The Express Tools BURST command explodes one level and converts that level’s attributes. Blocks nested inside then have to be burst again, and again, as many times as the nesting goes deep. DEEPBURST does the whole tree in a single operation.
What DeepBurst does
Select the blocks, run the command, and every level comes apart with attributes converted to text as it goes. No repeated passes, no wondering whether you got the last one.
It also fixes BURST’s worst habit. Express BURST reveals invisible attributes — fields the block author deliberately hid, which suddenly appear as text all over the drawing. This routine checks each attribute’s invisible flag and discards hidden ones rather than converting them, so what you see after bursting is what you saw before.
How to use it
- Load the routine with
APPLOAD, or put it on your support file search path. - Type
DEEPBURST. - Select the blocks to burst. Xrefs cannot be selected — they are excluded by the filter.
- Every level comes apart in one pass.
How it works
Properties are resolved before the reference is deleted. Objects inside a block drawn on layer 0, or with BYBLOCK colour or linetype, inherit those properties from the block reference. Once the block is gone there is nothing left to inherit from. So before deleting the reference, its layer, colour and linetype are pushed onto any child that was relying on them.
Without that step, bursting silently changes the appearance of the drawing — everything that was on layer 0 inside the block lands on layer 0 in the drawing, in the wrong colour, and nobody notices until it plots.
Non-uniformly scaled blocks are handled by a fallback. The ActiveX Explode method refuses a block scaled differently in X, Y and Z. Those are dealt with by running the EXPLODE command on a temporary copy instead, with QAFLAGS cleared so the command does not prompt.
Xrefs are excluded at selection. The filter builds a list of every xref-dependent block definition and leaves them out, since an xref cannot be exploded.
Notes and limits
- Invisible attributes stay invisible — they are discarded rather than converted. This is the main behavioural difference from Express BURST.
- Bursting is not reversible in practice. One undo will put it back, but once you have carried on working, rebuilding the block structure is manual.
- Xrefs cannot be burst and are filtered out at selection.
- Dynamic blocks burst to their current state, which is the geometry you can see, not the base definition.
Download and details
| Name | DeepBurst — burst blocks through every level of nesting in one pass |
| Download | DeepBurst.lsp — 15.3 KB, plain AutoLISP source. Downloads directly; no zip to unpack. |
| Type | AutoCAD AutoLISP routine |
| Command | DEEPBURST |
| Requires | the ActiveX layer (loaded automatically by the routine with vl-load-com). |
| Source size | 15.3 KB, 375 lines |
| Error handling | Yes — a *error* handler restores every system variable it changed, on cancel as well as on error. |
| Undo | The whole operation is wrapped in a single undo group, so one U reverses all of it. |
| Compatibility | AutoCAD (any release with AutoLISP), BricsCAD, ZWCAD, GstarCAD, ProgeCAD and other IntelliCAD-based platforms. AutoCAD LT needs a LISP enabler. No .NET, no ObjectARX, no installer. |
| Author | YZ, August 2026 |
| Licence | Free to use, supplied “as is” with no warranty. |
References
- SMARTBURST LISP — bursting with more control over what the attributes become
- BLOCKPROMOTE LISP — extract one nested block instead of bursting everything
- DEEPCOUNT LISP — count what is nested before you take it apart
- More free AutoLISP routines on this site
- Wisey’s Steel Shapes — free structural steel section drawing program