#!/usr/bin/env python # List Cryptech git repository URLs by scraping them from the # automatically generated Trac page. Yes, we know too much about how # the generation script and Trac format this page, c'est la vie. from urllib import urlopen from xml.etree.ElementTree import ElementTree from tempfile import NamedTemporaryFile from os import environ, fchmod url = "https://wiki.cryptech.is/wiki/GitRepositories" cacert = '''\ -----BEGIN CERTIFICATE----- MIIERzCCAy+gAwIBAgIBADANBgkqhkiG9w0BAQUFADB1MQswCQYDVQQGEwJVUzEL MAkGA1UECBMCTUExEDAOBgNVBAcTB1JlYWRpbmcxITAfBgNVBAoTGEdydW5jaHdl YXRoZXIgQXNzb2NpYXRlczEkMCIGCSqGSIb3DQEJARYVcG9zdG1hc3RlckBoYWN0 cm4ubmV0MCAXDTEwMDkyNTA1MzQ0MloYDzIxMDkwOTI1MDUzNDQyWjB1MQswCQYD VQQGEwJVUzELMAkGA1UECBMCTUExEDAOBgNVBAcTB1JlYWRpbmcxITAfBgNVBAoT GEdydW5jaHdlYXRoZXIgQXNzb2NpYXRlczEkMCIGCSqGSIb3DQEJARYVcG9zdG1h c3RlckBoYWN0cm4ubmV0MIIBIjANBgkqhkiG9w0BAQEFAAOCAQ8AMIIBCgKCAQEA wf723VEaWozuyJCytEgFkEMXDd17RTChb4RCpqFrlswbKEk4/Vw7IvBeLYNfl5F8 1hx6ca7CU4qD+4tVkanJ7dJhT9a1jWPKjx/Xz8rpZBSDzgiDZNlgX4giDHjG3Ju8 QcT9Elpx4SEBxeaVEl2AXXsShv1PrSpCt9WUdS+0NQkA7z3cXASNPOBFoITM37pC 21nuQuocpjFTaGXw5s3/FhVgcKSUHmQN5UdYK5N2w7pcn7UV2v92UFYSG9cEjQL5 DfPqzJeLemzmUl+XBqSfY/gPCyJdAPOudflnFyptVv3BrmMQYEyJBeRnt2o5aYWd q1w2wB2HPsEglm1K19gEoQIDAQABo4HfMIHcMAwGA1UdEwQFMAMBAf8wCwYDVR0P BAQDAgEGMB0GA1UdDgQWBBQkrUZV1MeYpBcTnpsuT60UPRjCZjCBnwYDVR0jBIGX MIGUgBQkrUZV1MeYpBcTnpsuT60UPRjCZqF5pHcwdTELMAkGA1UEBhMCVVMxCzAJ BgNVBAgTAk1BMRAwDgYDVQQHEwdSZWFkaW5nMSEwHwYDVQQKExhHcnVuY2h3ZWF0 aGVyIEFzc29jaWF0ZXMxJDAiBgkqhkiG9w0BCQEWFXBvc3RtYXN0ZXJAaGFjdHJu Lm5ldIIBADANBgkqhkiG9w0BAQUFAAOCAQEAMeBzm4x6hlwbcrR8AcjSP4S/8+pW 4WsmkuKlZMyP/E4rc2DpVBi9WRnLtQoQ+FDU47VkG1D97i+HYx1Ky7W8oU4YB3Xr ptoAenVmiQOVnbMluuQlxBa8JbOizt+BzL0NJEC6gHvCYT5SmFFsJgXBQZNPpRJb cDzuilf3b6vCFtRL+hD75VFLkljQWUWt59cNPGhtBItsq3q72LZ2ftPgk9ufC2h3 whlvaaZldSFGnPRcFgXYYUootSh0yUW2lIHn52s/tPRN6XXnrdz+cLUCCchSkPOi Q9q/UFgaXUetsqlb56PINOrLZUpAfqC4xFBVVo2oI0ubFq173QlATQpn/w== -----END CERTIFICATE----- ''' try: ssl_cert_file = NamedTemporaryFile("w") ssl_cert_file.write(cacert) ssl_cert_file.flush() try: fchmod(ssl_cert_file.fileno(), 0600) except: pass environ["SSL_CERT_FILE"] = ssl_cert_file.name # OpenSSL looks for this environ["GIT_SSL_CAINFO"] = ssl_cert_file.name # Git looks for this for x in ElementTree(file = urlopen(url)).iter("{http://www.w3.org/1999/xhtml}tt"): if x.text.startswith("https://git.cryptech.is/") and x.text.endswith(".git"): print x.text finally: ssl_cert_file.close()