Dali Test Examples

The Dali tests exercise the /dali endpoint directly, bypassing the WMS/WMTS layers. Test inputs are .get files in test/input/ and expected outputs in test/output/. Product JSON configurations are under test/dali/customers/.

Each request URL is decomposed into a table showing every query parameter and its effect. Layer-specific URL overrides follow the pattern l.{qid}.{setting} and view overrides follow v{n}.{setting}.

Contents

• Output Formats
• Isoband and Isoline
• Pressure smoothing comparison
• Grid smoother comparison
• Grid smoother comparison — low cost
• TFP — Thermal-Front Parameter diagnostics
• Isoband Labels
• Isolabel examples
• Location Label Placement
• SVG Customisation
• Clipping to Geographic Boundaries
• Pressure Level and Elevation Selection
• Multiple Views and Side-by-Side Products
• Time Layer and Timezone
• Observation Numbers
• Observation Positions and Keyword Selection
• Weather Symbols
• Flash and Lightning Symbols
• Wind Arrows
• Wind Barbs and Wind Roses
• WAFS Aviation Charts
• Observations with Wind and Temperature
• Heatmap
• Hovmoeller Diagrams
• Graticule
• Circles Layer
• Fronts
• PostGIS Layer
• WKT Layer
• METAR Layer
• Fire Weather
• Text Layer
• Map simplification
• World Map Products

---

Output Formats

Dali can produce SVG (default), PNG, PDF, GeoJSON, KML, GeoTIFF, MVT, and DataTile from the same product definition.

t2m_p — SVG (default)

Input: test/input/t2m_p.get

GET /dali?customer=test&product=t2m_p&time=200808050300 HTTP/1.0

Parameter	Value	Description
customer	test	Customer directory
product	t2m_p	Product JSON: test/dali/customers/test/products/t2m_p.json
time	200808050300	Valid time: 2008-08-05 03:00 UTC

Renders Scandinavian temperature isobands and isolines from the kap producer in the data's native CRS (500×500 px). The product uses a defs block to define SVG symbols and CSS styles. Output is SVG (default when no type is specified).

Output:

---

png — PNG output

Input: test/input/png.get

GET /dali?customer=test&product=t2m_p&type=png&time=200808050300 HTTP/1.0

Parameter	Value	Description
customer	test	Customer directory
product	t2m_p	Same product as above
type	png	Forces 8-bit palette PNG raster output
time	200808050300	Valid time: 2008-08-05 03:00 UTC

Any product can be rasterised to PNG by supplying type=png. Output is 500×500 px, 8-bit colour-mapped PNG (default ≤256 colours).

Output:

---

png_truecolor — 24-bit PNG

Input: test/input/png_truecolor.get

GET /dali?customer=test&product=t2m_p&type=png&time=200808050300&png.truecolor=1 HTTP/1.0

Parameter	Value	Description
type	png	PNG output
png.truecolor	1	Forces 24-bit true-colour PNG instead of palette

Output:

---

png_maxcolors_20 — Limited palette PNG

Input: test/input/png_maxcolors_20.get

GET /dali?customer=test&product=t2m_p&type=png&time=200808050300&png.maxcolors=20 HTTP/1.0

Parameter	Value	Description
type	png	PNG output
png.maxcolors	20	Caps the palette at 20 colours; reduces file size but introduces posterisation

Output:

---

pdf — PDF output

Input: test/input/pdf.get

GET /dali?customer=test&product=t2m_p&time=200808050300&type=pdf HTTP/1.0

Parameter	Value	Description
type	pdf	Produces a vector PDF document

Output is a scalable PDF (no rasterisation).

---

t2m_p_geojson — GeoJSON output

Input: test/input/t2m_p_geojson.get

GET /dali?customer=test&product=t2m_p&time=200808050300&type=geojson HTTP/1.0

Parameter	Value	Description
type	geojson	Exports isoband/isoline geometries as GeoJSON

Polygon and line features are reprojected to WGS84 (EPSG:4326) in the output.

---

t2m_p_kml — KML output

Input: test/input/t2m_p_kml.get

GET /dali?customer=test&product=t2m_p&time=200808050300&type=kml HTTP/1.0

Parameter	Value	Description
type	kml	Exports isoband/isoline geometries as KML (Google Earth)

Note: listed in .testignore — output precision can vary between RHEL 7 and RHEL 8 environments.

---

datatile_temperature — DataTile: single-band temperature

Input: test/input/datatile_temperature.get

GET /dali?customer=grid&type=datatile&product=datatile_temperature&time=200808050800 HTTP/1.0

Parameter	Value	Description
customer	grid	Grid data customer directory
type	datatile	RGBA-encoded float data PNG (bypasses SVG pipeline)
product	datatile_temperature	Product JSON: test/dali/customers/grid/products/datatile_temperature.json
time	200808050800	Valid time: 2008-08-05 08:00 UTC

Returns a 64x64 PNG tile where each pixel encodes a temperature value (parameter T-K) as a 16-bit quantised value in the R and G channels. The A channel is 255 for valid pixels, 0 for missing data. Scale and offset metadata are embedded in PNG tEXt chunks (datatile:min, datatile:max).

Output: test/output/datatile_temperature.get — PNG (binary, not a visual image)

---

datatile_wind — DataTile: dual-band wind direction + speed

Input: test/input/datatile_wind.get

GET /dali?customer=grid&type=datatile&product=datatile_wind&time=200808050800 HTTP/1.0

Parameter	Value	Description
customer	grid	Grid data customer directory
type	datatile	RGBA-encoded float data PNG
product	datatile_wind	Product JSON: test/dali/customers/grid/products/datatile_wind.json
time	200808050800	Valid time: 2008-08-05 08:00 UTC

Returns a 64x64 PNG tile with two bands encoded in the RGBA channels: wind direction (DD-D) in R+G and wind speed (FF-MS) in B+A. This is the dual-band encoding where valid values occupy [1, 65535] and 0 indicates missing data. Metadata chunks include datatile:min1/max1 (direction) and datatile:min2/max2 (speed).

Output: test/output/datatile_wind.get — PNG (binary)

---

datatile_precipitation — DataTile: single-band precipitation (RasterLayer)

Input: test/input/datatile_precipitation.get

GET /dali?customer=grid&type=datatile&product=datatile_precipitation&time=200808050800 HTTP/1.0

Parameter	Value	Description
customer	grid	Grid data customer directory
type	datatile	RGBA-encoded float data PNG
product	datatile_precipitation	Product JSON: test/dali/customers/grid/products/datatile_precipitation.json
time	200808050800	Valid time: 2008-08-05 08:00 UTC

Returns a 64x64 PNG tile with 1-hour precipitation (PRECIP1H-KGM2) encoded as a single-band datatile via the raster layer type.

Output: test/output/datatile_precipitation.get — PNG (binary)

---

autoclass — Automatic CSS class names

Input: test/input/autoclass.get

GET /dali?customer=test&product=autoclass&time=200808050300&type=svg HTTP/1.0

Parameter	Value	Description
product	autoclass	Product JSON: test/dali/customers/test/products/autoclass.json
type	svg	Forces SVG output

The autoclass feature generates CSS class names automatically from isoband/isoline limit values using a template such as "Temperature_{}_{}"; the lower and upper limits substitute the placeholders (using None for open-ended bands). This eliminates the need to assign class names manually in every isoband specification.

Output:

---

autoqid — Automatic query IDs

Input: test/input/autoqid.get

GET /dali?customer=test&product=autoqid&time=200808050300&format=svg HTTP/1.0

Parameter	Value	Description
product	autoqid	Product JSON: test/dali/customers/test/products/autoqid.json
format	svg	Alternative to type=svg for specifying SVG output

The autoqid feature auto-generates qid values for each isoband/isoline from the band limits using a template such as "temperature_{}_{}"; the resulting qid can then be referenced in URL parameter overrides. Unlike autoclass (which generates CSS class names), autoqid generates the query IDs used for URL-based overrides.

Output:

---

disable_svg — Conditionally disabled layer

Input: test/input/disable_svg.get

GET /dali?customer=test&product=disable_svg&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	disable_svg	Product JSON: test/dali/customers/test/products/disable_svg.json

A layer carrying "disable": "svg" is skipped entirely when rendering to SVG but rendered for PNG/PDF. Useful for layers (e.g. pressure isolines) that are visually acceptable as vector but are too slow or heavy in SVG.

Output:

---

enable_png — Conditionally enabled layer

Input: test/input/enable_png.get

GET /dali?customer=test&product=enable_png&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	enable_png	Product JSON: test/dali/customers/test/products/enable_png.json

A layer carrying "enable": ["png","pdf"] is rendered only for those output formats and suppressed for SVG. The product adds pressure isolines only for raster/PDF output.

Output:

---

Isoband and Isoline

feelslike — FeelsLike parameter

Input: test/input/feelslike.get

GET /dali?customer=test&product=feelslike&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	feelslike	Product JSON: test/dali/customers/test/products/feelslike.json

Renders isobands and isolines of the FeelsLike apparent-temperature parameter from the pal_skandinavia producer, using the same colour scale as regular temperature.

Output:

---

precipitation — Precipitation isobands

Input: test/input/precipitation.get

GET /dali?customer=test&product=precipitation&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	precipitation	Product JSON: test/dali/customers/test/products/precipitation.json

Renders precipitation isobands and isolines from pal_skandinavia in the data's native CRS.

Output:

---

precipitation_warm — Precipitation warm season

Input: test/input/precipitation_warm.get

GET /dali?customer=test&product=precipitation_warm&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	precipitation_warm	Product JSON: test/dali/customers/test/products/precipitation_warm.json

Uses a warm-season precipitation colour scale with finer class resolution in the low-intensity range.

Output:

---

precipitation_minarea — Suppress small polygons (geographic area)

Input: test/input/precipitation_minarea.get

GET /dali?customer=test&product=precipitation&time=200808050300&l1.minarea=5000&l2.minarea=5000 HTTP/1.0

Parameter	Value	Description
l1.minarea	5000	Drops isoband polygons smaller than 5000 m² for layer l1
l2.minarea	5000	Same for layer l2

The minarea setting removes noise from scattered small polygons. The default unit is m².

Output:

---

precipitation_minarea_px — Suppress small polygons (pixel area)

Input: test/input/precipitation_minarea_px.get

GET /dali?customer=test&product=precipitation&time=200808050300&l1.minarea=1000&l2.minarea=1000&l1.areaunit=px^2&l2.areaunit=px^2 HTTP/1.0

Parameter	Value	Description
l1.minarea	1000	Minimum polygon area threshold for layer l1
l1.areaunit	px^2	Interprets the area threshold in pixels² (screen-space) instead of m²

Using areaunit=px^2 makes the filter resolution-independent: useful when the product may be rendered at different map scales.

Output:

---

radar — Radar precipitation

Input: test/input/radar.get

GET /dali?customer=test&product=radar&time=20130910T1000 HTTP/1.0

Parameter	Value	Description
product	radar	Product JSON: test/dali/customers/test/products/radar.json
time	20130910T1000	Valid time: 2013-09-10 10:00 UTC

Renders radar-derived precipitation rate (PrecipitationRate) from the tutka_suomi_rr producer with a blue sea background and country-fill. Demonstrates isoband rendering on a 500×800 px product in the data's native CRS.

Output:

---

radar_subdivide — Bilinear cell subdivision comparison (K = 0, 2, 4, 8)

Input: test/input/radar_subdivide.get

GET /dali?customer=test&product=radar_subdivide&time=20130910T1000 HTTP/1.0

Parameter	Value	Description
product	radar_subdivide	Product JSON: test/dali/customers/test/products/radar_subdivide.json
time	20130910T1000	Valid time: 2013-09-10 10:00 UTC

Four-panel comparison over the SW Finland / Åland archipelago zoom at the same radar timestamp, each panel using a different value of the isoband layer's subdivide key. The top-left panel (K = 0) is the classic linear marching-squares output — note the sharp diamond-shaped lone-pixel polygons. With increasing K every interior level-curve segment is split into K shorter chords on the true bilinear level curve between the cell-edge intersections, turning the diamonds into progressively rounded four-lobed blobs. The cell-edge intersections themselves are identical across all four panels, which is what lets adjacent cells keep stitching together cleanly regardless of the K value. K = 8 is close to the analytical bilinear shape; K = 10 is the hard cap and produces only a sub-1 % further change.

See the subdivide row in the isoband layer reference and the Bilinear cell subdivision section of the trax docs for the underlying math and limitations.

Output:

---

tmax_smooth — Smoothed maximum temperature

Input: test/input/tmax_smooth.get

GET /dali?customer=test&product=tmax_smooth&time=200101011200 HTTP/1.0

Parameter	Value	Description
product	tmax_smooth	Product JSON: test/dali/customers/test/products/tmax_smooth.json

Applies a contouring smoother ("smoother": {"degree": 1, "size": 1}) to maximum temperature before rendering isobands; reduces jagged edges in coarse-resolution data.

Output:

---

t2m_p_smooth — Smoothed temperature

Input: test/input/t2m_p_smooth.get

GET /dali?customer=test&product=t2m_p_smooth&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	t2m_p_smooth	Product JSON: test/dali/customers/test/products/t2m_p_smooth.json

Same as t2m_p but with a stronger smoother applied to the temperature field, producing fewer and smoother contours.

Output:

---

Pressure smoothing comparison

The same European pressure field is rendered with several smoothing pipelines so the visual effect of each can be compared side by side.

Each test is a /dali?customer=test&product=pressure_europe[_variant]&time=200808050300 request. All variants share the same product structure — only the filter block on the isoband/isoline layers changes.

Test	Filter	Product JSON	Output
pressure_europe.get	None — raw isobands and isolines	pressure_europe.json
pressure_europe_bezier.get	Bezier curve fitting on the contour polylines, accuracy 5.0 px (no pre-smoothing of the field)	pressure_europe_bezier.json
pressure_europe_bezier_compare.get	Side-by-side comparison: no Bezier vs accuracy 2/3/5 px	pressure_europe_bezier_compare.json
pressure_europe_gaussian.get	Gaussian filter applied to the field before contouring	pressure_europe_gaussian.json
pressure_europe_gaussian_bezier.get	Gaussian filter on the field + Bezier fitting on the contours	pressure_europe_gaussian_bezier.json
pressure_europe_tukey.get	Tukey biweight filter (robust to outliers) on the field	pressure_europe_tukey.json

The Bezier accuracy parameter is the maximum allowed deviation (in pixels) of the fitted cubic curve from the source polyline. The minimum allowed value is 2 px — below that the moment-matching fitter starts producing extreme control points that project far outside the view. Higher accuracy values give smoother, more compact paths at the cost of (mostly imperceptible) visual fidelity. The _compare test renders the same field at four accuracy settings in a 2×2 grid for direct visual comparison.

To eliminate gaps between adjacent isobands' shared edges, the bezier code keeps a per-request cache of fitted cubics keyed by canonical-direction polyline. The cache works for whole rings (rotation- and direction-invariant via canonical-form lookup) and for sub-segments split at counter-flagged corners and at view-boundary vertices. Isolines drawn over isobands hit the same cache. Sub-pixel residual differences may remain where contour ghostlines diverge between rings, but they fall well below the chosen accuracy threshold.

The fitting algorithm is a C++ port of Raph Levien's moment-matching cubic Bezier fitter from the Rust kurbo library, which itself implements ideas from his 2009 UC Berkeley PhD thesis From Spiral to Spline: Optimal Techniques in Interactive Curve Design. The algorithm computes the signed area and first moment of the source curve segment and solves a quartic polynomial for cubic control points whose moments match. The original Rust implementation is dual-licensed Apache-2.0 / MIT.

---

Grid smoother comparison

Input: test/input/smoother_compare.get

GET /dali?customer=test&product=smoother_compare&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	smoother_compare	Product JSON: test/dali/customers/test/products/smoother_compare.json
time	200808050300	Valid time: 2008-08-05 03:00 UTC

Five panels over the same pal_skandinavia temperature field: the unsmoothed original followed by the four grid smoothers applied at a matched scale so the comparison is about each filter's character rather than how hard it happens to be pushed. The Savitzky-Golay size 3 (7×7) window sets the reference scale; box, median and morphology are tuned to remove the same spatial scale (box radius 1, passes 3; median radius 2; morphology radius 2) — note that at equal radius the rank/morphological filters would remove far more than the polynomial fit, which is what made the earlier per-method gallery misleading.

At this matched scale the differences are qualitative: Savitzky-Golay keeps the most fine detail and the strongest extrema (and costs the most); box removes the same scale of structure but, as a linear low-pass, attenuates the cold pockets the most; median gives sharp, overshoot-free edges (every output value existed in the input) and holds features broader than its window; morphology (openclose) keeps the magnitude of the broad extrema but leaves visibly blocky, axis-aligned boundaries. See the Smoother structure section of the reference for the full parameter list and the Trax grid smoother docs for the algorithms.

Output:

---

Grid smoother comparison — low cost

Input: test/input/smoother_compare_small.get

GET /dali?customer=test&product=smoother_compare_small&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	smoother_compare_small	Product JSON: test/dali/customers/test/products/smoother_compare_small.json
time	200808050300	Valid time: 2008-08-05 03:00 UTC

The companion to the comparison above, but with every method at its minimal setting (Savitzky-Golay reduced to a 5×5 window; box, median and morphology to a single 3×3-class pass). The previous example is tuned to equal smoothing quality, which makes Savitzky-Golay look like the obvious choice — but it is the slowest filter, whereas box and morphology cost O(N) per pass independent of radius and median grows with radius. This panel answers the practical question on a loaded server: which is the cheapest filter that smooths well enough? A single cheap box r1 p1 pass already removes the grid speckle nearly as well as the much more expensive Savitzky-Golay fit. Prefer box for plain smoothing, median/morphology for their edge/speckle behaviour, and reserve Savitzky-Golay for when exact extremum-height preservation is genuinely required. See the Smoother structure section of the reference for the per-method cost notes.

Output:

---

TFP — Thermal-Front Parameter diagnostics

The TFP (Thermal-Front Parameter) diagnostic highlights baroclinic zones — boundaries where temperature/humidity changes sharply. These three tests apply TFP isobands to different upper-level fields to reveal jet-stream edges, moisture boundaries, and frontal surfaces. Each test is a /dali?customer=test&product=tfp_*&type=png&time=200809101200 request. See the TFP metaparameter section in the reference for the underlying scalar field options.

Test	Field / Level	Product JSON	Output
tfp_humidity.get	Humidity at 850 hPa (moisture boundaries)	tfp_humidity.json
tfp_theta.get	Potential temperature at 850 hPa (frontal surface marker)	tfp_theta.json
tfp_wind.get	Wind speed at 300 hPa (jet-stream edges)	tfp_wind.json

All three use the ecmwf_skandinavia_painepinta pressure-level producer, render to PNG (700×500 px) over Northern Europe in Web Mercator.

---

crs_bbox — CRS bounding box handling

Input: test/input/crs_bbox.get

GET /dali?customer=test&product=crs_bbox&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	crs_bbox	Product JSON: test/dali/customers/test/products/crs_bbox.json

Tests that the bounding box specified in the product is interpreted correctly in the product's CRS. The product uses "crs": "data" with a smoother applied.

Output:

---

t2m_p_noqid — Layers without qid

Input: test/input/t2m_p_noqid.get

GET /dali?customer=test&product=t2m_p_noqid&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	t2m_p_noqid	Product JSON: test/dali/customers/test/products/t2m_p_noqid.json

Like t2m_p but the layers omit qid fields. Without a qid, layers cannot be overridden via URL parameters but the product still renders correctly.

Output:

---

Isoband Labels

Labels can be placed directly in the isoband definition with the t2m_isoband_labels test, or as a separate IsolabelLayer — see the Isolabel examples section below.

t2m_isoband_labels — Labels embedded in isoband layer

Input: test/input/t2m_isoband_labels.get

GET /dali?customer=test&product=t2m_isoband_labels&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	t2m_isoband_labels	Product JSON: test/dali/customers/test/products/t2m_isoband_labels.json

Demonstrates value labels embedded directly in the isoband layer specification. Labels are drawn along the contour lines at a configured spacing.

Output:

---

Isolabel examples

Labels along isolines using the dedicated IsolabelLayer. The placement algorithm — multi-angle local-extremum scan, divisibility-weighted MST selection, data-probed orientation — is documented in isolabel_algorithms.md. The tests below exercise each tunable in turn (angle handling, cut-through style, multi-style, wide labels, concentric/dense/crossing isolines).

isolabel — Basic isolabels

Input: test/input/isolabel.get

GET /dali?customer=test&product=isolabel&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	isolabel	Product JSON: test/dali/customers/test/products/isolabel.json

Uses the dedicated isolabel layer type to draw contour labels at a fixed spacing along isoline paths.

Output:

---

isolabel_angles — Angle-following labels

Input: test/input/isolabel_angles.get

GET /dali?customer=test&product=isolabel_angles&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	isolabel_angles	Product JSON: test/dali/customers/test/products/isolabel_angles.json

Labels are rotated to follow the tangent direction of the isoline.

Output:

---

isolabel_horizontal — Horizontal labels only

Input: test/input/isolabel_horizontal.get

GET /dali?customer=test&product=isolabel_horizontal&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	isolabel_horizontal	Product JSON: test/dali/customers/test/products/isolabel_horizontal.json

All labels are rendered horizontally regardless of isoline orientation; suitable when readability is prioritised over alignment with the contour.

Output:

---

isolabel_cut — Cut-through labels

Input: test/input/isolabel_cut.get

GET /dali?customer=test&product=isolabel_cut&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	isolabel_cut	Product JSON: test/dali/customers/test/products/isolabel_cut.json

The isoline is cut at each label position so the label appears embedded in the line rather than on top of it.

Output:

---

isolabel_styles — Multiple label styles

Input: test/input/isolabel_styles.get

GET /dali?customer=test&product=isolabel_styles&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	isolabel_styles	Product JSON: test/dali/customers/test/products/isolabel_styles.json

Demonstrates per-isoline label styling: each contour level can have its own font size, colour, and background.

Output:

---

isolabel_wide — Wide formatted labels

Input: test/input/isolabel_wide.get

GET /dali?customer=test&product=isolabel_wide&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	isolabel_wide	Product JSON: test/dali/customers/test/products/isolabel_wide.json

Sets label.precision: 1 and label.suffix: " hPa" so labels render as 1015.0 hPa. Each candidate's collision box is sized from the actual rendered text, so wider labels reserve more space and the MST drops nearby competitors that would have fit at the default width.

Output:

---

isolabel_concentric — Concentric closed contours

Input: test/input/isolabel_concentric.get

GET /dali?customer=test&product=isolabel_concentric&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	isolabel_concentric	Product JSON: test/dali/customers/test/products/isolabel_concentric.json

Tightens the spacing thresholds (min_distance_self: 80, min_distance_same: 30, min_distance_other: 15) so closely nested closed pressure rings can each carry a label. The MST keeps the ladder placement readable instead of collapsing labels into clusters.

Output:

---

isolabel_crossing — Two parameters labelled together

Input: test/input/isolabel_crossing.get

GET /dali?customer=test&product=isolabel_crossing&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	isolabel_crossing	Product JSON: test/dali/customers/test/products/isolabel_crossing.json

Labels two parameters at once — pressure plus temperature. The SAT-based overlap test catches places where the two label sets cross at large angles; without SAT a corner-to-corner-only test would miss those collisions.

Output:

---

isolabel_dense — Dense isovalue grid

Input: test/input/isolabel_dense.get

GET /dali?customer=test&product=isolabel_dense&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	isolabel_dense	Product JSON: test/dali/customers/test/products/isolabel_dense.json

Generates pressure isolines at every 1 hPa via isovalues: { start: 990, stop: 1010, step: 1 }. Every divisibility class — multiples of 10, 5, 2, and others — competes for placement; the divisibility weighting keeps the round numbers visible while the rest fill in only where there is room.

Output:

---

Location Label Placement

The LocationLayer places labels next to city markers using one of several algorithms. The full algorithm reference is in docs/labeling_algorithms.md; the tests below exercise each placement strategy and the cross-cutting options (free-space bias, population bucketing, pan invariance).

location_labels_fixed — Fixed NE position

Input: test/input/location_labels_fixed.get

GET /dali?customer=test&product=location_labels_fixed&type=svg HTTP/1.0

Parameter	Value	Description
product	location_labels_fixed	Product JSON: test/dali/customers/test/products/location_labels_fixed.json

Every label is placed at the NE corner with no conflict avoidance. Labels that would cover a higher-priority marker are dropped (and so is their own marker), but labels are allowed to overlap each other.

Output:

---

location_labels_greedy — Greedy placement

Input: test/input/location_labels_greedy.get

GET /dali?customer=test&product=location_labels_greedy&type=svg HTTP/1.0

Parameter	Value	Description
product	location_labels_greedy	Product JSON: test/dali/customers/test/products/location_labels_greedy.json

For each candidate (in geonames priority order) the algorithm tries the eight Imhof positions in priority order and keeps the first one that fits without overlapping a previously placed label or another candidate's marker. Recommended default for most weather maps.

Output:

---

location_labels_priority_greedy — Population-sorted greedy

Input: test/input/location_labels_priority_greedy.get

GET /dali?customer=test&product=location_labels_priority_greedy&type=svg HTTP/1.0

Parameter	Value	Description
product	location_labels_priority_greedy	Product JSON: test/dali/customers/test/products/location_labels_priority_greedy.json

Same as greedy, but candidates are pre-sorted by population descending before placement. Use this when geonames autocomplete order does not already match the population ranking you want.

Output:

---

location_labels_priority_greedy_bucketed — Population buckets + shorter-name tiebreak

Input: test/input/location_labels_priority_greedy_bucketed.get

GET /dali?customer=test&product=location_labels_priority_greedy_bucketed&type=svg HTTP/1.0

Parameter	Value	Description
product	location_labels_priority_greedy_bucketed	Product JSON: test/dali/customers/test/products/location_labels_priority_greedy_bucketed.json

Adds priority_bucket_ratio: 2.0 so cities with similar populations land in the same log-bucket and the shorter label wins ties. On dense maps this typically places more labels overall — short names slot into prime NE positions and free up room for neighbours.

Output:

---

location_labels_sa — Simulated annealing

Input: test/input/location_labels_sa.get

GET /dali?customer=test&product=location_labels_sa&type=svg HTTP/1.0

Parameter	Value	Description
product	location_labels_sa	Product JSON: test/dali/customers/test/products/location_labels_sa.json

Uses Christensen-Marks-Shieber simulated annealing to minimise total label-overlap energy. Converges toward a near-global optimum that pure greedy can miss in dense urban areas where several labels need to coordinate to fit. The random seed is fixed so results are deterministic.

Output:

---

location_labels_freespace — Greedy with free-space bias

Input: test/input/location_labels_freespace.get

GET /dali?customer=test&product=location_labels_freespace&type=svg HTTP/1.0

Parameter	Value	Description
product	location_labels_freespace	Product JSON: test/dali/customers/test/products/location_labels_freespace.json

Adds free_space_weight: 1.5, free_space_radius: 120 to the greedy run. Each candidate scores positions by alignment with the direction away from neighbouring markers and labels, so coastal cities push their labels out into the empty water without needing a sea mask.

Output:

---

location_labels_sa_freespace — Simulated annealing with free-space bias

Input: test/input/location_labels_sa_freespace.get

GET /dali?customer=test&product=location_labels_sa_freespace&type=svg HTTP/1.0

Parameter	Value	Description
product	location_labels_sa_freespace	Product JSON: test/dali/customers/test/products/location_labels_sa_freespace.json

The same free-space bias on top of SA's overlap optimisation: the bias term is added to the position penalty and recomputed on every move so the bias reflects the current assignment.

Output:

---

location_labels_pan_invariant — Pan-invariant greedy

Input: test/input/location_labels_pan_invariant.get

GET /dali?customer=test&product=location_labels_pan_invariant&type=svg HTTP/1.0

Parameter	Value	Description
product	location_labels_pan_invariant	Product JSON: test/dali/customers/test/products/location_labels_pan_invariant.json

Sets pan_invariant: true and centres on Jyväskylä. Candidates are filtered to a margin-buffered bbox so cities just beyond the visible edge still influence the placement of visible cities; anchors outside the original bbox are skipped at render time.

Output:

---

location_labels_pan_invariant_shifted — Same view, panned 0.5° east

Input: test/input/location_labels_pan_invariant_shifted.get

GET /dali?customer=test&product=location_labels_pan_invariant_shifted&type=svg HTTP/1.0

Parameter	Value	Description
product	location_labels_pan_invariant_shifted	Product JSON: test/dali/customers/test/products/location_labels_pan_invariant_shifted.json

Same projection as location_labels_pan_invariant but with cx shifted by +0.5°. Cities visible in both panned views land at identical offsets from their markers — pan invariance verified per-city by the test harness.

Output:

---

location_labels_country_off / _on — Country constraint

This pair demonstrates the country_constraint option, zoomed to the Finland–Russia border around Lappeenranta / Imatra (Finland) and Viipuri / Vyborg (Russia). Finland is tinted blue and Russia red so it is obvious which side of the border each label lands on. The keyword ajax_fi_all is a worldwide place set, so non-Finnish names (e.g. Viipuri) appear alongside the Finnish towns. Both products are identical except for the single "country_constraint": true flag.

Input: test/input/location_labels_country_off.get · test/input/location_labels_country_on.get

GET /dali?customer=test&product=location_labels_country_off&type=svg HTTP/1.0
GET /dali?customer=test&product=location_labels_country_on&type=svg HTTP/1.0

Parameter	Value	Description
product	location_labels_country_off	Product JSON: test/dali/customers/test/products/location_labels_country_off.json — greedy placement, no constraint
product	location_labels_country_on	Product JSON: test/dali/customers/test/products/location_labels_country_on.json — same, with "country_constraint": true

Watch the border-hugging towns Nuijamaa, Vainikkala, Hiivaniemi and Ahola. Without the constraint their labels lean east, across the border into Russia (red). With the constraint enabled every label is forced to stay within its own country's polygon, so those labels flip west into Finland (blue). Towns far from the border — Viipuri (RU), Lappeenranta and Imatra (FI) — are unaffected: the constraint only intervenes where a label would otherwise cross a border.

Output — without country_constraint (left) vs with it (right):

country_constraint: off

country_constraint: on

---

SVG Customisation

t2m_p_shadow — SVG drop shadow

Input: test/input/t2m_p_shadow.get

GET /dali?customer=test&product=t2m_p_shadow&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	t2m_p_shadow	Product JSON: test/dali/customers/test/products/t2m_p_shadow.json

Adds an SVG <filter> drop-shadow (feDropShadow) to the product's root element via the top-level "attributes": {"filter": "url(#shadow)"} setting.

Output:

---

t2m_p_display_none — Hidden layer with optimizesize

Input: test/input/t2m_p_display_none.get

GET /dali?customer=test&product=t2m_p&time=200808050300&l3.attributes.display=none&optimizesize=1 HTTP/1.0

Parameter	Value	Description
l3.attributes.display	none	Sets display=none on layer l3 via URL override, making it invisible
optimizesize	1	When set, layers with display=none are omitted from the output entirely (reducing file size)

The optimizesize flag strips out hidden layers rather than leaving them as invisible elements in the SVG.

Output:

---

t2m_pattern — SVG pattern fill

Input: test/input/t2m_pattern.get

GET /dali?customer=test&product=t2m_pattern&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	t2m_pattern	Product JSON: test/dali/customers/test/products/t2m_pattern.json

Uses SVG <pattern> elements defined in the defs section to fill isoband polygons with a hatching or repeat pattern rather than a solid colour.

Output:

---

defs_circle — SVG defs with reusable shapes

Input: test/input/defs_circle.get

GET /dali?customer=test&product=defs_circle&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	defs_circle	Product JSON: test/dali/customers/test/products/defs_circle.json

Shows how defs.layers can define reusable SVG elements (filters, named circles, clip paths, text nodes) that are referenced later via xlink:href.

Output:

---

clip — Explicit clipPath layer

Input: test/input/clip.get

GET /dali?customer=test&product=clip&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	clip	Product JSON: test/dali/customers/test/products/clip.json

Uses a tag: "clipPath" layer containing a tag: "circle" to create an SVG <clipPath> element inline. A tag: "g" wrapper layer with clip-path: url(#mycircle) then applies it to the isoband layers, while the product-level filter: url(#shadow) adds a drop shadow.

Output:

---

circle — Circular clip with isobands

Input: test/input/circle.get

GET /dali?customer=test&product=circle&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	circle	Product JSON: test/dali/customers/test/products/circle.json
time	200808050300	Valid time: 2008-08-05 03:00 UTC

Projects Scandinavian temperature data using the data's own native CRS, renders temperature isobands and isolines, and clips the output with an SVG <circle> element centred at (250, 250) with radius 250.

Output:

---

Clipping to Geographic Boundaries

t2m_inside_finland — Clip to country polygon

Input: test/input/t2m_inside_finland.get

GET /dali?customer=test&product=t2m_inside_finland&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	t2m_inside_finland	Product JSON: test/dali/customers/test/products/t2m_inside_finland.json

Uses "inside" on the isoband layer referencing a PostGIS query for the Finland border (iso_a2='FI') to clip the rendered isobands to inside Finland.

Output:

---

t2m_inside_simplified_finland — Clip to simplified country polygon

Input: test/input/t2m_inside_simplified_finland.get

GET /dali?customer=test&product=t2m_inside_simplified_finland&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	t2m_inside_simplified_finland	Product JSON: test/dali/customers/test/products/t2m_inside_simplified_finland.json

Same as above but uses a mindistance tolerance to simplify the clip polygon, producing a smoother (less precise) boundary and faster rendering.

Output:

---

t2m_outside_finland — Clip to outside country

Input: test/input/t2m_outside_finland.get

GET /dali?customer=test&product=t2m_outside_finland&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	t2m_outside_finland	Product JSON: test/dali/customers/test/products/t2m_outside_finland.json

Uses "outside" to clip isobands to the area outside Finland — the inverse of t2m_inside_finland.

Output:

---

t2m_inside_rain — Clip isobands to rain area

Input: test/input/t2m_inside_rain.get

GET /dali?customer=test&product=t2m_inside_rain&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	t2m_inside_rain	Product JSON: test/dali/customers/test/products/t2m_inside_rain.json

Clips temperature isobands to areas where precipitation exceeds 0 mm/h (the rain area). The inside polygon is derived dynamically from another data layer rather than from a static geographic boundary.

Output:

---

t2m_lines_inside_rain — Isolines clipped to rain area

Input: test/input/t2m_lines_inside_rain.get

GET /dali?customer=test&product=t2m_lines_inside_rain&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	t2m_lines_inside_rain	Product JSON: test/dali/customers/test/products/t2m_lines_inside_rain.json

Temperature isolines (not isobands) are clipped to the rain area. Shows that the inside clipping mechanism works equally for line layers.

Output:

---

frost_inside_simplified_finland — Frost probability inside Finland

Input: test/input/frost_inside_simplified_finland.get

GET /dali?customer=test&product=frost_inside_simplified_finland&time=201604140000 HTTP/1.0

Parameter	Value	Description
product	frost_inside_simplified_finland	Product JSON: test/dali/customers/test/products/frost_inside_simplified_finland.json
time	201604140000	Valid time: 2016-04-14 00:00 UTC

Renders FrostProbability isobands from the ground producer at 1200×1200 px in EPSG:4326. Uses "inside": "ref:refs.finland" where the refs.finland entry queries the Natural Earth countries table with iso_a2 IN ('FI','AX') and a mindistance simplification tolerance.

Output:

---

Pressure Level and Elevation Selection

t2m_level_* — Named pressure levels

Input files: test/input/t2m_level_1000.get, test/input/t2m_level_850.get, test/input/t2m_level_300.get

GET /dali?customer=test&product=t2m_pressurelevel&time=20080909T1200&level=1000 HTTP/1.0
GET /dali?customer=test&product=t2m_pressurelevel&time=20080909T1200&level=850  HTTP/1.0
GET /dali?customer=test&product=t2m_pressurelevel&time=20080909T1200&level=300  HTTP/1.0

Parameter	Value	Description
product	t2m_pressurelevel	Product JSON: test/dali/customers/test/products/t2m_pressurelevel.json
time	20080909T1200	Valid time: 2008-09-09 12:00 UTC
level	1000 / 850 / 300	Selects the pressure level in hPa from the ecmwf_skandinavia_painepinta producer

The level parameter selects a pressure level slice; the producer must have data for that level.

Level	Output
1000 hPa
850 hPa
300 hPa

---

t2m_elevation_* — Elevation in metres

Input files: test/input/t2m_elevation_1000.get, test/input/t2m_elevation_850.get, test/input/t2m_elevation_300.get

GET /dali?customer=test&product=t2m_pressurelevel&time=20080909T1200&elevation=1000 HTTP/1.0
GET /dali?customer=test&product=t2m_pressurelevel&time=20080909T1200&elevation=850  HTTP/1.0
GET /dali?customer=test&product=t2m_pressurelevel&time=20080909T1200&elevation=300  HTTP/1.0

Parameter	Value	Description
elevation	1000 / 850 / 300	Selects the level using the elevation parameter (in metres or hPa depending on data convention)

The elevation parameter is an alternative to level; which one is applicable depends on the data source's level type convention.

Elevation	Output
1000 m
850 m
300 m

---

high_resolution — High resolution rendering

Input: test/input/high_resolution.get

GET /dali?customer=test&product=high_resolution&time=202006051200 HTTP/1.0

Parameter	Value	Description
product	high_resolution	Product JSON: test/dali/customers/test/products/high_resolution.json
time	202006051200	Valid time: 2020-06-05 12:00 UTC

Tests rendering of a high-resolution model grid (e.g. HARMONIE/AROME). Listed in .testignore as a debugging test case with non-deterministic output in CI.

No reference output image available.

---

Multiple Views and Side-by-Side Products

t2m_twice — Two views side by side

Input: test/input/t2m_twice.get

GET /dali?customer=test&product=t2m_twice&time=200808050300&v2.time_offset=1440 HTTP/1.0

Parameter	Value	Description
product	t2m_twice	Product JSON: test/dali/customers/test/products/t2m_twice.json
v2.time_offset	`1440$	\text{Advances} \text{the} \text{valid} \text{time} \text{of} \text{view} 2 \text{by} 1440 \text{minutes} (24 \text{hours})

\text{The} \text{product} \text{defines} \text{two} 500 \times 500 \text{px} \text{views} ($v1, v2) side by side in a 1030×520 canvas. Each view includes temperature isobands, isolines, and a time label. The URL overrides v2.time_offset` to show T+24 h in the second panel.

Output:

---

t2m_twice_altered — Transformed second view

Input: test/input/t2m_twice_altered.get

GET /dali?customer=test&product=t2m_twice&time=200808050300
    &v2.time_offset=1440
    &v2.attributes.transform=translate%28500,1%29+rotate%2830%29+scale%280.75%29
    &v1.attributes.filter=url%28#shadow%29
    &v2.attributes.filter=url%28#shadow%29 HTTP/1.0

Parameter	Value	Description
v2.attributes.transform	translate(500,1) rotate(30) scale(0.75)	Applies an SVG transform to view 2: shifted, rotated 30°, scaled 75%
v1.attributes.filter	url(#shadow)	Adds a drop-shadow filter to view 1
v2.attributes.filter	url(#shadow)	Adds a drop-shadow filter to view 2

Demonstrates that SVG attributes (including transform and filter) can be set on any view via URL overrides. The URL-encoded values %28 = (, %29 = ).

Output:

---

t2m_twice_margins — Views with margins (unclipped)

Input: test/input/t2m_twice_margins.get

GET /dali?customer=test&product=t2m_twice&time=200808050300
    &v2.time_offset=1440
    &v1.clip=0&v2.clip=0
    &v1.margin=5&v2.margin=10
    &projection.place=Helsinki&projection.resolution=1 HTTP/1.0

Parameter	Value	Description
v1.clip	0	Disables clipping for view 1 (data may extend beyond the view boundary)
v2.clip	0	Disables clipping for view 2
v1.margin	5	Adds a 5-pixel margin around view 1's data extent
v2.margin	10	Adds a 10-pixel margin around view 2's data extent
projection.place	Helsinki	Centres the projection on Helsinki
projection.resolution	1	Sets projection resolution to 1 km/px

Output:

---

t2m_twice_margins_clipped — Views with margins (clipped)

Input: test/input/t2m_twice_margins_clipped.get

GET /dali?customer=test&product=t2m_twice&time=200808050300
    &v2.time_offset=1440
    &v1.clip=1&v2.clip=1
    &v1.margin=5&v2.margin=10
    &projection.place=Helsinki&projection.resolution=1 HTTP/1.0

Parameter	Value	Description
v1.clip	1	Enables clipping for view 1
v2.clip	1	Enables clipping for view 2

Same as t2m_twice_margins but clipping is re-enabled. Layers are clipped to the view boundary even though a margin is specified.

Output:

---

resolution — View-level projection resolution

Input: test/input/resolution.get

GET /dali?customer=test&product=resolution&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	resolution	Product JSON: test/dali/customers/test/products/resolution.json

Two 500×500 views centred on Finland are placed side by side, each with a different resolution setting in their projection block (2.5 and 1.5 km/px). Map layers use minresolution and maxresolution to select detail levels: coarse data is shown in black for coarse resolutions, fine data in red for fine resolutions.

Output:

---

ely_overlay — Multi-customer overlay

Input: test/input/ely_overlay.get

GET /dali?customer=ely&product=temperatureoverlay&time=200808080000&type=svg HTTP/1.1

Parameter	Value	Description
customer	ely	Selects the ely customer directory (separate from test)
product	temperatureoverlay	Product JSON: test/dali/customers/ely/products/temperatureoverlay.json
type	svg	SVG output

A second customer (ely) demonstrating that multiple customer configurations can coexist. This overlay product uses a number layer with a masked legend background.

Output:

---

Time Layer and Timezone

timelayer — Time annotation

Input: test/input/timelayer.get

GET /dali?customer=test&product=timelayer&time=200808051200 HTTP/1.0

Parameter	Value	Description
product	timelayer	Product JSON: test/dali/customers/test/products/timelayer.json
time	200808051200	Valid time: 2008-08-05 12:00 UTC

The time layer renders the product's valid time as formatted text (e.g. "%Y-%m-%d %H:%M") at a specified position in the SVG. The product uses "timestamp": "validtime" and "timezone": "UTC".

Output:

---

timezone — Time in local timezone

Input: test/input/timezone.get

GET /dali?customer=test&product=timelayer&time=200808051500&tz=Europe/Helsinki HTTP/1.0

Parameter	Value	Description
time	200808051500	Valid time: 2008-08-05 15:00 UTC
tz	Europe/Helsinki	Overrides the display timezone for the time layer; 15:00 UTC → 18:00 EEST

The tz URL parameter overrides the product's timezone for all time layer annotations in the response.

Output:

---

Observation Numbers

temperature_fmisid — Station temperatures by FMI ID

Input: test/input/temperature_fmisid.get

GET /dali?customer=test&product=temperature_fmisid&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	temperature_fmisid	Product JSON: test/dali/customers/test/products/temperature_fmisid.json
time	200808050300	Valid time: 2008-08-05 03:00 UTC

Plots temperature observations at a hand-picked list of FMI station IDs (fmisid). Individual stations may override the default label offset (dx, dy). Background: white fill + Natural Earth country borders.

Output:

---

opendata_temperature_fmisid — Opendata station temperatures

Input: test/input/opendata_temperature_fmisid.get

GET /dali?customer=test&product=opendata_temperature_fmisid&time=20130805T1500 HTTP/1.0

Parameter	Value	Description
product	opendata_temperature_fmisid	Product JSON: test/dali/customers/test/products/opendata_temperature_fmisid.json
time	20130805T1500	Valid time: 2013-08-05 15:00 UTC (ISO 8601 compact format)

Same concept as temperature_fmisid but using the opendata producer from the Finnish Meteorological Institute.

Output:

---

opendata_temperature_fmisid_shift — Station numbers with label offset

Input: test/input/opendata_temperature_fmisid_shift.get

GET /dali?customer=test&product=opendata_temperature_fmisid_shift&time=20130805T1500 HTTP/1.0

Parameter	Value	Description
product	opendata_temperature_fmisid_shift	Product JSON: test/dali/customers/test/products/opendata_temperature_fmisid_shift.json

Station labels are systematically shifted (dx, dy) to reduce overlap with symbols.

Output:

---

opendata_temperature_grid — Grid-spaced observation numbers

Input: test/input/opendata_temperature_grid.get

GET /dali?customer=test&product=opendata_temperature_grid&time=20130805T1500 HTTP/1.0

Parameter	Value	Description
product	opendata_temperature_grid	Product JSON: test/dali/customers/test/products/opendata_temperature_grid.json

Uses "positions": {"layout": "grid", "dx": 30, "dy": 30} to place observation values at regular pixel intervals, thinning dense station networks.

Output:

---

opendata_temperature_keyword — Keyword-selected stations

Input: test/input/opendata_temperature_keyword.get

GET /dali?customer=test&product=opendata_temperature_keyword&time=20130805T1500 HTTP/1.0

Parameter	Value	Description
product	opendata_temperature_keyword	Product JSON: test/dali/customers/test/products/opendata_temperature_keyword.json

Selects stations by keyword (administrative classification such as county or municipality) rather than by explicit FMI station IDs.

Output:

---

ecmwf_data_numbers — ECMWF grid point values

Input: test/input/ecmwf_data_numbers.get

GET /dali?customer=test&product=ecmwf_data_numbers&time=201503131200 HTTP/1.0

Parameter	Value	Description
product	ecmwf_data_numbers	Product JSON: test/dali/customers/test/products/ecmwf_data_numbers.json
time	201503131200	Valid time: 2015-03-13 12:00 UTC

Plots temperature values at regular grid intervals from the ECMWF surface analysis (ecmwf_maailma_pinta) over Europe.

Output:

---

ecmwfpoint_data_numbers — ECMWF point values

Input: test/input/ecmwfpoint_data_numbers.get

GET /dali?customer=test&product=ecmwfpoint_data_numbers&time=200808051200 HTTP/1.0

Parameter	Value	Description
product	ecmwfpoint_data_numbers	Product JSON: test/dali/customers/test/products/ecmwfpoint_data_numbers.json
time	200808051200	Valid time: 2008-08-05 12:00 UTC

Plots ECMWF model values at specific geographic point locations rather than a regular grid.

Output:

---

ecmwf_world_numbers — ECMWF global grid values

Input: test/input/ecmwf_world_numbers.get

GET /dali?customer=test&product=ecmwf_world_numbers&time=201503131200 HTTP/1.0

Parameter	Value	Description
product	ecmwf_world_numbers	Product JSON: test/dali/customers/test/products/ecmwf_world_numbers.json

Plots ECMWF global surface temperature values at grid intervals on a world map.

Output:

---

aviation_numbers — Aviation temperature numbers

Input: test/input/aviation_numbers.get

GET /dali?customer=test&product=aviation_numbers&time=201503131200 HTTP/1.0

Parameter	Value	Description
product	aviation_numbers	Product JSON: test/dali/customers/test/products/aviation_numbers.json
time	201503131200	Valid time: 2015-03-13 12:00 UTC

Renders ECMWF temperature isobands over the North Atlantic aviation area (EPSG:4326, 62.5–75°N, 15°W–37.5°E) and overlays integer temperature values at a 20×20 px grid with alternating 10-pixel row offsets (ddx).

Output:

---

aviation_numbers_multiple5 — Rounded-to-5 numbers

Input: test/input/aviation_numbers_multiple5.get

GET /dali?customer=test&product=aviation_numbers_multiple5&time=201503131200 HTTP/1.0

Parameter	Value	Description
product	aviation_numbers_multiple5	Product JSON: test/dali/customers/test/products/aviation_numbers_multiple5.json

Like aviation_numbers but values are rounded to the nearest 5 (ICAO convention: temperatures reported in multiples of 5°C on SIGWX charts).

Output:

---

synop_numbers — SYNOP observation values

Input: test/input/synop_numbers.get

GET /dali?customer=test&product=synop_numbers&time=20081201T1200 HTTP/1.0

Parameter	Value	Description
product	synop_numbers	Product JSON: test/dali/customers/test/products/synop_numbers.json
time	20081201T1200	Valid time: 2008-12-01 12:00 UTC

Plots SYNOP observation values from the pal_skandinavia producer at station locations.

Output:

---

monitoring — Health monitoring text

Input: test/input/monitoring.get

GET /dali?customer=test&product=monitoring HTTP/1.0

Parameter	Value	Description
product	monitoring	Product JSON: test/dali/customers/test/products/monitoring.json

A minimal product that outputs a plain-text "OK" health-check response without requiring any weather data. Used for liveness/readiness probes.

Output:

---

Observation Positions and Keyword Selection

latlon_positions — Lat/lon coordinate list

Input: test/input/latlon_positions.get

GET /dali?customer=test&product=latlon_positions&time=201503131200 HTTP/1.0

Parameter	Value	Description
product	latlon_positions	Product JSON: test/dali/customers/test/products/latlon_positions.json

Plots observation values at positions specified as explicit latitude/longitude coordinate pairs in the product JSON.

Output:

---

latlon_positions_shifted — Positions with per-station label offset

Input: test/input/latlon_positions_shifted.get

GET /dali?customer=test&product=latlon_positions_shifted&time=201503131200 HTTP/1.0

Parameter	Value	Description
product	latlon_positions_shifted	Product JSON: test/dali/customers/test/products/latlon_positions_shifted.json

Each coordinate entry includes an individual dx/dy offset for its label, allowing crowded stations to be legible without overlapping.

Output:

---

keyword — Keyword-based station selection

Input: test/input/keyword.get

GET /dali?customer=test&product=keyword&time=201503131200 HTTP/1.0

Parameter	Value	Description
product	keyword	Product JSON: test/dali/customers/test/products/keyword.json

Selects stations using a keyword field (e.g. a municipality or region name) instead of numeric IDs. The server resolves the keyword to matching stations.

Output:

---

keyword_shifted — Keyword stations with shifted labels

Input: test/input/keyword_shifted.get

GET /dali?customer=test&product=keyword_shifted&time=201503131200 HTTP/1.0

Parameter	Value	Description
product	keyword_shifted	Product JSON: test/dali/customers/test/products/keyword_shifted.json

Same as keyword but with per-station label offsets (dx, dy) defined for each keyword entry.

Output:

---

Weather Symbols

opendata_temperature_symbols — SmartMet symbols

Input: test/input/opendata_temperature_symbols.get

GET /dali?customer=test&product=opendata_temperature_symbols&time=20130805T1500 HTTP/1.0

Parameter	Value	Description
product	opendata_temperature_symbols	Product JSON: test/dali/customers/test/products/opendata_temperature_symbols.json
time	20130805T1500	Valid time: 2013-08-05 15:00 UTC

Plots SmartMet weather symbols at station locations combined with temperature numbers.

Output:

---

opendata_temperature_symbols_oddmitime — Odd-minute valid time

Input: test/input/opendata_temperature_symbols_oddmitime.get

GET /dali?customer=test&product=opendata_temperature_symbols&time=20130805T1502 HTTP/1.0

Parameter	Value	Description
time	20130805T1502	Valid time: 15:02 UTC — not on the hour; tests nearest-time matching

Tests that the system correctly snaps to the nearest available observation when the requested time is between observation steps.

Output:

---

opendata_temperature_symbols_oddmitime_interval_back — Time interval backward

Input: test/input/opendata_temperature_symbols_oddmitime_interval_back.get

GET /dali?customer=test&product=opendata_temperature_symbols&time=20130805T1502&interval_start=2 HTTP/1.0

Parameter	Value	Description
time	20130805T1502	Non-round valid time
interval_start	2	Accept data up to 2 minutes before the requested time

The interval_start parameter widens the time-matching window backward, useful for observations that arrive with a slight delay.

Output:

---

opendata_temperature_symbols_oddmitime_interval_forward — Time interval forward

Input: test/input/opendata_temperature_symbols_oddmitime_interval_forward.get

GET /dali?customer=test&product=opendata_temperature_symbols&time=20130805T1455&interval_end=5 HTTP/1.0

Parameter	Value	Description
time	20130805T1455	Non-round valid time: 5 minutes before 15:00
interval_end	5	Accept data up to 5 minutes after the requested time

The interval_end parameter widens the window forward, allowing a request before the hour to still display the on-the-hour data.

Output:

---

smartsymbol — SmartMet numeric symbol codes

Input: test/input/smartsymbol.get

GET /dali?customer=test&product=smartsymbol&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	smartsymbol	Product JSON: test/dali/customers/test/products/smartsymbol.json

Renders the SmartSymbol parameter (integer code) as SVG symbol graphics at station locations. Each code maps to an SVG symbol ID loaded from the symbol set.

Output:

---

smartsymbolnumber — Symbol codes as numbers

Input: test/input/smartsymbolnumber.get

GET /dali?customer=test&product=smartsymbolnumber&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	smartsymbolnumber	Product JSON: test/dali/customers/test/products/smartsymbolnumber.json

Plots the numeric SmartSymbol code as a text label (for debugging).

Output:

---

weather — Standard weather symbols

Input: test/input/weather.get

GET /dali?customer=test&product=weather&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	weather	Product JSON: test/dali/customers/test/products/weather.json

Renders a set of weather symbols loaded from SVG symbol files. The product combines a map background with symbol layer.

Output:

---

weather-cssdef — Inline CSS override via URL

Input: test/input/weather-cssdef.get

GET /dali?customer=test&product=weather&time=200808050300
    &defs.css.test%2Fsymbols%2Fdot=data:,.Dot%20%7B%20fill%3A%20yellow%3B%20stroke%3A%20none%20%7D%0A HTTP/1.0

Parameter	Value (URL-decoded)	Description
defs.css.test/symbols/dot	data:,.Dot { fill: yellow; stroke: none }	Inlines a CSS rule for the dot symbol class via a data-URL; overrides the file-based CSS

The defs.css.{name}=data:,... pattern allows injecting arbitrary CSS rules through the URL without modifying product files.

Output:

---

weather-symbol-dataurl-get — Inline SVG symbol override

Input: test/input/weather-symbol-dataurl-get.get

GET /dali?customer=test&product=weather-base64-symbol&time=200808050300
    &defs.symbols.weather%2F1_II%2F1="data:;base64,PHN5bWJvbC..." HTTP/1.0

Parameter	Value (URL-decoded)	Description
defs.symbols.weather/1_II/1	data:;base64,<base64-SVG>	Replaces the SVG symbol weather/1_II/1 with an inline base64-encoded SVG symbol (an orange circle)

Allows runtime replacement of individual symbols through the URL, e.g. for A/B testing or custom client-specific iconography.

Output:

---

weather-base64json-symbol — Base64 symbol from JSON

Input: test/input/weather-base64json-symbol.get

GET /dali?customer=test&product=weather-base64-symbol&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	weather-base64-symbol	Product JSON: test/dali/customers/test/products/weather-base64-symbol.json

The product JSON itself specifies symbols using data:;base64,… data URLs (base64-encoded SVG), bypassing the filesystem symbol store. Tests that the server can decode and embed inline symbols from the product definition.

Output:

---

symbols_coloured — Weather symbols in multiple projections

The same symbols_coloured product is rendered in four different projections using the projection.crs URL override:

Inputs:

• test/input/symbols_coloured_native.get — data native CRS
• test/input/symbols_coloured_fin.get — projection.crs=EPSG:3067
• test/input/symbols_coloured_webmercator.get — projection.crs=EPSG:3857
• test/input/symbols_coloured_wgs84.get — projection.crs=WGS84

Parameter	Value	Description
product	symbols_coloured	Product JSON: test/dali/customers/test/products/symbols_coloured.json
projection.crs	(varies)	Overrides the product's CRS on-the-fly

Demonstrates that projection.crs can reproject any product at request time without modifying the JSON.

Projection	Output
Native CRS
EPSG:3067 (ETRS-TM35FIN)
EPSG:3857 (WebMercator)
WGS84

---

Flash and Lightning Symbols

flash_symbols — Combined flash symbols

Input: test/input/flash_symbols.get

GET /dali?customer=test&product=flash_symbols&time=20130805T1500 HTTP/1.0

Parameter	Value	Description
product	flash_symbols	Product JSON: test/dali/customers/test/products/flash_symbols.json
time	20130805T1500	Valid time: 2013-08-05 15:00 UTC

Plots all flash (lightning) strike locations (cloud-to-ground and cloud-to-cloud) as symbols from the flash producer.

Output:

---

flash_cloud_symbols — Cloud flash only

Input: test/input/flash_cloud_symbols.get

GET /dali?customer=test&product=flash_cloud_symbols&time=20130805T1500 HTTP/1.0

Parameter	Value	Description
product	flash_cloud_symbols	Product JSON: test/dali/customers/test/products/flash_cloud_symbols.json

Plots only intra-cloud (IC) lightning flashes, filtering out cloud-to-ground (CG) strikes.

Output:

---

flash_ground_symbols — Ground flash only

Input: test/input/flash_ground_symbols.get

GET /dali?customer=test&product=flash_ground_symbols&time=20130805T1500 HTTP/1.0

Parameter	Value	Description
product	flash_ground_symbols	Product JSON: test/dali/customers/test/products/flash_ground_symbols.json

Plots only cloud-to-ground (CG) lightning strikes.

Output:

---

flash_cloud_and_ground_symbols — Separate symbol types

Input: test/input/flash_cloud_and_ground_symbols.get

GET /dali?customer=test&product=flash_cloud_and_ground_symbols&time=20130805T1500 HTTP/1.0

Parameter	Value	Description
product	flash_cloud_and_ground_symbols	Product JSON: test/dali/customers/test/products/flash_cloud_and_ground_symbols.json

Uses two symbol layers (one for IC, one for CG) with distinct symbol shapes and colours in a single product.

Output:

---

Wind Arrows

wind — Basic wind arrows

Input: test/input/wind.get

GET /dali?customer=test&product=wind&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	wind	Product JSON: test/dali/customers/test/products/wind.json

Renders wind direction arrows from pal_skandinavia using "direction" and "speed" parameters. Arrow length scales with wind speed.

Output:

---

wind_margin — Wind arrows with map margin

Input: test/input/wind_margin.get

GET /dali?customer=test&product=wind_margin&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	wind_margin	Product JSON: test/dali/customers/test/products/wind_margin.json

Same wind arrows but with a map area margin setting that extends the data area outside the nominal bounding box.

Output:

---

wind_minrotationspeed — Arrow rotation threshold

Input: test/input/wind_minrotationspeed.get

GET /dali?customer=test&product=wind&time=200808050300&l3.minrotationspeed=10 HTTP/1.0

Parameter	Value	Description
l3.minrotationspeed	10	Arrows for layer l3 are only rotated (to point in the wind direction) when wind speed exceeds 10 m/s; calmer arrows remain pointing north

The minrotationspeed threshold prevents cluttered arrow directions at very low wind speeds.

Output:

---

wind_scaled — Scaled wind arrows

Input: test/input/wind_scaled.get

GET /dali?customer=test&product=wind_scaled&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	wind_scaled	Product JSON: test/dali/customers/test/products/wind_scaled.json

Uses an explicit "scale" factor to map wind speed to arrow length.

Output:

---

wind_without_speed — Direction-only arrows

Input: test/input/wind_without_speed.get

GET /dali?customer=test&product=wind_without_speed&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	wind_without_speed	Product JSON: test/dali/customers/test/products/wind_without_speed.json

Arrows show direction only; all arrows are rendered at the same fixed length since no speed parameter is provided.

Output:

---

hirlam_wind — HIRLAM wind (direction + speed)

Input: test/input/hirlam_wind.get

GET /dali?customer=test&product=hirlam_wind&time=201804060600 HTTP/1.0

Parameter	Value	Description
product	hirlam_wind	Product JSON: test/dali/customers/test/products/hirlam_wind.json
time	201804060600	Valid time: 2018-04-06 06:00 UTC

Arrow layer using meteorological "direction": "WindDirection" and "speed": "WindSpeedMS" parameters from the HIRLAM NWP model.

Output:

---

hirlam_uv — HIRLAM wind (U/V components)

Input: test/input/hirlam_uv.get

GET /dali?customer=test&product=hirlam_uv&time=201804060600 HTTP/1.0

Parameter	Value	Description
product	hirlam_uv	Product JSON: test/dali/customers/test/products/hirlam_uv.json

Arrow layer using Cartesian wind components "u": "WindUMS" and "v": "WindVMS" instead of direction/speed. The server converts U/V to direction and magnitude internally.

Output:

---

Wind Barbs and Wind Roses

windbarb — Classic wind barbs

Input: test/input/windbarb.get

GET /dali?customer=test&product=windbarb&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	windbarb	Product JSON: test/dali/customers/test/products/windbarb.json

Renders standard meteorological wind barbs (staff + barbs/pennants encoding speed in knots).

Output:

---

windrose — Wind roses at stations

Input: test/input/windrose.get

GET /dali?customer=test&product=windrose&time=2013-08-05 HTTP/1.0

Parameter	Value	Description
product	windrose	Product JSON: test/dali/customers/test/products/windrose.json
time	2013-08-05	Valid date: 2013-08-05 (ISO 8601 date-only format)

Renders wind roses at named coastal stations showing frequency distribution of wind direction and speed classes. Each rose uses starttimeoffset/endtimeoffset to aggregate over a 24-hour window. Station titles and connectors are customisable via CSS classes.

Output:

---

WAFS Aviation Charts

WAFS (World Area Forecast System) products use ICAO aviation weather data at pressure levels.

wafs_temperature — Upper-level temperature

Input: test/input/wafs_temperature.get

GET /dali?customer=test&product=wafs_temperature&time=201601191800 HTTP/1.0

Parameter	Value	Description
product	wafs_temperature	Product JSON: test/dali/customers/test/products/wafs_temperature.json
time	201601191800	Valid time: 2016-01-19 18:00 UTC

Temperature isobands at a WAFS pressure level (flight level) over the North Atlantic aviation area.

Output:

---

wafs_wind — Upper-level wind arrows

Input: test/input/wafs_wind.get

GET /dali?customer=test&product=wafs_wind&time=201601191800 HTTP/1.0

Parameter	Value	Description
product	wafs_wind	Product JSON: test/dali/customers/test/products/wafs_wind.json

Wind arrows at WAFS pressure level overlaid on a base map.

Output:

---

wafs_leveldata — Level data display

Input: test/input/wafs_leveldata.get

GET /dali?customer=test&product=wafs_leveldata&time=201601191800 HTTP/1.0

Parameter	Value	Description
product	wafs_leveldata	Product JSON: test/dali/customers/test/products/wafs_leveldata.json

Displays numeric values at grid points for WAFS level data.

Output:

---

wafs_windbarb — WAFS wind barbs

Input: test/input/wafs_windbarb.get

GET /dali?customer=test&product=wafs_windbarb&time=201601191800 HTTP/1.0

Parameter	Value	Description
product	wafs_windbarb	Product JSON: test/dali/customers/test/products/wafs_windbarb.json

Standard wind barbs at a WAFS flight level.

Output:

---

wafs_windbarb_shifted — Wind barbs with label offset

Input: test/input/wafs_windbarb_shifted.get

GET /dali?customer=test&product=wafs_windbarb_shifted&time=201601191800 HTTP/1.0

Parameter	Value	Description
product	wafs_windbarb_shifted	Product JSON: test/dali/customers/test/products/wafs_windbarb_shifted.json

WAFS wind barbs with a global label offset (dx, dy) applied to all barb positions.

Output:

---

wafs_windbarb_graticulefill — Wind barbs with graticule fill

Input: test/input/wafs_windbarb_graticulefill.get

GET /dali?customer=test&product=wafs_windbarb_graticulefill&time=201601191800 HTTP/1.0

Parameter	Value	Description
product	wafs_windbarb_graticulefill	Product JSON: test/dali/customers/test/products/wafs_windbarb_graticulefill.json

Adds temperature-coloured filled rectangles in the graticule grid cells behind the wind barbs — a compact SIGWX-style layout.

Output:

---

Observations with Wind and Temperature

opendata_meteorological_wind_keyword — Wind by keyword

Input: test/input/opendata_meteorological_wind_keyword.get

GET /dali?customer=test&product=opendata_meteorological_wind_keyword&time=20130805T1500 HTTP/1.0

Parameter	Value	Description
product	opendata_meteorological_wind_keyword	Product JSON: test/dali/customers/test/products/opendata_meteorological_wind_keyword.json

Wind arrows at keyword-selected stations from the opendata producer.

Output:

---

opendata_meteorological_wind_keyword_plustemperatures — Wind + temperature

Input: test/input/opendata_meteorological_wind_keyword_plustemperatures.get

GET /dali?customer=test&product=opendata_meteorological_wind_keyword_plustemperatures&time=20130805T1500 HTTP/1.0

Parameter	Value	Description
product	opendata_meteorological_wind_keyword_plustemperatures	Product JSON: test/dali/customers/test/products/opendata_meteorological_wind_keyword_plustemperatures.json

Combines wind arrows and temperature numbers at keyword-selected stations.

Output:

---

opendata_meteorological_wind_grid — Wind on grid

Input: test/input/opendata_meteorological_wind_grid.get

GET /dali?customer=test&product=opendata_meteorological_wind_grid&time=20130805T1500 HTTP/1.0

Parameter	Value	Description
product	opendata_meteorological_wind_grid	Product JSON: test/dali/customers/test/products/opendata_meteorological_wind_grid.json

Wind arrows placed at regular pixel-grid intervals (layout: "grid"), thinning the station network for legibility.

Output:

---

opendata_meteorological_wind_data_plustemperatures — Wind + temperature data

Input: test/input/opendata_meteorological_wind_data_plustemperatures.get

GET /dali?customer=test&product=opendata_meteorological_wind_data_plustemperatures&time=20130805T1500 HTTP/1.0

Parameter	Value	Description
product	opendata_meteorological_wind_data_plustemperatures	Product JSON: test/dali/customers/test/products/opendata_meteorological_wind_data_plustemperatures.json

Combines wind arrows and temperature numbers sourced from the same OpenData observation query.

Output:

---

opendata_meteorological_wind_data_plustemperatures_aggregate_max — Time-aggregated max wind

Input: test/input/opendata_meteorological_wind_data_plustemperatures_aggregate_max.get

GET /dali?customer=test&product=opendata_meteorological_wind_data_plustemperatures_aggregate_max&time=20130805T1500 HTTP/1.0

Parameter	Value	Description
product	opendata_meteorological_wind_data_plustemperatures_aggregate_max	Product JSON: test/dali/customers/test/products/opendata_meteorological_wind_data_plustemperatures_aggregate_max.json

Uses SmartMet time-aggregation function syntax to compute maximum wind speed over a trailing 12-hour window: "speed": "nanmax_t(WindSpeedMS:12h:0h)". The nanmax_t function aggregates over the time range ignoring NaN values.

Output:

---

Heatmap

heatmap — Kernel-density heatmap

Input: test/input/heatmap.get

GET /dali?customer=test&product=heatmap&time=20130805T1500&l.heatmap.kernel=exp&l.heatmap.radius=40&l.heatmap.deviation=4 HTTP/1.0

Parameter	Value	Description
product	heatmap	Product JSON: test/dali/customers/test/products/heatmap.json
time	20130805T1500	Valid time: 2013-08-05 15:00 UTC
l.heatmap.kernel	exp	Kernel function for density estimation (exp = exponential decay)
l.heatmap.radius	40	Override: kernel radius in pixels (product default is 20)
l.heatmap.deviation	4	Override: kernel deviation parameter (product default is 10.0)

The heatmap is generated by an isoband layer with a "heatmap" sub-block specifying resolution, radius, kernel, and deviation. A kernel-density surface is computed from the observation point locations and then rendered as isobands.

Output:

---

Hovmoeller Diagrams

Hovmoeller diagrams show a meteorological parameter plotted over two axes, one of which is time.

hovmoeller_geoph500 — Geopotential: time × longitude

Input: test/input/hovmoeller_geoph500.get

GET /dali?customer=test&product=hovmoeller_geoph500 HTTP/1.0

Parameter	Value	Description
product	hovmoeller_geoph500	Product JSON: test/dali/customers/test/products/hovmoeller_geoph500.json

"direction": "time_lon" — time on Y-axis, longitude on X-axis at fixed "latitude": 60.0. Plots 500 hPa geopotential height isobands over lon_min/lon_max = 5°–35°E.

Output:

---

hovmoeller_geoph500_time_lat — Geopotential: time × latitude

Input: test/input/hovmoeller_geoph500_time_lat.get

GET /dali?customer=test&product=hovmoeller_geoph500_time_lat HTTP/1.0

Parameter	Value	Description
product	hovmoeller_geoph500_time_lat	Product JSON: test/dali/customers/test/products/hovmoeller_geoph500_time_lat.json

"direction": "time_lat" — time on Y-axis, latitude on X-axis at fixed "longitude": 20.0.

Output:

---

hovmoeller_temperature_time_level — Temperature: time × pressure level

Input: test/input/hovmoeller_temperature_time_level.get

GET /dali?customer=test&product=hovmoeller_temperature_time_level HTTP/1.0

Parameter	Value	Description
product	hovmoeller_temperature_time_level	Product JSON: test/dali/customers/test/products/hovmoeller_temperature_time_level.json

"direction": "time_level" — time on X-axis, pressure level on Y-axis at a fixed geographic point. Shows the vertical temperature structure over time (a time-height cross-section).

Output:

---

Graticule

graticule — Basic lat/lon grid

Input: test/input/graticule.get

GET /dali?customer=test&product=graticule HTTP/1.0

Parameter	Value	Description
product	graticule	Product JSON: test/dali/customers/test/products/graticule.json

Two overlapping grids: a fine 1° grid (light, 0.2 px) skipping 10° multiples, and a coarse 10° grid (dark, 0.5 px). Projection is inferred from the kap producer's native CRS.

Output:

---

graticule_goode — Graticule in Goode projection

Input: test/input/graticule_goode.get

GET /dali?customer=test&product=graticule_goode HTTP/1.0

Parameter	Value	Description
product	graticule_goode	Product JSON: test/dali/customers/test/products/graticule_goode.json

A 10° graticule in the Goode Interrupted Homolosine projection (+proj=igh). Tests correct handling across the projection's interior discontinuities.

Output:

---

graticule_num_center — Centred labels

Input: test/input/graticule_num_center.get

GET /dali?customer=test&product=graticule_num_center HTTP/1.0

Parameter	Value	Description
product	graticule_num_center	Product JSON: test/dali/customers/test/products/graticule_num_center.json

Graticule with "labels": {"layout": "center"} — degree values placed at the centres of grid lines with a rectangle background.

Output:

---

graticule_num_center_edge — Edge-centred labels

Input: test/input/graticule_num_center_edge.get

GET /dali?customer=test&product=graticule_num_edge_center HTTP/1.0

Parameter	Value	Description
product	graticule_num_edge_center	Product JSON: test/dali/customers/test/products/graticule_num_edge_center.json

Graticule with "layout": "edge_center" — labels placed at the mid-point of each grid line where it intersects the map edge (frame), with a bordered rectangle background.

Output:

---

graticule_num_cross — Crossing-point labels

Input: test/input/graticule_num_cross.get

GET /dali?customer=test&product=graticule_num_cross HTTP/1.0

Parameter	Value	Description
product	graticule_num_cross	Product JSON: test/dali/customers/test/products/graticule_num_cross.json

Graticule with "layout": "cross" — labels placed at every intersection of latitude and longitude grid lines.

Output:

---

graticule_ticks — Tick marks at frame

Input: test/input/graticule_ticks.get

GET /dali?customer=test&product=graticule_ticks HTTP/1.0

Parameter	Value	Description
product	graticule_ticks	Product JSON: test/dali/customers/test/products/graticule_ticks.json

Uses "layout": "ticks" to draw only short tick marks at the map border where grid lines intersect, without drawing the full grid lines across the map.

Output:

---

Circles Layer

The circle layer type draws SVG circles at named geographic locations.

circles — Concentric circles at cities

Input: test/input/circles.get

GET /dali?customer=test&product=circles&time=201503131200 HTTP/1.0

Parameter	Value	Description
product	circles	Product JSON: test/dali/customers/test/products/circles.json
time	201503131200	Valid time: 2015-03-13 12:00 UTC

Draws concentric rings around named cities (Helsinki, Tampere, Turku) over a Natural Earth map in EPSG:3035 (ETRS-LAEA). Each circle layer entry specifies a name (resolved to coordinates via a geocoder) and a radius.

Output:

---

circles_labels_sw — Circle labels (SW anchor)

Input: test/input/circles_labels_sw.get

GET /dali?customer=test&product=circles_labels_sw&time=201503131200 HTTP/1.0

Parameter	Value	Description
product	circles_labels_sw	Product JSON: test/dali/customers/test/products/circles_labels_sw.json

Adds city name labels to the circles, anchored at the south-west corner of each station.

Output:

---

circles_labels_tr — Circle labels (TR anchor)

Input: test/input/circles_labels_tr.get

GET /dali?customer=test&product=circles_labels_tr&time=201503131200 HTTP/1.0

Parameter	Value	Description
product	circles_labels_tr	Product JSON: test/dali/customers/test/products/circles_labels_tr.json

Same as circles_labels_sw but labels are anchored at the top-right corner.

Output:

---

Fronts

fronts — Weather fronts

Input: test/input/fronts.get

GET /dali?customer=test&product=fronts&type=png&time=201503131200 HTTP/1.0

Parameter	Value	Description
product	fronts	Product JSON: test/dali/customers/test/products/fronts.json
type	png	Rasterised PNG output (700×500 px, 8-bit palette)

Renders weather fronts (cold, warm, occluded, stationary) in EPSG:3857 over a background map. Uses the fronts layer type with CSS styling.

Output:

---

PostGIS Layer

ice — Ice chart from PostGIS

Input: test/input/ice.get

GET /dali?customer=test&product=ice HTTP/1.0

Parameter	Value	Description
product	ice	Product JSON: test/dali/customers/test/products/ice.json

Renders Baltic Sea ice chart polygons from the icemap PostGIS database using "layer_type": "postgis". The PostGIS layer can read any geometry type (Polygon, MultiPolygon, LineString) and apply CSS fill/stroke styling.

Output:

---

WKT Layer

wkt — Well-Known Text geometry

Input: test/input/wkt.get

GET /dali?customer=test&product=wkt&time=201503131200 HTTP/1.0

Parameter	Value	Description
product	wkt	Product JSON: test/dali/customers/test/products/wkt.json
time	201503131200	Valid time: 2015-03-13 12:00 UTC

The wkt layer type renders arbitrary WKT (Well-Known Text) geometry strings (e.g. polygons, lines) stored in the product definition. Geometries are projected to the map CRS and styled with CSS.

Output:

---

METAR Layer

metar_basic — Aviation METAR observations

Input: test/input/metar_basic.get

GET /dali?customer=test&product=metar_basic&time=201511170020&format=svg HTTP/1.0

Parameter	Value	Description
product	metar_basic	Product JSON: test/dali/customers/test/products/metar_basic.json
time	201511170020	Valid time: 2015-11-17 00:20 UTC
format	svg	SVG output format

Uses "layer_type": "metar" with "message_type": "METAR" and "message_format": "TAC" to decode raw METAR reports stored as TAC (Traditional Alphanumeric Code) weather tokens. Controls which elements to display: show_station, show_wind, show_visibility, show_present_weather, show_sky_condition, show_temperature.

No reference output image available (no test output file generated in CI).

---

Fire Weather

forestfire — Forest Fire Index isobands

Input: test/input/forestfire.get

GET /dali?product=forestfire&time=20190207T0600 HTTP/1.0

Parameter	Value	Description
product	forestfire	No customer specified — uses the default customer directory
time	20190207T0600	Valid time: 2019-02-07 06:00 UTC

Renders ForestFireIndex isobands using "extrapolation": 2 to fill in missing data near the domain boundaries. The product uses the forestfire producer (separate from general synoptic producers).

Output:

---

grassfire — Grassfire risk by named region

Input: test/input/grassfire.get

GET /dali?product=grassfire&time=20181212T0600 HTTP/1.0

Parameter	Value	Description
product	grassfire	Uses default customer
time	20181212T0600	Valid time: 2018-12-12 06:00 UTC

Renders grassfire risk levels using the grassfire layer type. Regions are matched by named feature attributes (e.g. "value": "Uusimaa" matches the Uusimaa administrative area).

Output:

---

grassfire_automatic — Grassfire risk by attribute range

Input: test/input/grassfire_automatic.get

GET /dali?product=grassfire_automatic&time=20181212T0600 HTTP/1.0

Parameter	Value	Description
product	grassfire_automatic	Uses default customer

Like grassfire but regions are matched by numeric attribute ranges instead of named values — more flexible for data with numeric risk-class codes.

No reference output image available.

---

Text Layer

text — SVG text layer

Input: test/input/text.get

GET /dali?customer=test&product=text&time=200808050300 HTTP/1.0

Parameter	Value	Description
product	text	Product JSON: test/dali/customers/test/products/text.json

The text layer type renders static or parameterised text strings as SVG <text> elements. The product also demonstrates how tag: "clipPath" and tag: "g" layers are used to create inline clip regions.

Output:

---

Map simplification

The MapLayer honours four optional settings that thin the polygons returned by PostGIS before they are clipped, projected and rendered:

• an amalgamator that merges nearby polygons via constrained Delaunay triangulation,
• a Visvalingam-Whyatt simplifier with a pixel-based tolerance (Douglas-Peucker is no longer supported — see the Map structure documentation for why),
• the legacy minarea (km²) and mindistance (km) filters.

The GIS engine applies them in the order amalgamator → minarea → mindistance → simplifier, so that minarea and the new simplifier operate on the merged outline. See the Map structure documentation for the algorithmic background and a "Choosing an algorithm" comparison.

The seven examples below all render the same Northern-Baltic view (lon 18°–27°, lat 59°–65°, 900×600 px) using the gshhg.gshhs_h_l1 table (the GSHHG high-resolution L1 land polygons, ≈ 200 m vertex spacing). A bbox prefilter (the_geom && ST_MakeEnvelope(17, 58, 28, 66, 4326)) is added to every test so that the engine reads only the ~14 000 polygons in the area instead of all 144 000 globally — without that filter the amalgamator would attempt a Delaunay triangulation of every island in the world. They are directly comparable so that the visual effect of each option is easy to read.

The four amalgamation examples (map_amalgamate, map_amalgamate_wide, map_amalgamate_extreme, map_amalgamate_pathological) form a deliberate calibration sequence at amalgamation_length 0.01 → 0.02 → 0.05 → 0.1, so that the visual cost of pushing the gap distance past a sensible value for the rendering pixel scale is easy to read off the gallery PNGs. The pathological case is included to show that the algorithm degrades gracefully — it produces valid topology even at clearly absurd distances.

Apart from the baseline (which keeps every vertex), every other example applies minarea=2 (km²) after simplification / amalgamation. This is the standard "drop islands that are below ~2 px² at the rendered scale" filter; the point of running the amalgamator first is that nearby small islands first merge into one larger landmass and only then face the area filter, so an archipelago survives as a chunky shape instead of being thrown away entirely.

Units. amalgamation_length is in CRS coordinate units — at this latitude (≈ 60° N), 1° lat ≈ 111 km and 1° lon ≈ 56 km, so amalgamation_length=0.01 ≈ 0.6 km. The simplifier's tolerance is in pixels at the active projection. minarea is in km².

map_baseline — No simplification (reference image)

Input: test/input/map_baseline.get

GET /dali?customer=test&product=map_baseline HTTP/1.0

Parameter	Value	Description
product	map_baseline	test/dali/customers/test/products/map_baseline.json

The unfiltered source data from gshhg.gshhs_h_l1: every coastline vertex from the high-resolution GSHHG product is kept (≈ 45 000 vertices in the rendered output). The Stockholm archipelago, Finland's southwestern archipelago and Åland are visible as dense clusters of individual skerries — exactly the kind of data the simplifier and amalgamator are designed to thin. This image is the reference against which the simplification settings below should be compared.

Output:

---

map_simplifier_vw — Visvalingam-Whyatt simplification

Input: test/input/map_simplifier_vw.get

GET /dali?customer=test&product=map_simplifier_vw HTTP/1.0

Parameter	Value	Description
product	map_simplifier_vw	test/dali/customers/test/products/map_simplifier_vw.json

Same data as the baseline with simplifier="visvalingam_whyatt", tolerance=5.0 pixels, and minarea=2 (km²). About 89 % of the source vertices are removed (≈ 5 200 vertices remain). VW iteratively removes the vertex whose triangle with its neighbours has the smallest area, which preserves the overall shape character — bays and peninsulas remain recognisable, and the silhouette is blockier rather than ragged. Note that without amalgamation, minarea deletes every individual island that is below the threshold; large stretches of the archipelago that would survive amalgamation are lost.

Output:

---

map_amalgamate — Polygon amalgamation

Input: test/input/map_amalgamate.get

GET /dali?customer=test&product=map_amalgamate HTTP/1.0

Parameter	Value	Description
product	map_amalgamate	test/dali/customers/test/products/map_amalgamate.json

Same data as the baseline with amalgamation_length=0.01 (≈ 0.6 km gap-bridging), amalgamation_mainland_area=1000 km² with amalgamation_mainland_amalgamate=true, and minarea=2 (km²). The amalgamator triangulates the gaps between polygons via constrained Delaunay; gap triangles whose edges are all shorter than amalgamation_length are accepted as part of the merged outline. Mainland polygons (Finnish, Estonian, Latvian coasts in this view) are above the 1000 km² threshold and would otherwise dominate the CDT input vertex count, so they are routed to a per-polygon CDT pass instead — far cheaper than including them in the global cluster CDT, but it still closes their own bays/inlets below the gap-triangle threshold. The visible result is a coastline that reads as solid rather than fingered.

Compared to the simplifier-only examples above, far more of the archipelago survives: nearby skerries that are individually smaller than 2 km² merge into a single landmass that exceeds the threshold and is therefore kept. This is a topology change that GEOS' SimplifyPreserveTopology cannot perform — that simplifier would leave each tiny island as its own (simplified) polygon.

Output:

---

map_amalgamate_wide — Doubled gap distance

Input: test/input/map_amalgamate_wide.get

GET /dali?customer=test&product=map_amalgamate_wide HTTP/1.0

Parameter	Value	Description
product	map_amalgamate_wide	test/dali/customers/test/products/map_amalgamate_wide.json

Same as map_amalgamate with amalgamation_length doubled from 0.01 to 0.02 (≈ 1.2 km gap-bridging at 60° N). Useful for calibrating how aggressive the gap distance should be at a given rendering pixel scale: at this view the silhouette still reads cleanly along the Finnish coast, but the Stockholm and Åland archipelagos dissolve further into solid blobs and clusters of small skerries in the southern bbox start to merge into single landmasses. Pushing the distance further would eventually produce a coastline that no longer looks like the source dataset — at twice the default the result is still recognisable but visibly more abstract than map_amalgamate.

Output:

---

map_amalgamate_extreme — Excessive gap distance (do not exceed)

Input: test/input/map_amalgamate_extreme.get

GET /dali?customer=test&product=map_amalgamate_extreme HTTP/1.0

Parameter	Value	Description
product	map_amalgamate_extreme	test/dali/customers/test/products/map_amalgamate_extreme.json

Same as map_amalgamate with amalgamation_length raised to 0.05 (≈ 3 km gap-bridging at 60° N) — five times the recommended default. Included as a deliberate "do not exceed" reference: the Stockholm archipelago has dissolved into the Swedish mainland, Åland has become a single solid blob no longer recognisable as an archipelago, the Estonian and Latvian coasts have lost most of their detail, and shallow Finnish bays have filled in. The output is faster to render than the default (less geometry to draw) but is no longer a faithful representation of the source dataset. Useful for calibrating an upper limit when picking amalgamation_length for a new view or zoom level.

Output:

---

map_amalgamate_pathological — Algorithm breakdown at 10× the recommended distance

Input: test/input/map_amalgamate_pathological.get

GET /dali?customer=test&product=map_amalgamate_pathological HTTP/1.0

Parameter	Value	Description
product	map_amalgamate_pathological	test/dali/customers/test/products/map_amalgamate_pathological.json

amalgamation_length=0.1 (≈ 6 km gap-bridging at 60° N), 10 × the recommended default and well past the source data's natural island spacing. Mainland coastlines reduce to cartoonish silhouettes, the Stockholm archipelago is fully fused with the Swedish mainland, and the gap-triangle pass starts producing scattered "ghost" white spots in the Bothnian Sea and Gulf of Finland — narrow interior holes inside the merged region that survive the boundary walk. The algorithm is degrading gracefully (it still produces valid topology, just nonsensical output) rather than failing. Included to make the upper edge of the calibration sequence visible.

Output:

---

map_amalgamate_simplified — Amalgamation followed by Visvalingam-Whyatt

Input: test/input/map_amalgamate_simplified.get

GET /dali?customer=test&product=map_amalgamate_simplified HTTP/1.0

Parameter	Value	Description
product	map_amalgamate_simplified	test/dali/customers/test/products/map_amalgamate_simplified.json

Same amalgamation, mainland-amalgamate flag and minarea as map_amalgamate with simplifier="visvalingam_whyatt" and tolerance=1.5 pixels chained on. This combination is the recommended pipeline for archipelago and dense-coastline datasets: amalgamation gets the silhouette right by merging nearby islands and closing mainland bays, minarea drops the merged blobs that are still too small to render meaningfully, and the simplifier then thins out the redundant vertices on what's left.

Output:

---

World Map Products

These products render global-scale model data in various projections. Each family (ECMWF, GFS) covers a standard set of projections; Pacific-centred views are available for equidistant cylindrical and WGS84 projections.

ECMWF global temperature

All EC world products use the `ecmwf_maailma_pinta$ \text{producer} \text{and} \text{a} 1440 \times 720 \text{px} \text{canvas}:

\text{Test}	\text{Projection}	\text{CRS}	\text{Input}
$ec_world_eqc`	Equidistant Cylindrical (Atlantic)	EPSG:4087	ec_world_eqc.get
ec_world_eqc_pacific	Equidistant Cylindrical (Pacific, lon₀=180°)	+init=epsg:4087 +lon_0=180	ec_world_eqc_pacific.get
ec_world_webmercator	Web Mercator	EPSG:3857	ec_world_webmercator.get
ec_world_wgs84_atlantic	Geographic WGS84 (Atlantic)	+init=epsg:4326	ec_world_wgs84_atlantic.get
ec_world_wgs84_pacific	Geographic WGS84 (Pacific, lon_wrap=180)	+init=epsg:4326 +lon_wrap=180	ec_world_wgs84_pacific.get

The +lon_wrap=180 modifier shifts the longitude seam to the antimeridian so that the Pacific Ocean is unbroken.

Projection	Output
EQC Atlantic
EQC Pacific
Web Mercator
WGS84 Atlantic
WGS84 Pacific

---

GFS global temperature

All GFS world products use the gfs producer. In addition to the five standard projections, GFS also includes polar, Goode, Natural Earth, and Near-side Perspective views:

Test	Projection	CRS
gfs_world_eqc	EQC Atlantic	+init=epsg:4087
gfs_world_eqc_pacific	EQC Pacific (lon₀=180°)	+init=epsg:4087 +lon_0=180
gfs_world_webmercator	Web Mercator (Atlantic)	EPSG:3857
gfs_world_webmercator_pacific	Web Mercator (Pacific bbox)	EPSG:3857
gfs_world_wgs84_atlantic	Geographic WGS84 Atlantic	+init=epsg:4326
gfs_world_wgs84_pacific	Geographic WGS84 Pacific	+init=epsg:4326 +lon_wrap=180
gfs_world_arctic	Polar Stereographic (N)	+proj=stere +lat_0=90
gfs_world_antarctic	Polar Stereographic (S)	+proj=stere +lat_0=-90
gfs_world_goode	Goode Homolosine	+proj=igh +lon_0=0
gfs_world_natural_earth	Natural Earth	+proj=natearth
gfs_world_perspective	Near-side Perspective (h=3000 km)	+proj=nsper +h=3000000 +lat_0=60 +lon_0=25
gfs_world_stereographic	Polar Stereographic (Europe)	+proj=stere +lat_0=90 +lat_ts=60 +lon_0=20

Projection	Output
EQC Atlantic
EQC Pacific
Web Mercator
Web Mercator Pacific
WGS84 Atlantic
WGS84 Pacific
Arctic
Antarctic
Goode Homolosine
Natural Earth
Perspective (h=3000 km)
Stereographic (Europe)

---