improvements to chitanka.info and e-knigni.net

src/calibre/gui2/store/stores/chitanka_plugin.py

@@ -6,7 +6,8 @@ __license__ = 'GPL 3'
 __copyright__ = '2011, Alex Stanev <alex@stanev.org>'
 __docformat__ = 'restructuredtext en'
 
-import urllib
+import re
+import urllib2
 from contextlib import closing
 
 from lxml import html
@@ -39,13 +40,19 @@ class ChitankaStore(BasicStoreConfig, StorePlugin):
             d.exec_()
 
     def search(self, query, max_results=10, timeout=60):
+        # check for cyrilic symbols before performing search
+        uquery = unicode(query.strip(), 'utf-8')
+        reObj = re.search(u'^[а-яА-Я\\d]{4,}[а-яА-Я\\d\\s]*$', uquery)
+        if not reObj:
+            return
 
         base_url = 'http://chitanka.info'
-        url = base_url + '/search?q=' +  urllib.quote(query)
+        url = base_url + '/search?q=' +  urllib2.quote(query)
         counter = max_results
 
         # search for book title
         br = browser()
+        try:
             with closing(br.open(url, timeout=timeout)) as f:
                 f = unicode(f.read(), 'utf-8')
                 doc = html.fromstring(f)
@@ -71,10 +78,17 @@ class ChitankaStore(BasicStoreConfig, StorePlugin):
                     s.downloads['TXT'] = base_url + ''.join(data.xpath('.//a[@class="dl dl-txt"]/@href')).strip().replace('.zip', '')
                     s.formats = 'FB2, EPUB, TXT, SFB'
                     yield s
+        except urllib2.HTTPError, e:
+            if e.code == 404:
+                return
+            else:
+                raise
 
         # search for author names
-        for data in doc.xpath('//ul[@class="superlist"][1]/li'):
+        for data in doc.xpath('//ul[@class="superlist"][1]/li/dl/dt'):
             author_url = ''.join(data.xpath('.//a[contains(@href,"/person/")]/@href'))
+            if author_url == '':
+                continue
             if counter <= 0:
                 break
 
@@ -94,12 +108,17 @@ class ChitankaStore(BasicStoreConfig, StorePlugin):
                     if not id:
                         continue
 
+                    title = ''.join(data.xpath('.//a[@class="booklink"]/i/text()')).strip()
+                    author = ''.join(data.xpath('.//span[@class="bookauthor"]/a/text()')).strip()
+                    if title.lower().find(query.lower()) == -1 and author.lower().find(query.lower()) == -1:
+                        continue
+
                     counter -= 1
 
                     s = SearchResult()
                     s.cover_url = ''.join(data.xpath('.//a[@class="booklink"]/img/@src')).strip()
-                    s.title = ''.join(data.xpath('.//a[@class="booklink"]/i/text()')).strip()
+                    s.title = title
-                    s.author = ''.join(data.xpath('.//span[@class="bookauthor"]/a/text()')).strip()
+                    s.author = author
                     s.detail_item = id
                     s.drm = SearchResult.DRM_UNLOCKED
                     s.downloads['FB2'] = base_url + ''.join(data.xpath('.//a[@class="dl dl-fb2"]/@href')).strip().replace('.zip', '')

src/calibre/gui2/store/stores/eknigi_plugin.py

@@ -6,6 +6,7 @@ __license__ = 'GPL 3'
 __copyright__ = '2011, Alex Stanev <alex@stanev.org>'
 __docformat__ = 'restructuredtext en'
 
+import re
 import random
 import urllib2
 from contextlib import closing
@@ -45,8 +46,14 @@ class eKnigiStore(BasicStoreConfig, StorePlugin):
             d.exec_()
 
     def search(self, query, max_results=10, timeout=60):
+        # check for cyrilic symbols before performing search
+        uquery = unicode(query.strip(), 'utf-8')
+        reObj = re.search(u'^[а-яА-Я\\d]{2,}[а-яА-Я\\d\\s]*$', uquery)
+        if not reObj:
+            return
+
         base_url = 'http://e-knigi.net'
-        url = base_url + '/virtuemart?page=shop.browse&search_category=0&search_limiter=anywhere&limitstart=0&limit=' + str(max_results) + '&keyword=' + urllib2.quote(query)
+        url = base_url + '/virtuemart?page=shop.browse&search_category=0&search_limiter=anywhere&keyword=' + urllib2.quote(query)
 
         br = browser()
 
@@ -75,12 +82,18 @@ class eKnigiStore(BasicStoreConfig, StorePlugin):
                 if not id:
                     continue
 
+                title = ''.join(data.xpath('.//a[@class="gk_vm_product_image"]/img/@title')).strip()
+                author = ''.join(data.xpath('.//div[@style="float:left;width:90%"]/b/text()')).strip().replace('Автор: ', '')
+
+                if title.lower().find(query.lower()) == -1 and author.lower().find(query.lower()) == -1:
+                    continue
+
                 counter -= 1
 
                 s = SearchResult()
                 s.cover_url = ''.join(data.xpath('.//a[@class="gk_vm_product_image"]/img/@src')).strip()
-                s.title = ''.join(data.xpath('.//a[@class="gk_vm_product_image"]/img/@title')).strip()
+                s.title = title
-                s.author = ''.join(data.xpath('.//div[@style="float:left;width:90%"]/b/text()')).strip().replace('Автор: ', '')
+                s.author = author
                 s.price = ''.join(data.xpath('.//span[@class="productPrice"]/text()')).strip()
                 s.detail_item = base_url + id
                 s.drm = SearchResult.DRM_UNLOCKED

src/calibre/gui2/store/stores/gutenberg_plugin.py

@@ -45,7 +45,7 @@ class GutenbergStore(BasicStoreConfig, StorePlugin):
         counter = max_results
         with closing(br.open(url, timeout=timeout)) as f:
             doc = html.fromstring(f.read())
-            for data in doc.xpath('//ol[@class="results"]//li[contains(@class, "icon_title")]'):
+            for data in doc.xpath('//ol[@class="results"]//li[contains(@class, "icon_title") and not(contains(@class, "toplink"))]'):
                 if counter <= 0:
                     break
 
@@ -69,7 +69,7 @@ class GutenbergStore(BasicStoreConfig, StorePlugin):
                 yield s
 
     def get_details(self, search_result, timeout):
-        url = url_slash_cleaner('http://m.gutenberg.org/' + search_result.detail_item + '.mobile')
+        url = url_slash_cleaner('http://m.gutenberg.org/' + search_result.detail_item)
         
         br = browser()
         with closing(br.open(url, timeout=timeout)) as nf: