A1 Swap Mod Packer

Current version: v0.5.0

A1 Swap Mod Packer is an open-source implementation 3MF packer for Bambu Lab A1 SwapMod workflows.

It takes one or more A1-sliced .3mf files, repeats their plate G-code according to copy counts, inserts an external SwapMod ejection/swap G-code block, and writes a new packed .3mf that can be sent to the printer.

Notes

• This project contains no proprietary code. All functionalities were implemented independently by comparing the states before and after 3MF file generation.
• Currently, this project has only been tested and verified on Windows 10.
• Always test a generated file on a safe/simple print before using it in production.

Key features

• GUI for drag-and-drop batch packing.
• CLI for repeatable automation.
• Multiple input .3mf files.
• Per-file copy count.
• Automatic time and filament summary from Metadata/slice_info.config.
• External SwapMod G-code template folder.
• Editable G-code patch file.
• Optional remaining-time plate-number encoding through M73 R....
• Selected-input thumbnail preview in the GUI.
• Packed preview image composition from up to 9 unique input files, with a short {plates} P label.
• Combined mode: all input rows become one packed 3MF.
• Individual batch mode: every input row becomes its own packed 3MF.
• zlib-ng Deflate ZIP compression level, defaulting to Level 7.
• GUI settings are saved to the program folder as settings.json.

Installation for source use

Python 3.10 or newer is recommended.

Install dependencies:

pip install -r requirements.txt

Current runtime dependencies:

Dependency	Used for
PySide6	GUI application, drag-and-drop table, file dialogs, thumbnail preview, and settings UI
Pillow	Reading and composing PNG preview images inside the output 3MF, including the {plates} P label
zlib-ng	Required ZIP Deflate backend for writing compressed 3MF files and honoring the compression level

No Java runtime, Bambu Studio SDK, external archive tool, or encrypted/template decoder is required by the current Python source version.

Run the GUI:

python -m a1_swap_mod_packer.gui

or

python run_gui.py

Run the CLI:

python -m a1_swap_mod_packer.cli --help

or

python run_cli.py

Show the version:

python -m a1_swap_mod_packer.cli --version

Build Windows executables

The repository includes a Windows build script:

build_win.cmd

The script uses Nuitka onefile mode and creates this portable release folder:

build/onefile/

Expected output:

build/onefile/
  a1packer.exe
  a1packer-cli.exe
  gcode_patches.ini
  swap_gcode/
  settings.json      Optional; created by the GUI after settings are saved

External resources are intentionally kept outside the executables. Do not pack these files into the onefile binary:

swap_gcode/
gcode_patches.ini
settings.json

The application resolves these paths from the executable folder when running as a packaged exe.

Windows build requirements

Tested build environment:

• Windows 10.
• Python 3.10 or newer.
• A virtual environment at .venv/.
• Project runtime dependencies from requirements.txt.
• Nuitka with onefile support.
• Microsoft C++ Build Tools / Visual Studio Build Tools with the MSVC compiler and Windows SDK.

Install the Python dependencies in a fresh virtual environment:

py -3.12 -m venv .venv
.venv\Scripts\python.exe -m pip install --upgrade pip
.venv\Scripts\python.exe -m pip install -r requirements.txt
.venv\Scripts\python.exe -m pip install "Nuitka[onefile]"

Then run:

build_win.cmd

The build script performs two Nuitka builds:

• GUI: run_gui.py -> build/onefile/a1packer.exe
• CLI: run_cli.py -> build/onefile/a1packer-cli.exe

The GUI build enables the PySide6 plugin and includes the Qt plugin groups needed by the current interface:

platforms,imageformats,styles,iconengines

After building

Before publishing the folder, verify the CLI can see the external resources:

build\onefile\a1packer-cli.exe --version
build\onefile\a1packer-cli.exe list-swap-gcode

The second command should list files from:

build/onefile/swap_gcode/

Also start build/onefile/a1packer.exe once and confirm that:

• The Swap G-code dropdown lists the copied templates.
• Editing GUI options creates or updates build/onefile/settings.json.
• The app still runs if settings.json is absent before first launch.

Swap G-code templates

The GUI automatically scans this fixed folder:

swap_gcode/

Plain UTF-8 text files are read directly. The current scanner accepts these suffixes:

.gcode
.nc
.ngc
.txt

The current source version does not decode encrypted or vendor-template G-code archives.

In the GUI:

1. Put template files in swap_gcode/.
2. Click Refresh next to Swap G-code.
3. Select the template from the dropdown.

Editable G-code patches

The GUI and CLI read the fixed patch file:

gcode_patches.ini

Default rule:

[patch.a1_start_y]
enabled = true
flag = G0 X128 F30000
find = G0 Y254 F3000
replace = G0 Y250 F3000 ; Patched
max_count = 1

Meaning:

• Search after the optional flag line.
• Replace the first matching find line with replace.
• Apply at most max_count replacements.

You can disable this in the GUI with G-code patches, or edit the INI file and keep the option enabled.

The swap insertion marker can also be edited:

[swap]
insert_before_marker = ;=====printer finish  sound=========

GUI guide

Input 3MF files

The input table supports:

• Add 3MF: choose one or more .3mf files.
• Drag and drop .3mf files into the table.
• Drag and drop a folder into the table. The GUI adds top-level .3mf files from that folder.
• Remove: remove selected rows.
• Remove All: clear the full input list.
• Move Up / Move Down: change input order.
• Apply Default Copies to Selected: overwrite selected rows with the current default copy count.

Columns:

• 3MF path: source file path.
• Copies: how many times this source should be repeated.
• Time: estimated print time multiplied by copies.
• Filament: estimated filament multiplied by copies.

The time and filament values are read from:

Metadata/slice_info.config

If the source 3MF lacks this metadata, the GUI shows Unknown.

The summary line under the input list shows the total plate count, estimated time, and filament for the current table.

The right-side thumbnail panel shows the selected input file's active plate preview when one is available in the 3MF.

Build 3MF button

The Build 3MF button is placed below the input list on the right side for fast batch operation.

In normal combined mode, one click creates one output 3MF containing all input rows.

In individual batch mode, one click creates one output 3MF per input row.

Swap G-code

Selects the ejection/swap G-code block inserted into each repeated plate.

The dropdown is generated from files in:

swap_gcode/

Buttons:

• Refresh: rescan the folder.
• Open Folder: open the template folder.

Default copies for new inputs

Sets the default copy count applied to newly added or dragged-in files.

This does not automatically change existing rows. Use Apply Default Copies to Selected for existing rows.

Bed cooldown

Controls whether the tool inserts a bed temperature wait before the SwapMod G-code block.

Enabled example:

M190 S45

If disabled, no M190 line is added by the packer.

Wait after ejection

Adds a dwell after the SwapMod G-code block.

Example for 45 seconds:

G4 P45000

Remaining-time plate number

When enabled, the packer offsets M73 ... R... remaining-time values by:

plate_number × 100 hours × 60 minutes

This allows the A1 remaining-time hour hundreds digit to display the current plate number.

Example concept:

• Plate 1: +6000 minutes
• Plate 2: +12000 minutes
• Plate 3: +18000 minutes

Final swap

When enabled, the swap/ejection block is also inserted after the final repeated plate.

When disabled, the final repeated plate ends normally without running the swap G-code block.

G-code patches

Applies rules from:

gcode_patches.ini

The Open Config button opens that fixed file for editing.

3MF metadata

Options:

• Keep source prediction and weight
  Leaves slice_info.config prediction and weight values from the base 3MF unchanged. This best matches the observed vendor packer behavior.

• Sum prediction and filament
  Updates the first plate metadata using the sum of all repeated source plates. This is more logically correct for statistics, but may differ from the vendor software.

Preview image handling

By default, the output 3MF keeps the base archive preview members for the active output plate and rewrites those PNG previews with:

• a composite of up to 9 unique input-file preview images;
• a short green label such as 5 P, applied once to the final composite.

If preview images are missing or cannot be read, the packer keeps the available base preview and still tries to apply the short plate label.

The CLI can disable preview rewriting with:

--no-preview-label

Batch mode

Combined mode

Default behavior.

All input rows are packed into one output file.

Example:

A.3mf copies 2
B.3mf copies 3

Output:

one packed 3MF containing A, A, B, B, B

Individual batch mode

When enabled, the GUI shows an explanation popup.

Every input row is treated as an independent build.

Example:

A.3mf copies 5
B.3mf copies 5
C.3mf copies 5

Output:

5 Plates - A.3mf
5 Plates - B.3mf
5 Plates - C.3mf

This is intended for quickly converting a large batch of independent single-plate 3MF files into multi-copy SwapMod packs.

The GUI builds these independent outputs in parallel, capped by CPU count and a small safety limit.

It does not combine all inputs into one file.

Input handling

Options:

• Skip duplicate file paths when adding inputs
  Prevents accidentally adding the same path multiple times.

• Clear input list after successful build
  Clears the table after a successful build. In individual batch mode, the list is cleared only if all outputs are built successfully.

Output directory

If this field is empty:

• Combined mode writes next to the first input file.
• Individual batch mode writes each output next to its own input file.

If a directory is selected, all outputs go to that directory.

Output filename rule

Default:

{plates} Plates - {sources}.3mf

Click the ? button next to the rule field to show token help.

Available tokens:

Token	Meaning
{source}	First input file stem, without .3mf
{sources}	Source summary. One source uses its stem; multiple sources become first_source_and_N_more
{plates}	Total plate count in this output
{copies}	Total copy count in this output
{date}	Current date as YYYYMMDD
{time}	Current time as HHMMSS

In individual batch mode, tokens are calculated per input row.

Examples:

{plates} Plates - {sources}.3mf
SwapMod - {source} - x{copies}.3mf
{date}_{time}_{source}.3mf

GUI settings

The GUI writes settings to the program folder:

settings.json

This is intentional so a portable extracted folder or future packaged .exe build keeps its options next to the application.

Saved options include:

• Selected Swap G-code file.
• Default copies.
• Bed cooldown settings.
• Wait after ejection.
• Remaining-time plate-number switch.
• Final swap switch.
• G-code patch switch.
• Metadata mode.
• ZIP compression level.
• Individual batch mode.
• Input handling options.
• Output directory.
• Output filename rule.

CLI examples

List available Swap G-code files:

python -m a1_swap_mod_packer.cli list-swap-gcode

Build one source with five copies:

python -m a1_swap_mod_packer.cli build \
  --item "SC05720_01.gcode(1).3mf" 5 \
  --swap-gcode "a1_swap.gcode" \
  --cool-bed 45 \
  --wait 45 \
  --show-plate-number \
  -o "5 Plates - SC05720_01.gcode(1).3mf"

Build multiple sources into one output:

python -m a1_swap_mod_packer.cli build \
  --item "A.3mf" 2 \
  --item "B.3mf" 3 \
  --swap-gcode "a1_swap.gcode" \
  -o "5 Plates - A_and_B.3mf"

Use summed metadata instead of source metadata:

python -m a1_swap_mod_packer.cli build \
  --item "A.3mf" 5 \
  --swap-gcode "a1_swap.gcode" \
  --metadata-mode sum \
  -o "5 Plates - A.3mf"

Disable editable G-code patches:

python -m a1_swap_mod_packer.cli build \
  --item "A.3mf" 5 \
  --swap-gcode "a1_swap.gcode" \
  --no-gcode-patches \
  -o "5 Plates - A.3mf"

Build positional inputs with the same copy count:

python -m a1_swap_mod_packer.cli build \
  "A.3mf" "B.3mf" \
  --copies 2 \
  --swap-gcode "a1_swap.gcode" \
  -o "4 Plates - A_and_B.3mf"

Use an explicit Swap G-code folder and ZIP compression level:

python -m a1_swap_mod_packer.cli build \
  --item "A.3mf" 5 \
  --swap-gcode "a1_swap.gcode" \
  --swap-gcode-dir "D:\SwapMod\swap_gcode" \
  --zip-level 9 \
  -o "5 Plates - A.3mf"

Disable final swap and preview label rewriting:

python -m a1_swap_mod_packer.cli build \
  --item "A.3mf" 5 \
  --swap-gcode "a1_swap.gcode" \
  --no-swap-after-final \
  --no-preview-label \
  -o "5 Plates - A.3mf"

Useful CLI options:

Option	Meaning
--version	Print the current version
list-swap-gcode	List files found in the Swap G-code folder
--item PATH COPIES	Add one input with its own copy count; can be repeated
positional inputs + --copies N	Add multiple inputs with the same copy count
--swap-gcode-dir DIR	Use a custom Swap G-code folder
--no-bed-cooldown	Do not insert M190 before the swap block
--no-swap-after-final	Do not run the swap block after the final plate
`--line-ending lf	crlf`
--zip-level 1-9	zlib-ng Deflate compression level; default is 7
--no-preview-label	Do not rewrite the preview image label/composite
--no-gcode-patches	Do not apply rules from gcode_patches.ini