Releases: YusufEminoglu/zero2viz
v0.10.2
v0.9.2 - Plotly/Vega-Lite WebKit fallback
Plotly & Vega-Lite: no more blank panel on QGIS without QtWebEngine
When a QGIS build ships no QtWebEngine, the embedded chart panel falls back to the legacy QtWebKit viewer. That old viewer cannot run the modern JavaScript (ES6+) that Plotly.js and Vega-Lite require, so their charts stayed blank in the dock — even though they exported and opened correctly in a browser. ECharts and the matplotlib image engine were unaffected (they run on QtWebKit).
What changed: the dock now detects the QtWebKit viewer and, for Plotly/Vega-Lite, shows a short explainer in the panel instead of a silent blank. Your chart is still built — click ↗ to open it in your system browser, or use Export HTML. For a chart that renders inside the dock, switch to the ECharts engine, or install QtWebEngine for QGIS.
Root-caused in real QGIS: under QtWebKit the engines fail with ReferenceError: Can't find variable: Plotly / vega; under QtWebEngine (QGIS 4 / Qt6) all engines render. Chart output, the engines and exports are unchanged.
Install: download zero2viz.zip below → QGIS → Plugins → Install from ZIP. Also on the QGIS Plugin Hub.
v0.9.1 - QGIS 4 (Qt6) dock readability
- Readable on QGIS 4 (Qt6). The studio panel now pins its own light colours for every control, so it reads identically under any QGIS theme. On QGIS 4 the dark application palette bled into the parts the panel didn't explicitly colour: the drop-down lists rendered on a near-black background and the field labels and check-box text faded into the white cards. Combo pop-ups, line edits, spin boxes, the diagram field list, form labels, check boxes and tool-tips are now explicitly styled — white fields, dark text, teal selection — independent of the host palette. QGIS 3.x looked correct before (its palette is light) and is unchanged.
- The fix is purely the dock's stylesheet; the chart engines, rendering and exports are untouched. Re-verified end-to-end on QGIS 4.0.2 and QGIS 3.44: all four engines render every supported type (ECharts 17, Plotly 17, Vega-Lite 14, matplotlib 11) plus the 12 animated pages, with zero JS exceptions in headless Chrome; the panel and an open drop-down were also grabbed under a simulated dark palette to confirm white fields and dark, legible text.
Install: download zero2viz.zip below → QGIS → Plugins → Install from ZIP. Also available on the QGIS Plugin Hub.
v0.8.0 - three-tab studio, labels & matplotlib engine
02viz v0.8.0 — three-tab studio, labels & an optional matplotlib engine
New
- One dock, three tabs — Charts · Map diagrams · Labels, sharing a single layer selector. Three ways to turn a layer into an elegant visual.
- Labels tab — one click turns a field into well-placed, publication-grade labels (clean / strong-halo / bold / plain presets), built on native QGIS labeling.
- Embedded colour editor — palette swatches are now inline in the Charts tab: click to recolour, +/- to resize. No modal. Plus a chart-title override.
- Optional matplotlib / seaborn engine — a fourth engine renders the spec to publication-grade static figures (11 chart types), embedded as a PNG so the viewer and one-file export work unchanged. It installs on demand with your consent (auto-detect missing libraries → one-click pip install --user into the QGIS Python). The three vendored JS engines stay zero-dependency and offline.
- Zero2Visual — the motto ("from zero to elegant visuals, fast") now reads in the dock and About. Hub name and "02viz" branding unchanged.
Verified
86 checks on real QGIS Python (matplotlib engine across 11 types, on-canvas labels, dependency detector) + 50 interactive chart pages in headless Chrome, zero JS exceptions.
Install: download zero2viz.zip below → QGIS Plugins → Install from ZIP. Requires QGIS 3.28+. The matplotlib engine is optional and installs from inside the plugin.
v0.7.0 — map diagrams, custom palettes & Tableau-grade polish
02viz v0.7.0 — map diagrams, custom palettes, render fix & Tableau-grade polish
✨ New
- Map diagrams (on-canvas charts) — draw native QGIS pie / bar / stacked-bar / text diagrams on every feature, directly on the map canvas, coloured with the studio palette. They print and export to layout like any symbology (
Data → Map diagrams…). - Custom colour palettes — a Custom… swatch editor lets you define your own series colours, applied identically across ECharts, Plotly and Vega-Lite.
- Engine-first workflow — pick the renderer, then choose from the chart types it can draw.
🐛 Fixed
- Charts blank in the dock but fine on export — the embedded viewer now gives the chart a definite size and re-fits once laid out. Added an ↗ open-in-browser fallback and surfaced the active web backend in the status line.
- Explore noise — identifier columns (
fid,id,gid,uuid, primary keys) are no longer profiled into meaningless charts.
💅 Polish
- Publication-grade visual pass across all three engines: shared typography, soft dashed gridlines (no vertical chart-junk), clean card tooltips, rounded bars and minimal value axes.
✅ Verified
73 checks on real QGIS Python + 50 chart pages in headless Chrome, zero JS exceptions.
Install: download zero2viz.zip below → QGIS Plugins → Install from ZIP. Requires QGIS 3.28+.
v0.6.0
02viz v0.6.0
Six advanced chart types (17 total): Mean ± σ band, Mean ± σ bars, Density (KDE), Violin, Radar/spider and Pareto (80/20) — all pure-Python statistics rendered by Apache ECharts, Plotly.js and Vega-Lite from one spec contract (radar is greyed out for Vega-Lite). New Colors selector with 8 curated palettes (including a colorblind-safe one) that override the theme palette in every engine and in Explore dashboards alike.
Install: download zero2viz.zip below → QGIS Plugins → Install from ZIP. Requires QGIS 3.28+.