Fix a bug in mutable publish that could cause an IndexError when a writer is removed...

diff --git a/src/allmydata/mutable/publish.py b/src/allmydata/mutable/publish.py
index 3b7c6e0595da443a62f6fbd841c74d28abf7d8d8..c87e89781b40782e412f5f3a4536d2ac29cce31a 100644 (file)
--- a/src/allmydata/mutable/publish.py
+++ b/src/allmydata/mutable/publish.py
@@ -253,7 +253,9 @@ class Publish:
         # updating, we ignore damaged and missing shares -- callers must
         # do a repair to repair and recreate these.
         self.goal = set(self._servermap.get_known_shares())
         # updating, we ignore damaged and missing shares -- callers must
         # do a repair to repair and recreate these.
         self.goal = set(self._servermap.get_known_shares())

-        self.writers = {}
+
+        # shnum -> set of IMutableSlotWriter
+        self.writers = DictOfSets()

 
         # SDMF files are updated differently.
         self._version = MDMF_VERSION
 
         # SDMF files are updated differently.
         self._version = MDMF_VERSION


@@ -278,7 +280,7 @@ class Publish:
                                   self.segment_size,
                                   self.datalength)
 
                                   self.segment_size,
                                   self.datalength)
 

-            self.writers.setdefault(shnum, []).append(writer)
+            self.writers.add(shnum, writer)

             writer.server = server
             known_shares = self._servermap.get_known_shares()
             assert (server, shnum) in known_shares
             writer.server = server
             known_shares = self._servermap.get_known_shares()
             assert (server, shnum) in known_shares


@@ -294,7 +296,7 @@ class Publish:
         # after we are done writing share data and have started to write
         # blocks. In the meantime, we need to know what to look for when
         # writing, so that we can detect UncoordinatedWriteErrors.
         # after we are done writing share data and have started to write
         # blocks. In the meantime, we need to know what to look for when
         # writing, so that we can detect UncoordinatedWriteErrors.

-        self._checkstring = self.writers.values()[0][0].get_checkstring()
+        self._checkstring = self._get_some_writer().get_checkstring()

 
         # Now, we start pushing shares.
         self._status.timings["setup"] = time.time() - self._started
 
         # Now, we start pushing shares.
         self._status.timings["setup"] = time.time() - self._started


@@ -452,7 +454,10 @@ class Publish:
 
         # TODO: Make this part do server selection.
         self.update_goal()
 
         # TODO: Make this part do server selection.
         self.update_goal()

-        self.writers = {}
+
+        # shnum -> set of IMutableSlotWriter
+        self.writers = DictOfSets()
+

         if self._version == MDMF_VERSION:
             writer_class = MDMFSlotWriteProxy
         else:
         if self._version == MDMF_VERSION:
             writer_class = MDMFSlotWriteProxy
         else:


@@ -476,7 +481,7 @@ class Publish:
                                    self.total_shares,
                                    self.segment_size,
                                    self.datalength)
                                    self.total_shares,
                                    self.segment_size,
                                    self.datalength)

-            self.writers.setdefault(shnum, []).append(writer)
+            self.writers.add(shnum, writer)

             writer.server = server
             known_shares = self._servermap.get_known_shares()
             if (server, shnum) in known_shares:
             writer.server = server
             known_shares = self._servermap.get_known_shares()
             if (server, shnum) in known_shares:


@@ -495,7 +500,7 @@ class Publish:
         # after we are done writing share data and have started to write
         # blocks. In the meantime, we need to know what to look for when
         # writing, so that we can detect UncoordinatedWriteErrors.
         # after we are done writing share data and have started to write
         # blocks. In the meantime, we need to know what to look for when
         # writing, so that we can detect UncoordinatedWriteErrors.

-        self._checkstring = self.writers.values()[0][0].get_checkstring()
+        self._checkstring = self._get_some_writer().get_checkstring()

 
         # Now, we start pushing shares.
         self._status.timings["setup"] = time.time() - self._started
 
         # Now, we start pushing shares.
         self._status.timings["setup"] = time.time() - self._started


@@ -521,6 +526,8 @@ class Publish:
 
         return self.done_deferred
 
 
         return self.done_deferred
 

+    def _get_some_writer(self):
+        return list(self.writers.values()[0])[0]

 
     def _update_status(self):
         self._status.set_status("Sending Shares: %d placed out of %d, "
 
     def _update_status(self):
         self._status.set_status("Sending Shares: %d placed out of %d, "


@@ -622,9 +629,8 @@ class Publish:
         # Can we still successfully publish this file?
         # TODO: Keep track of outstanding queries before aborting the
         #       process.
         # Can we still successfully publish this file?
         # TODO: Keep track of outstanding queries before aborting the
         #       process.

-        all_shnums = filter(lambda sh: len(self.writers[sh]) > 0,
-                            self.writers.iterkeys())
-        if len(all_shnums) < self.required_shares or self.surprised:
+        num_shnums = len(self.writers)
+        if num_shnums < self.required_shares or self.surprised:

             return self._failure()
 
         # Figure out what we need to do next. Each of these needs to
             return self._failure()
 
         # Figure out what we need to do next. Each of these needs to


@@ -835,7 +841,7 @@ class Publish:
         uncoordinated writes. SDMF files will have the same checkstring,
         so we need not do anything.
         """
         uncoordinated writes. SDMF files will have the same checkstring,
         so we need not do anything.
         """

-        self._checkstring = self.writers.values()[0][0].get_checkstring()
+        self._checkstring = self._get_some_writer().get_checkstring()

 
 
     def _make_and_place_signature(self):
 
 
     def _make_and_place_signature(self):


@@ -844,7 +850,7 @@ class Publish:
         """
         started = time.time()
         self._status.set_status("Signing prefix")
         """
         started = time.time()
         self._status.set_status("Signing prefix")

-        signable = self.writers.values()[0][0].get_signable()
+        signable = self._get_some_writer().get_signable()

         self.signature = self._privkey.sign(signable)
 
         for (shnum, writers) in self.writers.iteritems():
         self.signature = self._privkey.sign(signable)
 
         for (shnum, writers) in self.writers.iteritems():


@@ -881,7 +887,7 @@ class Publish:
 
 
     def _record_verinfo(self):
 
 
     def _record_verinfo(self):

-        self.versioninfo = self.writers.values()[0][0].get_verinfo()
+        self.versioninfo = self._get_some_writer().get_verinfo()

 
 
     def _connection_problem(self, f, writer):
 
 
     def _connection_problem(self, f, writer):


@@ -891,7 +897,7 @@ class Publish:
         """
         self.log("found problem: %s" % str(f))
         self._last_failure = f
         """
         self.log("found problem: %s" % str(f))
         self._last_failure = f

-        self.writers[writer.shnum].remove(writer)
+        self.writers.discard(writer.shnum, writer)

 
 
     def log_goal(self, goal, message=""):
 
 
     def log_goal(self, goal, message=""):