Ignition Vision: Print All Windows as Single Multi-Page PDF

Overview

system.print.createPrintJob() wraps a single Swing component into one print job — calling it in a loop spawns one dialog per window. To produce a single multi-page PDF (or print job) covering an entire Ignition Vision project, you must either (a) assemble a java.awt.print.Book of Printable pages from each root container, or (b) rasterize each window into a BufferedImage and stitch the images into a PDF using a library such as Apache PDFBox bundled in your Ignition gateway.

Prerequisites

Method A — Java AWT Book: Single Print Dialog, Multiple Pages

This approach uses java.awt.print.Book (implements Pageable) to collect one Printable per window, then submits the whole book in a single PrinterJob. Selecting "Print to PDF" or "Microsoft Print to PDF" in the OS dialog writes all pages to one file.

# Button actionPerformed — Ignition Vision 8.1
# Jython 2.7
from java.awt.print import PrinterJob, Book, PageFormat, Printable
from java.awt import RenderingHints
import math

# ── 1. Define which windows to capture ──────────────────────────────────────
# Option A: all windows in the project
targetWindows = system.gui.getWindowNames()

# Option B: explicit subset (recommended for production)
# targetWindows = ['Overview', 'Process/Reactor', 'Alarms/ActiveAlarms']

# ── 2. Open every window (no sleep — EDT must stay free) ─────────────────────
openedRoots = []
for name in targetWindows:
    try:
        win = system.nav.openWindow(name)   # opens if not already open
        openedRoots.append(win.getRootContainer())
    except Exception as e:
        system.util.getLogger('PDFExport').warn('Skipping %s: %s' % (name, str(e)))

# ── 3. Build a Printable wrapper for each root container ─────────────────────
class RootPrintable(Printable):
    def __init__(self, component):
        self.component = component

    def print(self, g2d, pf, pageIndex):
        if pageIndex > 0:
            return Printable.NO_SUCH_PAGE
        g2d.translate(pf.getImageableX(), pf.getImageableY())
        iw = pf.getImageableWidth()
        ih = pf.getImageableHeight()
        scaleX = iw / float(self.component.getWidth())
        scaleY = ih / float(self.component.getHeight())
        scale  = min(scaleX, scaleY)
        g2d.scale(scale, scale)
        g2d.setRenderingHint(RenderingHints.KEY_ANTIALIASING,
                             RenderingHints.VALUE_ANTIALIAS_ON)
        self.component.paint(g2d)
        return Printable.PAGE_EXISTS

# ── 4. Defer actual print call until EDT finishes painting all windows ─────────
def submitPrintJob():
    pj = PrinterJob.getPrinterJob()
    book = Book()
    pf = pj.defaultPage()
    pf.setOrientation(PageFormat.LANDSCAPE)

    for rc in openedRoots:
        book.append(RootPrintable(rc), pf)

    pj.setPageable(book)
    pj.setJobName('IgnitionProjectSnapshot')
    if pj.printDialog():        # shows ONE dialog for all pages
        pj.print()

system.util.invokeLater(submitPrintJob)

Method B — BufferedImage Rasterization to Disk

Use this when you need PNG snapshots for archiving or want to drive a Python PDF library rather than the OS print dialog. Each window is painted into a BufferedImage and saved to a timestamped folder accessible from the gateway file system or a network share.

from java.awt.image import BufferedImage
from java.awt import RenderingHints
from javax.imageio import ImageIO
from java.io import File
import system, os

# Destination directory (gateway-accessible path via system.file or UNC)
OUT_DIR = 'C:/IgnitionSnapshots'
ts = system.date.format(system.date.now(), 'yyyyMMdd_HHmmss')
outPath = os.path.join(OUT_DIR, ts)
os.makedirs(outPath)

targetWindows = ['Overview', 'Process/Boiler', 'Alarms/ActiveAlarms']

def captureWindows():
    for name in targetWindows:
        try:
            win = system.nav.openWindow(name)
            rc  = win.getRootContainer()
            w, h = rc.getWidth(), rc.getHeight()
            img  = BufferedImage(w, h, BufferedImage.TYPE_INT_RGB)
            g    = img.createGraphics()
            g.setRenderingHint(RenderingHints.KEY_ANTIALIASING,
                               RenderingHints.VALUE_ANTIALIAS_ON)
            rc.paint(g)
            g.dispose()
            safeName = name.replace('/', '_').replace(' ', '_')
            ImageIO.write(img, 'PNG', File('%s/%s.png' % (outPath, safeName)))
        except Exception as e:
            system.util.getLogger('SnapExport').error('%s failed: %s' % (name, str(e)))

system.util.invokeLater(captureWindows)
system.gui.messageBox('Snapshots saved to:\n' + outPath)

Window Selection and Parameter Handling

Windows with required navigation parameters throw exceptions on openWindow(name) if no parameters are passed. Use system.nav.openWindow(name, params) where params is a PyDictionary. Maintain a configuration dictionary at the project script level:

# Project script: shared.pdf.windowConfig
WINDOW_MAP = {
    'Overview':            {},
    'Process/Reactor':     {'unitID': 1, 'mode': 'view'},
    'Process/Boiler':      {'unitID': 2, 'mode': 'view'},
    'Alarms/ActiveAlarms': {},
    'Reports/ShiftSummary': {'shift': 'Day'},
}

# In your button script, replace the openWindow loop:
for name, params in shared.pdf.windowConfig.WINDOW_MAP.items():
    win = system.nav.openWindow(name, params)
    openedRoots.append(win.getRootContainer())

Verification

1. Add a system.util.getLogger('PDFExport').info('Opened: ' + name) inside the open loop. Check Gateway → Diagnostics → Logs filtered on PDFExport to confirm each window opened without exceptions.
2. For Method A: after the print dialog appears, verify the page count in the dialog matches len(targetWindows). A mismatch indicates a window failed to open and was skipped.
3. For Method B: navigate to OUT_DIR and confirm one PNG per window name, file size > 50 KB (near-zero size = paint called before component rendered — add a second invokeLater nesting).
4. Verify no "red overlay" placeholders appear in captured images — this means a required window parameter was missing. Update WINDOW_MAP with correct defaults.
5. Test with a non-admin user account to confirm all target windows are within the user's role permissions before scheduling.

FAQ