GIMP-style resize handles in the GUI#34
Merged
Merged
Conversation
Drag the rendered art directly instead of typing numbers: right edge = width, bottom edge = height stretch/squish, corner = free resize with Shift for aspect lock; live cols x rows readout; double-click resets height to auto. The sandboxed preview reports its pixel box via an injected postMessage snippet (handles the video <img> being CSS-scaled through its natural size), and the server response now carries the post-scaling art grid so drags convert exactly to cells. Server enablers: exact/max sizing is honored on the colorize-off path (previously silently ignored), and video gains exact rows via cell-grid frame pre-resize (braille 2x4 / glyph 8x16 px cells make the rows formula land exactly), wired as --rows on the CLI and video_rows in the API/GUI. Height handles hide when a caption is stitched into the block (the pixel math would conflate caption rows with art rows). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- A block-level pre stretches to the full iframe width, so the ring spanned the whole preview and every re-render snapped it back; measure the pre's CONTENT with a DOM Range instead (widest text line). - Fast drags lost the handle the moment the cursor crossed the iframe, which swallows pointer events: setPointerCapture on the handle keeps the drag, iframe pointer-events disabled while dragging, touch-action none on handles. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
The server now reports how many rows the caption occupies within the rendered block (cap_rows + position; zero for the animation player, whose caption is a separate element). The resize ring subtracts that strip and wraps just the art, so vertical drags convert cleanly to art rows again — the caption stays outside the selection, GIMP-style. Cell math simplified to displayed-box / grid-count, which also handles the CSS-downscaled video GIF without natural-size juggling. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- The caption gets its own ring (round handle, dashed outline) wrapping just the caption lines: dragging its corner maps width to the caption Size knob as a fraction of art width (snapped to the slider's 0.05 steps), with a percentage readout; double-click resets to 60%. Shown only for scalable styles (block/small/shadow/figlet). Server art dims now report cap_lines/cap_gap/cap_style so both rings place exactly. - Figlet corruption fix: the size ladder measured candidates rendered AT the art width, where pyfiglet wraps large fonts and smushes the wrapped blocks into each other (letters truncated, fragments of neighbors bleeding in - 'I see you' lost its I). Candidates now render unwrapped for measurement so a mangled wrap can never be selected; when nothing fits on one line, the smallest font wraps cleanly into stacked blocks. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- Center/right alignment padded each ROW independently, shearing multi-row letterforms apart (short rows of a glyph drifted toward the middle while long rows stayed put) - the real culprit behind mangled centered captions. The block now shifts as a unit. - The figlet ladder had big gaps (43 -> 61 -> 135 wide for a short phrase), so at common art widths half the Size range picked the same font and the caption drag felt dead. Added chunky/banner3/epic/ larry3d/roman/univers rungs; the sweep now moves through distinct sizes across the range. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
…drag The caption ring's discrete font-ladder mapping felt dead between rungs. Captions now take explicit cols/rows (CaptionOptions.cols/rows, threaded through static, animation, and video paths): the closest natural rendering is grid-transformed to the exact size, with a missing dimension derived from aspect. New Width/Height fields in the Caption panel and --caption-cols/--caption-rows on the image, colorize, and video commands; the Auto-size slider applies only when the exact fields are empty. The caption ring gains E/S/SE handles committing exact chars x rows continuously (Shift-corner keeps aspect; double-click resets to auto), now hugs the caption's ink instead of the padded row (server reports ink width and offset), and shows a cols x rows readout. All caption styles can free-transform, so the style gate is gone. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Grab-and-stretch sizing on the preview, like GIMP's Scale tool: right edge = width, bottom edge = stretch/squish height, corner = free resize (Shift = keep aspect), live
cols × rowsreadout, double-click to reset height to auto. Images/text re-render on release; video sets its knobs and waits for Render.Mechanics: the sandboxed preview (opaque origin) reports its art's pixel box via an injected
postMessagesnippet — including natural dimensions for the video<img>, which browsers CSS-downscale — and the API response now includes the post-scaling art grid (art: {cols, rows}), so drags convert exactly to character cells with no font-metric guessing.Server enablers (each independently useful):
--rowson the CLI,video_rowsin the API, a Rows field in the Video tabHeight handles hide when a caption is enabled (caption rows would pollute the pixel→rows math); width dragging stays.
Test plan
#m, skipping the caption pre), and video img🤖 Generated with Claude Code