Debian Codegraph @

doc/release_notes/release_1.00.md

raw
1 # Version 1.0 #
2
3 ## DataLab Version 1.0.4 (2026-02-02) ##
4
5 ### πŸ› οΈ Bug Fixes since version 1.0.3 ###
6
7 **Centroid computation compatibility:**
8
9 * Fixed centroid computation failing with NumPy 2.4.0 and scikit-image 0.26.0 due to new einsum optimization that doesn't accept masked arrays
10 * Image data is now explicitly converted to a standard NumPy array before computing the centroid
11 * This closes [Issue #289](https://github.com/DataLab-Platform/DataLab/issues/289) - Compatibility issue with NumPy 2.4.0 and scikit-image 0.26.0
12
13 ## DataLab Version 1.0.3 (2025-12-19) ##
14
15 ### πŸ› οΈ Bug Fixes since version 1.0.2 ###
16
17 **Loading images with a single row or column fails:**
18
19 * Fixed loading image files with a single row or column (e.g., some SIF spectroscopy files with shape `(1, N)`) causing an `IndexError` and preventing the image from being displayed
20 * The underlying issue was in PlotPy's coordinate-to-bin conversion function which assumed at least 2 points
21 * This fix is included in PlotPy 2.8.3
22
23 **LUT range incorrectly copied when processing images:**
24
25 * Fixed processed images inheriting the LUT range (display contrast settings) from the source image, causing incorrect visualization when data values change significantly
26 * When applying operations like offset correction, the result image now auto-scales its display range to fit the new data values instead of using the original image's LUT range
27 * Previously, users had to manually adjust the LUT to see processed images correctly (they could appear completely black or with wrong contrast)
28 * This closes [Issue #288](https://github.com/DataLab-Platform/DataLab/issues/288) - LUT range incorrectly copied when processing images
29 * This fix is included in Sigima 1.0.4
30
31 **Separate plot windows - Incorrect aspect ratio for non-uniform coordinates:**
32
33 * Fixed separate plot windows (e.g., "View in a new window", ROI editors, profile dialogs) not using the same aspect ratio configuration as the integrated plot handler
34 * When displaying images with non-uniform coordinates, the main plot correctly unlocks the aspect ratio for proper display, but separate windows were not inheriting this setting
35 * All image dialogs now properly inherit the aspect ratio settings, ensuring consistent display behavior across all plot windows
36 * This closes [Issue #287](https://github.com/DataLab-Platform/DataLab/issues/287) - Separate plot windows don't use same aspect ratio configuration as integrated plot
37
38 **Remove all results - AttributeError when ROI was removed:**
39
40 * Fixed "Remove all results" action failing with `AttributeError: 'NoneType' object has no attribute 'get_single_roi_title'` when results contain ROI information but the ROI was subsequently removed from the object
41 * The issue occurred when:
42 1. Running an analysis with a ROI (e.g., Centroid on an image with rectangular ROI)
43 2. Removing the ROI from the object
44 3. Trying to delete all results via "Analysis > Remove results > Remove all results"
45 * The fix adds a proper None check before accessing the ROI's title
46 * This closes [Issue #286](https://github.com/DataLab-Platform/DataLab/issues/286) - "Remove all results" fails with AttributeError when ROI was removed
47
48 **Analysis auto-recompute - Stale parameters after deleting results:**
49
50 * Fixed analysis parameters not being cleared when deleting analysis results, which could cause unexpected auto-recompute behavior when modifying ROIs
51 * After deleting results (via "Delete all results" or individual result deletion), changing the ROI would trigger recomputation of the deleted analysis because the stored parameters remained in object metadata
52 * The fix ensures analysis parameters are properly cleared alongside the results, preventing unwanted automatic recomputation
53 * This closes [Issue #285](https://github.com/DataLab-Platform/DataLab/issues/285) - Analysis parameters not cleared after deleting results
54
55 **Backwards-drawn rectangular ROI causes NaN statistics:**
56
57 * Fixed rectangular ROI statistics returning NaN values when the ROI was drawn "backwards" (from bottom-right to top-left instead of top-left to bottom-right)
58 * The issue occurred because the ROI coordinate conversion produced negative width/height values, causing the mask generation to fail
59 * ROI mask generation now works correctly regardless of the direction in which the rectangle was drawn
60 * This closes [Issue #284](https://github.com/datalab-platform/datalab/issues/284) - Backwards-drawn rectangular ROI causes NaN statistics
61 * This fix is included in Sigima 1.0.4
62
63 **Plot refresh - ROIs and annotations persist after "Remove all":**
64
65 * Fixed ROIs and annotations occasionally remaining visible after executing "Remove all" action from the Edit menu
66 * The issue occurred because the plot widget wasn't explicitly told to redraw after clearing all plot items and shape items
67 * The plot now correctly refreshes its display after clearing all objects, ensuring a clean view without requiring panel switching
68 * The bug was intermittent and hard to reproduce systematically, making it appear as if objects were "stuck" on the plot
69
70 **Result labels - False "omitted" message for single-row results:**
71
72 * Fixed result labels (e.g., pulse features) incorrectly showing "X more rows omitted" message when displaying single-row results with no actual truncation
73 * The bug occurred because `len(adapter.result)` was used to get the row count, but this actually returns the number of column headers, not data rows
74 * For pulse features with 13 columns, this caused the label to think there were 13 rows when only 1 row existed, leading to a false "12 more rows omitted" message
75 * The fix ensures accurate row counting using `len(df)` instead, preventing false truncation notices
76 * This closes [Issue #281](https://github.com/datalab-platform/datalab/issues/281) - Result Labels False "Omitted" Message
77
78 **Signal view - XRangeSelection items persist after removing signals:**
79
80 * Fixed XRangeSelection and DataInfoLabel items (created by PlotPy's curve statistics tools) remaining visible on the plot after removing all signals
81 * When activating the XRangeSelection tool from the PlotPy toolbar and then removing the signal(s), these tool-created graphical objects were not cleaned up
82 * The cleanup logic now correctly triggers when there are zero selected objects (after deleting the last one), not just when exactly one object is selected
83 * This closes [Issue #280](https://github.com/datalab-platform/datalab/issues/280) - XRangeSelection items persist after removing all signals
84
85 **Remote control - Crash when adding object via proxy:**
86
87 * Fixed `AttributeError: 'NoneType' object has no attribute 'setText'` crash when adding objects via `RemoteProxy.add_object()` in macros
88 * The crash occurred in `update_tree()` when the tree view became temporarily out of sync with the object model (e.g., when adding objects with specific metadata configurations or ROIs via remote proxy)
89 * The tree view now rebuilds automatically if inconsistencies are detected, ensuring robust object addition in all scenarios
90 * This closes [Issue #279](https://github.com/datalab-platform/datalab/issues/279) - AttributeError in update_tree() when adding object via RemoteProxy
91
92 **Icons - Qt SVG warnings on Linux:**
93
94 * Fixed "qt.svg: Invalid path data; path truncated" warnings appearing 4 times at startup on Linux/Ubuntu
95 * Cleaned SVG icon files to remove Inkscape/Sodipodi metadata and fixed path syntax in two icon files that used implicit cubic BΓ©zier command continuation (not supported by Qt's SVG renderer on Linux)
96 * This closes [Issue #278](https://github.com/datalab-platform/datalab/issues/278) - Fix "qt.svg: Invalid path data; path truncated" warnings on Linux
97
98 **ROI extraction - KeyError when extracting ROIs:**
99
100 * Fixed `KeyError` when extracting ROIs from images with ROIs generated by blob detection or other analysis functions
101 * The issue occurred because the ROI editor was trying to access plot items that hadn't been created yet (e.g., when "auto refresh" was disabled or in certain selection scenarios)
102 * The ROI extraction dialog now handles missing plot items gracefully
103 * This closes [Issue #276](https://github.com/datalab-platform/datalab/issues/276) - KeyError when extracting ROIs from images with blob detection results
104
105 **Plugin development - Infinite loop when iterating over group objects:**
106
107 * Fixed infinite loop occurring in plugins when iterating over object IDs returned by `ObjectGroup.get_object_ids()` while simultaneously adding new objects to the same group
108 * The method was returning a direct reference to the internal list instead of a copy, causing the iteration to include newly added objects and loop indefinitely
109 * This particularly affected plugin workflows that loop over signals/images in a group and add results to the same or another group
110 * The fix ensures `get_object_ids()` now returns a copy of the list, making iteration safe even when modifying the group
111 * This closes [Issue #274](https://github.com/datalab-platform/datalab/issues/274) - Infinite loop when iterating over group objects while adding new objects
112
113 **Marker visibility:**
114
115 * Fixed cross marker (shown when pressing Alt key on plot) colors inherited from PlotPy being too pale on white background, and other marker properties explicitly configured for better contrast
116 * Fixed default annotation colors inherited from PlotPy being too pale on white background - segment and drag shape annotations now use brighter green (`#00ff55`) for better visibility
117
118 **Macro panel layout:**
119
120 * Fixed macro console taking excessive vertical space on first open - the script editor now properly gets 70% of space and console 30% by default, ensuring comfortable editing without manual resizing
121
122 **Macro naming:**
123
124 * Fixed automatic macro naming generating duplicate names after reloading workspace from HDF5 files
125 * When macros were deserialized from saved workspaces or imported from files, the internal counter used for generating default names (e.g., "macro_01", "macro_02") was not updated, causing the next new macro to potentially reuse an existing name
126 * The counter now synchronizes with existing macro names after loading, ensuring unique sequential naming
127 * Default macro names simplified from translated "Untitled XX" to language-neutral "macro_XX" format for better consistency across locales
128
129 **Macro execution:**
130
131 * Fixed syntax errors when using f-strings with nested quotes in macros (e.g., `f'text {func("arg")}'` now works correctly)
132 * Fixed corrupted Unicode characters in macro console output on Windows - special characters like βœ…, πŸ’‘, and β†’ now display correctly instead of showing garbled text
133
134 **Internal console - HDF5 workspace operations:**
135
136 * Fixed application freeze when calling `open_h5_files()` from the internal console - the console runs in a separate thread and creating Qt GUI elements (progress dialogs) from non-main threads causes deadlocks
137 * Added new headless API methods `load_h5_workspace()` and `save_h5_workspace()` that can be safely called from the internal console or any script context without GUI dependencies
138 * These methods are also available through the remote control API for consistency
139 * Note: `load_h5_workspace()` only supports native DataLab HDF5 files; for importing arbitrary HDF5 files, use macros with `RemoteProxy`
140 * This closes [Issue #275](https://github.com/datalab-platform/datalab/issues/275) - Console freezes when calling `open_h5_files()` from internal console
141
142 **Signal cursor dialogs - Decimal input with regional settings:**
143
144 * Fixed decimal value input failing in signal cursor dialogs when using regional settings with comma as decimal separator (e.g., French, German locales)
145 * The Y input field in "First abscissa at y=...", "Ordinate at x=...", and "Full width at y=..." dialogs now consistently accepts dot as decimal separator regardless of system locale
146 * This closes [Issue #276](https://github.com/datalab-platform/datalab/issues/276) - Decimal input fails with regional settings using comma as decimal separator
147
148 **Radial profile title:**
149
150 * Fixed duplicate suffix in result image title when extracting radial profile from an image (e.g., `radial_profile(i019)|center=(192.500, 192.500)|center=(192.500, 192.500)` instead of `radial_profile(i019)|center=(192.500, 192.500)`)
151 * This fix is provided by Sigima 1.0.4
152
153 **Grid ROI - Missing spacing parameters for non-uniform grids:**
154
155 * Fixed grid ROI feature not working correctly for images where subimages don't fill the entire image area (e.g., laser spot arrays with gaps between spots)
156 * Added missing `xstep` and `ystep` parameters to control horizontal and vertical spacing between ROI centers, as a percentage of the automatically computed cell width/height (default 100% = evenly distributed grid)
157 * Previously, the grid was always assumed to be evenly distributed across the entire image, which failed when there was significant offset or gaps between features
158 * Users can now adjust ROI spacing independently from ROI size to accurately extract grids from real-world data like laser spot arrays, diffraction patterns, or any regularly spaced features with gaps
159 * This fix is provided by Sigima 1.0.4
160 * This closes [Issue #282](https://github.com/datalab-platform/datalab/issues/282) - Grid ROI Missing Spacing Parameters
161
162 **Analysis auto-recompute - Redundant O(nΒ²) calculations when adding ROI:**
163
164 * Fixed severe performance issue where adding ROI to N selected images with analysis results (e.g., Statistics, Centroid) triggered NΒ² calculations instead of N
165 * When 10 images were selected with pre-computed statistics, adding a ROI would show 10 progress bars of 10 calculations each (100 total) instead of a single progress bar with 10 calculations
166 * The issue occurred because `auto_recompute_analysis()` was calling `compute_1_to_0()` which by default processes all selected objects, not just the specific object being recomputed
167 * Added `target_objs` parameter to `compute_1_to_0()` to allow specifying which objects to process, and updated `auto_recompute_analysis()` to use it
168 * This closes [Issue #283](https://github.com/datalab-platform/datalab/issues/283) - Redundant O(nΒ²) calculations when adding ROI to multiple images with analysis results
169
170 **Result visualization - Analysis result segments hard to see:**
171
172 * Fixed analysis result markers (FWHM, pulse features, etc.) being difficult or impossible to see on signal plots
173 * These measurement indicators were appearing black on white backgrounds or white on dark backgrounds, making them blend into the plot
174 * They now display in bright green with thicker lines, making them clearly visible in all color themes
175 * Also removed the filled area that was obscuring the underlying signal data - markers now show only the boundary lines indicating the measured region
176
177 ## DataLab Version 1.0.2 (2025-12-03) ##
178
179 ### πŸ› οΈ Bug Fixes since version 1.0.1 ###
180
181 **Signal axis calibration - Replace X by other signal's Y:**
182
183 * Added new "Replace X by other signal's Y" operation in Processing > Axis transformation menu for signal calibration workflows
184 * Addresses critical missing functionality reported by users who needed to apply wavelength calibration or similar transformations to spectroscopy data
185 * The operation combines two signals: uses Y values from one signal as the new X coordinates for another signal's Y values
186 * Unlike X-Y mode (which resamples and interpolates), this operation directly uses Y arrays without interpolation, preserving exact calibration values
187 * Requires both signals to have the same number of points - raises clear error message if sizes don't match
188 * Automatically transfers metadata: X axis label/unit taken from calibration signal's Y label/unit
189 * Menu location: Processing > Axis transformation > "Replace X by other signal's Y"
190 * This closes [Issue #273](https://github.com/datalab-platform/datalab/issues/273) - Missing signal axis calibration: no way to replace X with Y from another signal
191
192 **X-Y mode:**
193
194 * The X-Y mode processing operation for signals has been moved to Processing > Axis transformation > "X-Y mode" for better discoverability
195 * The nuance between X-Y mode (which resamples/interpolates) and the new "Replace X by other signal's Y" operation has been clarified in documentation
196
197 **Lock LUT setting persistence:**
198
199 * Fixed "Lock LUT range when updating data" setting not persisting in Settings > Visualization > Images > Default Image visualization settings
200 * The `keep_lut_range` parameter was not being saved to configuration, causing the checkbox to systematically uncheck itself after validation (added missing `ima_def_keep_lut_range` option in configuration)
201 * This closes [Issue #270](https://github.com/datalab-platform/datalab/issues/270) - Lock LUT setting not persisting in image visualization defaults
202
203 **Custom signal creation:**
204
205 * Fixed `AttributeError: 'NoneType' object has no attribute 'T'` error when creating a custom signal from the menu
206 * This closes [Issue #269](https://github.com/datalab-platform/datalab/issues/269) - Custom Signal Creation: `AttributeError` when creating signal from menu
207
208 **Macro execution:**
209
210 * Fixed `UnicodeEncodeError` when executing macros that print Unicode characters (e.g., arrows `β†’`) on Windows systems with certain locales, e.g. cp1252 (closes [Issue #263](https://github.com/datalab-platform/datalab/issues/263))
211 * The macro subprocess now automatically uses UTF-8 encoding for stdout and stderr, eliminating the need to manually add `sys.stdout.reconfigure(encoding='utf-8')` at the beginning of each macro.
212
213 **ROI coordinate precision:**
214
215 * ROI coordinates are now automatically rounded to appropriate precision when defining ROIs interactively from geometrical shapes, avoiding excessive decimal places while maintaining reasonable precision relative to data sampling (1/10th of sampling period for signals, 1/10th of pixel spacing for images)
216 * ROI coordinates are also rounded when displaying them in the "Edit numerically" dialog, preventing floating-point arithmetic errors from showing excessive decimal places (e.g., 172.29999999999995 is now displayed as 172.3)
217 * This closes [Issue #266](https://github.com/datalab-platform/datalab/issues/266) - Excessive decimal places in ROI coordinates
218
219 **Polygonal ROI handling:**
220
221 * Fixed `ValueError: Buffer has wrong number of dimensions` error when creating masks from polygonal ROIs in the ROI editor
222 * The PolygonalTool was incorrectly initializing ROI coordinates as a nested list instead of a flat list, causing mask computation to fail
223
224 **HDF5 file opening dialog:**
225
226 * Fixed bug where user's choice in the "clear workspace" confirmation dialog was ignored when opening HDF5 files
227 * When the user clicked "No" in the dialog, the default configuration setting was applied instead of respecting the user's choice
228 * This closes [Issue #267](https://github.com/datalab-platform/datalab/issues/267) - HDF5 file opening dialog ignores user's choice
229
230 **Creation tab axis update:**
231
232 * Fixed plot not updating when modifying only xmin/xmax parameters for distribution signals (Zero, Normal, Poisson, Uniform) in the Creation tab
233 * The issue occurred because the data hash calculation only considered Y values, so changes to X axis bounds were not detected
234 * Plot now properly refreshes when any axis parameter changes, even if Y values remain identical
235 * This closes [Issue #268](https://github.com/datalab-platform/datalab/issues/268) - Creation tab axis not updating for distribution signals
236
237 **ROI statistics with out-of-bounds ROI:**
238
239 * Fixed `ValueError: zero-size array to reduction operation minimum which has no identity` error when computing statistics on images with ROI extending beyond canvas boundaries
240 * The issue occurred when a ROI partially or completely extended outside the image bounds, resulting in empty array slices during statistics computation
241 * ROI bounding boxes are now properly clipped to image boundaries, and fully out-of-bounds ROIs return NaN statistics values
242 * This fix is implemented in Sigima library (see [Issue #1](https://github.com/DataLab-Platform/Sigima/issues/1) - `ValueError` when computing statistics on ROI extending beyond image boundaries)
243
244 **Object property panel tab selection:**
245
246 * Fixed tab selection behavior in object properties panel to be more predictable and user-friendly
247 * Properties tab is now always shown by default when switching between objects, providing consistent navigation
248 * Creation, Processing, and Analysis tabs now appear automatically only once after their respective triggering events (object creation, 1-to-1 processing, or analysis computation), then revert to Properties tab for subsequent selections
249 * This eliminates the confusing behavior where the tab would arbitrarily persist or change based on previous selections
250 * This closes [Issue #271](https://github.com/datalab-platform/datalab/issues/271) - Improve object property panel tab selection behavior
251
252 ## DataLab Version 1.0.1 (2025-11-18) ##
253
254 This major release represents a significant milestone for DataLab with numerous enhancements across all areas. The changes are organized by category for easier navigation.
255
256 > **Note**: Version 1.0.1 is a patch release that addresses Windows installer limitations in version 1.0.0 (released 2025-11-16), where it was not possible to install DataLab V1.0 alongside earlier major versions (V0.20 and earlier) despite the intended coexistence support. This release contains no new features or functional changes.
257
258 ### 🎨 User Interface & Workflow ###
259
260 **Menu reorganization:**
261
262 * **New "Create" menu**: Separated object creation functionality from the "File" menu, placed between "File" and "Edit" menus for clearer organization
263 * All "File > New [...]" actions moved to "Create" menu
264 * **Migration note**: Find signal/image creation actions in the new "Create" menu
265 * **New "ROI" menu**: Dedicated menu for Region of Interest management, positioned between "Edit" and "Operations" menus
266 * **New "Annotations" submenu**: Consolidated annotation operations in the "Edit" menu
267 * **New "Metadata" submenu**: Grouped all metadata operations in the "Edit" menu
268
269 **Interactive object editing:**
270
271 * **Interactive object creation**: Creation parameters can be modified after object creation via new "Creation" tab in properties panel
272 * Apply changes without creating new objects, preserving subsequent processing
273 * Available for parametric generators (Gaussian, sine, etc.)
274 * **Interactive 1-to-1 processing**: Processing parameters can be adjusted and reapplied via new "Processing" tab
275 * Update result objects in-place with modified parameters
276 * Only for parametric operations (filters, morphology, etc.)
277 * **Recompute feature**: New "Recompute" action (Ctrl+R) to quickly reprocess objects with stored parameters
278 * Works with single or multiple objects
279 * Automatically updates results when source data changes
280 * **Automatic ROI analysis update**: Analysis results automatically recalculate when ROI is modified
281 * Works for all analysis operations (statistics, centroid, etc.)
282 * Silent background recomputation for immediate feedback
283 * **Select source objects**: Navigate to source objects used in processing via new "Edit" menu action
284 * Handles all processing patterns (1-to-1, 2-to-1, n-to-1)
285 * Shows informative messages if sources no longer exist
286 * **Processing history**: New "History" tab displays object processing lineage as hierarchical tree
287 * Shows complete processing chain from creation to current state
288 * Selectable text for documentation purposes
289
290 **Multi-object property editing:**
291
292 * Apply property changes to multiple selected objects simultaneously
293 * Only modified properties are applied, preserving unchanged individual settings
294 * Typical use case: Adjust LUT boundaries or colormap for multiple images at once
295
296 **Dialog sizing improvements:**
297
298 * Processing dialogs now intelligently resize based on main window size
299 * Never exceed main window dimensions for better user experience
300
301 **Internal console indicator:**
302
303 * Status bar indicator shows console status when hidden
304 * Turns red on errors/warnings to alert users
305 * Click to open console
306
307 ### 🎬 New Object Creation Features ###
308
309 **Parametric signal generators:**
310
311 * Linear chirp, logistic function, Planck function
312 * Generate signals with Poisson noise
313
314 **Parametric image generators:**
315
316 * **Checkerboard**: Calibration pattern with configurable square size and light/dark values
317 * **Sinusoidal grating**: Frequency response testing with independent X/Y spatial frequencies
318 * **Ring pattern**: Concentric circular rings for radial analysis and PSF testing
319 * **Siemens star**: Resolution testing with radial spokes and configurable radius
320 * **2D sinc**: Cardinal sine function for PSF modeling and diffraction simulation
321 * **2D ramp**: New ramp image generator
322 * **Poisson noise**: Generate images with Poisson noise distribution
323
324 **Signal/image creation from operations:**
325
326 * Create complex-valued signal/image from real and imaginary parts
327 * Create complex-valued signal/image from magnitude and phase (closes [Issue #216](https://github.com/DataLab-Platform/DataLab/issues/216))
328 * Extract phase (argument) information from complex signals or images
329
330 ### πŸ”¬ Data Processing & Analysis ###
331
332 **Signal processing:**
333
334 * **Enhanced curve fitting**: Significantly improved parameter estimation for all curve types (Gaussian, Lorentzian, Voigt, exponential, sinusoidal, Planckian, asymmetric peaks, CDF)
335 * Smarter initial parameter guesses for robust convergence
336 * **Locked parameter support**: Lock individual parameters during optimization (requires PlotPy v2.8.0)
337 * **X-array compatibility checking**: Comprehensive validation for multi-signal operations with automatic interpolation options
338 * New configuration setting: Ask user or interpolate automatically
339 * Prevents unexpected results from incompatible signal arrays
340 * **Zero padding enhancements**: Support for prepending and appending zeros, default strategy now "Next power of 2"
341 * **Ideal frequency domain filter**: "Brick wall filter" for signals (closes [Issue #215](https://github.com/DataLab-Platform/DataLab/issues/215))
342 * **Bandwidth at -3dB**: Enhanced to support passband bandwidth
343 * **Pulse features extraction**: Comprehensive pulse analysis for step and square signals
344 * Automated shape recognition and polarity detection
345 * Measures amplitude, rise/fall time, FWHM, timing parameters, baseline ranges
346 * **Frequency domain filters**: Deconvolution and Gaussian filter (closes [Issue #189](https://github.com/DataLab-Platform/DataLab/issues/189), [Issue #205](https://github.com/DataLab-Platform/DataLab/issues/205))
347 * **Coordinate transformations**: Convert to Cartesian/polar coordinates
348 * **X-Y mode**: Simulate oscilloscope X-Y mode (plot one signal vs. another)
349 * **Find operations**: First abscissa at y=..., ordinate at x=..., full width at y=...
350 * **`1/x` operation**: Reciprocal operation with NaN handling for zero denominators
351
352 **Image processing:**
353
354 * **2D resampling**: Resample images to new coordinate grids with multiple interpolation methods (closes [Issue #208](https://github.com/DataLab-Platform/DataLab/issues/208))
355 * **Convolution**: 2D convolution operation
356 * **Erase area**: Erase image areas defined by ROI (closes [Issue #204](https://github.com/DataLab-Platform/DataLab/issues/204))
357 * **Horizontal/vertical projections**: Sum of pixels along axes (closes [Issue #209](https://github.com/DataLab-Platform/DataLab/issues/209))
358 * **Improved centroid computation**: More accurate in challenging cases (truncated/asymmetric images) (see [Issue #251](https://github.com/DataLab-Platform/DataLab/issues/251))
359 * **Flip diagonally**: New geometric transformation
360 * **`1/x` operation**: Reciprocal operation for images
361
362 **Cross-panel operations:**
363
364 * **Signals to image conversion**: Combine multiple signals into 2D images
365 * Two orientation modes: as rows (spectrograms) or columns (waterfall displays)
366 * Optional normalization (Z-score, Min-Max, Maximum)
367 * Typical use cases: heatmaps, spectrograms, multi-channel data visualization
368
369 **Common features:**
370
371 * **Standard deviation**: Calculate across multiple signals/images (closes [Issue #196](https://github.com/DataLab-Platform/DataLab/issues/196))
372 * **Add noise**: Add Gaussian, Poisson, or uniform noise (closes [Issue #201](https://github.com/DataLab-Platform/DataLab/issues/201))
373 * **Add metadata**: Add custom metadata with pattern support (`{title}`, `{index}`, etc.) and type conversion
374
375 ### πŸ“ ROI & Annotation Management ###
376
377 **ROI features:**
378
379 * **ROI clipboard operations**: Copy/paste ROIs between objects
380 * **ROI import/export**: Save/load ROIs as JSON files
381 * **Individual ROI removal**: Remove ROIs selectively via "Remove" submenu
382 * **ROI title editing**: Set titles during interactive creation and in confirmation dialog
383 * **Create ROI grid**: Generate grid of ROIs with configurable rows, columns, and spacing
384 * Import/export grid configurations
385 * Preview before creation
386 * **Inverse ROI logic**: Select area outside defined shapes (images only)
387 * **Coordinate-based ROI creation**: Manual input of coordinates for rectangular and circular ROIs
388 * **Multi-object ROI editing**: Edit ROIs on multiple objects simultaneously
389
390 **Annotation features:**
391
392 * **Annotation clipboard**: Copy/paste annotations between objects
393 * **Edit annotations**: Interactive editor dialog with PlotPy tools
394 * **Import/export annotations**: Save/load as .dlabann JSON files with versioning
395 * **Delete annotations**: Remove from single or multiple objects with confirmation
396 * **Annotations independent from ROI**: Can coexist on same object
397
398 **Detection with ROI creation:**
399
400 * All 7 detection algorithms now support automatic ROI creation:
401 * Peak detection, contour shape, blob detection (DOG/DOH/LOG/OpenCV), Hough circle
402 * **ROI geometry choice**: Rectangular or circular ROIs
403 * **2D peak detection**: Option to choose ROI geometry (closes related requirements)
404
405 ### πŸ“Š Visualization & Display ###
406
407 **Performance & display limits:**
408
409 * **Configurable result display limits**: Prevent UI freezing with large result sets
410 * `max_shapes_to_draw` (default: 1,000), `max_cells_in_label` (default: 100), `max_cols_in_label` (default: 15)
411 * Settings documented with performance implications
412 * **Faster contour rendering**: Over 5x performance improvement for contour display
413 * **Signal rendering optimization**: Smart linewidth clamping for large datasets
414 * New setting: "Line width performance threshold" (default: 1,000 points)
415 * Prevents 10x slowdown from Qt raster engine limitation
416
417 **Result visualization:**
418
419 * **Merged result labels**: All analysis results consolidated in single read-only label
420 * Reduces visual clutter, auto-updates, horizontally divided results
421 * **Result label visibility control**: Toggle visibility via Properties panel checkbox
422 * Default visibility configurable in Settings
423 * **Results group organization**: Plot results automatically organized in dedicated "Results" group
424 * **Comprehensive result titles**: Include source object identifiers (e.g., "FWHM (s001, s002, s003)")
425 * **Individual result deletion**: Remove results selectively via Analysis menu
426 * **Customizable shape/marker styles**: Four new style configuration buttons in Settings
427 * Separate styles for signals and images
428 * Interactive editor with comprehensive options
429 * Persistent configuration with refresh on change
430
431 **Enhanced profile extraction:**
432
433 * Click directly on X/Y profile plots to switch extraction direction
434 * No need to open parameters dialog for direction changes
435 * Improves workflow efficiency (closes [Issue #156](https://github.com/DataLab-Platform/DataLab/issues/156))
436
437 **DateTime signal support:**
438
439 * Automatic datetime detection in CSV files
440 * **Datetime axis formatting**: Human-readable timestamps on X-axis
441 * **Configurable formats**: Separate formats for standard and sub-second units
442 * Supports various time units (seconds, milliseconds, microseconds, minutes, hours)
443 * Closes [Issue #258](https://github.com/DataLab-Platform/DataLab/issues/258)
444
445 **Settings:**
446
447 * **Autoscale margins**: Configurable margins (0-50%) for signal/image plots
448 * **Lock image aspect ratio**: Option for 1:1 aspect ratio (default: use physical pixel size) (closes [Issue #244](https://github.com/DataLab-Platform/DataLab/issues/244))
449 * **Show console on error**: Configurable behavior (default: off)
450
451 **Image extent parameters:**
452
453 * New "Extent" group box showing computed Xmin, Xmax, Ymin, Ymax
454 * Automatically calculated from origin, pixel spacing, and dimensions
455
456 ### πŸ“ Import/Export & File Handling ###
457
458 **New file format support:**
459
460 * **FT-Lab signals and images**: CEA binary formats (.sig, .ima) (closes [Issue #211](https://github.com/DataLab-Platform/DataLab/issues/211))
461 * **Coordinated text files**: Real and complex-valued images with error images (similar to Matris format)
462 * Automatic NaN handling, metadata with units and labels
463
464 **Enhanced HDF5 support:**
465
466 * **Custom file extensions**: Intelligent HDF5 detection by content (not just extension)
467 * Extension-based detection for dialogs, content-based for drag-and-drop
468 * "All files (*)" option in file dialogs
469 * **HDF5 Browser improvements**: Default collapsed tree view for better navigation
470 * **Workspace clearing options**: Configurable behavior with "Ignore" option (closes [Issue #146](https://github.com/DataLab-Platform/DataLab/issues/146))
471
472 **Text file improvements:**
473
474 * **CSV delimiter handling**: Better support for various whitespace separators
475 * **Locale decimal separator**: Support for comma as decimal separator (closes [Issue #124](https://github.com/DataLab-Platform/DataLab/issues/124))
476 * **Encoding error tolerance**: Ignore errors for files with special characters
477 * **Header detection**: Automatic detection and skipping of data headers
478
479 **Other I/O features:**
480
481 * **Save to directory**: New feature (closes [Issue #227](https://github.com/DataLab-Platform/DataLab/issues/227))
482 * **Open from directory**: Recursively open multiple files with folder drag-and-drop support
483 * **File ordering**: Consistent alphabetical sorting across platforms
484
485 ### πŸ”§ Advanced Features ###
486
487 **Non-uniform coordinate support:**
488
489 * Images now support non-uniform pixel spacing
490 * "Set uniform coordinates" feature for conversion
491 * **Polynomial calibration**: Up to cubic order for X, Y, Z axes
492 * Creates non-uniform coordinates for X/Y, transforms values for Z
493 * HDF5 and text file formats preserve non-uniform information
494
495 **Group management:**
496
497 * **Panel-specific short IDs**: `gs` prefix for signals, `gi` prefix for images
498 * Avoids ambiguity in cross-panel operations
499 * Fixed group numbering for new groups
500
501 **Configuration:**
502
503 * **Version-specific folders**: Major version coexistence (`.DataLab_v1`, `.DataLab_v2`, etc.)
504 * Allows v0.x and v1.x to run simultaneously
505
506 **Public API & Remote Control:**
507
508 * Enhanced metadata handling with function name context
509 * `add_group`, `add_signal`, `add_image` methods with `group_id` and `set_current` arguments
510 * `get_object_uuids` with optional `group` filter
511 * Multiple API improvements for better programmability
512
513 **Processing infrastructure (for developers):**
514
515 * New `@computation_function` decorator for computation functions
516 * Renamed computation functions (removed "compute_" prefix)
517 * Refactored `BaseProcessor` methods with clear naming:
518 * `compute_1_to_1`, `compute_1_to_0`, `compute_1_to_n`, `compute_n_to_1`, `compute_2_to_1`
519 * **No backward compatibility maintained** for these internal changes (closes [Issue #180](https://github.com/DataLab-Platform/DataLab/issues/180))
520
521 ### πŸ› οΈ Bug Fixes since version 0.20 ###
522
523 **Performance fixes:**
524
525 * **Switching between images with many results**: Dramatic improvement (66s β†’ <1ms) when navigating images with hundreds of shapes
526 * **Plot cleanup robustness**: Fixed errors when removing analysis results
527 * **Critical fix**: Result labels now properly excluded from cleanup to prevent disappearance
528
529 **Cross-panel & group handling:**
530
531 * **Cross-panel computation groups**: Fixed inconsistent group organization for image-to-signal operations
532 * **File import ordering**: Consistent alphabetical sorting across all platforms
533
534 **Action state updates:**
535
536 * Fixed action enable states not updating after annotation/metadata operations
537 * UI now immediately reflects current object state
538
539 **Result management:**
540
541 * Fixed result label deletion to permanently remove associated metadata
542 * Fixed duplicate results parameter metadata cleanup
543 * **Result coordinate fixes**: Corrected shifted results on images with ROIs and shifted origin ([Issue #106](https://github.com/DataLab-Platform/DataLab/issues/106))
544 * **Profile extraction indices**: Fixed wrong indices with ROI ([Issue #107](https://github.com/DataLab-Platform/DataLab/issues/107))
545
546 **ROI-related fixes:**
547
548 * Fixed multi-image ROI extraction not saving ROI in first object ([Issue #120](https://github.com/DataLab-Platform/DataLab/issues/120))
549 * Fixed AttributeError when extracting multiple ROIs on single image with multiple selections ([Issue #121](https://github.com/DataLab-Platform/DataLab/issues/121))
550 * Fixed mask refresh issues ([Issue #122](https://github.com/DataLab-Platform/DataLab/issues/122), [Issue #123](https://github.com/DataLab-Platform/DataLab/issues/123))
551 * Fixed ROI editor on multiple signals/images ([Issue #135](https://github.com/DataLab-Platform/DataLab/issues/135))
552 * Fixed ROI clearing affecting only first image ([Issue #160](https://github.com/DataLab-Platform/DataLab/issues/160))
553 * Fixed ROI editor showing first instead of last image ([Issue #158](https://github.com/DataLab-Platform/DataLab/issues/158))
554
555 **Text Import Wizard:**
556
557 * Fixed preservation of user-defined titles and units ([Issue #239](https://github.com/DataLab-Platform/DataLab/issues/239))
558 * Fixed preservation of data types ([Issue #240](https://github.com/DataLab-Platform/DataLab/issues/240))
559 * Fixed comma decimal separator support ([Issue #186](https://github.com/DataLab-Platform/DataLab/issues/186))
560 * Fixed trailing delimiter issue ([Issue #238](https://github.com/DataLab-Platform/DataLab/issues/238))
561
562 **Curve fitting & analysis:**
563
564 * Improved initial frequency estimate for sinusoidal fitting
565 * Fixed parameter display formatting for extreme values
566 * Fixed FWHM computation exception handling
567 * Fixed curve marker style changing unexpectedly ([Issue #184](https://github.com/DataLab-Platform/DataLab/issues/184))
568 * Fixed hard crash on zero signal with curve stats tool ([Issue #233](https://github.com/DataLab-Platform/DataLab/issues/233))
569
570 **Image handling:**
571
572 * Fixed aspect ratio not updating when switching images
573 * Fixed shape unpacking issues ([Issue #246](https://github.com/DataLab-Platform/DataLab/issues/246), [Issue #247](https://github.com/DataLab-Platform/DataLab/issues/247))
574 * Fixed colormaps not stored in metadata (PlotPy v2.6.3+ issue) ([Issue #138](https://github.com/DataLab-Platform/DataLab/issues/138))
575 * Fixed amplitude calculation for non-integer data types
576
577 **Signal processing:**
578
579 * Fixed ifft1d x-axis computation when shift=False ([Issue #241](https://github.com/DataLab-Platform/DataLab/issues/241))
580 * Fixed moving median crash on Linux with mirror mode ([Issue #117](https://github.com/DataLab-Platform/DataLab/issues/117)) - SciPy bug
581 * Fixed magnitude spectrum with logarithmic scale ([Issue #169](https://github.com/DataLab-Platform/DataLab/issues/169))
582 * Fixed pairwise operation mode for asymmetric functions ([Issue #157](https://github.com/DataLab-Platform/DataLab/issues/157))
583
584 **Analysis & results:**
585
586 * Fixed NaN value handling in statistics and normalization ([Issue #141](https://github.com/DataLab-Platform/DataLab/issues/141), [Issue #152](https://github.com/DataLab-Platform/DataLab/issues/152), [Issue #153](https://github.com/DataLab-Platform/DataLab/issues/153))
587 * Fixed analysis results kept from original after processing ([Issue #136](https://github.com/DataLab-Platform/DataLab/issues/136))
588 * Fixed duplicate results with no ROI defined
589 * Fixed "One curve per result title" mode ignoring ROIs ([Issue #132](https://github.com/DataLab-Platform/DataLab/issues/132))
590 * Added result validation for array-like results
591
592 **Plot & visualization:**
593
594 * Fixed profile plots not refreshing when moving/resizing ([Issue #172](https://github.com/DataLab-Platform/DataLab/issues/172)) - PlotPy fix
595 * Fixed empty average profile display outside image area ([Issue #168](https://github.com/DataLab-Platform/DataLab/issues/168)) - PlotPy fix
596 * Fixed average profile extraction ValueError with oversized rectangle ([Issue #144](https://github.com/DataLab-Platform/DataLab/issues/144))
597 * Disabled generic "Axes" tab in parameter dialogs
598
599 **Other fixes:**
600
601 * Fixed proxy `add_object` method not supporting metadata ([Issue #111](https://github.com/DataLab-Platform/DataLab/issues/111))
602 * Fixed RemoteClient method calls without optional arguments ([Issue #113](https://github.com/DataLab-Platform/DataLab/issues/113))
603 * Fixed KeyError when removing group after opening HDF5 ([Issue #116](https://github.com/DataLab-Platform/DataLab/issues/116))
604 * Fixed KeyError in "View in new window" with multiple images after HDF5 open ([Issue #159](https://github.com/DataLab-Platform/DataLab/issues/159))
605 * Fixed long object titles display ([Issue #128](https://github.com/DataLab-Platform/DataLab/issues/128))
606 * Fixed file name titles showing relative paths ([Issue #165](https://github.com/DataLab-Platform/DataLab/issues/165))
607 * Fixed unexpected group names in "Open from directory" ([Issue #177](https://github.com/DataLab-Platform/DataLab/issues/177))
608 * Fixed one group per folder expectation ([Issue #163](https://github.com/DataLab-Platform/DataLab/issues/163))
609 * Fixed unsupported files in recursive loading ([Issue #164](https://github.com/DataLab-Platform/DataLab/issues/164))
610
611 **Removed features:**
612
613 * Removed "Use reference image LUT range" setting (confusing behavior)
614 * **Migration**: Use multi-selection property editing instead
615
616 ### πŸ”’ Security Fixes ###
617
618 * **CVE-2023-4863**: Fixed vulnerability in opencv-python-headless
619 * Updated minimum requirement from 4.5.4.60 to 4.8.1.78
620 * See [DataLab security advisory](https://github.com/DataLab-Platform/DataLab/security/dependabot/1)
621
622 ### ℹ️ Other Changes ###
623
624 * Bumped minimum `plotpy` requirement to V2.8
625 * Bumped minimum `guidata` requirement to V3.13
626 * Using new `guidata` translation utility based on `babel`
627 * Python 3.13 now supported (via scikit-image V0.25)