Changeset 32583 for main/trunk/greenstone2/perllib/plugins/GreenstoneSQLPlugin.pm

Timestamp:

2018-11-08T17:22:04+13:00 (5 years ago)

Author:

ak19

Message:

1. Some tidying up of the code. 2. Removing unnecessary calls to force_disconnect_from_db (now made 'private') after die() since on die() perl calls object destructor which will ensure disconnection anyway.

File:

• main/trunk/greenstone2/perllib/plugins/GreenstoneSQLPlugin.pm (modified) (7 diffs)

main/trunk/greenstone2/perllib/plugins/GreenstoneSQLPlugin.pm

@@ -50 +50 @@
 
 # DONE:
+# + TODO: For on cancel, add a SIGTERM handler or so to call end()
+# or to explicitly call gs_sql->close_connection if $gs_sql def
+#
 # + TODO: Incremental delete can't work until GSSQLPlugout has implemented build_mode = incremental
 # (instead of tossing away db on every build)
@@ -126 +129 @@
 
     return q^(?i)docsql(-\d+)?\.xml$^; # regex based on this method in GreenstoneXMLPlugin
-    #return q^(?i)docsql(-.+)?\.xml$^; # no longer storing the OID embedded in docsql .xml filename
 }
 
@@ -191 +193 @@
 
 
-# TODO: For on cancel, add a SIGTERM handler or so to call end()
-# or to explicitly call gs_sql->close_connection if $gs_sql def
+###### Methods called during buildcol and import #######
 
 sub new {
@@ -217 +218 @@
 }
 
-###### Called during import.pl
+# GS SQL Plugin::init() (and deinit()) is called by import.pl and also by buildcol.pl
+# This means it connects and deconnects during import.pl as well. This is okay
+# as removeold, which should drop the collection tables, happens during the import phase,
+# calling GreenstoneSQLPlugin::and therefore also requires a db connection.
+# TODO: Eventually can try moving get_gssql_instance into gssql.pm? That way both GS SQL Plugin
+# and Plugout would be using one connection during import.pl phase when both plugs exist.
+
+# Call init() not begin() because there can be multiple plugin passes and begin() called for
+# each pass (one for doc level and another for section level indexing), whereas init() should
+# be called before any and all passes.
+# This way, we can connect to the SQL database once per buildcol run.
+sub init {
+    my ($self) = shift (@_);
+    ##print STDERR "@@@@@@@@@@ INIT CALLED\n";
+    
+    $self->SUPER::init(@_); # super (GreenstoneXMLPlugin) will not yet be trying to read from doc.xml (docsql .xml) files in init().
+
+    ####################
+#    print "@@@ SITE NAME: ". $self->{'site_name'} . "\n" if defined $self->{'site_name'};
+#    print "@@@ COLL NAME: ". $ENV{'GSDLCOLLECTION'} . "\n";
+
+#    print STDERR "@@@@ db_pwd: " . $self->{'db_client_pwd'} . "\n";
+#    print STDERR "@@@@ user: " . $self->{'db_client_user'} . "\n";
+#    print STDERR "@@@@ db_host: " . $self->{'db_host'} . "\n";
+#    print STDERR "@@@@ db_driver: " . $self->{'db_driver'} . "\n";
+    ####################
+
+    # create gssql object.
+    # collection name will be used for naming tables (site name will be used for naming database)
+    my $gs_sql = new gssql({
+    'collection_name' => $ENV{'GSDLCOLLECTION'},
+    'verbosity' => $self->{'verbosity'} || 0
+               });
+    
+    # if autocommit is set, there's no rollback support
+    my $autocommit = ($self->{'rollback_on_cancel'} eq "false") ? 1 : 0;
+
+    # try connecting to the mysql db, die if that fails
+    if(!$gs_sql->connect_to_db({
+    'db_driver' => $self->{'db_driver'},
+    'db_client_user' => $self->{'db_client_user'},
+    'db_client_pwd' => $self->{'db_client_pwd'},
+    'db_host' => $self->{'db_host'},
+    'autocommit' => $autocommit
+                   })
+    )
+    {
+    # This is fatal for the plugout, let's terminate here
+    # PrintError would already have displayed the warning message on connection fail    
+    die("Could not connect to db. Can't proceed.\n");
+    }
+    
+    my $db_name = $self->{'site_name'} || "greenstone2"; # one database per GS3 site, for GS2 the db is called greenstone2
+
+    # Attempt to use the db, create it if it doesn't exist (but don't create the tables yet)
+    # Bail if we can't use the database
+    if(!$gs_sql->use_db($db_name)) {
+    
+    # This is fatal for the plugout, let's terminate here after disconnecting again
+    # PrintError would already have displayed the warning message on load fail
+    # And on die() perl will call gssql destroy which will ensure a disconnect() from db
+    #$gs_sql->force_disconnect_from_db();
+    die("Could not use db $db_name. Can't proceed.\n");
+    }
+    
+    
+    # store db handle now that we're connected
+    $self->{'gs_sql'} = $gs_sql;    
+}
+
+
+# This method also runs on import.pl if gs_sql has a value. But we just want to run it on buildcol
+# Call deinit() not end() because there can be multiple plugin passes:
+# one for doc level and another for section level indexing
+# and deinit() should be called before all passes
+# This way, we can close the SQL database once per buildcol run.
+sub deinit {
+    my ($self) = shift (@_);
+    
+    ##print STDERR "@@@@@@@@@@ GreenstoneSQLPlugin::DEINIT CALLED\n";
+    
+    if($self->{'gs_sql'}) {
+
+    # Important to call finished():
+    # it will disconnect from db if this is the last gssql instance,
+    # and it will commit to db before disconnecting if rollbback_on_cancel turned on
+    $self->{'gs_sql'}->finished();
+
+    # Clear gs_sql (setting key to undef has a different meaning from deleting:
+    # undef makes key still exist but its value is unded whereas delete deletes the key)
+    # So all future use has to make the connection again
+    delete $self->{'gs_sql'};
+    }
+
+    $self->SUPER::deinit(@_);
+}
+
+
+
+###### Methods only called during import.pl #####
 
 # This is called once if removeold is set with import.pl. Most plugins will do
@@ -246 +346 @@
     }
 
-    # UNNECESSARY
-    # The removeold related DB transaction (deleting collection tables) is complete
-    # Don't let GS SQL PlugIN interfere with GS SQL PlugOUT's database transactions
-    # during import.pl hereafter. Finish up.
-    #$gs_sql->do_commit_if_on();
 }
 
@@ -294 +389 @@
 }
 
-#### Called during buildcol
+##### Methods called only during buildcol #####
 
 sub xml_start_tag {
@@ -417 +512 @@
 }
 
-#### Called during buildcol and import
-
-# GS SQL Plugin::init() (and deinit()) is called by import.pl and also by buildcol.pl
-# This means it connects and deconnects during import.pl as well. This is okay
-# as removeold, which should drop the collection tables, happens during the import phase,
-# calling GreenstoneSQLPlugin::and therefore also requires a db connection.
-# TODO: Eventually can try moving get_gssql_instance into gssql.pm? That way both GS SQL Plugin
-# and Plugout would be using one connection during import.pl phase when both plugs exist.
-
-# Call init() not begin() because there can be multiple plugin passes and begin() called for
-# each pass (one for doc level and another for section level indexing), whereas init() should
-# be called before any and all passes.
-# This way, we can connect to the SQL database once per buildcol run.
-sub init {
-    my ($self) = shift (@_);
-    ##print STDERR "@@@@@@@@@@ INIT CALLED\n";
-    
-    $self->SUPER::init(@_); # super (GreenstoneXMLPlugin) will not yet be trying to read from doc.xml (docsql .xml) files in init().
-
-    ####################
-#    print "@@@ SITE NAME: ". $self->{'site_name'} . "\n" if defined $self->{'site_name'};
-#    print "@@@ COLL NAME: ". $ENV{'GSDLCOLLECTION'} . "\n";
-
-#    print STDERR "@@@@ db_pwd: " . $self->{'db_client_pwd'} . "\n";
-#    print STDERR "@@@@ user: " . $self->{'db_client_user'} . "\n";
-#    print STDERR "@@@@ db_host: " . $self->{'db_host'} . "\n";
-#    print STDERR "@@@@ db_driver: " . $self->{'db_driver'} . "\n";
-    ####################
-
-    # create gssql object.
-    # collection name will be used for naming tables (site name will be used for naming database)
-    my $gs_sql = new gssql({
-    'collection_name' => $ENV{'GSDLCOLLECTION'},
-    'verbosity' => $self->{'verbosity'} || 0
-               });
-    
-    # if autocommit is set, there's no rollback support
-    my $autocommit = ($self->{'rollback_on_cancel'} eq "false") ? 1 : 0;
-
-    # try connecting to the mysql db, die if that fails
-    if(!$gs_sql->connect_to_db({
-    'db_driver' => $self->{'db_driver'},
-    'db_client_user' => $self->{'db_client_user'},
-    'db_client_pwd' => $self->{'db_client_pwd'},
-    'db_host' => $self->{'db_host'},
-    'autocommit' => $autocommit
-                   })
-    )
-    {
-    # This is fatal for the plugout, let's terminate here
-    # PrintError would already have displayed the warning message on connection fail    
-    die("Could not connect to db. Can't proceed.\n");
-    }
-    
-    my $db_name = $self->{'site_name'} || "greenstone2"; # one database per GS3 site, for GS2 the db is called greenstone2
-
-    # Attempt to use the db, create it if it doesn't exist (but don't create the tables yet)
-    # Bail if we can't use the database
-    if(!$gs_sql->use_db($db_name)) {
-    
-    # This is fatal for the plugout, let's terminate here after disconnecting again
-    # PrintError would already have displayed the warning message on load fail
-    $gs_sql->force_disconnect_from_db();
-    die("Could not use db $db_name. Can't proceed.\n");
-    }
-    
-    
-    # store db handle now that we're connected
-    $self->{'gs_sql'} = $gs_sql;    
-}
-
-
-# This method also runs on import.pl if gs_sql has a value. But we just want to run it on buildcol
-# Call deinit() not end() because there can be multiple plugin passes:
-# one for doc level and another for section level indexing
-# and deinit() should be called before all passes
-# This way, we can close the SQL database once per buildcol run.
-sub deinit {
-    my ($self) = shift (@_);
-    
-    ##print STDERR "@@@@@@@@@@ GreenstoneSQLPlugin::DEINIT CALLED\n";
-    
-    if($self->{'gs_sql'}) { # only want to work with sql db if buildcol.pl, gs_sql won't have
-    # a value except during buildcol, so when processor =~ m/buildproc$/.
-    $self->{'gs_sql'}->finished();
-
-    # Clear gs_sql (setting key to undef has a different meaning from deleting:
-    # undef makes key still exist but its value is unded whereas delete deletes the key)
-    # So all future use has to make the connection again
-    delete $self->{'gs_sql'};
-    }
-
-    $self->SUPER::deinit(@_);
-}
-
-
-
-
+
+1;