def do_fetch(self, statefile, logger):
# Get and parse repomd.xml
repomd_url = self.url + 'repodata/repomd.xml'
logger.Log('fetching metadata from ' + repomd_url)
repomd_content = fetch(repomd_url, check_status=True).text
repomd_xml = xml.etree.ElementTree.fromstring(repomd_content)
repodata_url = self.url + repomd_xml.find(
'{http://linux.duke.edu/metadata/repo}data[@type="primary"]/{http://linux.duke.edu/metadata/repo}location'
).attrib['href']
logger.Log('fetching ' + repodata_url)
data = fetch(repodata_url).content
logger.GetIndented().Log('size is {} byte(s)'.format(len(data)))
if repodata_url.endswith('gz'):
logger.GetIndented().Log('decompressing with gzip')
data = gzip.decompress(data)
elif repodata_url.endswith('xz'):
logger.GetIndented().Log('decompressing with xz')
data = lzma.LZMADecompressor().decompress(data)
logger.GetIndented().Log(
'size after decompression is {} byte(s)'.format(len(data)))
logger.GetIndented().Log('saving')
statefile.write(data)
def do_fetch(self, statedir, logger):
for letter in ['0-9'] + [l for l in ascii_uppercase]:
page = 1
numpages = 1
while True:
logger.Log('fetching {} page {}'.format(letter, page))
pageurl = '{}/{}/page/{}/'.format(self.url, letter, page)
# fetch HTML
response = fetch(pageurl)
response.encoding = 'utf-8' # is not detected properly
text = response.text
# get number of pages, if there are more than 1 of them
if numpages == 1:
for pagebutton in lxml.html.document_fromstring(
text).xpath('.//nav[@class="page-selector"]/a'):
numpages = max(numpages, int(pagebutton.text))
# save HTML
with open(os.path.join(statedir,
'{}-{}.html'.format(letter, page)),
'w',
encoding='utf-8') as pagefile:
pagefile.write(text)
# end if that was last (or only) page
if page >= numpages:
break
# proceed with the next page
page += 1
def do_fetch(self, statefile, logger):
fetching_what = [self.url]
if isinstance(self.post, dict):
fetching_what.append('{} fields of form data'.format(len(self.post)))
elif self.post:
fetching_what.append('{} bytes of post data'.format(len(self.post)))
if self.headers:
fetching_what.append('{} extra headers'.format(len(self.headers)))
logger.Log('fetching ' + ', with '.join(fetching_what))
data = fetch(self.url, post=self.post, headers=self.headers).content
logger.GetIndented().Log('size is {} byte(s)'.format(len(data)))
if self.compression == 'gz':
logger.GetIndented().Log('decompressing with gzip')
data = gzip.decompress(data)
elif self.compression == 'bz2':
logger.GetIndented().Log('decompressing with bz2')
data = bz2.decompress(data)
elif self.compression == 'xz':
logger.GetIndented().Log('decompressing with xz')
data = lzma.LZMADecompressor().decompress(data)
if self.compression:
logger.GetIndented().Log('size after decompression is {} byte(s)'.format(len(data)))
logger.GetIndented().Log('saving')
statefile.write(data)
def _load_spec(self, package, statedir, logger):
specurl = self.giturl + '/{0}.git/plain/{0}.spec'.format(package)
logger.GetIndented().Log('getting spec from {}'.format(specurl))
r = fetch(specurl, check_status=False)
if r.status_code != 200:
deadurl = self.giturl + '/{0}.git/plain/dead.package'.format(
package)
dr = fetch(deadurl, check_status=False)
if dr.status_code == 200:
logger.GetIndented(2).Log('dead: ' +
';'.join(dr.text.split('\n')))
else:
logger.GetIndented(2).Log('failed: {}'.format(
r.status_code)) # XXX: check .dead.package, instead throw
return
with open(os.path.join(statedir, package + '.spec'), 'wb') as file:
file.write(r.content)
def do_fetch(self, statedir, logger):
packages_url = self.url + 'packages.gz'
logger.GetIndented().Log('fetching package list from ' + packages_url)
data = fetch(packages_url).text # autogunzipped?
package_names = []
for line in data.split('\n'):
line = line.strip()
if line.startswith('#') or line == '':
continue
package_names.append(line)
if not package_names:
raise RuntimeError(
'Empty package list received, refusing to continue')
logger.GetIndented().Log('{} package name(s) parsed'.format(
len(package_names)))
pagesize = 100
for page in range(0, len(package_names) // pagesize + 1):
ifrom = page * pagesize
ito = (page + 1) * pagesize
url = '&'.join([
'arg[]=' + urllib.parse.quote(name)
for name in package_names[ifrom:ito]
])
url = self.url + '/rpc/?v=5&type=info&' + url
logger.GetIndented().Log('fetching page {}/{}'.format(
page + 1,
len(package_names) // pagesize + 1))
with open(os.path.join(statedir, '{}.json'.format(page)),
'wb') as statefile:
statefile.write(fetch(url, timeout=self.fetch_timeout).content)
if self.fetch_delay:
time.sleep(self.fetch_delay)
def do_fetch(self, statedir, logger):
page = 1
while True:
pageurl = self.apiurl + 'packages/?page={}'.format(page)
logger.Log('getting page {} from {}'.format(page, pageurl))
pagedata = json.loads(fetch(pageurl).text)
for package in pagedata['packages']:
self._load_spec(package['name'], statedir, logger)
page += 1
if page > pagedata['page_total']:
break
def fetch(self, statepath, update=True, logger=NoopLogger()):
if os.path.isfile(statepath) and not update:
logger.Log('no update requested, skipping')
return
state = {}
if os.path.isfile(statepath):
with open(statepath, 'r', encoding='utf-8') as oldstatefile:
state = json.load(oldstatefile)
logger.Log('loaded old state, {} entries'.format(len(state)))
else:
logger.Log('starting with empty state')
newdata = json.loads(fetch(self.url).text)
if not newdata['releases']:
raise RuntimeError(
'Empty freshcode package list received, refusing to go on')
# add new entries in reversed order, oldest first so newest
# have higher priority; may also compare versions here
for entry in newdata['releases']:
if 'name' not in entry:
logger.Log('skipping entry with no name')
continue
if entry['name'] in state:
oldentry = state[entry['name']]
if version_compare(entry['version'], oldentry['version']) > 0:
logger.Log(
'replacing entry "{}", version changed {} -> {}'.
format(entry['name'], oldentry['version'],
entry['version']))
state[entry['name']] = entry
else:
logger.Log('adding entry "{}", version {}'.format(
entry['name'], entry['version']))
state[entry['name']] = entry
with state_file(statepath, 'w', encoding='utf-8') as statefile:
json.dump(state, statefile)
logger.Log('saved new state, {} entries'.format(len(state)))
def do_fetch(self, statedir, logger):
numpage = 1
while True:
url = self.url + '?page={}&per_page={}&sort=alpha'.format(
numpage, self.per_page)
logger.Log('getting ' + url)
text = fetch(url, timeout=self.fetch_timeout).text
with open(os.path.join(statedir, '{}.json'.format(numpage)),
'w',
encoding='utf-8') as pagefile:
pagefile.write(text)
# parse next page
if not json.loads(text)['crates']:
logger.Log('last page detected')
return
numpage += 1
if self.fetch_delay:
time.sleep(self.fetch_delay)
def do_fetch(self, statedir, logger):
numpage = 0
nextpageurl = self.url + 'Packages()?$filter=IsLatestVersion'
while True:
logger.Log('getting ' + nextpageurl)
text = fetch(nextpageurl, timeout=self.fetch_timeout).text
with open(os.path.join(statedir, '{}.xml'.format(numpage)),
'w',
encoding='utf-8') as pagefile:
pagefile.write(text)
# parse next page
logger.Log('parsing ' + nextpageurl)
root = xml.etree.ElementTree.fromstring(text)
next_link = root.find(
'{http://www.w3.org/2005/Atom}link[@rel="next"]')
if next_link is None:
break
nextpageurl = next_link.attrib['href']
numpage += 1