(kserve)@haven:~/kserve/fresh/kserver$ find lib/ controllers/ model/ -type f -name “*.py”  | grep -v template.py | grep -v auth.py | grep -v error.py | grep -v __init__ |grep -v app_globals.py| grep -v base.py |xargs cat  | wc -l
1337

l33t.. no bs.

Unfortunately, it is not done. I am going to to have to code 30,000 more lines now.. damnit!

Jrod.

that triggers this exception! Woe to them indeed!

DeclarativeBase = declarative_base()
class TableClass(DeclarativeBase):
    __tablename__ = 'table_name'
    __table_args__ =  {'mysql_engine':'InnoDB', 'mysql_charset':'utf8'}
    col_id = Column(Integer, primary_key=True)
    ...

In order to create/drop a table using an SQLAlchemy object, you need access to the table object, the table metadata, and a database engine object. When extending DeclarativeBase, the table object is exposed by the __table__ attribute of every model class. For instance:

class Distro(DeclarativeBase):
__tablename__ = ‘distro’
id = Column(Integer, primary_key=True, unique=True)
label = Column(Unicode(35), nullable=False)

In [1]: from yourapp.model import *

In [2]: import transaction

In [3]: distro_obj = Distro()

In [4]: print distro_obj.__table__.__doc__
——> print(distro_obj.__table__.__doc__)
Represent a table in a database.

e.g.::

mytable = Table(“mytable”, metadata,
Column(‘mytable_id’, Integer, primary_key=True),
Column(‘value’, String(50))
)

As you can see, you may access the methods of Table() from the DeclarativeBase __table__ attribute. In order to use the model to create/drop the table it describes, you need to bind the table metadata to an instantiated SQLAlchemy engine object. Pylons, the framework which Turbogears 2+ is based, stores the engine (defined in the paster .ini configuration) in the config global configuration dictionary, specifically:

config['pylons.app_globals'].sa_engine

Going back to the model we created, we can create the table using the Table() method create.

distro_obj.__table__.create(config['pylons.app_globals'].sa_engine)

Now that the table is created you may use the session object to interact with the table.

In [5]: distro_obj.label = 'rhel5_minimal'
In [6]: DBSession.add(distro_obj)
In [7]: transaction.commit()
In [8]: distro_label = DBSession.query(Distro.label).filter(Distro.label=='rhel5_minimal').one()[0]
In [9]: print distro_label
-------> print(distro_label)
rhel5_minimal

For more information, take a look at the documentation for TG2.1 and SQLAlchemy declarative syntax.

$ sudo apt-get install libmysqlclient-dev python-dev
$ easy_install SQLAlchemy mysql-python

Not sure why the package is labeled mysql-python. Also, you need the development headers because the MySQLdb module is written in c.

while(fgets(line, sizeof(line)-1, fp)) {
  char *hash; //ptr to hash mark denoting a comment
  if(line[0] == '#' || line[0] == '\n')
    continue;
  if((hash = strchr(line,'#')) != NULL)
  line[hash - line] = '\0'; //line + hash = memory location of hash
  printf("%s", line);
}

Now that I am beginning to get a handle on python, I am really starting to appreciate its elegance. This line of code accomplishes the same task:

line.strip()[:line.find("#")].strip()

I have not compared the performance of the above line to my c code , but I would assume using all the nested objects would incur a performance hit. I may test this later, but I am pretty sure my c code would be much faster.