USER_GUIDE.md

Foam Dictionary Editor (FoDE)

FoDE — Foam Dictionary Editor (pronounced "foh-dee")

What is FoDE?

FoDE is a graphical editor for OpenFOAM case dictionary files. It lets you browse, edit, and manage dictionaries through a structured tree view or a plain-text editor — whichever suits the task. It is aimed at engineers and researchers who run OpenFOAM simulations and want a more convenient way to set up and modify case files.

About This Guide

This is the full feature reference for FoDE. It covers every panel, menu, dialog, and workflow in detail. For installation and a quick-start walkthrough, see README.md. For project structure, dev setup, and testing, see DEVELOPER.md.

Where to Find Things

I want to…	Section
Browse and manage case files	File list behavior
Add non-default files, whole directories, or result time steps to the file list	Adding extra directories
Open or reload a case	Reloading a case
Save or duplicate a case	Duplicating a case / Saving as a new case
Back up a file, or clean up old backups	Creating a backup / Cleaning up backup files
Compare two cases side-by-side	Case comparison
Use tutorial or template cases	Case Library
Try FoDE with a ready-made example case	Bundled example cases
Work with a multiRegion (CHT) case	MultiRegion cases
Understand tree ↔ text sync	Tree and text workflow
Edit values in the tree	Tree view context menu
Get help on a dictionary setting	Detail pane
View all boundary conditions at a glance without opening each field file	Boundary view
Delete or add a boundary condition across all field files at once	Deleting / Adding a boundary condition across all field files
Rename a boundary patch across blockMeshDict and all field files	Renaming a boundary patch
Copy the boundary table into a report (Markdown/CSV)	Copy Table
View blockMeshDict geometry in 3-D	BlockMesh panel
View topoSetDict geometry in 3-D	topoSetDict overlay
View snappyHexMeshDict surfaces/regions in 3-D	snappyHexMeshDict overlay
View setFieldsDict regions in 3-D	setFieldsDict overlay
See the tree and the 3-D view at the same time	Side-by-side mode
Tweak vertex coordinates and see the 3-D effect instantly	Vertices table
Export topoSet/snappyHexMesh/setFields shapes as STL, or overlay an STL/OBJ file	Geometry controls
View the generated mesh in ParaView	Open Mesh in ParaView
Filter tree keys	Tree key filter
Jump to a tree entry from the editor	Find in Tree
Find text in the editor	Editor toolbar
View keyboard shortcuts	Help > Keyboard Shortcuts…
Run OpenFOAM commands from FoDE	Terminal tab
Check whether the mesh is up to date	Mesh presence/staleness indicator
Run blockMesh or restore 0/ without leaving FoDE	Run blockMesh / Restore 0/ from 0.orig
Run snappyHexMesh or topoSet without leaving FoDE	Run snappyHexMesh / Run topoSet
Set initial field regions, or validate the mesh	Run setFields / Run checkMesh
Run the case's whole workflow, or clean it back to a pristine state	Run Allrun Script / Run Allclean Script / Clean Case
Plot residuals while a solver is running	foamMonitor launcher
Read a long log.snappyHexMesh log without scrolling through thousands of lines	View Log Summary
See how a keyword or function object is used in real cases	Find OpenFOAM Examples
Choose a variant (no-terminal, BlockMesh)	Variants
Add schema help for custom keys	Schema module configuration
Match syntax highlighting to my OpenFOAM version	Generate OpenFOAM Keywords
Configure application settings	Application settings
Switch the UI language (English / 日本語)	Settings > Language
Open help links or reference sites	Resources dialog
See annotated screenshots of the app	Screenshot gallery

Contents

Getting started

• Features
• UI layout
• Current case and file display

File & case management

• File list behavior
  • Directory header markers
  • Default target files
  • MultiRegion cases
  • Symbolic links
  • Adding files at runtime
  • Adding extra directories
  • Extra files and directories indicator
  • Removing extra files and directories
  • Numeric time directory indicator
  • Mesh presence/staleness indicator
  • Creating a backup
  • Creating a new file
  • Cleaning up backup files
  • Deleting a file
  • Duplicating a file
  • Duplicating a field directory
  • Deleting a field directory
  • Resetting the file list
• Case Library
• Bundled example cases
• Case comparison
• Reloading a case
• Saving as a new case
• Duplicating a case

Editing

• Tree and text workflow
• Tree view context menu
• Detail pane
• Boundary view
  • Table layout
  • Directory selector
  • Transpose
  • Single-click navigation
  • Editing a boundary condition
  • Creating a patch entry
  • Deleting a patch entry
  • Copy and paste
  • Copy Table
  • Renaming a boundary patch
  • Deleting a boundary condition across all field files
  • Adding a boundary condition across all field files
• BlockMesh panel
  • Side-by-side mode
  • Variable resolution
  • Geometry controls
  • topoSetDict overlay
  • snappyHexMeshDict overlay
  • setFieldsDict overlay
  • Overlay clipping
  • Vertices table
  • Mouse controls
• Behavior on parse failure

Navigation & search

• Tree key filter and editor sync
  • Tree key filter
  • Editor sync (Auto-scroll editor)
  • Find in Tree (editor → tree)
• Editor behavior

Configuration

• Terminal tab
• Variants
• Schema module configuration
• Generate OpenFOAM Keywords
• Application settings
• Resetting settings
• Resources dialog

Tools

• foamMonitor launcher
• Restore 0/ from 0.orig
• Run blockMesh
• Run snappyHexMesh
• Run topoSet
• Run setFields
• Run checkMesh
• Run Allrun Script
• Open Mesh in ParaView
• Run Allclean Script
• Clean Case (foamCleanTutorials)
• View Log Summary
• Find OpenFOAM Examples

Reference

• Supported syntax and node types
• Limitations
• Disclaimer
• Acknowledgements

---

Features

• Open an OpenFOAM case directory and list common dictionary files from the case.
• Load and edit a wide range of common dictionary files including controlDict, fvSchemes, fvSolution, blockMeshDict, snappyHexMeshDict, transportProperties, and many more when present.
• Automatically list all files found under the 0 and 0.orig field directories.
• Automatically detect multiRegion case structures (e.g. chtMultiRegionFoam, chtMultiRegionSimpleFoam) and list per-region files under separate group headers such as system/fluid and constant/heater.
• Automatically collect phase-variant files such as thermophysicalProperties.air and turbulenceProperties.water that are present in constant/ or per-region constant directories.
• Automatically list the case-root All* scripts (Allrun, Allrun.pre, Allclean, …) under a case root group so the scripts run from the Tools menu can be inspected and edited; they open as text with shell syntax highlighting (no tree view), keep their executable permission when saved, and are included when the case is duplicated. Other case-root files (log.*, results, *.foam) stay hidden.
• Display symbolic links in the file list with a ⇢ marker and italic text; hovering shows the link target in the tooltip.
• Add files not in the default list to the file panel at runtime via right-click on a directory header.
• Add entire directories to the file list so that all files inside are scanned automatically, just like 0/ and 0.orig/. Each directory can be scanned flat (direct files only, default) or recursively (all subdirectories). Useful for custom field directories (initial/), restart time steps (0.5/), or deep subdirectories (lagrangian/chemkin/) and cases that include a validation/ tree. Extra directories are shown with a distinct (purple) header. Right-click the Results indicator to add a numeric time directory directly; use Settings > Manage Extra Files & Directories… or the indicator button for arbitrary directories.
• Create a new file from a FoamFile template in any directory group via right-click on the directory header.
• Save per-case extra file and directory selections to .foam-editor-files.json inside the case directory.
• Show an indicator at the top of the file list when extra files or directories are registered for the current case, with a link to the management dialog.
• Remove extra files individually via right-click, or manage files and directories in bulk via the management dialog.
• Create a timestamped backup of any file via right-click on a file in the file list.
• Delete any file from disk via right-click on a file in the file list. A confirmation dialog offers three choices: create a backup and delete, delete without a backup, or cancel. The deletion cannot be undone.
• Clean up accumulated backup files for the current case via Case > Clean Backup Files.... A dialog lists all .bak_* files found in the case directory with checkboxes (all selected by default) and Select All / Deselect All buttons for quick bulk selection.
• Duplicate any file in the file list via right-click; a dialog prompts for a new name and the object field in the FoamFile header is updated automatically to the new name. If the source file has unsaved changes, the application asks whether to save first or duplicate with the unsaved state.
• Duplicate the 0/ or 0.orig/ directory to create the missing counterpart via right-click on the group header (shown only when exactly one of the two directories exists).
• Delete the 0/ directory from disk via right-click on the 0 group header. Shown only when 0.orig exists. A confirmation dialog shows the full path; any open files inside 0/ are closed and the deletion cannot be undone.
• View all boundary conditions across field variables at a glance in the Boundary tab. Rows are field files and columns are patches by default; a Transpose checkbox swaps the orientation. Double-click a cell to edit its content, or to create a new entry for a – cell. Right-click any cell to edit, create, delete, copy, paste, or rename a boundary condition. Right-click a patch column/row header to rename, delete, or add a boundary condition across all field files. Use the Copy Table button to copy the entire table to the clipboard as Markdown or CSV.
• Show the parsed dictionary structure in a tree view. Filter visible keys by typing in the filter bar above the tree; matching is case-insensitive and recursive (parent nodes remain visible when a descendant matches).
• Highlight unknown_raw_entry nodes (entries the parser could not fully interpret) in amber so they stand out from normal entries.
• Indicate the presence of numeric time directories (OpenFOAM calculation results such as 0.5, 1, 10) with a bold Results: … row at the bottom of the file list. The indicator shows up to six directory names and notes the total count when more exist. Hovering shows the full count and range. Right-clicking the indicator offers a submenu to add any listed directory to the file list as an extra directory.
• Hide the Type column in the tree by default; re-enable it with View > Show Type Column.
• View rich schema information for each setting in the right-side detail pane: description, available choices, applicable solver versions, and notes. Schema annotations are provided for a subset of common settings; not all keys are covered.
• Edit node values directly from the detail pane. When a fractional value is entered for an int node, the node type is automatically promoted to scalar.
• Customize or extend schema information by writing your own schema modules.
• Edit raw file text directly in the lower text editor.
• Show line numbers and the current line position in the editor.
• Rebuild raw text from the parsed tree with Reload from Tree.
• Rebuild the tree from edited text with Apply Text to Tree.
• Save the current file from the raw text editor with Save File.
• Save all modified files at once with Save Case.
• Reload the current case from disk with the Reload Case button in the top bar or Case > Reload Case, discarding all in-memory edits. If there are unsaved changes, a confirmation dialog shows the number of affected files before proceeding.
• Duplicate the open case to a new directory with Case > Duplicate Case, choosing between a full directory copy or a copy of only the app-visible files.
• Save the currently open case as a new case with Case > Save as New Case..., which copies files from disk (all files or app-visible files only, selectable) and then writes any unsaved in-memory edits on top, then switches to the new case.
• Register reference case directories in the Case Library for quick access. The $FOAM_TUTORIALS directory is included automatically whenever the environment variable is set.
• Open a case directly from any Case Library directory with Case > Open from Case Library....
• Duplicate a case from the Case Library into the working directory with Case > Duplicate from Case Library..., with the Default Case Directory pre-filled as the destination parent.
• Show a warning when the selected directory contains neither system/ nor constant/, suggesting it may not be a valid OpenFOAM case, with an option to open it anyway.
• Open a case by dragging a directory from the file manager and dropping it anywhere on the application window. If there are unsaved changes, a confirmation dialog is shown first.
• Add, duplicate, comment out, restore, or delete tree nodes via the right-click context menu in the tree view.
• Copy and paste node values in the tree view via right-click context menu or Ctrl+C / Ctrl+V.
• Run shell commands in the integrated Terminal tab at the bottom of the window. The terminal automatically changes to the case directory when a case is opened.
• Undo, redo, cut, copy, paste, select all, and find text via the edit toolbar.
• Keep the text editor usable even if parsing fails.
• Configure schema modules and application settings at runtime via the Settings menu.
• Open Help > Resources... to access official OpenFOAM documentation links and manage personal reference links. The My Links tab lets you add, edit, reorder, and remove links; double-click any entry to open it in your browser. Links are saved to app_config.json.

Current UI layout

The main window is divided into two top-level columns separated by a horizontal splitter.

• Left column — file list for the selected OpenFOAM case (full window height).
• Right column — a vertical splitter with two rows:
  • Upper row — a tab widget with up to three tabs:
    • Tree tab — parsed dictionary tree (center) and detail editor (right). A ⊞ button in the tab bar's top-right corner activates side-by-side mode when blockMeshDict, topoSetDict, snappyHexMeshDict, or setFieldsDict is the active file (see BlockMesh panel — Side-by-side mode).
    • Boundary tab — boundary condition table for all field variables (see Boundary view).
    • BlockMesh tab — interactive 3-D viewer for blockMeshDict geometry (see BlockMesh panel). Visible only when the Simple terminal is active; can also be toggled via View > BlockMesh 3-D Panel.
  • Lower row — tabbed panel with two tabs:
    • Editor — plain-text editor with line numbers.
    • Terminal — integrated terminal with a mode toggle (see Terminal tab below).

The top action bar contains the frequently used commands and the current case and file display.

• Save File.
• Save Case.
• Apply Text to Tree.
• Reload from Tree.
• Case: current case name display.
• File: current file name display.

The Editor tab has its own toolbar row with text search operations.

• Find — opens a search dialog (Ctrl+F).
• Find Prev — wraps to the previous match (Shift+F3).
• Find Next — advances to the next match (F3).
• Find in Tree — selects the deepest tree node whose source span covers the current cursor line (Ctrl+Shift+T).
• Highlight — toggleable button; click to turn syntax highlighting on or off. The setting is saved to app_config.json and restored on the next launch.
• Line: N — current cursor line number (right side of the toolbar).

The menu bar provides a Case menu, a View menu, a Settings menu, a Tools menu, and a Help menu.

Case menu:

• Case > Open Case Ctrl+O.
• Case > Open from Case Library...
• Drag and drop: drag a case directory from your file manager onto any part of the application window to open it. If there are unsaved changes, a confirmation dialog is shown first.
• Case > Reload Case.
• Case > Duplicate Case...
• Case > Duplicate from Case Library...
• Case > Find OpenFOAM Examples… — the same action as Tools > Find OpenFOAM Examples…, listed here too because it can duplicate a tutorial case as the starting point for a new case. See Find OpenFOAM Examples.
• Case > Save as New Case...
• Case > Clean Backup Files...
• Case > Compare with Case...
• Case > Save File Ctrl+S.
• Case > Save Case Ctrl+Shift+S.
• Case > Exit Ctrl+Q.

View menu:

• View > Show Type Column (checkable; hidden by default).
• View > BlockMesh 3-D Panel (checkable; shows or hides the BlockMesh tab. Grayed out — with the label "BlockMesh 3-D Panel (unavailable: xterm active)" — while xterm terminal mode is active due to the GPU conflict).
• View > View Log Summary… — the same action as Tools > View Log Summary…, listed here too for discoverability. See View Log Summary.

Settings menu:

• Settings > Set Default Case Directory.
• Settings > Manage Case Library…
• Settings > Manage Extra Files & Directories…
• Settings > Reset File List.
• Settings > Manage Schema Modules.
• Settings > Reset Window Size.
• Settings > Reset All Settings…
• Settings > Language — select the UI language (English / 日本語). Takes effect after restarting the application.
• Settings > Generate OpenFOAM Keywords… — rebuild the syntax-highlighter keyword list from a selected OpenFOAM installation. See Generate OpenFOAM Keywords.

Tools menu:

• Tools > foamMonitor… — launch foamMonitor to plot residuals or other time-series data with gnuplot. See foamMonitor launcher.
• Tools > Restore 0/ from 0.orig — recreate 0/ from 0.orig/, discarding any edits made directly to 0/. See Restore 0/ from 0.orig.
• Tools > Run blockMesh… — choose options and run blockMesh in the case directory. See Run blockMesh.
• Tools > Run snappyHexMesh… — choose options and run snappyHexMesh in the case directory. See Run snappyHexMesh.
• Tools > Run topoSet… — choose options and run topoSet in the case directory. See Run topoSet.
• Tools > Run setFields… — choose options and run setFields in the case directory, offering to restore 0/ from 0.orig/ first. See Run setFields.
• Tools > Run checkMesh… — choose options and run checkMesh in the case directory. See Run checkMesh.
• Tools > Run Allrun Script — run the case's ./Allrun script, including the solver. See Run Allrun Script.
• Tools > Open Mesh in ParaView… — open the current case in ParaView. See Open Mesh in ParaView.
• Tools > Run Allclean Script — run the case's ./Allclean script. See Run Allclean Script.
• Tools > Clean Case (foamCleanTutorials) — clean the case with foamCleanTutorials. See Clean Case (foamCleanTutorials).
• Tools > View Log Summary… — show a condensed summary of a log.* file. See View Log Summary.
• Tools > Find OpenFOAM Examples… — search the OpenFOAM tutorials and etc/caseDicts templates for usage examples of a keyword; also listed in the Case menu. See Find OpenFOAM Examples.

Help menu:

• Help > About Foam Dictionary Editor (FoDE)... — shows the application version (with a short dev-build suffix when run from a git checkout), license, and acknowledgements.
• Help > Keyboard Shortcuts...
• Help > Resources...

Current case and file display

The top bar shows the current case as its directory name and the current file as parent-directory/file-name. For example, system/controlDict and constant/transportProperties are shown in that form.

If no case or file is loaded, the labels show -.

File list behavior

The file list is populated by services/case_loader.py. The implementation combines a fixed list of common dictionary files, per-case user-added files, and automatic enumeration of field files.

Directory header markers

Each directory group in the file list is shown as a bold header. A [+] suffix is added to a header when files exist in that directory on disk but are not currently in the file list. No marker is shown when all files in the directory are already listed.

Default target files

system/

• blockMeshDict
• changeDictionaryDict
• controlDict
• createBafflesDict
• createPatchDict
• decomposeParDict
• extrudeMeshDict
• fvOptions
• fvSchemes
• fvSolution
• meshQualityDict
• mirrorMeshDict
• refineMeshDict
• setFieldsDict
• snappyHexMeshDict
• surfaceFeatureExtractDict
• topoSetDict

constant/

• boundaryRadiationProperties
• dynamicMeshDict
• fvOptions
• g
• kinematicCloudProperties
• radiationProperties
• regionProperties
• thermophysicalProperties
• transportProperties
• turbulenceProperties

In addition, all files found directly inside the 0 and 0.orig directories of the case are listed automatically when those directories exist.

Phase-variant files matching thermophysicalProperties.* and turbulenceProperties.* (for example thermophysicalProperties.air) are collected automatically via glob from constant/ and from each per-region constant directory.

Case-root scripts

Script files matching All* directly in the case root (Allrun, Allrun.pre, Allclean, …) are listed automatically under a case root group, shown after all other groups. Since the Tools menu can run these scripts, listing them lets you inspect and edit exactly what will be executed. Other case-root files (log.*, *.foam, result directories) stay hidden by default to keep the list clean, and this group never shows a [+] marker (there are almost always unlisted logs at the root, so the marker would be permanently on). To list specific extra root files — a run log, a README, a custom script — right-click the case root header and choose Add files from 'case root'...; the same menu also offers New file in 'case root'..., consistent with the other directory headers. Hidden files (names starting with .) are never offered.

Scripts are shell files, not OpenFOAM dictionaries: opening one shows it in the text editor only — the tree view stays empty, Apply Text to Tree has no effect, and compare mode skips them. The editor automatically switches to a shell-highlighting mode (detected by the #! shebang): # comments, quoted strings, $variables, OpenFOAM run helpers such as runApplication and runParallel, and utility/solver names from the regular OpenFOAM keyword list are all coloured. Editing and saving work as for any other file, and the executable permission is preserved. The scripts are also included when the case is duplicated or saved as a new case, so the copy remains runnable from the Tools menu.

MultiRegion cases

When subdirectories are found inside system/, the editor treats them as region names and builds additional file groups. The following files are listed per region when present:

system/<region>/

• changeDictionaryDict
• decomposeParDict
• fvOptions
• fvSchemes
• fvSolution
• meshQualityDict

constant/<region>/

• boundaryRadiationProperties
• dynamicMeshDict
• fvOptions
• radiationProperties
• thermophysicalProperties
• turbulenceProperties

Each region directory appears as its own group header in the file list (for example system/fluid, constant/heater). Region groups are sorted after the top-level system and constant groups.

Field files inside 0/<region>/ and 0.orig/<region>/ are also listed automatically, grouped under headers such as 0/fluid and 0/solid. This covers cases like chtMultiRegionFoam where each region has its own initial-condition files (e.g. 0/heater/T, 0/bottomWater/p).

Symbolic links

Symbolic links in the file list are shown with a ⇢ marker appended to the file name and the item text is rendered in italic. Hovering over the item shows a tooltip containing the path and the link target. Editing a symlink writes through to the target file, exactly as the operating system resolves the link.

Adding files at runtime

Right-click any directory header in the file list — including the case root header — to open a context menu with the following actions.

• New file in '<dir>'... — create a new file from a FoamFile template (see Creating a new file).
• Add files from '<dir>'... — add existing files to the file list. A dialog lists all files in that directory that are not yet shown (hidden files starting with . are excluded). Select one or more and click OK to add them.
• Remove '<dir>' from file list — removes an extra directory from the file list. Shown only on extra-directory headers (see Adding extra directories).
• Duplicate '<dir>' → '<counterpart>'... — copy a 0/ or 0.orig/ directory to create the missing counterpart (see Duplicating a field directory). This action is shown only for 0 and 0.orig headers, and only when exactly one of the two directories is present.
• Delete '0' directory... — permanently delete the 0/ directory from disk (see Deleting a field directory). Shown only on the 0 header, and only when 0.orig exists.

Added files are saved to .foam-editor-files.json in the case directory and are restored the next time the case is opened. The config file is also copied when Duplicate Case uses the app-visible files only mode.

Extra files are displayed in a distinct colour in the file list so they can be told apart from the default target files at a glance.

Adding extra directories

Any directory inside the case root can be added to the file list so that all files inside it are scanned and listed automatically, in the same way as 0/ and 0.orig/. This is useful for:

• Custom field directories used by non-standard solvers (e.g. initial/ for laserBeamFoam).
• Restart time steps you want to browse and edit (e.g. 0.5/, 1/).
• Deep subdirectories with supplementary dictionaries (e.g. lagrangian/sprayFoam/aachenBomb/chemkin/).
• Result or validation trees (e.g. validation/) that contain files in subdirectories.

Extra directory group headers are shown in purple to distinguish them from built-in groups. The [+] unlisted-files marker is never shown for extra directories because all files are already loaded.

Hidden files and directories (names starting with ., such as the app's own .foam-editor-files.json or a .git/ tree) are always excluded from extra-directory scans.

Adding a time directory via the Results indicator:
Right-click the Results: … row at the bottom of the file list. A submenu lists each time directory; click one to add it. Directories already added are greyed out.

Adding an arbitrary directory via the management dialog:
Open Settings > Manage Extra Files & Directories… (or click the indicator button at the top of the file list). In the Extra Directories tab, click Add Directory…. A folder picker opens rooted at the case directory; select any subdirectory and click OK. The selected directory is added with flat (non-recursive) scanning by default and the file list refreshes.

Enabling recursive scan:
By default a directory is scanned flat — only files directly inside it are listed. To include files in subdirectories, check the entry in the Extra Directories tab and click Toggle Recursive. The entry label changes to <dir> [recursive] to indicate the mode. Click Toggle Recursive again on the same entry to revert to flat scanning.

Removing an extra directory:
Right-click the (purple) directory header in the file list and select Remove '<dir>' from file list, or use the Extra Directories tab in the management dialog (check the entry, then click Remove Selected).

Files created or duplicated inside an extra directory are not added to the extra-files list — they are already visible because the whole directory is scanned.

Run logs and large non-dictionary files in extra directories:
log.* run logs (e.g. log.blockMesh, log.simpleFoam) appearing via an extra directory are shown dimmed (grey) in the file list and always open as plain text — the dictionary parser is never run on them, so there is no tree view, Apply Text to Tree has no effect, and case comparison skips them (no ≠ marker). Opening a log larger than 100 KB still shows a confirmation dialog first, since loading a very large text file can take a moment. Other large non-dictionary files get the same confirmation dialog and are likewise skipped during diff computation.

Extra files and directories indicator

A Manage extra files… button is always shown at the top of the file list panel when a case is loaded. When extra files or directories are registered it shows the count instead (e.g. Extra files: 2, directories: 1 — Manage…). Clicking it opens the Manage Extra Files & Directories dialog, which has two tabs.

• Extra Files — lists all individually registered extra files. Select one or more and click Remove Selected to remove them.
• Extra Directories — lists all registered extra directories. Entries with recursive scanning enabled are shown as <dir> [recursive]. Click Add Directory… to add a new one (flat by default). Check one or more entries and click Toggle Recursive to flip their scan mode, or Remove Selected to remove them.

Changes take effect immediately when the dialog is closed.

Removing extra files and directories

Extra files:

• Right-click an extra file (shown in blue) in the file list and select Remove from extra files. The file is removed immediately.
• Open the management dialog and use the Extra Files tab to remove one or more files at once.

Extra directories:

• Right-click the purple directory header and select Remove '<dir>' from file list. All files from that directory disappear from the list immediately.
• Open the management dialog and use the Extra Directories tab to remove one or more directories at once.

Removing an extra file or directory from the list does not delete anything from disk. It only removes it from the per-case configuration so it no longer appears in the file list.

Numeric time directory indicator

When numeric directories (OpenFOAM calculation results such as 0.5, 1, 10) are found at the case root — excluding 0, 0.orig, and any directories already added as extra directories — a non-selectable Results: … row is appended at the bottom of the file list. The indicator is rendered in bold gray text and lists up to six directory names. When more than six exist, a (+N more) suffix is added. Hovering shows the total count and, for multiple directories, the first and last in the sorted range.

Right-click the indicator to add any of the listed directories to the file list as an extra directory (see Adding extra directories). Once a directory is added, it no longer appears in the Results indicator and instead shows as a full group header.

Mesh presence/staleness indicator

When constant/polyMesh/owner exists, a non-selectable indicator row is appended near the bottom of the file list (alongside the Results indicator) showing the cell count, e.g. "constant/polyMesh: 8,000 cells". The count is read from the note field in the file's FoamFile header (e.g. nPoints:9261 nCells:8000 nFaces:25200) rather than by parsing the mesh, so the check is cheap. If the count can't be read from the header, the indicator falls back to "constant/polyMesh (present)".

If system/blockMeshDict has been modified more recently than the mesh was generated, the indicator is shown in amber with a " — stale (blockMeshDict changed since last run)" suffix, and the tooltip explains why. Hovering over the indicator always shows the full point/cell/face counts when available.

The indicator updates automatically: it is recomputed on every file-list refresh, including after Tools > Run blockMesh, after saving blockMeshDict (Save File / Save Case), and after any change detected by the file-system watcher — no manual Reload Case is needed.

Creating a backup

Right-click any file in the file list and select Create Backup. A copy of the file is saved in the same directory with a timestamp suffix, for example controlDict.bak_20260502_143022.

The backup captures the current in-memory buffer if the file has been loaded, so unsaved edits are included. If the file has not been loaded yet, the on-disk version is copied. A confirmation message is shown in the status bar, and if the backup includes unsaved edits that note is appended to the message.

Creating a new file

Right-click a directory header in the file list and select New file in '<dir>'.... An input dialog prompts for a file name. The file is created at <case_dir>/<dir>/<name> with a minimal FoamFile dictionary header. The file is opened automatically in the editor after creation.

Files created in directories other than 0/ and 0.orig/ are registered in .foam-editor-files.json so they persist across sessions.

Cleaning up backup files

Select Case > Clean Backup Files... to remove accumulated .bak_* files from the current case. The dialog scans the entire case directory recursively and lists every file whose name matches the .bak_YYYYMMDD_HHMMSS pattern, showing each relative path and file size.

All files are checked by default. Use the checkboxes to select individual files, or use Select All / Deselect All to change the entire selection at once. The Delete Selected (N) button shows the current count of checked files and is disabled when nothing is selected.

Clicking Delete Selected deletes the checked files from disk immediately, with no further confirmation. If any deleted file was open in the editor, the editor and tree view are cleared. If any error occurs (e.g. a permission problem), a summary is shown after the operation.

If no backup files are found, the dialog shows a message and offers only a Close button.

Deleting a file

Right-click any file in the file list and select Delete file.... A confirmation dialog shows the file name and three choices.

• Backup && Delete — creates a timestamped backup in the same directory (for example controlDict.bak_20260506_143022) and then deletes the original. The backup captures the current in-memory buffer, so unsaved edits are included. If the backup write fails, the deletion is aborted.
• Delete — deletes the file immediately without creating a backup.
• Cancel (default) — aborts the operation.

If the file has unsaved changes, the dialog shows an additional warning. The deleted file is removed from the in-memory buffers, the dirty-state tracking, and the per-case extra files configuration. If the deleted file was open in the editor, the editor and tree view are cleared. The file list is refreshed after deletion.

This action deletes the file from disk and cannot be undone.

Duplicating a file

Right-click any file in the file list and select Duplicate.... An input dialog pre-fills the current file name and prompts for a new name. The duplicate is written to the same directory and the object field inside the FoamFile header is updated to match the new file name.

If the source file has unsaved changes, the application presents three choices.

• Save and Duplicate — save the file first, then duplicate the saved content.
• Duplicate with Unsaved Changes — duplicate the current in-memory buffer as-is.
• Cancel — abort the operation.

Files duplicated outside 0/ and 0.orig/ are registered in .foam-editor-files.json.

Duplicating a field directory

Right-click the 0 or 0.orig header in the file list and select Duplicate '<dir>' → '<counterpart>'.... This action copies the entire directory to create the missing counterpart (0/ → 0.orig/ or 0.orig/ → 0/). The action is shown only when exactly one of the two directories is present. A confirmation dialog is displayed before the copy begins.

Deleting a field directory

Right-click the 0 header in the file list and select Delete '0' directory.... This action permanently deletes the entire 0/ directory and all files inside it from disk. The action is shown only when 0.orig exists, so that initial field data is preserved.

A confirmation dialog shows the full directory path before deletion begins. If the currently open file is inside 0/, the editor and tree view are cleared. The file list and boundary panel are refreshed after deletion.

This action deletes files from disk and cannot be undone.

Resetting the file list

Select Settings > Reset File List to remove all user-added files and directories for the current case. This deletes .foam-editor-files.json from the case directory and reloads the file list with the default target files only.

Boundary view

The Boundary tab in the upper right panel shows a table of boundary conditions for all field variable files in the current field directory.

Table layout

Default orientation:

• Rows — one per field variable file, sorted by file name (epsilon, k, nut, p, U, etc.).
• Columns — one per boundary patch name, sorted alphabetically.

Transposed orientation (enable with the Transpose checkbox): rows become patches and columns become field files.

In both orientations:

• The column set is the union of all patch names across all field files; cells show – when a patch is not defined in a given file.
• Cell text — the type value for that patch (e.g. fixedValue, zeroGradient, noSlip). A † marker is appended when the patch contains large or binary data that cannot be shown in the edit dialog.
• Tooltip — hovering over a cell shows the full patch sub-dictionary text. For †-marked cells, the tooltip shows the type and a brief data description (large data or binary data) instead.
• Lines per cell — a spin box at the top right of the panel controls how many lines are shown per cell (1–10, default 1). When set above 1, additional key-value entries from the patch sub-dictionary are shown below the type line. Entries with large or complex values display as key ….

Directory selector

A drop-down at the top of the panel selects which field directory to display. It shows 0 and 0.orig when they exist. For multiRegion cases it also shows one entry per region subdirectory (e.g. 0/bottomWater, 0/heater, 0/topAir). Only directories that contain at least one parseable field file (i.e. a file with a boundaryField dictionary) appear in the list. When only one qualifying directory exists, the selector is disabled. The table is rebuilt automatically when the selection changes.

Transpose

Check the Transpose checkbox at the top of the panel to swap rows and columns. The table is rebuilt immediately. All other operations (edit, create, copy, paste) work the same in both orientations.

Single-click navigation

Clicking a cell opens its field file in the editor and scrolls to the corresponding patch entry in boundaryField.

The Auto-scroll editor checkbox at the top of the panel controls this behaviour:

• Checked (default) — clicking a cell loads the file (if it is not already open), updates the file list selection, and scrolls the editor to the patch name, highlighting the line in amber.
• Unchecked — single-click has no effect on the editor; only double-click (edit) and right-click (context menu) remain active.

When a cell shows – (no patch entry in that file), the file is still opened in the editor but no scroll is performed because there is no location to jump to.

Editing a boundary condition

Double-click a cell (or right-click → Edit) to open the Edit boundary dialog. The dialog header shows the variable name and patch name in read-only form.

Normal mode (no † marker) — the full content of the patch sub-dictionary is shown in an editable text area. Edit any lines, including type, and click OK to apply. If the content is unchanged, OK has no effect and the file is not marked as modified.

Complex mode († marker) — the patch contains a large nonuniform field or binary data. Only the type field is editable here; use the Text Editor tab to modify the full value.

Creating a patch entry

When a cell shows –, the patch is not yet defined in that field file. Double-click the cell (or right-click → Create Entry) to open an empty Edit boundary dialog. Enter the patch content and click OK to create the entry. The new entry is appended to boundaryField in that file, and the cell updates immediately.

Deleting a patch entry

Right-click a cell with a defined patch and select Delete Entry to remove that patch from the specific field file. The row or column for that patch is removed from the table if the patch no longer exists in any file.

Copy and paste

Right-click any cell with a defined patch and select Copy to store that patch's full content in an internal clipboard. Then right-click any other cell (including – cells) and select Paste to apply the copied content. Pasting into a – cell creates the entry; pasting into an existing cell replaces its content. The clipboard persists within the application session.

Copy Table

Click the Copy Table button in the panel toolbar to copy the entire boundary condition table to the system clipboard. A drop-down menu offers two formats:

• Copy as Markdown — produces a pipe-delimited Markdown table with field variable names as row headers and patch names as column headers (or the transposed equivalent). Multi-line cells (when Lines per cell is above 1) use <br> tags, which render as real line breaks in GitHub-Flavored Markdown.
• Copy as CSV — produces an RFC 4180-compliant CSV. Multi-line cell content is preserved inside quoted fields and displays correctly in spreadsheet applications (Excel, LibreOffice Calc).

Both formats include the row header column (field variable names or patch names, depending on the current orientation). The – marker is preserved as-is.

Renaming a boundary patch

Right-click a cell, a patch column header (non-transposed view), or a patch row header (transposed view) and select Rename Boundary.... The same action is available by right-clicking a boundary_entry node or a patch dictionary node inside boundaryField in the tree view.

A dialog shows the current patch name, a text field for the new name, and a checklist of all loaded files in which that patch name was found — including both blockMeshDict boundary entries and boundaryField patch keys in field files. All matching files are pre-selected.

• Uncheck any file you want to leave unchanged.
• Use Select All / Deselect All to toggle the whole list.
• The Rename button is enabled only when the new name is non-empty, different from the current name, and at least one file is selected. It shows the count of selected files.

Clicking Rename applies the change atomically: each selected file is updated in memory and marked dirty. If the currently open file is among the renamed files, the tree view and text editor are refreshed immediately.

Deleting a boundary condition across all field files

Right-click a patch column header (non-transposed view) or a patch row header (transposed view) and select Delete BoundaryField '<patch>'. A confirmation dialog lists the affected field files. Clicking Yes removes that patch entry from every field file in the current directory in one step.

Adding a boundary condition across all field files

Right-click a patch column header or patch row header and select Add BoundaryField.... Enter the new patch name. A confirmation dialog states how many field files will receive an empty entry. Clicking Yes appends an empty <patch> {} block to each file that has a boundaryField dictionary but does not yet contain that patch. Edit individual cells afterwards to fill in the boundary condition content.

No-type cells

Cells whose patch sub-dictionary has no type key are displayed in italic text. All other interactions (edit, delete, copy, paste) work identically.

Changes from all boundary view operations are immediately reflected in the tree view and text editor. Files modified through the boundary view are marked dirty (*) in the title bar and file list, the same as any other edit.

BlockMesh panel

The BlockMesh tab in the upper panel provides an interactive 3-D view of the geometry defined in blockMeshDict. It is powered by pyVista / VTK and is available only when pyvista and pyvistaqt are installed. If they are absent, the tab shows an install prompt.

Availability: The BlockMesh tab is shown when the blockmesh feature flag is enabled. In the standard variant it is visible only when the Terminal is in Simple mode (switching to xterm mode hides it — see Terminal tab). In the no-terminal-blockmesh variant it is always visible because there is no terminal to conflict with. In the no-terminal variant the tab is absent entirely. See Variants.

The tab can also be toggled at any time via View > BlockMesh 3-D Panel (checkable). When xterm terminal mode is active the action is grayed out and its label changes to "BlockMesh 3-D Panel (unavailable: xterm active)" to explain why. Switch the terminal to Simple mode first to re-enable the panel.

The panel is updated automatically whenever blockMeshDict is loaded or edited. A Refresh button forces a manual update.

Loading or editing topoSetDict also updates the panel: action geometry sources (box, rotated box, sphere, cylinder, cone, point-carrying sources such as nearestToCell, and planeToFaceZone) are overlaid on the mesh. The overlay is shown even if blockMeshDict has not been loaded. Use the "Show topoSet geometry" toggle in the topoSet ▾ menu to toggle it.

Loading or editing snappyHexMeshDict overlays its geometry block (box/sphere/cylinder/cone/triSurfaceMesh/collection), classified as surface, region, or geometry-only via refinementSurfaces/refinementRegions, plus locationInMesh/locationsInMesh keep-point markers — see snappyHexMeshDict overlay. This overlay is also shown independently of blockMeshDict/topoSetDict.

Loading or editing setFieldsDict overlays its regions ( … ) list the same way, each region labelled with its fieldValues summary — see setFieldsDict overlay.

Overlay shapes that extend beyond the block mesh are clipped in the view so the mesh stays visible — see Overlay clipping.

Side-by-side mode

A ⊞ toggle button appears in the top-right corner of the upper tab widget when blockMeshDict, topoSetDict, snappyHexMeshDict, or setFieldsDict is the active file and the BlockMesh panel is available. Clicking it places the 3-D viewer in a horizontal splitter to the right of the Tree tab so both are visible at the same time. The separate BlockMesh tab is removed while side-by-side mode is on, and restored when it is turned off.

The 3-D view is not updated automatically while side-by-side mode is on — click Refresh after editing the tree to see the updated geometry.

The splitter handle between the tree and the 3-D view can be dragged freely: neither pane snaps closed, and the 3-D view can be narrowed down to a small minimum width, with its toolbar wrapping onto extra lines as needed.

The ⊞ button is disabled while the xterm terminal is active (GPU/OpenGL conflict prevents the BlockMesh panel from being open at the same time).

Variable resolution

Variable definitions at the top level of blockMeshDict are automatically resolved before the geometry is extracted:

• Direct values — xMin -0.5;, length 10; etc. are used as-is.
• Macro references — nx $nCell; is resolved to whatever nCell evaluates to, including through chains of arbitrary depth.
• Negated macro references — xMin -$xMax; (a leading minus sign before a $reference) is resolved once xMax is known.
• $varName and ${varName} references inside vertices and blocks are substituted with the resolved values.
• #eval{ expr } — arithmetic expressions are evaluated after variable substitution. Supported operators: +, −, *, /, parentheses. Example: zMax #eval{ $length / $nCell };.

If a reference cannot be resolved (the variable is defined in an external file, for instance), any vertex or block that contains it is silently skipped.

Geometry controls

All controls sit in a single toolbar at the top of the panel. The toolbar wraps: when the panel is narrow (for example in side-by-side mode) the buttons flow onto additional lines instead of dictating a minimum panel width.

Control	Description
Vertices ▾	Drop-down menu with three checkable items: Vertices (show vertices as red spheres), Vertex labels (overlay vertex index numbers), Vertices table (show or hide the vertex coordinate table on the right).
Blocks ▾	Drop-down menu with four checkable items: Block edges (wireframe), Block labels (index at centroid), Color blocks (distinct colour per block from the tab10 palette), Solid blocks (semi-transparent solid faces at opacity 0.25, shares colour with Color blocks).
Boundary faces	Show boundary patch faces, colour-coded by type. Exterior block faces not claimed by any patch — blockMesh's implicit defaultFaces — are drawn in a fainter grey; without them a quasi-2-D case like damBreak (whose big front/back faces are never listed under boundary) would appear to have no boundary at all.
topoSet ▾	Drop-down menu for the topoSetDict overlay: a master Show topoSet geometry toggle, Show all shapes / Hide all shapes actions, an action-colour legend, one checkable toggle per renderable shape, and a Non-geometric sources (N) submenu collecting the entries that carry no drawable geometry (see topoSetDict overlay).
snappyHexMesh ▾	Drop-down menu for the snappyHexMeshDict overlay, structured like topoSet ▾: a master toggle, Show all shapes / Hide all shapes, a category-colour legend, per-shape and keep-point toggles, and a Non-geometric sources (N) submenu (see snappyHexMeshDict overlay).
setFields ▾	Drop-down menu for the setFieldsDict overlay, structured like topoSet ▾: a master Show setFields regions toggle, Show all shapes / Hide all shapes, the region colour legend, one checkable toggle per renderable region, and a Non-geometric sources (N) submenu (see setFieldsDict overlay).
sample ▾	Drop-down menu for the sampling overlay — probe points, sample lines, and sample planes from controlDict's functions {} block or a standalone sampling dict — structured like topoSet ▾ (see Sampling overlay).
Refresh	Re-extract geometry from the current tree and redraw. In Preview mode, also discards all preview edits and resets vertex coordinates to the tree values.
Preview	Appears only when the vertices block contains variable references ($varName). This button is shown inside the Vertices panel (not in the main toolbar). Click to enter Preview mode: table cells become editable and each change immediately updates the 3-D view, but the tree and file are not modified. A yellow banner is shown while Preview mode is active. Click Refresh to leave Preview mode and restore the original values.
STL ▾	Drop-down menu: Load STL / OBJ… loads an STL or OBJ file and displays it as a translucent grey overlay (multiple files can be loaded); gzip-compressed files (.stl.gz, .obj.gz) load directly, no manual gunzip needed; Clear STL removes all loaded overlays (greyed out when none are loaded); Export Shapes as STL… opens a dialog listing every renderable topoSetDict/snappyHexMeshDict/setFieldsDict shape currently loaded, with a checkbox per shape (defaulting to its current visibility in the 3-D view) and a Select All / Deselect All pair. Choose an output folder and click Export to write each checked shape as its own .stl file, named after its label/name (greyed out when no topoSet/snappyHexMesh/setFields geometry is loaded). Point-marker shapes (e.g. nearestToCell) and planeToFaceZone discs are not offered — they have no meaningful STL surface. Exported geometry is always the full, unclipped shape, even when the 3-D view shows it clipped.

Both the traditional 4-vertex face notation (v0 v1 v2 v3) and the newer compact notation (blockIndex faceIndex) are supported and can coexist in the same file. The compact form is automatically expanded to 4-vertex lists using the standard hex block face numbering before display.

topoSetDict overlay

When topoSetDict is loaded or edited, the geometry source of each action entry is drawn as a semi-transparent shape overlaid on the mesh. The "Show topoSet geometry" master toggle in the topoSet ▾ menu toggles the whole overlay; each renderable shape also gets its own checkable toggle, and Show all shapes / Hide all shapes switch every toggle at once. Sources that carry no drawable geometry (e.g. cellToFace, patchToCell) are collected under a Non-geometric sources (N) submenu so a source-rich dictionary stays readable.

Supported source types:

Source family	Examples	Shape
boxToCell/Face/Point	box (x0 y0 z0) (x1 y1 z1), or min/max keys, or a multi-box boxes ( … ) list	Wireframe + surface box (a boxes list draws every box)
rotatedBoxToCell/Face/Point	origin, i, j, k	Oriented (rotated) box
sphereToCell/Face/Point	origin/centre, radius, optional innerRadius	Sphere (hollow when innerRadius is given — the inner shell shows through)
cylinderToCell/Face/Point, cylinderAnnulusToCell	point1/p1, point2/p2, radius	Cylinder
coneToCell/Face/Point, coneAnnulusToCell	point1/p1, point2/p2, radius1, radius2	Cone (truncated; a true cone when radius2 is 0)
nearestToCell, nearestToPoint	points ( … )	Labelled point markers
regionToCell / regionToFace	insidePoints/insidePoint / nearPoint	Labelled point markers
planeToFaceZone	point, normal	Disc in the given plane, sized to the scene bounds (a plane is infinite, so the extent is display-only)

Colour coding by action:

Action	Colour
new	Steel blue
add	Green
subtract / delete	Red
subset	Purple
invert	Gold

If an entry's geometry cannot be fully resolved (e.g. because of an unresolvable $variable), that entry is silently skipped. The overlay is shown even when blockMeshDict is not loaded.

Variable resolution — Top-level scalar/integer definitions in topoSetDict (e.g. xMin -0.01;) are resolved before extraction, including macro chains (y $x;) and #eval{ expr } expressions. This is the same mechanism used in blockMeshDict. Variables from other files are not in scope.

snappyHexMeshDict overlay

When snappyHexMeshDict is loaded or edited, its geometry {} block is drawn as a set of semi-transparent shapes overlaid on the mesh, each classified as a surface, a region, or plain geometry by cross-referencing castellatedMeshControls.refinementSurfaces / refinementRegions. The "Show snappyHexMesh geometry" master toggle in the snappyHexMesh ▾ menu toggles the whole overlay; the menu also lists a checkable toggle per individual shape (with Show all shapes / Hide all shapes to switch them all at once), plus a Non-geometric sources (N) submenu collecting the entries for geometry that could not be drawn.

Supported geometry entry types:

type	Fields	Shape
box	min, max	Wireframe + surface box
sphere	centre/origin, radius	Sphere; if radius is given as a vector (rx ry rz) (e.g. an igloo-shaped dome), an ellipsoid is drawn instead of a true sphere
cylinder	point1/p1, point2/p2, radius	Cylinder
cone	point1/p1, point2/p2, radius1, radius2	Cone (truncated; a true cone when radius2 is 0)
triSurfaceMesh / distributedTriSurfaceMesh	file (or the entry's own key, if it ends in .stl/.stlb/.obj)	The referenced mesh, loaded automatically from constant/triSurface/ relative to the case directory. Found even when shipped gzip-compressed (e.g. motorBike.obj.gz), whether the dict references the plain name or the .gz name directly — no manual gunzip needed. If the file cannot be found, the entry is listed greyed-out instead of drawn.
collection (searchableSurfaceCollection)	named sub-entries, each with surface <name>, scale, transform { origin; rotation none; } or transform { origin; e1; e3; }	Each sub-entry referencing a box-type base surface is drawn as its own scaled/rotated/translated box, named <collection>.<member> (e.g. twoFridgeFreezers.seal). Members referencing a non-box base, or using an unsupported/missing transform, are not drawn; if none of a collection's members could be resolved, the collection itself is listed greyed-out.

Colour coding by category:

Category	Colour	How it's determined
Surface	Teal	The geometry entry's name (or a name override) matches a refinementSurfaces entry — either exactly, or via a regex-pattern key such as "iglo.*"
Region	Purple	Matches a refinementRegions entry (by exact name only); a shape that matches both is shown as a surface
Geometry	Grey	Defined in geometry {} but not referenced by either control block

Each drawn shape is labelled with its name and, when available, its refinementSurfaces level (min max) or its refinementRegions mode (inside/outside/distance).

Keep points — castellatedMeshControls.locationInMesh (a single point) or locationsInMesh (a list of ((x y z) regionName) pairs, used by multi-region cases) are drawn as labelled point markers, independent of the shape overlay above.

If a geometry entry's fields cannot be fully resolved (e.g. an unresolvable $variable), it is listed greyed-out rather than drawn. The overlay is shown even when blockMeshDict/topoSetDict are not loaded.

Not visualized — per-region overrides inside a refinementSurfaces entry's regions { … } sub-block, patchInfo, castellatedMeshControls.features (feature-edge .eMesh references), and refinementRegions.levelIncrement (directional refinement) are read by the parser/schema but have no effect on this overlay.

setFieldsDict overlay

When setFieldsDict is loaded or edited, each entry of its regions ( … ) list is drawn as a semi-transparent orange shape overlaid on the mesh. Region sources use the same geometry keywords as topoSetDict, so the same shape families render: boxToCell (including the min/max and multi-box boxes ( … ) forms), rotatedBoxToCell, sphereToCell (with optional innerRadius), cylinderToCell/cylinderAnnulusToCell, and coneToCell/coneAnnulusToCell — see the source table in topoSetDict overlay. $variable and #eval{ expr } references are resolved the same way.

Each shape is labelled with a summary of the entry's fieldValues (e.g. alpha.water=1), so you can see at a glance which region sets which field. The "Show setFields regions" master toggle in the setFields ▾ menu toggles the whole overlay; each region also gets its own checkable toggle, with Show all shapes / Hide all shapes to switch them all at once. Sources that carry no drawable geometry (e.g. zoneToCell, surfaceToCell) are collected under a Non-geometric sources (N) submenu. defaultFieldValues applies to the whole mesh and is not drawn.

The overlay is shown even when blockMeshDict is not loaded. Renderable regions are also offered by STL ▾ → Export Shapes as STL… as a third group.

Sampling overlay

When controlDict — or one of the standalone sampling dicts system/sample, system/probes, system/surfaces, system/singleGraph (all four now appear in the file list when present) — is loaded or edited, its sampling definitions are drawn in teal on the 3-D view:

• Probe points — probes / patchProbes / boundaryProbes function objects: each probeLocations set is drawn as point markers.
• Sample lines — sets-type entries: each member with a start/end pair (lineUniform, lineCell, uniform, face, …) is drawn as a thin tube; cloud-type members with a points list are drawn as point markers. Both member-list syntaxes work — the dictionary form sets { … } and the classic parenthesised list form sets ( y0.1 { … } … ); — as does the .org singleGraph style, whose start/end sit at the file's top level.
• Sample planes — surfaces-type entries: plane and cuttingPlane members (direct point/normal, basePoint/normalVector, or a pointAndNormalDict {}) are drawn as discs whose extent is display-only, like planeToFaceZone in the topoSet overlay.

Definitions can come from several files at once (e.g. probes in controlDict plus a system/sample file); the sample ▾ menu shows the union, and each row is tagged with its source file in brackets. Entries that carry no drawable geometry — patch-type surfaces, members without resolvable coordinates, or a sets/surfaces entry whose member list is missing — are collected under Non-geometric sources (N). Non-sampling function objects (forces, fieldAverage, …) are ignored entirely. Sampling shapes are not offered for STL export: points and lines have no surface, and a plane's drawn extent is display-only.

Overlay clipping

setFieldsDict regions in particular are often written far larger than the domain (e.g. damBreak's box (0 0 -1) (0.1461 0.292 1) on a ~0.6 m tank) — drawn at full size they dwarf the block mesh. Whenever a block mesh is loaded, every overlay shape (topoSet, snappyHexMesh, setFields, and sampling) is therefore clipped, for display only, to the block-mesh bounding box expanded by 10% per axis:

• A shape that fits inside the expanded bounds is drawn unchanged.
• A shape that pokes out is cut down to the bounds and its scene label gains a "✂ clipped" mark, so the cut is not mistaken for the shape's real extent.
• A shape lying entirely outside the block mesh is kept unclipped and labelled "⚠ outside block mesh".
• A shape whose volume encloses the whole scene (no part of its surface falls inside the bounds) is drawn as its bounding-box overlap instead of disappearing.

The clipping never modifies the dictionary or the extracted geometry: Export Shapes as STL… always writes the full shape, and no clipping is applied when no block mesh is loaded.

Variable resolution — Top-level scalar/integer definitions in snappyHexMeshDict are resolved the same way as in blockMeshDict/topoSetDict, including macro chains and #eval{ expr } expressions.

Boundary face colours

Patch type	Colour
wall	Orange
patch	Blue
empty	Light grey
symmetry / symmetryPlane	Green
wedge	Purple
cyclic / cyclicAMI	Gold
inlet	Light blue
outlet	Red
Other / unknown	Blue

Vertices table

A scrollable table on the right side of the panel lists every vertex from the vertices block as a row with columns #, X, Y, Z (coordinates are shown after applying the scale/convertToMeters factor).

• Click a row to highlight the corresponding vertex in the 3-D view as a larger cyan sphere.
• Double-click an X, Y, or Z cell to edit the coordinate value inline (only available when vertices use literal numbers — see below). Entering a non-numeric value cancels the edit and restores the previous value. After a valid commit the FoamNode tree and text editor are updated immediately and the file is marked dirty. The 3-D view updates after clicking Refresh.
• When vertices contain variables ($varName references), the X/Y/Z cells are read-only by default. A ⚙ Variable-based chip and a Preview button appear at the top of the Vertices panel. Click Preview to unlock editing. In Preview mode changes update the 3-D view immediately without touching the tree or file. Click Refresh to exit Preview mode and restore the original values.
• The # index column is read-only. Reordering vertices would silently break all hex and faces index references.
• For meshes with more than 500 vertices only the first 500 rows are shown, with a truncation notice at the bottom.
• The Vertices table item in the Vertices ▾ menu collapses the table pane entirely, giving more space to the 3-D view.

Scale and label controls

Control	Description
Scale ▾	Drop-down menu with three checkable items: Axes (corner XYZ orientation arrows), Grid (coordinate grid with tick labels), Dimensions (wireframe bounding box and X/Y/Z extents overlay; also shows the scale factor when convertToMeters or scale is not 1).
Label size	Spin box (range 6–32, default 10) that sets the font size for both vertex labels and block labels simultaneously.
View: +X −X +Y −Y +Z −Z Iso	Snap the camera to a standard direction. +X looks from the +X side toward the origin (shows the YZ plane), −X from the opposite side, and so on for Y and Z. Iso uses an isometric view.

Mouse controls

A compact hint bar at the bottom of the BlockMesh panel shows the key bindings at a glance. Hover the hint to see the full tooltip.

Action	Mouse / key
Rotate	Left drag
Pan	Shift + left drag
Zoom	Scroll wheel or right drag
Reset camera	R
Fly to point	F

The full reference is also available in Help > Keyboard Shortcuts… under "BlockMesh 3-D viewer (mouse)".

Schema module configuration

Schema definitions for each dictionary type are managed in the schemas/ directory. The runtime schema registry is handled by schemas/registry.py, and the default built-in configuration is defined in schemas/builtin.py.

Built-in schema modules:

• schemas/control_dict.py — schema for controlDict. Covers startFrom, stopAt, writeControl, writeFormat, writeCompression, timeFormat, graphFormat, runTimeModifiable, adjustTimeStep, and purgeWrite.
• schemas/fv_schemes.py — schema for fvSchemes.
• schemas/fv_solution.py — schema for fvSolution.
• schemas/block_mesh_dict.py — schema for blockMeshDict. Covers the top-level scaling keys (convertToMeters, scale), mergeType, and verbose.
• schemas/snappy_hex_mesh_dict/ — schema for snappyHexMeshDict (a package split by subdomain into submodules, merged in __init__.py). Covers the top-level phase-toggle keys (castellatedMesh, snap, addLayers), mergeTolerance, and debug; common settings inside castellatedMeshControls, snapControls, addLayersControls (including layer-quality and medial-axis parameters), and meshQualityControls (including the relaxed sub-dict); per-surface keys inside refinementSurfaces entries (level, faceZone, cellZone, cellZoneInside, faceType); per-region keys inside refinementRegions entries (mode, levels); per-patch nSurfaceLayers inside layers entries; and patchInfo.type with patch-type choices.

Which schema modules to load is controlled at runtime via schema_config.json. You can add or remove schema modules without modifying the source code.

For the complete list of internal node type strings, see DEVELOPER.md.

Writing a custom schema module

Each schema module must define two module-level names.

• TARGET_FILE — the dictionary filename the module applies to (e.g. "controlDict").
• SCHEMAS — a dict[str, KeySchema] mapping entry keys to their schema definitions.

KeySchema has the following fields (defined in schemas/_base.py):

Field	Type	Description
key	str	The dictionary key this entry describes (matches the SCHEMAS dict key for plain entries)
label	str	Short human-readable label shown in the Detail pane header
description	str	Longer explanation shown in the Key Help field
supported_in	tuple[str, ...]	Version strings shown in the Key Supported In field
note	str	Additional remark shown in the Key Note field
choices	tuple[ChoiceItem, ...]	Valid or common values shown in the Choices drop-down

Each ChoiceItem in the choices tuple has:

Field	Type	Description
value	str	The literal string written to the dictionary
description	str	Explanation shown when the choice is selected in the drop-down
supported_in	tuple[str, ...]	Which distributions support this value
note	str	Optional additional remark

_base.py exports pre-built version strings for supported_in: FOUNDATION_V13, OPENCFD_V2312, OPENCFD_V2512, and OPENCFD_SERIES. A minimal custom module looks like this:

from schemas._base import ChoiceItem, KeySchema, FOUNDATION_V13, OPENCFD_SERIES

TARGET_FILE = "myDict"

SCHEMAS = {
    "myKey": KeySchema(
        key="myKey",
        label="My Key",
        description="Controls what myKey does.",
        supported_in=(FOUNDATION_V13, OPENCFD_SERIES),
        choices=(
            ChoiceItem("optionA", "Use option A."),
            ChoiceItem("optionB", "Use option B."),
        ),
    ),
}

Keys in SCHEMAS can be plain ("startFrom") or context-qualified with a dotted prefix to avoid collisions between identically named keys in different sub-dicts:

• "snapControls.nRelaxIter" — matched when the selected node's direct parent is snapControls
• "addLayersControls.nRelaxIter" — matched when the parent is addLayersControls

For blocks whose parent name is user-defined (such as named entries inside refinementSurfaces, refinementRegions, or layers), use the grandparent as the prefix instead:

• "refinementSurfaces.level" — matched when the grandparent node is refinementSurfaces, regardless of what the immediate parent (surface name) is
• "layers.nSurfaceLayers" — matched when the grandparent is layers

The lookup order is: parent-qualified match → grandparent-qualified match → plain key. Existing flat keys continue to work without any changes.

The registry imports each configured module and reads these two names. Any module that follows this convention can be added through Settings > Manage Schema Modules without further configuration.

schema_config.json

schema_config.json is created the first time the user modifies schema settings via Settings > Manage Schema Modules or triggers a settings reset. If the file does not exist, the built-in default modules are used in memory without writing to disk. The file records the list of schema modules to load.

{
  "schema_modules": [
    "schemas.control_dict",
    "schemas.fv_schemes",
    "schemas.fv_solution",
    "schemas.block_mesh_dict",
    "schemas.snappy_hex_mesh_dict"
  ]
}

To add a custom schema module, select Settings > Manage Schema Modules in the application, then use the Add Module from File button to select a Python file. The change is saved to schema_config.json and takes effect immediately.

Generate OpenFOAM Keywords

Select Settings > Generate OpenFOAM Keywords… to rebuild the keyword list used by the editor's syntax highlighter (see Editor behavior) from your local OpenFOAM installation.

Pick an installation from the Installation dropdown — installations are auto-discovered the same way as in Find OpenFOAM Examples (sourced environment first, then the usual install locations), and Browse… lets you point at any other directory, which is remembered across sessions. Click Generate. The scan reads the installation's etc/caseDicts/ templates, TypeName(...)/ClassName(...) macros and run-time-selection-table registration names in src/ headers, and dictionary-read calls (lookup("…"), get<…>("…"), readEntry("…"), …) in the src/ and applications/ sources — the last of these is what picks up free-form keys such as application or writePrecision in controlDict. The result is written to app_config/foam_keywords.json, which overrides the bundled default list. Progress is shown in the dialog's log area; click Cancel to stop an in-progress scan. The dialog can be closed once the scan finishes or is cancelled.

If no installation is selected and no OpenFOAM environment is sourced, the scan reports an error in the log instead of failing silently.

This is an optional, advanced action — the application ships with a pre-generated app_config/foam_keywords.default.json (built from OpenFOAM v2512; see the file's version field), so highlighting works out of the box without ever running this scanner. Version-to-version keyword differences are small and only cosmetic for highlighting, so one bundled default is enough — if you want a list matched exactly to your own version or distribution, regenerate it here. Deleting app_config/foam_keywords.json reverts to the bundled default.

Variants

FoDE ships in three configurations controlled by feature flags. Choose the one that matches your platform and workflow.

Variant	Terminal tab	BlockMesh tab	Launch command
standard	Yes (xterm / Simple toggle)	Yes (visible in Simple mode)	python3 main.py
no-terminal	No	No	python3 main.py --variant no-terminal
no-terminal-blockmesh	No	Yes (always visible)	python3 main.py --variant no-terminal-blockmesh

The --variant flag loads a preset from the presets/ directory and saves the chosen feature flags to app_config.json on exit. Subsequent launches without --variant use the saved flags automatically.

The no-terminal variants avoid the QtWebEngineWidgets dependency and are the recommended choice for Windows or environments where the xterm.js terminal is not needed.

Preset files are in presets/:

presets/standard.json
presets/no-terminal.json
presets/no-terminal-blockmesh.json

You can also set the flags directly in app_config.json:

{
  "features": {
    "terminal": false,
    "blockmesh": true
  }
}

Feature flags default to true when absent, so a plain app_config.json with no features key runs in standard mode.

Application settings

General application settings are stored in app_config.json, which is separate from schema_config.json. This file is created the first time a case is opened (which triggers an automatic save of the case directory). If the file does not exist, built-in defaults are used. Settings are managed internally by the app_config package (app_config/app_config_manager.py).

app_config.json

{
  "window_size": [1200, 800],
  "default_case_dir": "/path/to/cases",
  "case_library_dirs": ["/home/user/my_templates"],
  "user_links": [{"label": "My reference", "url": "https://example.com"}],
  "features": {"terminal": true, "blockmesh": true},
  "openfoam_dir": "/usr/lib/openfoam/openfoam2606"
}

Key	Description
window_size	Main window size on startup. Saved automatically when the application is closed.
default_case_dir	Initial directory shown when Case > Open Case is used. Updated automatically to the parent of the last opened case. Also used as the default destination parent in Case > Duplicate from Case Library... and Case > Save as New Case....
case_library_dirs	User-added Case Library directories. The $FOAM_TUTORIALS directory is not stored here; it is included dynamically from the environment variable.
user_links	User-defined reference links shown in Help > Resources... > My Links. Each entry is {"label": "…", "url": "…"}.
features	Feature flags set by --variant (see Variants). Omitting this key is equivalent to {"terminal": true, "blockmesh": true}.
openfoam_dir	OpenFOAM installation chosen via Browse… in Find OpenFOAM Examples. Only written after browsing; auto-discovered installations are not stored.

Setting the default case directory

Select Settings > Set Default Case Directory to set the directory that opens when Case > Open Case is used. The directory is also updated automatically to the parent of the last opened case, so the next session starts from the same location.

Window size

The main window opens at the size recorded in app_config.json. If the file does not exist, the default size of 1200×800 is used. The size is saved automatically on exit. To restore the default size, select Settings > Reset Window Size.

Resetting settings

Select Settings > Reset All Settings to reset app_config.json, schema_config.json, or both to their default values. A confirmation dialog is shown before any changes are made.

Terminal tab

The bottom panel contains an Editor tab and a Terminal tab. The Terminal tab has two modes that can be switched at runtime.

Mode toggle

A checkbox at the top of the Terminal tab reads "xterm terminal (hides BlockMesh 3-D panel)".

Mode	Description
xterm (checked, default on Linux/macOS)	Full PTY terminal powered by xterm.js v6 in a QWebEngineView. Full VT100/xterm emulation: colour output, cursor control, interactive programs (vim, htop, …). The BlockMesh tab is hidden while this mode is active, and View > BlockMesh 3-D Panel is grayed out.
Simple (unchecked)	QProcess-based terminal running a persistent shell (bash on Linux/macOS, cmd.exe on Windows). Output shown in a plain-text area; ANSI escape codes are stripped. The BlockMesh 3-D panel becomes visible in the upper tab widget and View > BlockMesh 3-D Panel is re-enabled.

The xterm checkbox is disabled on Windows or when QtWebEngineWidgets is not installed — Simple mode is used automatically in those cases. In the no-terminal and no-terminal-blockmesh variants the Terminal tab is absent entirely; see Variants.

Why are they mutually exclusive? The Qt WebEngine GPU process (used by xterm.js) and VTK (used by the BlockMesh viewer) compete for the same GPU/OpenGL context on Linux. Switching modes ensures only one of the two is active at a time.

When a case is opened, the terminal automatically changes its working directory to the case root so that OpenFOAM commands such as blockMesh or interFoam can be run directly.

xterm.js assets

The xterm.js library files are downloaded automatically from jsDelivr on first use and cached in ui/xterm/. No manual installation is required. If the download fails (e.g. no internet access), a message is displayed instead; place the following files in ui/xterm/ manually:

File	Source
xterm.js	https://cdn.jsdelivr.net/npm/@xterm/xterm@6.0.0/lib/xterm.js
xterm.css	https://cdn.jsdelivr.net/npm/@xterm/xterm@6.0.0/css/xterm.css
xterm-addon-fit.js	https://cdn.jsdelivr.net/npm/@xterm/addon-fit@0.11.0/lib/addon-fit.js

SimpleTerminalWidget

SimpleTerminalWidget supports basic command-line interaction:

• Type a command in the input field at the bottom and press Enter to run it.
• Use the ↑ and ↓ arrow keys to navigate command history.
• Click Clear to clear the output area.
• If the shell process exits unexpectedly, it is restarted automatically.
• The shell process is terminated cleanly when the application closes.

foamMonitor launcher

Tools > foamMonitor… launches foamMonitor to plot residuals or other time-series data with gnuplot — without leaving FoDE.

Launching

1. Open a case and start a solver in the Terminal tab.
2. Select Tools > foamMonitor…. A dialog opens with the following options:

Field	Description
File	Path to the file to monitor. Click Browse… to open a file picker rooted at the case directory. Typical paths: log.icoFoam, log.simpleFoam, postProcessing/residuals/0/residuals.dat.
Log scale (-l)	Plot the y-axis on a log scale.
Grid (-g)	Draw grid lines.
Refresh (-r)	How often gnuplot re-reads the file (seconds, default 10).
Idle timeout (-i)	foamMonitor stops itself if the file has not changed for this many seconds (default 60).
Extra	Any additional foamMonitor flags, entered as free text (e.g. -y [1e-8:1]).

3. Click Launch. A gnuplot window opens and refreshes automatically as the solver writes new data.

Stopping

While foamMonitor is running the menu item reads ■ foamMonitor. Select Tools > ■ foamMonitor to stop foamMonitor and close the gnuplot window.

foamMonitor also stops automatically when its idle timeout expires (the monitored file has not been updated for the configured number of seconds). When that happens the menu item reverts to foamMonitor… on its own.

Opening a different case while foamMonitor is running stops the current instance.

Notes

• foamMonitor is a Unix shell script (foamMonitor must be on PATH). The menu item has no effect on Windows.
• If the selected file does not exist, or foamMonitor exits with an error (e.g. gnuplot not installed), a warning dialog shows the error message.
• FoDE automatically patches the reread command (deprecated in newer gnuplot versions) so that the gnuplot window refreshes correctly regardless of the installed gnuplot version.

Restore 0/ from 0.orig

Select Tools > Restore 0/ from 0.orig to recreate the 0/ field directory from 0.orig/. A confirmation dialog warns that 0/ will be deleted and replaced with a fresh copy of 0.orig/, discarding any edits made directly to 0/; click Yes to proceed. The command runs in the Terminal tab (rm -rf 0 && cp -r 0.orig 0), which is switched to automatically. The action is available only when a case is open, a 0.orig/ directory exists, and a terminal panel is present (absent in the no-terminal variant).

Run blockMesh

Select Tools > Run blockMesh… to run blockMesh in the case directory. An options dialog opens first — all five "Run *" actions share the same dialog layout:

• Curated options — the tool's most common flags as checkboxes and fields. For blockMesh: an alternative dictionary (-dict, with a Browse… button that inserts a case-relative path) and a mesh region (-region).
• Extra options — a free-text field for any other flags (e.g. -verbose), appended to the command as typed.
• Command — a live preview of the exact command line that will be sent, so you always see (and learn) the real OpenFOAM invocation before running it.

Press Run (the default button — pressing Enter runs with the current options) to send the command to the Terminal tab, which is switched to automatically. The output is also saved to log.blockMesh in the case directory; the 2>&1 | tee log.blockMesh redirection is added automatically so View Log Summary always finds the log. Option values are remembered per tool for the rest of the session. If the case already has numeric result directories, the dialog lists them at the top and warns that re-running blockMesh may invalidate those results. The action is available only when a case is open and a terminal panel is present.

Run snappyHexMesh

Select Tools > Run snappyHexMesh… to run snappyHexMesh in the case directory. The options dialog offers -overwrite (checked by default — uncheck it to write the snapped mesh into new time directories instead of replacing constant/polyMesh), -dict, and -region, and warns at the top if the case already has numeric result directories that re-meshing may invalidate. The command runs in the Terminal tab (switched to automatically) and its output is also saved to log.snappyHexMesh. The action is available only when a case is open and a terminal panel is present.

Run topoSet

Select Tools > Run topoSet… to run topoSet in the case directory, reading system/topoSetDict. The options dialog offers -dict (e.g. a topoSetDict.refine variant) and -region, and warns at the top if the case already has numeric result directories that re-running may invalidate. The command runs in the Terminal tab (switched to automatically) and its output is also saved to log.topoSet. The action is available only when a case is open and a terminal panel is present.

Run setFields

Select Tools > Run setFields… to run setFields in the case directory, reading system/setFieldsDict. The options dialog offers -dict and -region, and the command runs in the Terminal tab (switched to automatically) with its output also saved to log.setFields.

setFields overwrites the field files in 0/ in place, so re-running it on already-set fields compounds the values (an already-filled alpha.water region stays filled even if the dictionary's box has moved) — the dialog says so at the top. When the case has a 0.orig/ backup, the dialog therefore also shows a Restore 0/ from 0.orig/ first checkbox, checked by default: it prepends rm -rf 0 && cp -r 0.orig 0 && to the command (visible in the preview), the standard tutorial workflow of running setFields on clean initial fields. Uncheck it to run on the current 0/ as-is. Without a 0.orig/ directory the warning simply notes that there is no backup to restore from. The action is available only when a case is open and a terminal panel is present.

Run checkMesh

Select Tools > Run checkMesh… to run checkMesh in the case directory — typically after Run blockMesh or Run snappyHexMesh — to validate mesh quality (non-orthogonality, skewness, aspect ratio, …). The options dialog offers -allGeometry and -allTopology (the extended non-standard check sets), -writeSets with a surface format (e.g. vtk, writing the faulty cell/face sets for inspection in ParaView), and -region; checkMesh only reads the mesh, so the dialog carries no pre-flight warning. The command runs in the Terminal tab (switched to automatically) and its output is also saved to log.checkMesh, where View Log Summary can pick it up. The action is available only when a case is open and a terminal panel is present.

Run Allrun Script

Select Tools > Run Allrun Script to run the case's ./Allrun script in the Terminal tab (switched to automatically) — the case's full workflow, typically meshing, field initialisation, and the solver itself. A warning dialog is shown if the case has no Allrun script. Since solver runs can take a long time, a confirmation dialog is shown first; note that in Simple terminal mode a running job cannot be interrupted (the xterm terminal supports Ctrl+C).

If the case already has log.* files, a different dialog appears instead: OpenFOAM's Allrun helper functions (runApplication/runParallel) skip any step whose log.* file already exists ("already run … remove log file to re-run"), so on an already-run case a plain ./Allrun may appear to do nothing. The dialog lists the existing log files and offers three choices: Clean, then run (sends foamCleanTutorials && ./Allrun — the standard ./Allclean && ./Allrun idiom, see Clean Case), Run anyway (sends ./Allrun unchanged, with already-run steps skipped), or Cancel. The action is available only when a case is open and a terminal panel is present.

Each Allrun step writes its own log.<application> file, so View Log Summary and foamMonitor work on the results as usual.

Open Mesh in ParaView

Select Tools > Open Mesh in ParaView… to view the actual generated mesh in ParaView, as an alternative to the BlockMesh 3-D panel's block-topology preview. FoDE launches paraFoam -case <case directory> if paraFoam is on PATH, falling back to a plain paraview launch if it is not. If neither is found, a warning dialog reports that neither could be located. The action is available whenever a case is open; it does not require blockMesh to have been run in FoDE.

Run Allclean Script

Select Tools > Run Allclean Script to run the case's ./Allclean script in the Terminal tab (switched to automatically). A warning dialog is shown if the case has no Allclean script; otherwise a confirmation dialog warns that the generated mesh, time directories, log files, and other results will be removed. The file list refreshes automatically once the deletions hit the disk. The action is available only when a case is open and a terminal panel is present.

Note that Clean Case (foamCleanTutorials) already runs ./Allclean when the case has one — this action exists for when you want the script itself, explicitly.

Clean Case (foamCleanTutorials)

Select Tools > Clean Case (foamCleanTutorials) to clean the case with OpenFOAM's standard foamCleanTutorials script, run in the Terminal tab (switched to automatically). It removes the generated mesh (constant/polyMesh), time directories, processor*/ decompositions, postProcessing/, log.* files, and other generated artifacts. Two case-specific behaviors, both stated in the confirmation dialog:

• If the case has its own Allclean (or Allwclean) script, foamCleanTutorials runs that script instead of its built-in cleaning.
• Otherwise, if a 0.orig/ directory exists, 0/ is also removed (foamCleanTutorials' default -auto mode — the standard tutorial convention; use Restore 0/ from 0.orig or Run Allrun Script to recreate it).

The file list and the mesh indicator refresh automatically after cleaning. The action is available only when a case is open and a terminal panel is present, and requires a sourced OpenFOAM environment in the terminal (same as Run blockMesh).

View Log Summary

Select Tools > View Log Summary… to open a condensed report of a log.* file instead of reading the raw text — most useful for log.snappyHexMesh, which can run to thousands of lines of repetitive refinement/snapping/layer-iteration output. The dialog defaults to the most-recently-modified log.* file in the case directory (use Browse… to pick a different one, including outside the case directory) and shows two tabs:

• Summary — for blockMesh: the Mesh Information and Patches blocks. For snappyHexMesh: final cell/face/point counts and iteration counts per castellation refinement category, the snapping relaxation iteration count, the final per-patch layer-thickness table, and the mesh-quality result. For topoSet: one condensed line per cell/face/point set, listing the sources applied and the final set size. For a solver log (log.simpleFoam, log.interFoam, … — recognised by its time-loop shape, so any solver works without being on a list): the number of time steps and the simulated time range, the final ExecutionTime/ClockTime, a convergence line if the run converged (SIMPLE solution converged in N iterations), and the last step's residuals — initial/final residual and iteration count per field, Courant number(s), and the cumulative continuity error. A run that neither converged nor reached End is reported as FAILED. Repeated warnings (e.g. per-iteration displacement warnings) are collapsed into one line with a count; fatal errors are always shown in full. Any other utility's log falls back to its header plus the last ~20 lines.
• Raw Log — the untouched file contents, in case the summary omits something you need.

Like Find OpenFOAM Examples — and unlike FoDE's other dialogs — this one is non-modal: it can stay open beside the main window while you keep working in the tree, editor, or terminal, so you can cross-reference dictionary settings against the log's results. The action is available whenever a case is open; it does not require the terminal panel or a prior FoDE-run mesh command — any existing log.* file works.

Find OpenFOAM Examples

Select Tools > Find OpenFOAM Examples… (also available as Case > Find OpenFOAM Examples…) to search a local OpenFOAM installation for real usage examples of a keyword or setting — the kind of question ("how is nOuterCorrectors actually used?", "what goes in a forces function object?") that otherwise requires find/grep skills. The dialog searches two sources, each toggleable with a checkbox:

• Tutorials — the ~550 example cases shipped under the installation's tutorials/ directory.
• caseDicts templates — the curated files under etc/caseDicts/, including the postProcessing/ function-object templates (usable in a controlDict functions {} block via #includeFunc <name>) and the fully annotated example dictionaries under annotated/.

Choosing an installation. The selector at the top lists the OpenFOAM installations found on your machine: the sourced environment (WM_PROJECT_DIR/FOAM_TUTORIALS/FOAM_ETC) if FoDE was started from a sourced shell, plus auto-discovered installs under /usr/lib/openfoam/, /opt/, and ~/OpenFOAM/, newest first. Use Browse… to point at any other directory (an install root or a bare tutorials/ directory); a browsed choice is saved to app_config.json (openfoam_dir key) and preselected next time.

Searching. Type a keyword (matched case-insensitively, line by line) and press Enter or click Search. The file-name dropdown narrows the search to one dictionary file (controlDict, fvSchemes, …) — leave it on "All files" to search everything. The search runs in the background and can be cancelled; results are capped at 200 files. Hits are grouped by tutorial case (and by template folder for caseDicts), each with the first matching line as a preview snippet.

Using a result. Selecting a hit shows the whole file in a read-only, syntax-highlighted preview, scrolled to the first matching line. From there:

• Copy File / Copy Selection — copy the whole file, or just the text you selected in the preview (the button enables once something is selected), e.g. to paste into the Editor tab of your own case.
• Compare with this case — enabled when the hit lies inside a tutorial case. One click loads that case as the reference of the existing Case comparison view, so you can see exactly where your case differs and adopt individual settings with the right-click Use this value action. FoDE asks you to open a case first if none is open.
• Duplicate this case… — enabled when the hit lies inside a tutorial case. Opens the same dialog as Duplicating a case with the tutorial as the source, so you can copy it to a writable location and start your own case from it — tutorial cases inside the installation should never be edited in place. The destination defaults to your default case directory (falling back to your home directory, never the read-only installation), and after copying FoDE offers to open the new case immediately.

The dialog is non-modal: keep it open beside the main window and copy from examples while editing. No case needs to be open just to search.

Tree key filter and editor sync

Tree key filter

A filter bar sits above the tree view. Type any substring to narrow the tree to rows whose key contains the text (case-insensitive). Matching is recursive: when a descendant matches, all its ancestors remain visible so the path to the match is always clear.

Clearing the filter restores the full tree. The filter is also cleared automatically whenever a new file is loaded.

unknown_raw_entry and directive_entry nodes have an empty key and are hidden by an active filter unless a sibling or descendant matches.

Editor sync (Auto-scroll editor)

Selecting a row in the tree highlights the corresponding lines in the text editor with an amber background, so you can immediately see the raw text for the selected entry.

The Auto-scroll editor checkbox sits to the right of the filter bar and controls whether the editor also scrolls to the highlighted lines:

• Checked (default) — the editor scrolls to the entry's first line and highlights its full span in amber. Useful for navigating through a file using the tree.
• Unchecked — the amber highlight appears but the editor does not scroll. Useful for keeping a specific region of the raw text visible while browsing the tree structure.

The checkbox label turns grey and reads Auto-scroll editor (stale) when the editor text has been modified since the last parse. In this state the stored line numbers no longer match the text, so both jump and highlight are suspended. The status bar shows a reminder: "Apply Text to Tree to re-enable jump-to-line". Jump and highlight resume automatically as soon as you press Apply Text to Tree, load a different file and return, or open a new case.

When a node was added or modified directly in the tree (rather than parsed from text), it has no associated source location. Selecting such a node leaves the editor position unchanged and shows "No source location — entry was added or modified in the tree" in the status bar.

Find in Tree (editor → tree)

Click Find in Tree in the Editor toolbar (or press Ctrl+Shift+T) to navigate in the opposite direction: the tree selects the deepest node whose source span covers the current editor cursor line.

• If source lines are stale (text edited since the last parse), a status bar message is shown instead. Press Apply Text to Tree first.
• If the cursor sits outside every node's span (for example, a blank line at the top of the file), "No tree entry found for line N" is shown.
• If the matched node is hidden by the current key filter, the nearest visible ancestor is selected instead. If no ancestor is visible, "Entry is hidden by the current filter" is shown.

Tree view context menu

Right-click any row in the center tree view to open a context menu. Items are grouped by function.

Undo and redo

Tree edits are undoable: press Ctrl+Z while the tree has keyboard focus to revert the most recent tree change made in the current file, and Ctrl+Shift+Z to re-apply it (both also appear in the context menu as Undo Tree Edit / Redo Tree Edit). This covers every tree-side mutation — inline cell edits, Detail-panel applies, Paste Value, Add/Duplicate/Delete/Comment Out/Restore, Apply Text to Tree, vertex drags in the BlockMesh panel, and the Boundary tab's patch operations. A boundary operation that changed several field files at once (e.g. renaming a patch across 0/) is undone as a whole — one Ctrl+Z restores every file it touched, and the status bar reports the extra files.

Undo is a single history for the whole session (up to 50 steps), cleared when you open another case: Ctrl+Z reverses the most recent tree change wherever it happened, switching the view to the affected file when it is not already on screen. The bottom text editor keeps its own native Ctrl+Z for free-form typing — the tree shortcut only fires while the tree has focus. Undoing reloads the tree, so expanded/selected rows are reset.

Value copy and paste

• Copy Value — copies the display text from the Value column of the selected row to the system clipboard. Ctrl+C when the tree has keyboard focus does the same.
• Paste Value — applies the clipboard text to the Value column of the selected row using the same parsing path as direct cell editing. Ctrl+V when the tree has focus does the same.

Paste Value is disabled when the selected node type does not support value editing (for example, dictionary nodes that show an entry count). If the pasted text cannot be parsed for the node type (for example, a non-numeric string pasted into a scalar node), the paste is silently rejected and a status-bar message is shown.

Ctrl+C and Ctrl+V are scoped to the tree widget and do not interfere with the text editor at the bottom. When a tree cell is in inline-edit mode, the shortcuts go to the inline editor as usual.

Adding and duplicating entries

• Add Entry After — inserts a new word-typed entry (newKey / newValue) as a sibling immediately after the selected node and opens the key cell for inline editing. Enabled when the parent is a dictionary or the root.
• Add Child Entry — inserts a new entry as the last child of the selected node. Enabled only when the selected node is a dictionary.
• Duplicate — deep-copies the selected node and its subtree, inserting the copy immediately after the original.

Commenting and deleting

• Comment Out — converts the selected entry into a commented-out unknown_raw_entry by prepending // to every non-blank line of its rendered text. The result is written back into the file as a block comment. Disabled when the entry is already commented out.
• Restore from Comment — reverses Comment Out: strips the // prefix from each line, re-parses the result, and replaces the unknown_raw_entry with the recovered node(s). Enabled only when every non-blank line of the entry starts with //.
• Delete — removes the selected node after a Yes/No confirmation dialog.

Bundled example cases

The tutorials/ directory in the repository root contains ready-to-open OpenFOAM cases:

Directory	Solver	What it demonstrates
tutorials/cavity/cavity/	icoFoam	Single-region end-to-end workflow
tutorials/cavity/cavityGrade/	icoFoam	Non-uniform grading (simpleGrading)
tutorials/cavity/cavityClipped/	icoFoam	Clipped geometry; mapFieldsDict
tutorials/snappyMultiRegionHeater/	chtMultiRegionFoam	Multi-region boundary view and region file listing
tutorials/damBreak/	interFoam	Two-phase flow; setFieldsDict and 0.orig/
tutorials/oneBlocks/	icoFoam	3-D single-block; blockMeshDict editing and 3-D mesh viewer
tutorials/oneBlocks-vars/	icoFoam	As oneBlocks with variable substitution and compact face notation
tutorials/nineBlocks/	icoFoam	3×3 multi-block; regex boundary patches
tutorials/nineBlocks-vars/	icoFoam	As nineBlocks with variable substitution and compact face notation

Open any case directly with Case > Open Case and navigate to the case subdirectory, or duplicate it to a working directory with Case > Duplicate from Case Library after adding tutorials/ to the Case Library.

These case files are licensed under the GPL-3.0, separate from the AGPL-3.0 that covers FoDE source code. See tutorials/README.md for full provenance and license details.

Case Library

The Case Library is a set of directories used as sources for browsing and duplicating reference cases. It is intended to provide quick access to $FOAM_TUTORIALS and other template case collections without navigating the filesystem from scratch each time.

Automatic inclusion of $FOAM_TUTORIALS

When the FOAM_TUTORIALS environment variable is set and points to an existing directory, that directory is automatically included at the top of the Case Library list on every application start. It does not need to be added manually and is not stored in app_config.json. If the variable is unset or the directory does not exist, it is simply absent from the list.

Managing user-added directories

Select Settings > Manage Case Library... to open the library management dialog. The dialog shows two sections.

• Auto-detected (read-only) — shows the $FOAM_TUTORIALS path if the environment variable is active. This entry cannot be removed here; it disappears automatically when the variable is unset.
• User-added directories — directories you have added manually. Use Add Directory... to browse for a directory to register. Check one or more entries and click Remove Selected (N) to remove them. Changes are saved to app_config.json.

Opening a case from the library

Select Case > Open from Case Library... to browse within a library directory and open a case from it. If multiple library directories are registered (including the auto-detected one), the application first asks which library to start from via a drop-down prompt, then opens a directory picker rooted at the chosen library.

Duplicating a case from the library

Select Case > Duplicate from Case Library... to copy a case from the library into your working directory. The workflow is as follows.

1. Choose a library directory (if more than one is available).
2. Navigate to and select the source case inside that library.
3. A duplicate dialog opens with the Default Case Directory pre-filled as the destination parent.
4. Adjust the destination and name if needed, then click OK to copy.
5. After a successful copy, the application offers to open the new case immediately.

The copy mode options (full copy vs. app-visible files only) are the same as for Case > Duplicate Case.

Resources dialog

Select Help > Resources... to open the Resources dialog. It has two tabs.

OpenFOAM tab

Displays links to official OpenFOAM documentation for the two main distributions — OpenCFD/ESI Group (openfoam.com) and the OpenFOAM Foundation (openfoam.org). These links are fixed and read-only. This application is not affiliated with either organisation.

My Links tab

Lets you manage a personal list of reference links.

• Add — opens a dialog to enter a label and a URL. The label is optional; if left blank, the URL is used as the display text.
• Edit — opens the same dialog pre-filled with the selected entry's label and URL.
• Remove — deletes the selected entry from the list.
• Move Up / Move Down — reorders the selected entry.
• Double-click an entry to open its URL in the default browser.

Changes are saved to app_config.json when the dialog is closed.

Saving as a new case

Select Case > Save as New Case... to write the current editor state to a new case directory and immediately switch to it. Unlike Duplicate Case (which copies files from disk), this operation captures whatever is in the in-memory buffers — including unsaved edits — so the new case reflects the exact state visible in the editor at the moment the command is used. The original case is not modified.

A dialog shows the following fields.

• Save in — the parent directory for the new case (defaults to the parent of the current case).
• New case name — the name for the new case directory (defaults to the current name with _new appended).
• Destination — a live preview of the full destination path.
• Copy mode — controls which files are copied from disk before unsaved edits are applied:
  • Copy app-visible files only (default) — copies only the files shown in the file panel. Lightweight; excludes mesh data, processor directories, and result time steps.
  • Copy all files — full directory copy using shutil.copytree. Includes everything: mesh data, processor directories, existing result time steps, logs, etc.

In both modes, any unsaved in-memory edits are written into the new case on top of the copied files. The original case is not modified.

After the copy, the application opens the new case automatically.

Duplicating a case

Select Case > Duplicate Case to save a copy of the current case under a new name. A dialog is shown with the following fields.

• Save in — the parent directory for the new case (defaults to the parent of the current case).
• New case name — the name for the new case directory (defaults to the current name with _copy appended).
• Destination — a live preview of the full destination path.
• Copy mode — choose how files are copied:
  • Copy all files (default) — performs a full directory copy using shutil.copytree. Every file and subdirectory under the case root is included, such as mesh data, time-step directories, and log files.
  • Copy app-visible files only — copies only the files listed in the file panel (system/controlDict, fvSchemes, fvSolution, etc., and all files under 0/ and 0.orig/). Files not shown by the application, such as constant/polyMesh, processor*/, and arbitrary time-step directories, are not copied.

If the destination already exists, a confirmation prompt is shown before overwriting. After a successful copy, the application offers to open the new case immediately.

If there are unsaved changes when Duplicate Case is pressed, the application asks whether to save all modified files before copying so that the duplicated case reflects the latest edits.

To duplicate a case from a reference directory rather than the currently open case, use Case > Duplicate from Case Library... (see Case Library). To duplicate an OpenFOAM tutorial case found by keyword search, use the Duplicate this case… button in Find OpenFOAM Examples.

Case comparison

Select Case > Compare with Case... to pick a reference case directory and compare it against the currently open case.

Once a reference is selected, a diff bar appears below the action bar. It shows a colour legend, the reference case name and full path, a Side by side toggle, and a Clear button to exit compare mode.

Side-by-side view

When a reference case is selected the centre panel splits horizontally: the left pane shows the current case's editable tree as usual, and a new Reference pane opens on the right showing the corresponding file from the reference case in a read-only tree. A green header bar at the top of the reference pane displays the reference case name.

Both trees are annotated with the same colour scheme (see Tree overlay below). The reference pane additionally uses light green for keys that exist only in the reference case but are absent from the current file.

Applying a value from the reference case:
Right-click any leaf node in the reference pane and select Use this value. The value is applied to the matching node in the current case's tree immediately, or inserted if the key is absent — including when its enclosing block does not exist yet: missing parent dictionaries (e.g. a functions {} block) are created automatically. Unnamed entries such as #includeFunc … directives are appended into the target block (and skipped with a message if an identical entry is already there). The only case that is refused is a name clash where the enclosing key exists in the current case but is not a dictionary. The diff highlighting updates automatically after the change.

Toggling side-by-side:
Check or uncheck Side by side in the diff bar to show or hide the reference pane without leaving compare mode. The reference pane is hidden (not merely collapsed) when unchecked, so no splitter gap appears. The current case's tree always remains visible and editable.

Tree overlay

While compare mode is active, both the main tree and the reference tree annotate rows with background colours.

Current case tree (left):

Background	Meaning
Light yellow	Key exists in both cases; value or type differs.
Light blue	Key exists in this case but is absent in the reference.
No colour	Key is identical, or the node type is not compared.

Reference tree (right):

Background	Meaning
Light yellow	Key exists in both cases; value or type differs.
Light green	Key exists in the reference but is absent from the current file.
No colour	Key is identical.

Hovering over a highlighted row shows a tooltip with additional context:

• Changed rows: Ref: <value> showing the reference value (left pane), or the current-case value (right pane).
• Only in current rows end with (not in reference case).
• Only in reference rows end with (only in reference case).

Comparison recurses into structural blocks (dictionary, boundary_block, boundary_entry, region_block, region_entry, field_value_block), matching children by key name. Positional list items (e.g. vertices, blocks) and anonymous nodes (#include, $macro) are not annotated.

If the corresponding file does not exist in the reference case, no overlay is applied and the status bar shows a "not found in reference case" message.

File list markers

When a reference case is selected, diff counts are computed progressively in the background — markers appear in the file list as each file is processed. Large files without a FoamFile header (e.g. log files, residual outputs) are skipped and will show no marker.

Suffix	Colour	Meaning
(none)	—	Not yet computed, or skipped (large non-dictionary file).
≠0	Gray	Identical to the reference.
≠N	Amber	N differences found.
≠50+	Amber	More than 50 differences (display cap).

Changed files only filter:
A Changed files only checkbox appears in the file list panel while compare mode is active. When checked, files with zero differences are hidden, leaving only files that differ from the reference case. Uncheck to restore the full list. The filter is reset automatically when compare mode is cleared.

Clearing compare mode

Click Clear in the diff bar to exit compare mode. This removes all ≠ markers from the file list, all background highlights from both trees, closes the reference pane, and hides the Changed files only filter. Selecting Case > Compare with Case... again replaces the current reference with a new one.

Reloading a case

Click the Reload Case button in the top bar or select Case > Reload Case to discard all in-memory edits and reload the current case from disk. All file buffers and dirty state are cleared, and the file list and boundary panel are refreshed exactly as if the case had just been opened.

If there are unsaved changes, a confirmation dialog shows the number of affected files and asks whether to proceed. Clicking Yes discards all changes; clicking No leaves the current state unchanged.

This is useful when you want to undo edits across multiple files without closing and reopening the case.

Tree and text workflow

The editor is designed so that raw text editing remains available even when the parser cannot fully interpret a file. The intended workflow is:

1. Open a case and select a file from the file list.
2. Load the raw text into the bottom editor immediately.
3. Parse the text and update the tree when parsing succeeds.
4. Use the tree for structure browsing and simple value editing. Selecting a row highlights its source lines in the editor (see Editor sync).
5. Use the lower text editor for direct manual editing.
6. Press Apply Text to Tree when you want to parse the edited text back into the tree. This also restores editor sync if the source lines had become stale.
7. Press Reload from Tree when you want to regenerate the text from the current parsed tree.
8. Press Save File to save the current raw text to disk, or Save Case to save all modified files at once.

Detail pane

The right-side detail pane shows contextual information for the selected node and lets you edit its value.

When a schema is defined for the selected key, the pane displays:

• Key Help — a description of what the setting controls.
• Key Supported In — the solver types or configurations where the setting applies.
• Key Note — additional remarks or caveats.
• Choices — a drop-down list of valid or common values. Selecting a choice updates the help fields below to show the description, supported-in information, and notes specific to that value.

These help fields are populated from schema modules. Because OpenFOAM is highly flexible and supports a large number of settings and solvers, the built-in schemas cover only a selection of common keys — not every possible entry. For keys without a schema annotation, the detail pane still shows the key name, type, and value, but the help fields are left blank. You can fill in the gaps by writing your own schema modules, which are plain Python files you can load at runtime via Settings > Manage Schema Modules.

For ordinary nodes, the pane also shows Key, Type, and an editable Value field. For field_value nodes, the pane shows Field Type, Field Name, and Value.

Numeric type promotion

OpenFOAM does not strictly distinguish between integer and floating-point values in most contexts. When you enter a fractional value (for example 0.5) for a node whose current type is int, the editor automatically promotes the node to scalar without an error or confirmation prompt. The Type column updates immediately to reflect the new type. Entering an integer value in a scalar node keeps the type as scalar.

Edit validation

If the value you enter is not valid for the node's current type — for example, a non-numeric string for a scalar node, or a word other than true/false/on/off/yes/no for a bool node — the edit is rejected silently and a message is shown in the status bar describing the problem. The node value is left unchanged.

Behavior on parse failure

If parsing fails after loading or after editing, the application shows a warning and keeps the raw text editor usable. In this state, the last successfully parsed tree remains available until a later parse succeeds.

When parsing succeeds but some entries could not be fully interpreted, those entries are preserved as unknown_raw_entry nodes in the tree and a warning is shown in the status bar reporting how many unrecognized entries were found (for example, "Parsed: system/controlDict — 2 unrecognized entries"). The file is still usable; the raw text is preserved verbatim for those entries.

Editor behavior

The bottom editor is based on the custom CodeEditor widget and is used for direct plain-text editing of OpenFOAM dictionary files. It shows the current line number in the lower area of the editor panel.

The editor applies syntax highlighting to OpenFOAM dictionary tokens: comments (// and /* */) appear grey italic, string literals green, #directives purple bold, $macro references orange, reserved keywords (FoamFile, true/false, uniform, nonuniform, etc.) blue bold, and numbers teal. Use the Highlight button in the editor toolbar to toggle highlighting on or off; the preference is saved across sessions.

The window title shows a * suffix when the editor content has unsaved changes, and clears it after save, reload from tree, or a successful apply-to-tree.

Code folding

The line-number gutter on the left of the editor contains a narrow fold column. For every { … } block that spans more than one line, a small triangle is drawn in the fold column on the opening-brace line:

Triangle	Meaning
▾ (down)	Block is expanded — click to collapse
▶ (right)	Block is collapsed — click to expand

Clicking the triangle hides or reveals the lines between the { and the matching }. The FoamFile { … } header block is collapsed automatically each time a file is loaded to keep the working dictionary entries at the top of the viewport.

Folding is visual only: the underlying text is not modified. Apply Text to Tree, Save File, and all copy operations work on the full, unmodified text regardless of the current fold state.

Supported syntax and node types

FoDE targets common OpenFOAM dictionary syntax with practical, tolerant handling rather than full specification coverage. The string shown in the Type column of the tree view is the exact node_type name for that row (see DEVELOPER.md for the internal Python representation of each type). The parser recognises the following node types:

Type	Description
word	A plain keyword–value pair (single unquoted token, fallback type)
string	A double-quoted string value, e.g. "my value"
macro	A $variable reference used as a value, e.g. $nu in nu $nu;
compound	Multiple space-separated tokens that do not form a list, e.g. uniform 0 or uniform (0 0 0)
scalar	A floating-point value
int	An integer value (auto-promoted to scalar when a fractional value is entered)
bool	A boolean value (true/false/yes/no/on/off)
vector	A parenthesised three-component value, e.g. (0 0 9.81)
box_pair	Two (x y z) vectors defining a bounding box; produced only for the box key
int_list	A parenthesised list of integers, e.g. (0 1 2); stored internally as int_list
scalar_list	A parenthesised list of floats (not exactly 3 items), e.g. (0.1 0.5 1.0 2.0); stored internally as scalar_list
raw_list	A parenthesised list with mixed or nested content (e.g. vertices, blocks); stored as raw text internally as raw_list
nonuniform_list	A nonuniform List<T> N (…) field value (e.g. internalField in 0/U); shown as a count summary in the tree and stored as raw text — not editable inline
dictionary	A sub-dictionary block
field_value_block	A block with internalField/boundaryField structure (field files such as 0/U)
field_value	An individual field value entry within a field_value_block
region_block	A named region block (used in setFieldsDict)
region_entry	An entry within a region_block
boundary_block	A boundary ( … ); block in blockMeshDict; each patch is a boundary_entry child
boundary_entry	A named patch entry within a boundary_block
directive_entry	A #include, #inputMode, or other pre-processor directive
macro_entry	A $variable macro expansion
unknown_raw_entry	Any entry the parser could not fully interpret; stored and written back as raw text. Displayed in amber in the tree to distinguish it from normal entries

Unrecognised syntax is preserved as unknown_raw_entry nodes and written back verbatim, so partially-parsed files are not corrupted.

field_value nodes in the tree — The Key column displays the field_name (e.g. U) and the Value column shows the field_type followed by the formatted value (e.g. volVectorField uniform (0 0 0)). In the Detail pane, these nodes show separate Field Type, Field Name, and Value fields. The value portion is editable; the type classification follows the same rules as for ordinary nodes.

Limitations

This project is a practical editor template, not a complete OpenFOAM parser. Known limitations at this stage include these points:

• Full lossless round-trip behavior is not guaranteed for every possible OpenFOAM syntax variant.
• Formatting and comment placement can change for modified nodes when text is regenerated from the tree.

Disclaimer

This offering is not approved or endorsed by OpenCFD Limited, producer and distributor of the OpenFOAM software via www.openfoam.com, and owner of the OPENFOAM® and OpenCFD® trade marks.

Acknowledgements

• PySide6 (Qt for Python) — GUI framework (LGPL v3).
• pyVista / VTK — 3-D viewer for blockMeshDict (BSD-3-Clause, optional).
• xterm.js — Terminal emulator used in the Terminal panel (MIT). Files are downloaded automatically from jsDelivr on first launch and cached in ui/xterm/.
• pytest / pytest-qt — Test framework (development only).
• PyInstaller — Used to build standalone executables.

Special thanks to the OpenFOAM Foundation and OpenCFD / ESI Group and all contributors for developing and maintaining OpenFOAM as free, open-source CFD software.

---

Copyright (C) 2025-2026 Shinji NAKAGAWA — AGPL-3.0-or-later