geo-rep: Fix changelog encoding to encode only space and newline

libgfchangelog was encoding path using spec rfc3986, but encoding only
required for SPACE and NEWLINE chars since the NEWLINE char is used as
record separator and SPACE as field separator in the parsed changelogs
output.

Changed the encoding function to encode only SPACE and NEWLINE.

BUG: 1451724
Change-Id: I1936efad31788a9e636f912c832ed7d7efea4fe2
Signed-off-by: Aravinda VK <avishwan@redhat.com>
Reviewed-on: https://review.gluster.org/17787
Reviewed-by: Prashanth Pai <ppai@redhat.com>
Reviewed-by: Kotresh HR <khiremat@redhat.com>
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
Smoke: Gluster Build System <jenkins@build.gluster.org>

2 files changed, 33 insertions, 13 deletions

diff --git a/geo-replication/syncdaemon/master.py b/geo-replication/syncdaemon/master.py
index 9bd6721eab5..372717e36f2 100644
--- a/geo-replication/syncdaemon/master.py
+++ b/geo-replication/syncdaemon/master.py
@@ -22,8 +22,8 @@ from errno import ENOENT, ENODATA, EEXIST, EACCES, EAGAIN, ESTALE
 from threading import Condition, Lock
 from datetime import datetime
 from gconf import gconf
-from syncdutils import Thread, GsyncdError, boolify, escape
-from syncdutils import unescape, gauxpfx, md5hex, selfkill
+from syncdutils import Thread, GsyncdError, boolify, escape_space_newline
+from syncdutils import unescape_space_newline, gauxpfx, md5hex, selfkill
 from syncdutils import lstat, errno_wrap, FreeObject, lf
 from syncdutils import NoStimeAvailable, PartialHistoryAvailable
 
@@ -852,7 +852,8 @@ class GMasterChangelogMixin(GMasterCommon):
                 self.update_fop_batch_stats(ec[self.POS_TYPE])
 
                 # PARGFID/BNAME
-                en = unescape(os.path.join(pfx, ec[self.POS_ENTRY1]))
+                en = unescape_space_newline(
+                    os.path.join(pfx, ec[self.POS_ENTRY1]))
                 # GFID of the entry
                 gfid = ec[self.POS_GFID]
 
@@ -860,7 +861,8 @@ class GMasterChangelogMixin(GMasterCommon):
                     # The index of PARGFID/BNAME for UNLINK, RMDIR
                     # is no more the last index. It varies based on
                     # changelog.capture-del-path is enabled or not.
-                    en = unescape(os.path.join(pfx, ec[self.UNLINK_ENTRY]))
+                    en = unescape_space_newline(
+                        os.path.join(pfx, ec[self.UNLINK_ENTRY]))
 
                     # Remove from DATA list, so that rsync will
                     # not fail
@@ -918,7 +920,8 @@ class GMasterChangelogMixin(GMasterCommon):
                         if isinstance(rl, int):
                             rl = None
 
-                    e1 = unescape(os.path.join(pfx, ec[self.POS_ENTRY1 - 1]))
+                    e1 = unescape_space_newline(
+                        os.path.join(pfx, ec[self.POS_ENTRY1 - 1]))
                     entries.append(edct(ty, gfid=gfid, entry=e1, entry1=en,
                                         stat=st, link=rl))
                 else:
@@ -1583,8 +1586,10 @@ class GMasterXsyncMixin(GMasterChangelogMixin):
                 self.sync_done(self.stimes, False)
                 self.stimes = []
             if stat.S_ISDIR(mo):
-                self.write_entry_change("E", [gfid, 'MKDIR', str(mo),
-                    str(0), str(0), escape(os.path.join(pargfid, bname))])
+                self.write_entry_change("E",
+                                        [gfid, 'MKDIR', str(mo),
+                                         str(0), str(0), escape_space_newline(
+                                             os.path.join(pargfid, bname))])
                 self.write_entry_change("M", [gfid, "SETATTR", str(st.st_uid),
                                               str(st.st_gid), str(st.st_mode),
                                               str(st.st_atime),
@@ -1603,8 +1608,8 @@ class GMasterXsyncMixin(GMasterChangelogMixin):
                 self.stimes.append((e, stime_to_update))
             elif stat.S_ISLNK(mo):
                 self.write_entry_change(
-                    "E", [gfid, 'SYMLINK', escape(os.path.join(pargfid,
-                                                               bname))])
+                    "E", [gfid, 'SYMLINK', escape_space_newline(
+                        os.path.join(pargfid, bname))])
             elif stat.S_ISREG(mo):
                 nlink = st.st_nlink
                 nlink -= 1  # fixup backend stat link count
@@ -1615,12 +1620,12 @@ class GMasterXsyncMixin(GMasterChangelogMixin):
                     self.write_entry_change("E",
                                             [gfid, 'MKNOD', str(mo),
                                              str(0), str(0),
-                                             escape(os.path.join(
-                                                 pargfid, bname))])
+                                             escape_space_newline(
+                                                 os.path.join(pargfid, bname))])
                 else:
                     self.write_entry_change(
-                        "E", [gfid, 'LINK', escape(os.path.join(pargfid,
-                                                                bname))])
+                        "E", [gfid, 'LINK', escape_space_newline(
+                            os.path.join(pargfid, bname))])
                 self.write_entry_change("D", [gfid])
         if path == '.':
             stime_to_update = xtl
diff --git a/geo-replication/syncdaemon/syncdutils.py b/geo-replication/syncdaemon/syncdutils.py
index b5f09459c57..43b21668a46 100644
--- a/geo-replication/syncdaemon/syncdutils.py
+++ b/geo-replication/syncdaemon/syncdutils.py
@@ -70,6 +70,9 @@ CHANGELOG_AGENT_SERVER_VERSION = 1.0
 CHANGELOG_AGENT_CLIENT_VERSION = 1.0
 NodeID = None
 rsync_version = None
+SPACE_ESCAPE_CHAR = "%20"
+NEWLINE_ESCAPE_CHAR = "%0A"
+PERCENTAGE_ESCAPE_CHAR = "%25"
 
 
 def escape(s):
@@ -83,6 +86,18 @@ def unescape(s):
     return urllib.unquote_plus(s)
 
 
+def escape_space_newline(s):
+    return s.replace("%", PERCENTAGE_ESCAPE_CHAR)\
+            .replace(" ", SPACE_ESCAPE_CHAR)\
+            .replace("\n", NEWLINE_ESCAPE_CHAR)
+
+
+def unescape_space_newline(s):
+    return s.replace(SPACE_ESCAPE_CHAR, " ")\
+            .replace(NEWLINE_ESCAPE_CHAR, "\n")\
+            .replace(PERCENTAGE_ESCAPE_CHAR, "%")
+
+
 def norm(s):
     if s:
         return s.replace('-', '_')