#!/usr/bin/python2.2 # Make sure tabs expand to 8 spaces in vim # vim: expandtab # Copyright 2002 Raphaël Hertzog # This file is distributed under the terms of the General Public License # version 2 or (at your option) any later version. import os.path, rfc822, email, email.Utils, sys, string, re, cPickle from xml.dom import implementation, ext from config import dir, odir # address_from_string takes an address in RFC822 format # and turns it into a tuple of the form (real name, email). # The difference with email.Utils.parseaddr and rfc822.parseaddr # is that this routine allows unquoted comma's to appear in the real name # (in violation of RFC822). This is a hack to allow a Maintainer field to # be like 'Maintainer: John H. Robinson, IV '. [PvR] def address_from_string(content): hacked_content = string.replace(content, ",", "WEWANTNOCOMMAS") (name, mail) = email.Utils.parseaddr(hacked_content) return (string.replace(name,"WEWANTNOCOMMAS",","),string.replace(mail,"WEWANTNOCOMMAS",",")) # addresses_from_string takes a string with addresses in RFC822 format # and changes it into a list of tuples of the form (real name, email). # Just as address_from_string, it tries to be forgiving about unquoted # commas in addresses. [PvR] def addresses_from_string(content): pattern = re.compile("([^>]),") hacked_content = pattern.sub("\\1WEWANTNOCOMMAS", content) msg = email.message_from_string("Header: " + hacked_content) hacked_list = email.Utils.getaddresses(msg.get_all("Header", [])) list = map(lambda p: map(lambda s:string.replace(s,"WEWANTNOCOMMAS",","), p), hacked_list) return list def add_maintainer_info(child, name, mail, doc): text = doc.createTextNode(unicode(name,'UTF8','replace')) # Take care of non-ascii item_elt = doc.createElement("name") item_elt.appendChild(text) child.appendChild(item_elt) text = doc.createTextNode(mail) item_elt = doc.createElement("email") item_elt.appendChild(text) child.appendChild(item_elt) def update_sources_info(m, dist): """Update the XML information with the given Message (Package entry)""" global odir, old_done, new_done, new_dist_map package = m["Package"] hash = package[0] if package[0:3] == "lib": hash = package[0:4] # Check if the work has already been done key = "%s_%s_%s" % (m["Package"], m["Version"], dist) new_done[key] = 1 new_dist_map["%s_%s" % (m["Package"], dist)] = 1 if old_done.has_key(key) and os.path.isfile("%s/%s/%s/%s.xml" % (odir, hash, package, dist)): return # Make sure the directory exists if not os.path.isdir(odir + "/" + hash + "/" + package): if not os.path.isdir(odir + "/" + hash): os.mkdir(odir + "/" + hash) os.mkdir(odir + "/" + hash + "/" + package) # Create the XML DOM object doc = implementation.createDocument(None, None, None) root = doc.createElement("source") doc.appendChild(root) root.setAttribute("release", dist) if re.search("-\d+\.\d+(\.\d+)?$", m["version"]): root.setAttribute("nmu", "yes") for tag in m.keys(): child = doc.createElement(tag) root.appendChild(child) if tag == "binary" or tag[0:5] == "build": for item in re.split(", ?", m[tag]): # Take care of non-ascii, prevents troubles... text = doc.createTextNode(unicode(item,'ISO-8859-1','replace')) item_elt = doc.createElement("item") item_elt.appendChild(text) child.appendChild(item_elt) elif tag == "maintainer": (name, mail) = address_from_string(m[tag]) add_maintainer_info(child, name, mail, doc) elif tag == "uploaders": uploaders = addresses_from_string(m[tag]) for item in uploaders: item_elt = doc.createElement("item") (name,mail) = item add_maintainer_info(item_elt, name, mail, doc) child.appendChild(item_elt) elif tag == "files": for line in string.split(m[tag], "\n"): item_elt = doc.createElement("item") child.appendChild(item_elt) line = line.strip() fields = ["md5sum", "size", "filename"] values = string.split(line) for i in range(len(fields)): new_elt = doc.createElement(fields[i]) text = doc.createTextNode(values[i]) new_elt.appendChild(text) item_elt.appendChild(new_elt) else: text = doc.createTextNode(m[tag]) child.appendChild(text) # Print the DOM object to a file try: f = open("%s/%s/%s/%s.xml" % (odir, hash, package, dist), "w") ext.PrettyPrint(doc, f, "UTF-8") f.close() except: sys.stderr.write("Output problem for " + m["package"] + "\n"); def treat_sources_file(f, dist): """Scan the given Sources file and treat each Package entry""" while 1: try: m = rfc822.Message(f) if len(m) == 0: #eof break update_sources_info(m, dist) except EOFError: break # Load the list of sources generated the last time old_done = {} new_done = {} new_dist_map = {} if os.path.exists(odir + "/sources_done"): f = open(odir + "/sources_done", "r") old_done = cPickle.load(f) f.close() for comp in ["main", "contrib", "non-free"]: for dist in ["oldstable", "stable", "testing", "unstable"]: f = open(dir + "/Sources_%s_%s" % (dist, comp), "r") treat_sources_file(f, dist) f.close() # Experimental f = open(dir + "/Sources-experimental_" + comp, "r") treat_sources_file(f, "experimental") f.close() # s-p-u and t-p-u f = open(dir + "/Sources-spu_%s" % comp, "r") treat_sources_file(f, "s-p-u") f.close() f = open(dir + "/Sources-tpu_%s" % comp, "r") treat_sources_file(f, "t-p-u") f.close() # security f = open(dir + "/Sources-security-oldstable_%s" % comp, "r") treat_sources_file(f, "stable-security") f.close() f = open(dir + "/Sources-security-stable_%s" % comp, "r") treat_sources_file(f, "stable-security") f.close() #f = open(dir + "/Sources-security-testing_%s" % comp, "r") #treat_sources_file(f, "testing-security") #f.close() # Store the list of sources generated f = open(odir + "/sources_done", "w") cPickle.dump(new_done, f, 0) f.close() # Scan the old package/distribution that existed and check if they # still exist ... if they don't, remove the associated xml file. for key in old_done.keys(): (p, v, d) = key.split("_", 2) if not new_dist_map.has_key("%s_%s" % (p, d)): hash = p[0] if p[0:3] == "lib": hash = p[0:4] filename = "%s/%s/%s/%s.xml" % (odir, hash, p, d) filenamerebuild = "%s/%s/%s/force-rebuild" % (odir, hash, p) if os.path.exists(filename): os.unlink(filename) f = open(filenamerebuild, "w") f.close() # We're done