v0.1.1

pixelcoords 0.1.1 — crates.io README link fix

v0.1.0

Also on crates.io: cargo install pixelcoords

The initial release: a screen-coordinate toolchain built around a frozen
screen.

• The overlay. Freeze every monitor — or one (--monitor), or a
  single window (--target, and --pick on Wayland) — then mark
  regions with five tools: rectangle, ellipse, triangle, regular N-gon
  (3–9 sides), and freehand. Move, resize, rotate, label, alt-drag
  duplicate, cycle overlapped shapes, undo/redo, pixel nudging with
  arrows, a magnifier loupe on M, live cursor coordinates, and a
  movable control panel that remembers its parking spot. Every key is
  rebindable (config file or --bind).
• The outputs. A versioned session.json — monitor-local, global,
  and window-relative physical-pixel coordinates, per-monitor DPI
  scale, platform and capture provenance, an optional friendly name —
  plus full screenshots, per-selection PNG crops (non-rect shapes
  alpha-masked), and a frame-sized cutout pair per monitor: selections
  kept in place, and their exact complement. Saves never overwrite
  files pixelcoords did not write.
• The subcommands. assert scores a point against saved regions
  with the exit code as the API; emit prints ready-to-paste click
  snippets (pyautogui, cliclick, xdotool) with each tool's coordinate
  convention already applied; find re-locates regions in a fresh
  capture by their saved crops and reports the drift; resume reopens
  a session for editing and saves it in place; rename names a
  session for the interactive picker; shoot is the scripted
  no-overlay capture; windows lists targetable windows; doctor
  checks permissions, config, and monitors. --json everywhere a
  script would read the answer.
• Platforms. macOS, Windows, and Linux — X11 fully, Wayland
  snapshot and --pick — each verified on real hardware; the README's
  platform table records exactly what has been exercised where.