allow rtfd to build docs without needing to build larch

doc/conf.py

@@ -12,7 +12,6 @@
 # serve to show the default.
 
 import sys, os
-import larch
 
 # If extensions (or modules to document with autodoc) are in another directory,
 # add these directories to sys.path here. If the directory is relative to the
@@ -24,7 +23,7 @@
 .. |sqlite_version| replace:: %(sqlite)s
 .. |numpy_version| replace:: %(numpy)s
 .. |scipy_version| replace:: %(scipy)s
-""" % larch.versions
+""" % {'larch': '3.10a-DEC2014-0-ga2c83bd', 'pandas': '0.14.1', 'apsw': '3.8.7.3-r1', 'python': '3.4.0 final', 'scipy': '0.14.0c1', 'sqlite': '3.8.7.4', 'numpy': '1.8.1'}
 
 # -- General configuration -----------------------------------------------------
 

doc/data.rst

@@ -28,9 +28,30 @@ In addition to opening an existing SQLite database directly, there are a number
 methods available to create a :class:`DB` object without having it linked to an
 original database file.
 
-.. automethod:: DB.Copy
+.. method:: DB.Copy
+
+	:param source: The source database.
+	:type source:  str
+	:param destination: The destination database.
+	:type destination: str
+	:returns: A DB object with an open connection to the destination DB.
+
+	Create a copy of a database and link it to a DB object.
+	It is often desirable to work on a copy of your data, instead of working
+	with the original file. If you data file is not very large and you are 
+	working with multiple models, there can be significant speed advantages
+	to copying the entire database into memory first, and then working on it
+	there, instead of reading from disk every time you want data.
+
+
+.. method:: DB.Example
+
+	Generate an example data object in memory.
+	Larch comes with a few example data sets, which are used in documentation
+	and testing. It is important that you do not edit the original data, so
+	this function copies the data into an in-memory database, which you can
+	freely edit without damaging the original data.
 
-.. automethod:: DB.Example
 
 
 
@@ -40,11 +61,46 @@ Importing Data
 There are a variety of methods available to import data from external sources into
 a SQLite table for use with the larch DB facility.
 
-.. automethod:: DB.import_csv
+.. method:: DB.import_csv
+
+	Import raw csv or tab-delimited data into SQLite.
+
+	:param rawdata:     The absolute path to the raw csv or tab delimited data file.
+	:param table:       The name of the table into which the data is to be imported
+	:param drop_old:    Bool= drop old data table if it already exists?
+	:param progress_callback: If given, this callback function takes a single integer
+						as an argument and is called periodically while loading
+						with the current precentage complete.
+
+	:result:            A list of column headers from the imported csv file
+
 
-.. automethod:: DB.import_dataframe
+.. method:: DB.import_dataframe
 
-.. automethod:: DB.import_dbf
+	Imports data from a pandas dataframe into an existing larch DB.
+
+	:param rawdataframe: An existing pandas dataframe.
+	:param table:        The name of the table into which the data is to be imported
+	:param if_exists:    Should be one of {‘fail’, ‘replace’, ‘append’}. If the table
+						 does not exist this parameter is ignored, otherwise,
+						 *fail*: If table exists, raise a ValueError exception.
+						 *replace*: If table exists, drop it, recreate it, and insert data.
+						 *append*: If table exists, insert data.
+
+	:result:             A list of column headers from imported pandas dataframe
+
+
+.. method:: DB.import_dbf
+
+	Imports data from a DBF file into an existing larch DB.
+
+	:param rawdata:     The path to the raw DBF data file.
+	:param table:       The name of the table into which the data is to be imported
+	:param drop_old:    Bool= drop old data table if it already exists?
+
+	:result:            A list of column headers from imported csv file
+
+	Note: this method requires the dbfpy module (available using pip).