def parse_error(path):
html = open(path).read(10000)
soup = BeautifulSoup(html)
h1 = web.htmlunquote(soup.body.h1.string or "")
h2 = web.htmlunquote(soup.body.h2.string or "")
message = h1.split("at")[0].strip() + ": " + (h2 and h2.splitlines()[0])
code, url = [
web.htmlunquote(td.string) for td in soup.body.table.findAll("td")
]
# strip common prefixes
code = web.re_compile(
".*/(?:staging|production)/(openlibrary|infogami|web)").sub(
r"\1", code)
m = web.re_compile(r"(\d\d)(\d\d)(\d\d)(\d{6})").match(
web.numify(os.path.basename(path)))
hh, mm, ss, microsec = m.groups()
return web.storage(url=url,
message=message,
code=code,
time="%s:%s:%s" % (hh, mm, ss))
def parse_module(module_url):
module = web.storage(
url=module_url,
cached_xml="",
screenshot="",
title="",
title_url="",
directory_title="",
description="",
author="",
author_email="",
author_affiliation="",
author_location="",
render_inline="",
)
if not module_url.startswith("http://"):
raise "Ooops! Submission has failed – the URL seems to be invalid."
try:
html = utils.dnl(module_url)
html = web.htmlunquote(html) # this may confuse the parser
xml = utils.parse_xml(html)
except:
raise "Ooops! Submission has failed – the XML or HTML page could not be loaded successfully."
xnodes = xml.xpath("//ModulePrefs")
if not xnodes:
raise "Ooops! The XML is valid, but we cannot find the module."
xnodes = xnodes[0]
for attr in module:
module[attr] = xnodes.get(attr) or module[attr]
return module
def get_results(q, offset=0, limit=100, snippets=3, fragsize=200, hl_phrase=False):
m = re_bad_fields.match(q)
if m:
return { 'error': m.group(1) + ' search not supported' }
q = escape_q(q)
solr_params = [
('fl', 'ia,body_length,page_count'),
('hl', 'true'),
('hl.fl', 'body'),
('hl.snippets', snippets),
('hl.mergeContiguous', 'true'),
('hl.usePhraseHighlighter', 'true' if hl_phrase else 'false'),
('hl.simple.pre', '{{{'),
('hl.simple.post', '}}}'),
('hl.fragsize', fragsize),
('q.op', 'AND'),
('q', web.urlquote(q)),
('start', offset),
('rows', limit),
('qf', 'body'),
('qt', 'standard'),
('hl.maxAnalyzedChars', '-1'),
('wt', 'json'),
]
solr_select = solr_select_url + '?' + '&'.join("%s=%s" % (k, unicode(v)) for k, v in solr_params)
stats.begin("solr", url=solr_select)
json_data = urllib.urlopen(solr_select).read()
stats.end()
try:
return simplejson.loads(json_data)
except:
m = re_query_parser_error.search(json_data)
return { 'error': web.htmlunquote(m.group(1)) }
def do_search(param, sort, page=1, rows=100):
(reply, solr_select, q_list) = run_solr_query(param, rows, page, sort)
is_bad = False
if reply.startswith('<html'):
is_bad = True
if not is_bad:
try:
root = XML(reply)
except XMLSyntaxError:
is_bad = True
if is_bad:
m = re_pre.search(reply)
return web.storage(
facet_counts = None,
docs = [],
is_advanced = bool(param.get('q', 'None')),
num_found = None,
solr_select = solr_select,
q_list = q_list,
error = (web.htmlunquote(m.group(1)) if m else reply),
)
docs = root.find('result')
return web.storage(
facet_counts = read_facets(root),
docs = docs,
is_advanced = bool(param.get('q', 'None')),
num_found = (int(docs.attrib['numFound']) if docs is not None else None),
solr_select = solr_select,
q_list = q_list,
error = None,
)
def parse_module(module_url):
module = web.storage(
url=module_url, cached_xml='', screenshot='', title='', title_url='',
directory_title='', description='', author='', author_email='',
author_affiliation='', author_location='', render_inline='')
if not module_url.startswith('http://'):
raise 'Ooops! Submission has failed – the URL seems to be invalid.'
try:
html = utils.dnl(module_url)
html = web.htmlunquote(html) # this may confuse the parser
xml = utils.parse_xml(html)
except:
raise 'Ooops! Submission has failed – the XML or HTML page could not be loaded successfully.'
xnodes = xml.xpath('//ModulePrefs')
if not xnodes:
raise 'Ooops! The XML is valid, but we cannot find the module.'
xnodes = xnodes[0]
for attr in module:
module[attr] = xnodes.get(attr) or module[attr]
return module
def parse_error(path):
html = open(path).read(10000)
soup = BeautifulSoup(html)
h1 = web.htmlunquote(soup.body.h1.string or "")
h2 = web.htmlunquote(soup.body.h2.string or "")
message = h1.split('at')[0].strip() + ': ' + (h2 and h2.splitlines()[0])
code, url = [web.htmlunquote(td.string) for td in soup.body.table.findAll('td')]
# strip common prefixes
code = web.re_compile(".*/(?:staging|production)/(openlibrary|infogami|web)").sub(r'\1', code)
m = web.re_compile('(\d\d)(\d\d)(\d\d)(\d{6})').match(web.numify(os.path.basename(path)))
hh, mm, ss, microsec = m.groups()
return web.storage(url=url, message=message, code=code, time="%s:%s:%s" % (hh, mm, ss))
def find_matches(ia, q):
host, ia_path = ia_lookup('/download/' + ia)
url = 'http://' + host + '/fulltext/inside.php?item_id=' + ia + '&doc=' + ia + '&path=' + ia_path + '&q=' + web.urlquote(q)
ret = urllib.urlopen(url)
try:
return simplejson.load(ret)
except:
m = re_h1_error.search(ret)
return { 'error': web.htmlunquote(m.group(1)) }
def get_results(q,
offset=0,
limit=100,
snippets=3,
fragsize=200,
hl_phrase=False):
m = re_bad_fields.match(q)
if m:
return {'error': m.group(1) + ' search not supported'}
q = escape_q(q)
solr_params = [
('fl', 'ia,body_length,page_count'),
('hl', 'true'),
('hl.fl', 'body'),
('hl.snippets', snippets),
('hl.mergeContiguous', 'true'),
('hl.usePhraseHighlighter', 'true' if hl_phrase else 'false'),
('hl.simple.pre', '{{{'),
('hl.simple.post', '}}}'),
('hl.fragsize', fragsize),
('q.op', 'AND'),
('q', web.urlquote(q)),
('start', offset),
('rows', limit),
('qf', 'body'),
('qt', 'standard'),
('hl.maxAnalyzedChars', '-1'),
('wt', 'json'),
]
solr_select = solr_select_url + '?' + '&'.join(
"%s=%s" % (k, unicode(v)) for k, v in solr_params)
stats.begin("solr", url=solr_select)
json_data = urllib.urlopen(solr_select).read()
stats.end()
try:
results = simplejson.loads(json_data)
except:
m = re_query_parser_error.search(json_data)
return {'error': web.htmlunquote(m.group(1))}
ekey_doc = {}
for doc in results['response']['docs']:
ia = doc['ia']
q = {'type': '/type/edition', 'ocaid': ia}
ekeys = web.ctx.site.things(q)
if not ekeys:
del q['ocaid']
q['source_records'] = 'ia:' + ia
ekeys = web.ctx.site.things(q)
if ekeys:
ekey_doc[ekeys[0]] = doc
editions = web.ctx.site.get_many(ekey_doc.keys())
for e in editions:
ekey_doc[e['key']]['edition'] = e
return results
def parse_search_response(json_data):
try:
return json.loads(json_data)
except json.JSONDecodeError:
m = re_pre.search(json_data)
error = web.htmlunquote(m.group(1))
solr_error = 'org.apache.lucene.queryParser.ParseException: '
if error.startswith(solr_error):
error = error[len(solr_error):]
return {'error': error}
def parse_search_response(json_data):
try:
return json.loads(json_data)
except json.JSONDecodeError:
m = re_pre.search(json_data)
error = web.htmlunquote(m.group(1))
solr_error = 'org.apache.lucene.queryParser.ParseException: '
if error.startswith(solr_error):
error = error[len(solr_error):]
return {'error': error}
def get_results(q, offset=0, limit=100, snippets=3, fragsize=200):
q = escape_bracket(q)
solr_select = solr_select_url + "?fl=ia,body_length,page_count&hl=true&hl.fl=body&hl.snippets=%d&hl.mergeContiguous=true&hl.usePhraseHighlighter=false&hl.simple.pre={{{&hl.simple.post=}}}&hl.fragsize=%d&q.op=AND&q=%s&start=%d&rows=%d&qf=body&qt=standard&hl.maxAnalyzedChars=1000000&wt=json" % (snippets, fragsize, web.urlquote(q), offset, limit)
stats.begin("solr", url=solr_select)
json_data = urllib.urlopen(solr_select).read()
stats.end()
try:
return simplejson.loads(json_data)
except:
m = re_query_parser_error.search(json_data)
return { 'error': web.htmlunquote(m.group(1)) }
def find_matches(ia, q):
q = escape_q(q)
host, ia_path = ia_lookup('/download/' + ia)
url = 'http://' + host + '/fulltext/inside.php?item_id=' + ia + '&doc=' + ia + '&path=' + ia_path + '&q=' + web.urlquote(
q)
ret = urllib.urlopen(url)
try:
return simplejson.load(ret)
except:
m = re_h1_error.search(ret)
return {'error': web.htmlunquote(m.group(1))}
def find_matches(ia, q):
q = escape_q(q)
host, ia_path = ia_lookup('/download/' + ia)
doc = find_doc(ia, host, ia_path) or ia
url = 'http://' + host + '/fulltext/inside.php?item_id=' + ia + '&doc=' + doc + '&path=' + ia_path + '&q=' + web.urlquote(q)
ret = urllib.urlopen(url).read().replace('"matches": [],\n}', '"matches": []\n}')
try:
return simplejson.loads(ret)
except:
m = re_h1_error.search(ret)
return { 'error': web.htmlunquote(m.group(1)) }
def get_results(q, offset=0, limit=100, snippets=3, fragsize=200, hl_phrase=False):
m = re_bad_fields.match(q)
if m:
return { 'error': m.group(1) + ' search not supported' }
q = escape_q(q)
solr_params = [
('fl', 'ia,body_length,page_count'),
('hl', 'true'),
('hl.fl', 'body'),
('hl.snippets', snippets),
('hl.mergeContiguous', 'true'),
('hl.usePhraseHighlighter', 'true' if hl_phrase else 'false'),
('hl.simple.pre', '{{{'),
('hl.simple.post', '}}}'),
('hl.fragsize', fragsize),
('q.op', 'AND'),
('q', web.urlquote(q)),
('start', offset),
('rows', limit),
('qf', 'body'),
('qt', 'standard'),
('hl.maxAnalyzedChars', '-1'),
('wt', 'json'),
]
solr_select = solr_select_url + '?' + '&'.join("%s=%s" % (k, unicode(v)) for k, v in solr_params)
stats.begin("solr", url=solr_select)
json_data = urllib.urlopen(solr_select).read()
stats.end()
try:
results = simplejson.loads(json_data)
except:
m = re_query_parser_error.search(json_data)
return { 'error': web.htmlunquote(m.group(1)) }
ekey_doc = {}
for doc in results['response']['docs']:
ia = doc['ia']
q = {'type': '/type/edition', 'ocaid': ia}
ekeys = web.ctx.site.things(q)
if not ekeys:
del q['ocaid']
q['source_records'] = 'ia:' + ia
ekeys = web.ctx.site.things(q)
if ekeys:
ekey_doc[ekeys[0]] = doc
editions = web.ctx.site.get_many(ekey_doc.keys())
for e in editions:
ekey_doc[e['key']]['edition'] = e
return results
def parse_search_response(json_data):
"""Construct response for any input"""
if json_data is None:
return {'error': 'Error parsing empty search engine response'}
try:
return json.loads(json_data)
except json.JSONDecodeError:
logger.exception("Error parsing search engine response")
m = re_pre.search(json_data)
if m is None:
return {'error': 'Error parsing search engine response'}
error = web.htmlunquote(m.group(1))
solr_error = 'org.apache.lucene.queryParser.ParseException: '
if error.startswith(solr_error):
error = error[len(solr_error):]
return {'error': error}
def parse_search_response(json_data):
"""Construct response for any input"""
if json_data is None:
return {'error': 'Error parsing empty search engine response'}
try:
return json.loads(json_data)
except json.JSONDecodeError:
logger.exception("Error parsing search engine response")
m = re_pre.search(json_data)
if m is None:
return {'error': 'Error parsing search engine response'}
error = web.htmlunquote(m.group(1))
solr_error = 'org.apache.lucene.queryParser.ParseException: '
if error.startswith(solr_error):
error = error[len(solr_error):]
return {'error': error}
def do_search(param, sort, page=1, rows=100, spellcheck_count=None):
if sort:
sort = process_sort(sort)
(solr_result, solr_select, q_list) = run_solr_query(
param, rows, page, sort, spellcheck_count
)
is_bad = False
if not solr_result or solr_result.startswith(b'<html'):
is_bad = True
if not is_bad:
try:
root = XML(solr_result)
except XMLSyntaxError:
is_bad = True
if is_bad:
m = re_pre.search(solr_result)
return web.storage(
facet_counts=None,
docs=[],
is_advanced=bool(param.get('q')),
num_found=None,
solr_select=solr_select,
q_list=q_list,
error=(web.htmlunquote(m.group(1)) if m else solr_result),
)
spellcheck = root.find("lst[@name='spellcheck']")
spell_map = {}
if spellcheck is not None and len(spellcheck):
for e in spellcheck.find("lst[@name='suggestions']"):
assert e.tag == 'lst'
a = e.attrib['name']
if a in spell_map or a in ('sqrt', 'edition_count'):
continue
spell_map[a] = [i.text for i in e.find("arr[@name='suggestion']")]
docs = root.find('result')
return web.storage(
facet_counts=read_facets(root),
docs=docs,
is_advanced=bool(param.get('q')),
num_found=(int(docs.attrib['numFound']) if docs is not None else None),
solr_select=solr_select,
q_list=q_list,
error=None,
spellcheck=spell_map,
)
def do_search(param, sort, page=1, rows=100, spellcheck_count=None):
(reply, solr_select, q_list) = run_solr_query(
param, rows, page, sort, spellcheck_count)
is_bad = False
if not reply or reply.startswith('<html'):
is_bad = True
if not is_bad:
try:
root = XML(reply)
except XMLSyntaxError:
is_bad = True
if is_bad:
m = re_pre.search(reply)
return web.storage(
facet_counts = None,
docs = [],
is_advanced = bool(param.get('q')),
num_found = None,
solr_select = solr_select,
q_list = q_list,
error = (web.htmlunquote(m.group(1)) if m else reply),
)
spellcheck = root.find("lst[@name='spellcheck']")
spell_map = {}
if spellcheck is not None and len(spellcheck):
for e in spellcheck.find("lst[@name='suggestions']"):
assert e.tag == 'lst'
a = e.attrib['name']
if a in spell_map or a in ('sqrt', 'edition_count'):
continue
spell_map[a] = [i.text for i in e.find("arr[@name='suggestion']")]
docs = root.find('result')
return web.storage(
facet_counts = read_facets(root),
docs = docs,
is_advanced = bool(param.get('q')),
num_found = (int(docs.attrib['numFound']) if docs is not None else None),
solr_select = solr_select,
q_list = q_list,
error = None,
spellcheck = spell_map,
)
def get_results(q,
offset=0,
limit=100,
snippets=3,
fragsize=200,
hl_phrase=False):
m = re_bad_fields.match(q)
if m:
return {'error': m.group(1) + ' search not supported'}
q = escape_q(q)
solr_params = [
('fl', 'ia,body_length,page_count'),
('hl', 'true'),
('hl.fl', 'body'),
('hl.snippets', snippets),
('hl.mergeContiguous', 'true'),
('hl.usePhraseHighlighter', 'true' if hl_phrase else 'false'),
('hl.simple.pre', '{{{'),
('hl.simple.post', '}}}'),
('hl.fragsize', fragsize),
('q.op', 'AND'),
('q', web.urlquote(q)),
('start', offset),
('rows', limit),
('qf', 'body'),
('qt', 'standard'),
('hl.maxAnalyzedChars', '-1'),
('wt', 'json'),
]
solr_select = solr_select_url + '?' + '&'.join(
"%s=%s" % (k, unicode(v)) for k, v in solr_params)
stats.begin("solr", url=solr_select)
json_data = urllib.urlopen(solr_select).read()
stats.end()
try:
return simplejson.loads(json_data)
except:
m = re_query_parser_error.search(json_data)
return {'error': web.htmlunquote(m.group(1))}
solr_select = solr_select_url + "?" + urllib.urlencode(params)
stats.begin("solr", url=solr_select)
try:
json_data = urlopen(solr_select).read()
except IOError, e:
logger.error("Unable to query search inside solr", exc_info=True)
return {"error": web.htmlquote(str(e))}
finally:
stats.end()
try:
return simplejson.loads(json_data)
except:
m = re_query_parser_error.search(json_data)
return {'error': web.htmlunquote(m.group(1))}
def editions_from_ia(ia):
q = {
'type': '/type/edition',
'ocaid': ia,
'title': None,
'covers': None,
'works': None,
'authors': None
}
editions = web.ctx.site.things(q)
if not editions:
del q['ocaid']
q['source_records'] = 'ia:' + ia
solr_select = solr_select_url + "?" + urllib.urlencode(params)
stats.begin("solr", url=solr_select)
try:
json_data = urlopen(solr_select).read()
except IOError, e:
logger.error("Unable to query search inside solr", exc_info=True)
return {"error": web.htmlquote(str(e))}
finally:
stats.end()
try:
return simplejson.loads(json_data)
except:
m = re_query_parser_error.search(json_data)
return { 'error': web.htmlunquote(m.group(1)) }
def editions_from_ia(ia):
q = {'type': '/type/edition', 'ocaid': ia, 'title': None, 'covers': None, 'works': None, 'authors': None}
editions = web.ctx.site.things(q)
if not editions:
del q['ocaid']
q['source_records'] = 'ia:' + ia
editions = web.ctx.site.things(q)
return editions
def read_from_archive(ia):
meta_xml = 'http://archive.org/download/' + ia + '/' + ia + '_meta.xml'
stats.begin("archive.org", url=meta_xml)
xml_data = urllib.urlopen(meta_xml)
item = {}
