Should just change the default value of ToCsv() separator param to ',' 

http://tools.ietf.org/html/rfc4180


   4.  Within the header and each record, there may be one or more
       fields, separated by commas.  Each line should contain the same
       number of fields throughout the file.  Spaces are considered part
       of a field and should not be ignored.  The last field in the
       record must not be followed by a comma.  For example:

       aaa,bbb,ccc

Allows a DataTable to have a default columns_order and order_by, which are used 
if none is specified in one of the export calls.

While it's nice to have the `test` command directly available, I would much 
prefer to have test available as an option with possible dependencies such as 
`nosetest` and `coverage`. Overwriting method with `pass` as is currently the 
case is not best practice.

Regarding the tests themselves: unit tests should test individual methods and 
functions in isolation where possible with additional tests for different code 
paths. Currently _InnerAppendData has no test coverage and AppendData contains 
not only tests for different code paths into the method but also for detecting 
exceptions that will be raised by _InnerAppendData. I will submit a patch both 
for testing and for the setup.

What
Allow the developer to add not only int, long, float values but also 
decimal.Decimal values.

Why
When writing your own data source, you often pull data from a MySQL database. 
Any aggregation in the database (like AVG() or SUM()) returns a 
decimal.Decimal. It's bothersome to have to cast to int, long, or float and you 
have to take care of None values separately. It would be better if you could 
add any numbers from the database to a DataTable without having to know what 
SQL generated it.

[deleted issue]

What steps will reproduce the problem?
1. Add the trunk of this project to mine using svn:externals
2. Try to import gviz_api.py
3. Missing __init__.py causes fail

Checking in an __init__.py at the top-level trunk directory would be really 
cool.

I am sometimes sending a lot of float values over the wire however most of the 
time I am only interested in a the first 3 or 4 digits after the comma. 
However right now (even when the float is rounded) all digits are transmitted 
which increases the payload size significantly. 

It would be helpful to have a method that allows the user to specify the 
precision of the float. 

Right now I monkeypatched the ToString(value) method to return a float with 
precision 2 (hardcoded)

"%.2f" % value instead of just str(value).lower()

It would be nice if we could configure the precision somehow (maybe by column 
proprerties)?

What steps will reproduce the problem?
Adding a second visualization to a page fails if reqID=0
The supplied example will fail this test

What is the expected output? What do you see instead?
It would be nice to see the example improved to use reqID as per the 0.6 
standard.

What version of the product are you using? On what operating system?
1.7.0  for protocol ver 0.6

Please provide any additional information below.

What steps will reproduce the problem?

Let's say you want to build a data source showing your ice cream consumption 
per day. Users should be able to download the data in CSV format, so they can 
import it into their own spreadsheets. Let's say you ate one ice cream on Jan 1 
and two ice creams on Jan 2. Start the Python interpreter and run the following 
code:
------------------------------
import datetime
import gviz_api
data = []
row = {}
row['date'] = datetime.date(2011,1,1)
row['icecreams'] = 1
data.append(row)
row = {}
row['date'] = datetime.date(2011,1,2)
row['icecreams'] = 2
data.append(row)
description = {}
description["date"] = ("date", "Date")
description["icecreams"] = ("number", "Ice creams eaten")
data_table = gviz_api.DataTable(description)
data_table.LoadData(data)
print data_table.ToCsv()
------------------------------


What is the expected output? What do you see instead?

I expected this output:
"Date", "Ice creams eaten"
"2011-01-01", 1
"2011-01-02", 2

Instead I got this output:
"Date", "Ice creams eaten"
"new Date(2011,0,1)", 1
"new Date(2011,0,2)", 2

JavaScript formatting of dates makes sense in JSON output. In CSV output it 
doesn't. ISO-formatting, YYYY-MM-DD, would make more sense as it can be read by 
spreadsheet software.


What version of the product are you using? On what operating system?

gviz_api_py-1.7.0.tar.gz
Python 2.6.5
MacOS 10.5

Reproduce:
1. Create a table with a column who has a label "abc ' def"
2. Try to use the table's json.

Instead of showing the visualization, a JS error will be thrown,

DataTable.ToResponse

In certain cases the argument tqx end with a Semicolon causing the the 
following traceback

Traceback (most recent call last):
  File "/usr/lib/python2.6/site-packages/cherrypy/_cprequest.py", line 645, in respond
    response.body = self.handler()
  File "/usr/lib/python2.6/site-packages/cherrypy/lib/encoding.py", line 188, in __call__
    self.body = self.oldhandler(*args, **kwargs)
  File "/usr/lib/python2.6/site-packages/cherrypy/_cpdispatch.py", line 29, in __call__
    return self.callable(*self.args, **self.kwargs)
  File "/opt/sandbox/IDIPYT/trunk/src/py/_RECVIEW/Ajax.py", line 101, in timeLine
    return table.ToResponse(tqx = tqx)
  File "/usr/local/lib/python2.6/site-packages/gviz_api.py", line 1041, in ToResponse
    tqx_dict = dict(opt.split(":") for opt in tqx.split(";"))
ValueError: dictionary update sequence element #1 has length 1; 2 is required

adding at line 1033 solves the issue
if tqx.endswith(";") :
    tqx = tqx[:-1]

What steps will reproduce the problem?
1. Use library in Python <2.6
2. Get import error for "json"

What is the expected output? What do you see instead?
  Working module, import error.

What version of the product are you using? On what operating system?
  Python <2.6

Please provide any additional information below.
  Add import check against json and if failure, import simplejson.

As the project already contains the setup.py file, it should be fairly easy to 
add it to PyPI. This would ease automatic installation, uninstallation, 
dependency management etc. for the users of google-visualization-python.

http://pypi.python.org/

While I very much appreciate this module I found it extremely counterintuitive 
to use. Allowing different constructors not only for the description but 
depending on their structure for adding data doesn't make sense to me. Surely, 
the two should be decoupled?

Be that as it may. One thing I found I wanted to use but found lacking was an 
AppendRow() method as AppendData() works row by row anyway.

  def AppendRow(self, row, custom_properties):
    self._InnerAppendData(({}, custom_properties), row, 0)

NaN and +-Inf numbers are not allowed in JSON.  
They should be represented as "null" in JSON output.  
In CSV output, it's less clear since there is no standard, but empty string 
would work for me.

see http://tools.ietf.org/html/rfc4627 about NaN

see http://en.wikipedia.org/wiki/IEEE_754-2008 about JSON numbers
2.4.  Numbers
...
Numeric values that cannot be represented as sequences of digits
   (such as Infinity and NaN) are not permitted.

What steps will reproduce the problem?
1. python setup.py sdist
2. copy tarball elsewhere
3. try and install via easy_install

What is the expected output? What do you see instead?
I copied the sdist to our webserver, and then try to install it:
easy_install --find-links http://ooici.net/packages gviz_api.py

Searching for gviz-api.py
Reading http://ooici.net/packages
Best match: gviz-api.py 1.7.0
Downloading http://ooici.net/packages/gviz_api.py-1.7.0.tar.gz
Processing gviz_api.py-1.7.0.tar.gz
Running gviz_api.py-1.7.0/setup.py -q bdist_egg --dist-dir 
/var/folders/Sw/SwEOWKQ7FdihT03Yr5CBOU+++TI/-Tmp-/easy_install-IQGY_h/gviz_api.p
y-1.7.0/egg-dist-tmp-DH3Ktj
Traceback (most recent call last):
  File "/Users/hubbard/code/ve/junk/bin/easy_install", line 8, in <module>
    load_entry_point('setuptools==0.6c9', 'console_scripts', 'easy_install')()
  File "/Users/hubbard/code/ve/junk/lib/python2.6/site-packages/setuptools-0.6c9-py2.6.egg/setuptools/command/easy_install.py", line 1671, in main

  File "/Users/hubbard/code/ve/junk/lib/python2.6/site-packages/setuptools-0.6c9-py2.6.egg/setuptools/command/easy_install.py", line 1659, in with_ei_usage

  File "/Users/hubbard/code/ve/junk/lib/python2.6/site-packages/setuptools-0.6c9-py2.6.egg/setuptools/command/easy_install.py", line 1675, in <lambda>

  File "/System/Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/distutils/core.py", line 152, in setup
    dist.run_commands()
  File "/System/Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/distutils/dist.py", line 975, in run_commands
    self.run_command(cmd)
  File "/System/Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/distutils/dist.py", line 995, in run_command
    cmd_obj.run()
  File "/Users/hubbard/code/ve/junk/lib/python2.6/site-packages/setuptools-0.6c9-py2.6.egg/setuptools/command/easy_install.py", line 211, in run

  File "/Users/hubbard/code/ve/junk/lib/python2.6/site-packages/setuptools-0.6c9-py2.6.egg/setuptools/command/easy_install.py", line 446, in easy_install

  File "/Users/hubbard/code/ve/junk/lib/python2.6/site-packages/setuptools-0.6c9-py2.6.egg/setuptools/command/easy_install.py", line 476, in install_item

  File "/Users/hubbard/code/ve/junk/lib/python2.6/site-packages/setuptools-0.6c9-py2.6.egg/setuptools/command/easy_install.py", line 655, in install_eggs

  File "/Users/hubbard/code/ve/junk/lib/python2.6/site-packages/setuptools-0.6c9-py2.6.egg/setuptools/command/easy_install.py", line 930, in build_and_install

  File "/Users/hubbard/code/ve/junk/lib/python2.6/site-packages/setuptools-0.6c9-py2.6.egg/setuptools/command/easy_install.py", line 919, in run_setup

  File "/Users/hubbard/code/ve/junk/lib/python2.6/site-packages/setuptools-0.6c9-py2.6.egg/setuptools/sandbox.py", line 27, in run_setup
  File "/Users/hubbard/code/ve/junk/lib/python2.6/site-packages/setuptools-0.6c9-py2.6.egg/setuptools/sandbox.py", line 63, in run
  File "/Users/hubbard/code/ve/junk/lib/python2.6/site-packages/setuptools-0.6c9-py2.6.egg/setuptools/sandbox.py", line 29, in <lambda>
  File "setup.py", line 23, in <module>
ImportError: No module named gviz_api_test

The fix:
- Add gviz_api_test.py to MANIFEST
- Change py_modules stanza in setup.py to:

    py_modules=["gviz_api", 'gviz_api_test'],

Now works.

What steps will reproduce the problem?
1. Setup App Engine & Install google-visualization-python
2. Run the sample.py in the google App Enging
3.

What is the expected output? What do you see instead?
Google App Engine unable to find gviz_app libraries

What version of the product are you using? On what operating system?


Please provide any additional information below.

ERROR    2008-12-08 19:28:53,910 dev_appserver.py] Encountered error 
loading mod
ule "src.viz": <type 'exceptions.ImportError'>: Could not find module 
src.viz
Traceback (most recent call last):

File "C:\Users\Michael\DevTools\google_appengine\google\appengine\tools\dev
_ap
pserver.py", line 1607, in LoadTargetModule
    module_code = import_hook.get_code(module_fullname)

File "C:\Users\Michael\DevTools\google_appengine\google\appengine\tools\dev
_ap
pserver.py", line 847, in decorate
    return func(self, *args, **kwargs)

File "C:\Users\Michael\DevTools\google_appengine\google\appengine\tools\dev
_ap
pserver.py", line 1470, in get_code
    full_path, search_path, submodule = self.GetModuleInfo(fullname)

File "C:\Users\Michael\DevTools\google_appengine\google\appengine\tools\dev
_ap
pserver.py", line 847, in decorate
    return func(self, *args, **kwargs)

File "C:\Users\Michael\DevTools\google_appengine\google\appengine\tools\dev
_ap
pserver.py", line 1422, in GetModuleInfo
    submodule, search_path = self.GetParentSearchPath(fullname)

File "C:\Users\Michael\DevTools\google_appengine\google\appengine\tools\dev
_ap
pserver.py", line 847, in decorate
    return func(self, *args, **kwargs)

File "C:\Users\Michael\DevTools\google_appengine\google\appengine\tools\dev
_ap
pserver.py", line 1401, in GetParentSearchPath
    parent_package = self.GetParentPackage(fullname)

File "C:\Users\Michael\DevTools\google_appengine\google\appengine\tools\dev
_ap
pserver.py", line 847, in decorate
    return func(self, *args, **kwargs)

File "C:\Users\Michael\DevTools\google_appengine\google\appengine\tools\dev
_ap
pserver.py", line 1378, in GetParentPackage
    raise ImportError('Could not find module %s' % fullname)
ImportError: Could not find module src.viz
WARNING  2008-12-08 19:28:53,926 dev_appserver.py] Missing package 
initializatio
n files: C:\Users\Michael\workspace\helloworld\src\__init__.py
ERROR    2008-12-08 19:28:53,953 dev_appserver.py] Exception encountered 
handlin
g request
Traceback (most recent call last):

File "C:\Users\Michael\DevTools\google_appengine\google\appengine\tools\dev
_ap
pserver.py", line 2443, in _HandleRequest
    base_env_dict=env_dict)

File "C:\Users\Michael\DevTools\google_appengine\google\appengine\tools\dev
_ap
pserver.py", line 353, in Dispatch
    base_env_dict=base_env_dict)

File "C:\Users\Michael\DevTools\google_appengine\google\appengine\tools\dev
_ap
pserver.py", line 1869, in Dispatch
    self._module_dict)

File "C:\Users\Michael\DevTools\google_appengine\google\appengine\tools\dev
_ap
pserver.py", line 1787, in ExecuteCGI
    reset_modules = exec_script(handler_path, cgi_path, hook)

File "C:\Users\Michael\DevTools\google_appengine\google\appengine\tools\dev
_ap
pserver.py", line 1683, in ExecuteOrImportScript
    exec module_code in script_module.__dict__
  File "C:\Users\Michael\workspace\helloworld\src/viz.py", line 1, in 
<module>
    import gviz_api
ImportError: No module named gviz_api

var query = new google.visualization.Query(DATA_SOURCE_URL);
query.setQuery('select dept, sum(salary) group by dept');
query.send(handleQueryResponse);

             cpuData.addColumn('date', 'Date');
             cpuData.addColumn('number', 'CPU');
             cpuData.addRow([new Date(1252152000000),4])
             cpuData.addRow([new Date(1252152000123),5])
             cpuData.addRow([new Date(1252152000456),6])

function createCpuRamChart() {

    // A. Create new instance of DataTable to add our data to
    var cpuData = new google.visualization.DataTable();

    cpuData.addColumn('date', 'Date');
    cpuData.addColumn('number', '@Resources.Administration.ServerStatistics_CPU_ChartLabel');

    @if (Model.cpuUsage != null)
    {
        DateTime d1 = new DateTime(1970, 1, 1);
        foreach (var item in Model.cpuUsage)
        {
            var dateTime = (DateTime)item.GetValue(0); 
            DateTime d2 = dateTime.ToUniversalTime();
            TimeSpan ts = new TimeSpan(d2.Ticks - d1.Ticks);
            @: cpuData.addRow([new Date(@ts.TotalMilliseconds),@item.GetValue(1).ToString().Replace(",", ".")])
                }
    };

    //B. RAM data instance

    var ramData = new google.visualization.DataTable();

    // B. Create three columns with DataTable.addColumn(type, label)
    ramData.addColumn('date', 'Date');
    ramData.addColumn('number', '@Resources.Administration.ServerStatistics_RAM_ChartLabel');

    @if (Model.memUsage != null)
    {
        DateTime d1 = new DateTime(1970, 1, 1);
        foreach (var item in Model.memUsage)
        {
            var dateTime = (DateTime)item.GetValue(0);
            DateTime d2 = dateTime.ToUniversalTime();
            TimeSpan ts = new TimeSpan(d2.Ticks - d1.Ticks);
            @: ramData.addRow([new Date(@ts.TotalMilliseconds),@item.GetValue(1).ToString().Replace(",", ".")])
        }
    };


    // Create the main Chart
    var cpuChart = new google.visualization.ChartWrapper({
        chartType: 'LineChart',
        containerId: 'cpuUsageLineChart',
        options: {
            hAxis: {
                title: 'Period'
            },
            vAxis: {
                title: 'Percentage'
            },
            width: 800,
            height: 400,
            chartArea: { left: 70, top: 20, bottom: 30, width: "80%", height: "80%" }
        }
    });


    var ramChart = new google.visualization.ChartWrapper({
        'chartType': 'LineChart',
        'containerId': 'ramUsageLineChart',
        options: {
            //legend:'none',
            hAxis: {
                title: 'Period'
            },
            vAxis: {
                title: 'GB'
            },
            width: 800,
            height: 400,
            chartArea: { left: 70, top: 20, bottom: 30, width: "80%", height: "80%" },
        }
    });

    // Create a date range slider
    var cpuDateSlider = new google.visualization.ControlWrapper({
        'controlType': 'ChartRangeFilter',
        'containerId': 'cpuUsageDateRange',
        'options': {

            height: 150,
            // Filter by the date axis.
            filterColumnLabel: 'Date',
            backgroundColor: '#f1f8e9',
            ui: {
                //minRangeSize: 1800000,//not needed. You can set the range under the master dashboard
                chartOptions: {
                    chartArea: { width: '95%', height: '100' },
                },
            }
        },

    });

    var ramDateSlider = new google.visualization.ControlWrapper({
        'controlType': 'ChartRangeFilter',
        'containerId': 'ramUsageDateRange',
        'options': {
            // Filter by the date axis.
            filterColumnLabel: 'Date',
            backgroundColor: '#f1f8e9',
            ui: {
                //minRangeSize: 1800000, //not needed. You can set the range under the master dashboard
                chartOptions: {
                    chartArea: { width: '95%', height: '100' },
                },
            }
        },

    });
    var formatDate = new google.visualization.DateFormat(
       { prefix: 'Time: ', pattern: "dd MMM HH:mm", });
        //{ prefix: 'Time: ', formatType: 'medium', });

    formatDate.format(cpuData, 0);
    formatDate.format(ramData, 0);
    // Create a dashboard.
    var dash_container = new google.visualization.Dashboard(document.getElementById('dashboard'));
    dash_container.bind(cpuDateSlider, cpuChart);

    var dash_container2 = new google.visualization.Dashboard(document.getElementById('dashboard2'));
    dash_container2.bind(ramDateSlider, ramChart);

    dash_container.draw(cpuData);
    dash_container2.draw(ramData);

    // create the master control
    var controlMaster = new google.visualization.ControlWrapper({
        controlType: 'ChartRangeFilter',
        containerId: 'control_master2',
        dataTable: cpuData,
        options: {

            // Filter by the date axis.
            filterColumnLabel: 'Date',
            backgroundColor: '#f1f8e9',
            ui: {
                // 1 day in milliseconds = 24 * 60 * 60 * 1000 = 86,400,000
                minRangeSize: 1800000,// 30 min
                //chartView: {columns: [0, 1]},
                chartOptions: {
                    chartArea: { width: '95%', height: '80' },
                },
            }
        },
    });
    google.visualization.events.addListener(controlMaster, 'statechange', function () {
        var state = controlMaster.getState();
        cpuDateSlider.setState(state);
        ramDateSlider.setState(state);
        dash_container.draw(cpuData);
        dash_container2.draw(ramData);
    });

    controlMaster.draw();
}

Please consider adding support for a Unix Epoch form of Date (unixepoch), as 
used by sqlite3. This is very handy for sysadmin-type stuff, and I'm already 
using a modified version of the Python library supporting it.

Here is a context diff showing how to implement it:

*** google-visualization-python/gviz_api.py      Thu Oct  8 06:00:05 2009
--- with-unixepoch-support/gviz_api.py     Mon Oct 18 17:08:11 2010
***************
*** 185,192 ****

      Args:
        value: The value which should be converted
!       value_type: One of "string", "number", "boolean", "date", "datetime" or
!                   "timeofday".
        escape_func: The function to use for escaping strings.

      Returns:
--- 185,192 ----

      Args:
        value: The value which should be converted
!       value_type: One of "string", "number", "boolean", "date", "datetime",
!                   "timeofday" or "unixepoch".
        escape_func: The function to use for escaping strings.

      Returns:
***************
*** 264,269 ****
--- 264,274 ----
          raise DataTableException("Wrong type %s when expected time" % t_value)
        return "[%d,%d,%d]" % (value.hour, value.minute, value.second)

+     elif value_type == "unixepoch":
+       if not isinstance(value, (int, long)):
+         raise DataTableException("Wrong type %s when expected seconds since 
unix epoch" % t_value)
+       return "new Date(%d)" % (value*1000)
+
      elif value_type == "datetime":
        if not isinstance(value, datetime.datetime):
          raise DataTableException("Wrong type %s when expected datetime" %
***************
*** 335,341 ****
            if len(description) > 4:
              raise DataTableException("Description error: tuple of length > 4")
      if desc_dict["type"] not in ["string", "number", "boolean",
!                                  "date", "datetime", "timeofday"]:
        raise DataTableException(
            "Description error: unsupported type '%s'" % desc_dict["type"])
      return desc_dict
--- 340,346 ----
            if len(description) > 4:
              raise DataTableException("Description error: tuple of length > 4")
      if desc_dict["type"] not in ["string", "number", "boolean",
!                                  "date", "datetime", "timeofday", 
"unixepoch"]:
        raise DataTableException(
            "Description error: unsupported type '%s'" % desc_dict["type"])
      return desc_dict

What steps will reproduce the problem?
1. Create following code:
#In subclass of "webapp.RequestHandler"
cols = {('Col1', 'String'): {'Col2': 'string'}}
    rows = {'row1': {'Col2': 'Col2, row1'}, 
            'row2': {'Col2': 'Col2, row2'}}
    colOrder = ['Col1', 'Col2']
    data_table = gviz_api.DataTable(cols)
    data_table.AppendData(rows)
self.response.out.write(data_table.ToJSonResponse(columns_order=colOrder,order_b
y=colOrder[0]))

2. Make a simple JS embeded HTML to access this response:
<html xmlns="http://www.w3.org/1999/xhtml"> 
  <head> 
    <meta http-equiv="content-type" content="text/html; charset=utf-8"/> 
    <title> 
      Google Visualization API Sample
    </title> 
    <script type="text/javascript" src="http://www.google.com/jsapi"></script> 
    <script type="text/javascript"> 
    google.load('visualization', '1', {packages: ['table']});
    </script> 
    <script type="text/javascript"> 
    var visualization;

    function drawVisualization() {
      var query = new google.visualization.Query(
          'http://localhost:8080/query?name=gdp&dims=all&dfmt=%Y.%m');

      // Send the query with a callback function.
      query.send(handleQueryResponse);
    }

    function handleQueryResponse(response) {
      if (response.isError()) {
        alert('Error in query: ' + response.getMessage() + ' ' + response.getDetailedMessage());
        return;
      }

      var data = response.getDataTable();
      visualization = new google.visualization.Table(document.getElementById('mytable'));
      visualization.draw(data, null);
    }

    google.setOnLoadCallback(drawVisualization);
    </script> 
  </head> 
  <body style="font-family: Arial;border: 0 none;"> 
    <div id="mytable" style="height: 400px; width: 800px;"></div> 
  </body> 
</html> 

What is the expected output? What do you see instead?
Expected: The column2 value in row1 and row2 are "Col2, row1" and "Col2, row2"
Instead: The column2 value in row1 and row2 are "Col2", which is column2's name!

What version of the product are you using? On what operating system?
Version: gviz_api_py-1.1.0.tar
OS: WinXP

Please provide any additional information below.
If I simple add a non-value Col3 as following:

cols = {('Col1', 'String'): {'Col2': 'string', 'Col3': 'string'}}
rows = {'row1': {'Col2': 'Col2, row1'}, 
            'row2': {'Col2': 'Col2, row2'}}
colOrder = ['Col1', 'Col2', 'Col3']

The result shown will be correct.

DataTable.SingleValueToJS (version 1.7)
Using "datetime" as input the time resolution is limited to 1 second.
datetime.resolution is 1 microseconds and the javascript Date() allows also 
higher resolution (at least millisecs).

return statement of SingleValuetoToJS

millisecs = int(round(value.microsecond/1000.0, 0))
return "new Date(%d,%d,%d,%d,%d,%d,%d,%d)" % (value.year,
                                                 value.month - 1,
                                                 value.day,
                                                 value.hour,
                                                 value.minute,
                                                 value.second,
                                                 millisecs
                                               )

ToJSon returns valid syntax for a JavaScript object, but it is not parseable 
with many JSON implementations 
because it uses syntax that is not within the subset defined by the JSON 
standard.

The simplest solution is probably to rename this function or at least document 
that it doesn't really return 
JSON, but expects to be executed by a full JS interpreter.

For example, the result will not parse with Python 2.6's json package:
>>> d = gviz_api.DataTable(('a', 'string'), ['foo'])
>>> d.ToJSon()
"{cols: [{id:'a',label:'a',type:'string'}],rows: [{c:[{v:'foo'}]}]}"
>>> json.loads(d.ToJSon())
Traceback (most recent call last):
...
ValueError: Expecting property name: line 1 column 1 (char 1)

Some more basic issues are:
- JSON strings must use double-quotes
- JSON keys must be quoted following the string syntax
e.g. {"key": "value"}, not {key: 'value'}

Python's default unicode escaping is also not compatible with JSON strings.  
For example, it will use \x hex 
escape codes which do happen to work in JavaScript, but JSON only allows \u 
4-digit hex escapes, e.g.
Wrong:
>>> json.loads('"\\xfc"')
Traceback (most recent call last):
...
ValueError: Invalid \escape: line 1 column 1 (char 1)

Right:
>>> json.loads('"\\u00fc"')
u'\xfc'

Also, JSON does not define any syntax for Date types, so the "new Date()" 
syntax is not supported by most 
non-JS-native JSON parsers.

Since this data structure is primarily designed for consumption by the JS viz 
API, I don't have a problem with 
the syntax remaining as it is, just please don't misrepresent it as "JSON", 
which could cause confusion if 
people try to parse it with a JSON library.

This is a patch that adds support for warnings and errors in JSON responses.

It includes new unittests to test added functionality. The results were tested 
against Google's javascript library and they come through correctly.

No old unittests needed fixing, since the change is 100% backwards compatible.

Currently if a python datetime object is given, an appropriate javascript
date object will be created as follows:
new Date(%d,%d,%d,%d,%d,%d)" % (value.year, value.month ....

Please note that this will be local time for the js date, which may not
match the python datetime object. I'd suggest converting to a
unix-timestamp and creating the js date from the timestamp instead. e.g.:
new Date(1252152000000)

What steps will reproduce the problem?
1. visit http://code.google.com/p/google-visualization-python/
2. click http://google-visualization.appspot.com/python/dynamic_example


What is the expected output? What do you see instead?
some graph is expected, instead, I am looking at the following message:

google.visualization.Query.setResponse({'version':'0.5', 'reqId':'0', 
'status':'OK', 'table': {cols: 
[{id:'name',label:'Name',type:'string'},{id:'salary',label:'Salary',type:'n
umber'},{id:'full_time',label:'Full Time Employee',type:'boolean'}],rows: 
[{c:[{v:'Jim'},{v:800,f:'$800'},{v:false}]},{c:[{v:'Bob'},{v:7000,f:'$7,000
'},{v:true}]},{c:[{v:'Mike'},{v:10000,f:'$10,000'},{v:true}]},{c:[{v:'Alice
'},{v:12500,f:'$12,500'},{v:true}]}]}});

What version of the product are you using? On what operating system?
windows xp, chrome

Please provide any additional information below.

In the class DataTableJSONEncoder (line 62), the line

    return super(DataTableJSONEncoder, self).default(self, o)

should read:

    return super(DataTableJSONEncoder, self).default(o)

Example session:

>>> import gviz_api
>>> encoder = gviz_api.DataTableJSONEncoder()
>>> encoder.default(0)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "gviz_api.py", line 62, in default
    return super(DataTableJSONEncoder, self).default(self, o)
TypeError: default() takes exactly 2 arguments (3 given)

To reproduce the problem:
>>> dt = gviz_api.DataTable({"c1": ("string", "column 1")})
>>> dt.LoadData({"c1": "blah"})

Problem is, the case of one column is ambiguous with a different description 
case.