The goal of this app is to provide various graphs. The app uses nens-graph as a base for graphs, lizard-fewsnorm as the data source and the library timeseries for timeseries operations.

All graphs have a grid on the background and the time on the x-axis.

All graph types can be combined in a single graph. You can make a mess when combining everything, the user must review the usefulness himself. The graph types that can be produced using this app are described in the parts below.

• Buildout etc.
• Load the test fewsnorm database.
• Make a fewsnorm source.
• Run bin/django sync_fewsnorm on that fewsnorm source. Now the GeoLocationCache is filled.
• Run bin/django loaddata lizard_graph_test. It loads a fixture corresponding to the test database.
• Run the dev server and go to: /<lizard_graph>/examples/

Dump the data using natural keys. References are auto generated on target machines and pk's do not match.

>>> bin/django dumpdata lizard_graph --indent=2 --natural >lizard_graph.json

You can also dump a sub part only:

>>> bin/django dumpdata lizard_graph.graphitem --indent=2 --natural >lizard_graph.json

Predefined graphs are referenced by natural keys as well.

Make sure the target machine has lizard_fewsnorm 0.8 or higher, or the natural keys will not work (in that case, you need to sync lizard_fewsnorm as well with your source database).

>>> bin/django loaddata lizard_graph.json

The input is the base url, plus url parameters. An example is:

http://127.0.0.1:8000/graph/?dt_start=2011-02-11%2000:00:00&dt_end=2011-11-11%2000:00:00&item={%22fews_norm_source_slug%22:%22test%22,%22location%22:%22111.1%22,%22parameter%22:%22ALMR110%22,%22type%22:%22line%22}

Standard input is: - dt_start: datetime start, in iso8601 format. - dt_end: datetime end, in iso8601 format. - width: width of output image in pixels. - height: height of output image in pixels.

• aggregation-period: month*, day, quarter, year (for stacked-bar)
• aggregation: sum*, avg (for stacked-bar)
• reset-period: month*, day, quarter, year (for stacked-line-cumulative)
• now-line: False*, or True
• format: png*, csv.

All parameters can be omitted. For the datetime the session default will be taken. For the width and height also defaults will be taken.

Labels - title: default none - x-label: default none - y-label: default none

The standard graph has one or multiple lines. The values are on the y-axis. Provide an 'item', each one containing:

• type: line
• location
• parameter
• timestep if available
• qualifierset if available
• layout: color, line-width, line-style

You can use this list of characters for the line-style.

character description '-' solid line style '--' dashed line style '-.' dash-dot line style ':' dotted line style '.' point marker ',' pixel marker 'o' circle marker 'v' triangle_down marker '^' triangle_up marker '<' triangle_left marker '>' triangle_right marker '1' tri_down marker '2' tri_up marker '3' tri_left marker '4' tri_right marker 's' square marker 'p' pentagon marker '*' star marker 'h' hexagon1 marker 'H' hexagon2 marker '+' plus marker 'x' x marker 'D' diamond marker 'd' thin_diamond marker '|' vline marker '_' hline marker

A static horizontal line that is not linked to a timeseries. An example application is to display a goal value.

• type: horizontal-line
• value: xxx
• label (optional)
• layout (optional): color, line-width, line-style

A static vertical line that is not linked to a timeseries. An example application is to display 'today'.

• type: vertical-line
• value: 'today' or iso8601 date
• label (optional)
• layout (optional): color, line-width, line-style

Values are aggregated for certain periods, then displayed as bars. Color is the bar color. Color-outside is the border color. An example application is a waterbalance.

• type: stacked-bar
• location
• parameter
• value (optional): polarity "positive" (default), "negative"
• layout (optional): color, color-outside

Global parameters (outside item): - aggregation-period: month, day, quarter, year - aggregation: sum, avg

Same as stacked bar, except that the sign of the values determine whether each bar is stacked on the positive or negative side. You can use the polarity option to flip it upside-down.

• type: stacked-bar-sign
• location
• parameter
• value (optional): polarity "positive" (default), "negative"
• layout (optional): color, color-outside

Global parameters (outside item): - aggregation-period: month, day, quarter, year - aggregation: sum, avg

Values are displayed as lines with will. They are stacked. Color is the line color. Color-outside is the fill color. An example application is a fraction distribution.

• type: stacked-line
• location
• parameter
• layout (optional): color, color-outside

Cumulative values are displayed as lines with fill. The reset period determines when the cumulative value is reset. Color is the line color. Color-outside is the fill color. An example application is rainfall for an area with sub areas.

• type: stacked-line-cum
• location
• parameter
• layout (optional): color, color-outside

Global parameter (outside item): - reset-period: month, day, quarter, year

The url for a specific non predefined graph can be very long. Predefined graphs can be set up and you only need to provide a few parameters. A predefined graph is 'inserted' as any other graphtype.

This way you can combine multiple predefined graphs in a single graph. You can even define predefined graphs with other predefined graphs.

The parameters that can be provided: - type: predefined-graph - graph: slug of your predefined graph - location (optional, depends on configuration) - locations (optional, see below)

• title
• x-label
• y-label
• y-range-min
• y-range-max
• legend-location: 0..10 (see matplotlib docs) or -1 to turn it off
• aggregation-period: month
• aggregation: sum
• reset-period
• width: 1200
• height: 500
• flags: False

Predefined graphs are described with django models without location. The assumption here is that the same location can be applied to all parameters that occur in a single graph.

You can use predefined matplotlib colors: - blue/b - green/g - red/r - cyan/c - magenta/m - yellow/y - black/k - white/w

Or the html coding in #rrggbb: - #77ff77 - #ffff00

Or gray shades: - 0.75 - 0.5

Locations: dictionary with keyword items as keys. Overrides parameter location and GraphItem.location. For example:

location=naam3 locations={%22loc1%22:%22naam1%22,%22loc2%22:%22naam2%22}

This means: - loc1 = naam1 - loc2 = naam2 - default location = naam3

To be effective, this requires GraphItems with location_wildcard with something like:

"%loc1%_1234" -> this becomes "naam1_1234" and will be filled in as location_id

"%loc3%_asdf" -> loc3 does not exist in the input, so for this GraphItem it will take the predefined location.

"" -> nothing is filled in in location_wildcard, so for this GraphItem it will also take the predefined location.

If the predefined location is not filled in, the default location "naam3" will be used.

Most of the time you want to use a single predefined graph. Normally you would:

http://127.0.0.1:8000/graph/?dt_start=2011-02-11%2000:00:00&dt_end=2011-11-11%2000:00:00&item={%22type%22:%22predefined-graph%22,%22graph%22:%22test%22,%22location%22:%22111.1%22}&width=500&height=300

Shortcut to do the same:

http://127.0.0.1:8000/graph/?dt_start=2011-02-11%2000:00:00&dt_end=2011-11-11%2000:00:00&graph=test&location=111.1&width=500&height=300

And you can still use 'item' to add more stuff to your graph.

Horizontal bar graphs are different from other graphs. On the vertical axis each item has its own "row". Also, the vertical ticks display the item name.

• dt_start
• dt_end
• width
• height
• item

Each item has:

• label
• location
• parameter
• module
• goal: {year, value} (optional, multiple allowed)

• slug
• location (optional)

Provide a slug and optionally a location.

http://127.0.0.1:8000/graph/bar/?dt_start=2011-02-11%2000:00:00&dt_end=2011-11-11%2000:00:00&graph=test&location=111.1&width=500&height=300