# You should have received a copy of the GNU General Public License
# along with this program. If not, see <http://www.gnu.org/licenses/>.
+import re
import os
import sys
import glob
PKG_DIR = 'pkgs'
CHUNK_SIZE = 8192
+
+def get_pkg(json_file):
+ with open(json_file, mode='r') as f:
+ pkg = json.load(f)
+ basename = os.path.basename(json_file)
+ package_name = os.path.splitext(basename)[0]
+ pkg['package_name'] = package_name
+ f.close()
+ return pkg
+
+ return None
+
+
def process_all(path, cb):
listing = glob.glob(os.path.join(path, '*.json'))
for json_file in listing:
- with open(json_file, mode='r') as f:
- pkg = json.load(f)
- basename = os.path.basename(json_file)
- package_name = os.path.splitext(basename)[0]
- pkg['package'] = package_name
- cb(pkg)
- f.close()
+ pkg = get_pkg(json_file)
+ if not pkg:
+ sys.stderr.write("Error: cannot get a pkg for: %s\n" % json_file)
+ continue
+ cb(pkg)
+
def show(pkg):
- print 'Package: ', pkg['package']
+ print 'Package: ', pkg['package_name']
print 'Program: ', pkg['name']
print 'Homepage: ', pkg['homepage']
print
-def download(pkg):
- # the "%s" in URLs are meant to be replaced with LANGCODE
- try:
- url = pkg['URL'] % LANGCODE
- except:
- url = pkg['URL']
- pass
+def download_file(url, dest_dir):
response = urllib2.urlopen(url)
- filename = urllib2.unquote(os.path.basename(response.geturl()))
+ url = response.geturl()
+
+ filename = ""
+
+ # From http://paste.pound-python.org/show/9545/
+ # TODO: use a proper module to parse HTTP headers
+ if response.info().has_key('Content-Disposition') and len(response.info()['Content-Disposition'].split('filename=')) > 1:
+ # If the response has Content-Disposition, we take file name from it
+ filename = response.info()['Content-Disposition'].split('filename=')[1].decode('utf-8')
+ if filename[0] == '"' or filename[0] == "'":
+ filename = urllib2.unquote(filename.split('"')[1])
+ else:
+ filename = urllib2.unquote(url.split('/')[-1].decode('utf_8'))
+
if filename == "":
- sys.stderr.write("Debug (%s): filename: %s url: %s\n" %(pkg['package'], filename, response.geturl()))
+ sys.stderr.write("Debug (%s): filename: %s url: %s\n" % (pkg['package_name'], filename, url))
return
- destfile = os.path.join(OUTPUT_DIR, filename)
+ # TODO: Add some integrity verification of downloaded files (md5, sha256?)
+
+ destfile = os.path.join(dest_dir, filename)
if os.path.exists(destfile):
- sys.stderr.write("Warning (%s): %s exists!\n" % (pkg['package'], destfile))
+ # TODO: check if the file is a full download from previous run,
+ # if not download again discarding the existing file?
+ sys.stderr.write("Warning: %s exists!\n" % destfile)
return
outfile = open(destfile, mode='w')
total_size = response.info().getheader('Content-Length').strip()
total_size = int(total_size)
- widgets = [pkg['name'], ' ', Percentage(), ' ', Bar(marker='=', left='[', right=']'),
+ #widgets = [pkg['name'], ' ', Percentage(), ' ', Bar(marker='=', left='[', right=']'),
+ widgets = [filename, ' ', Percentage(), ' ', Bar(marker='=', left='[', right=']'),
' ', ETA(), ' ', FileTransferSpeed()]
pbar = ProgressBar(widgets=widgets, maxval=total_size).start()
pbar.update(bytes_so_far)
pbar.finish()
+ outfile.close()
+
+
+def download(pkg):
+ # the "%s" in URLs are meant to be replaced with LANGCODE
+ try:
+ pkg_url = pkg['URL'] % LANGCODE
+ except:
+ pkg_url = pkg['URL']
+ pass
+
+ download_file(pkg_url, OUTPUT_DIR)
+
+
def show_all():
process_all(PKG_DIR, show)
+
def download_all():
if os.path.exists(OUTPUT_DIR) == False:
os.mkdir(OUTPUT_DIR, 0755)
process_all(PKG_DIR, download)
+
def usage():
usage = "winfreed - download a selection of Free Software for MS Windows.\n\n"
usage += "usage: %s <COMMAND>\n\n" % sys.argv[0]
else:
usage()
sys.exit(1)
-
+
sys.exit(0)