Previous logic resolved duplicate connectionDesignation values by
appending sanitised description text or numeric suffixes, which was
fragile and produced names an electrician could not easily map back to
the physical wiring.
New scheme:
• terminalNr present → "XDI.2", "XRO1.3", "XPOW.1" …
• terminalNr absent → designation as-is ("L1/U1", "UDC+", "PE") —
these are busbar / power connections and are already globally unique
• collision (malformed data) → numeric suffix as safety net
The description (connectiondescription) remains as the human-readable
label beside the terminal symbol, exactly as suggested by plc-user.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
EdzPin gained a third field (group) but the fallback initializer at
line 70 still listed only two fields, triggering -Wmissing-field-initializers.
Added the explicit QString() for group.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
EPLAN 2022-style part.xml files (e.g. IFM AL1122) use a numeric
functiondefgroup attribute and do not carry the text functiondefinition
block name that the previous grouping logic relied on. Those parts
fell back to grouping by pin designation, producing symbols with all
pin "1"s stacked together, then all "2"s, etc. — the bug reported in
PR #513.
Fix: read terminalNr first (the physical M12/connector socket identifier,
e.g. "X01", "X31") as the primary group key; fall back to functiondefinition
text for older EPLAN formats that omit terminalNr. Pins within each
connector group are still sorted by designation using natural sort.
Also remove the dashed inter-group separator lines; the existing 5 px
gap between groups provides sufficient visual separation.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
The previous grouping compared `connectionDesignation` values (e.g. "1",
"2", "3") to detect group boundaries. Those values are pin position
numbers within a terminal block, not functional group identifiers, so
devices like the ABB ACS880 produced 11+ tiny single-pin "groups" instead
of the ~8 functional blocks (AC-IN, Motor-OUT, DC-Bus, Brake-Resistor,
Analog-I/O, Digital-I/O, ...) the reviewer identified.
Fix:
- Add `group` field to EdzPin, populated from the `functiondefinition`
attribute on <functiontemplate> (newer EPLAN) or its parent <function>
element (older EPLAN).
- Sort pins by group first (preserving XML appearance order per group),
then by designation within each group using natural sort.
- Use `groupKey()` — group when present, designation as fallback — for
the group-break detection that drives separator lines and Y-gaps.
Parts without any functiondefinition data retain the previous
designation-based behaviour unchanged.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Pins are already sorted by designation (see EdzPart::parse natural sort).
This change makes same-designation groups visually obvious in the element
symbol by:
- Inserting a 5 px gap between consecutive terminals whose designation
differs, so each group reads as a distinct block.
- Drawing a thin dashed horizontal line through each gap, mirroring
the grouped-I/O style shown in typical manufacturer datasheets.
The body rectangle and bounding box grow automatically to accommodate the
extra gaps, so no fixed sizes change. Unique terminal names (added in the
previous commit) are unaffected.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Two issues raised in PR review:
1. Copyright year: new files carried "2006" (the project's founding year).
Updated to "2006-2026" to reflect the actual authorship period.
2. Duplicate terminal names: EPLAN parts can have multiple connection
templates sharing the same connectionDesignation (e.g. a drive where
both wire entries of terminal "1" are labelled "1"). QET requires
unique terminal names for wiring and terminal-diagram generation.
Resolution order in EdzElementBuilder::build():
- Unique designation → used as-is.
- Duplicated designation with distinct sanitised description →
"designation_description" (e.g. "1_L_P").
- Otherwise → numeric suffix: "1", "1_2", "1_3", …
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- EdzArchive checks the archive magic up front: gives a clear message for
zip-format .edz (not yet supported) and unrecognised data, instead of an
opaque 7z decode error.
- Trim the vendored LZMA SDK headers to the decode closure actually used
(removes 21 unused encoder/multithread/Xz/Aes headers; 18 .c + 18 .h remain).
- Add sources/import/edz/README.md documenting the feature, the data mapping
and the bundled SDK.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The new import action and dialogs used French source strings (QET convention)
but had no English translation, so they showed French in the English UI. Add
the five strings to qet_en.ts (menu action, dialog title, file filter, error
box title and message).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Replaces the QProcess->7z shim with the public-domain LZMA SDK (23.01) 7-Zip
reader, vendored under sources/import/edz/lzma/ (decode-only subset). New
edzsevenzip.cpp wraps SzArEx_Open/SzArEx_Extract and writes entries via Qt, so
EdzArchive no longer needs an external 7-Zip at runtime. Enables the C language
in CMake for the vendored sources.
Verified: the bundled decoder extracts all three ifm sample .edz and the
generated elements still match the Python oracle exactly (byte-correct decode),
with no 7z on the path.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Wires EPLAN .edz import into the elements panel. EdzImporter orchestrates
EdzArchive -> EdzPart -> EdzElementBuilder and writes the generated .elmt into a
destination collection folder (named by order number). A right-click
"Importer une piece EPLAN (.edz)..." action on a writable collection directory
opens a file picker, runs the importer into that folder's fileSystemPath() and
reloads the panel; errors surface via QetMessageBox. Modeled on newElement().
EdzImporter verified headless against the Python oracle for KG6000/MFH200/
R1D200; the panel wiring is built/tested via the WSL Qt5+KF5 build.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Port of the edz2qet.py prototype's mapping to C++. EdzPart parses an EPLAN
part.xml into a portable model (identity/metadata, localized names, connection
list); EdzElementBuilder turns that into a QET element (generic symbol: body
rectangle + one west-facing terminal per pin, per-pin labels, localized <name>s,
and elementInformations for the BOM).
Pins are natural-sorted by designation so they stack 1,2,3,4 regardless of the
order EPLAN lists them (MFH200 lists 1,3,4,2). Output verified structurally
identical (uuids aside) to the Python oracle for three ifm samples — KG6000
(4-pin), MFH200 (4-pin, reordered) and R1D200 (5-pin, incl. Dutch name) — and
the generated element loads in the QET editor with correct UTF-8.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
First piece of native EPLAN Data Portal (.edz) import. A .edz is a 7-Zip
archive; QET previously had no archive handling. EdzArchive unpacks one to a
temporary directory and locates the contained part.xml.
The extraction backend is isolated behind extractWithSevenZipCli() so it can be
replaced with a bundled decompressor later without touching callers; this M0
step shells out to a 7-Zip CLI via QProcess. Verified on three ifm sample
parts (KG6000, MFH200, R1D200 — all 7z), plus the corrupt/missing-file paths.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The comments describing the terminal_names layout were inherited from a
previous version and no longer matched the actual assignment order:
terminal_names << nc_name << no_name << common_name;
i.e. [0]=NC, [1]=NO, [2]=Common
Update all affected comments to reflect the current storage order.
On macOS arm64 (Apple Silicon, Sequoia), exporting a PDF via
QPrintPreviewWidget leaves a black screen with only the mouse cursor
visible. Cmd+Tab restores the display; the exported PDF itself is
correct and clickable cross-reference links work fine.
Root cause
----------
requestPaint() is a slot connected to QPrintPreviewWidget::paintRequested.
Inside this slot the code was calling painter.end() manually, then
pdfConvertUriToGoTo(). On macOS arm64 the Qt5 paint cycle backed by
Metal/CALayer is asynchronous: closing the QPainter from *within* the
paintRequested slot interrupts the compositor before it has flushed the
backing store. The window goes black and never repaints because the
close() that follows immediately destroys it.
On x86_64 / older macOS (raster/CoreGraphics backend) the paint cycle is
synchronous, so the same code happened to work.
Fix
---
1. Remove the manual painter.end() and pdfConvertUriToGoTo() call from
requestPaint(). The QPainter is stack-allocated; it destructs normally
when the slot returns, which is the correct moment to flush the PDF.
2. In print(), capture the output file name before m_preview->print(),
then defer both pdfConvertUriToGoTo() and this->close() to the next
event-loop iteration via QTimer::singleShot(0, ...). This gives the
Metal compositor one full event-loop turn to finish compositing the
backing store before the window is torn down.
The fix is a no-op on all other platforms: QTimer::singleShot(0) posts
an event that fires in the very next iteration, so there is no perceptible
delay.
Tested
------
- macOS Sequoia 15.x, Apple M-series, Qt 5.15.x (arm64): black screen gone
- macOS 10.15 x86_64 VM, Qt 5.15.x: no regression
- Linux/Debian Qt 5.15.x: no regression
- PDF cross-reference links and GoTo/FitR destinations: unaffected
Fixes: black screen after PDF export on macOS arm64
directly to the related component on its folio, framing the target element.
When a project is exported to PDF, every cross-reference becomes an internal
link. Four kinds are covered:
- **Master → contact**: the contact list on a coil/relay (`CrossRefItem`)
- **Folio report → report**: report element labels (`DynamicElementTextItem`)
- **Slave → master**: the `(folio-position)` reference shown on a slave
(both standalone `DynamicElementTextItem` and grouped `ElementTextItemGroup`)
Clicking a link navigates **inside** the open document (no new viewer
instance) and zooms to frame the target element.
1. **Injection** (`printDiagram`, only when the paint engine is a `QPdfEngine`):
link rectangles are added with `QPdfEngine::drawHyperlink()`. The scene→page
mapping is rebuilt to match exactly what `QGraphicsScene::render()` does
(top-left anchored, `KeepAspectRatio`, **no centering**), and rectangles are
passed in device pixels — `pageMatrix()` already applies the 72/resolution
scale and Y-flip internally.
2. Each link URL encodes the target page and the target element's rectangle, in
PDF points on its own page: `#page=N&fitr=L_B_R_T`.
3. **Post-processing** (`pdfConvertUriToGoTo`, run after the painter is closed):
the `/S /URI` annotations are rewritten to native `/S /GoTo` actions with a
`/D [pageObj 0 R /FitR L B R T]` destination, and the xref table is rebuilt.
Pages are enumerated from the `/Pages /Kids` tree (reliable), not by scanning
for `/Type /Page` in raw bytes.
- `sources/print/projectprintwindow.{cpp,h}` — injection + post-processing
- `sources/qetgraphicsitem/crossrefitem.{cpp,h}` — `hoveredContactsMap()` accessor; store text rect for hit area
- `sources/qetgraphicsitem/dynamicelementtextitem.h` — `slaveXrefItem()` / `masterElement()` accessors
- `sources/qetgraphicsitem/elementtextitemgroup.h` — `slaveXrefItem()` accessor
- `qelectrotech.pro`, `cmake/qet_compilation_vars.cmake` — enable Qt gui-private headers (`<private/qpdf_p.h>`)
- **Fit-to-page mode only.** Links are not injected in tiled mode (multiple
pages per folio), which would require a per-tile transform.
- Uses Qt private API (`QPdfEngine::drawHyperlink`), stable since Qt 4 but not
part of the public API; the build links against `gui-private`.
- Page-tree enumeration assumes the flat `/Kids` array Qt produces (no nested
page trees).
- The frame zoom is controlled by two constants in `destRectPdf` (`pad`,
`minSide`) and can be tuned.
- Tested on Qt5; the `/Kids` parsing and `pageMatrix` behaviour are identical on
Qt6.
Fix a use-after-free crash (SIGSEGV in QRegion::begin, Qt5Gui+0x49af60)
confirmed by analysis of 19 coredumps. The crash was triggered when the
scene viewport clip region was freed during zoom/resize events while
QPicture::play() replayed drawPolyline commands through the scene painter.
Qt's raster engine then dereferenced a stale QRegionData pointer.
Root cause: CrossRefItem used three nested QPicture objects (m_drawing,
m_hdr_no_ctc, m_hdr_nc_ctc). The nested drawPicture() calls amplified
the use-after-free risk on any repaint event.
Fix: remove all QPicture from CrossRefItem entirely.
- updateLabel() now uses a QImage-backed dummy painter to compute
m_bounding_rect, m_shape_path and m_hovered_contacts_map geometry.
A bool m_update_map flag prevents the map from being overwritten
during paint().
- paint() calls drawAsCross()/drawAsContacts() directly on the scene
painter — no QPicture::play() anywhere in the class.
- buildHeaderContact() now draws NO/NC symbols directly onto the painter
instead of recording them into QPicture members.
Also fix mouseDoubleClickEvent: the element under the click is now found
directly from m_hovered_contacts_map using the event position, rather
than relying on m_hovered_contact which could be reset by hoverMoveEvent
between the two clicks of a double-click.
Also remove setBold(true) on terminal name labels: the Qt PDF/SVG/print
engine rendered bold at 4pt as extremely thick glyphs, making exports
unreadable. Normal weight at 4pt is correct and legible on all backends.
Fixes: SIGSEGV in CrossRefItem::paint() on zoom/resize
Fixes: double-click navigation unreliable on Xref contact symbols
Fixes: terminal name labels unreadable in PDF/SVG/print export
Add a new boolean property 'showTerminalName' (default: true) to
XRefProperties, with full persistence in XML and QSettings.
A new checkbox "Afficher les numéros de bornes dans les Xrefs" is
added to the XRefPropertiesWidget in the main display group (not in
the cross-only group), so it is active in both Cross and Contacts modes.
When unchecked, terminal names are hidden in all three rendering paths:
- drawContact() (Contacts mode: NO/NC/SW symbols)
- fillCrossRef() (Cross mode: NO and NC columns)
- setUpCrossBoundingRect() (Cross mode: bounding rect sizing)
Backward compatible: existing project files without the attribute
default to showTerminalName=true (no visual change).
Files changed:
sources/properties/xrefproperties.h
sources/properties/xrefproperties.cpp
sources/ui/xrefpropertieswidget.ui
sources/ui/xrefpropertieswidget.cpp
sources/qetgraphicsitem/crossrefitem.cpp
The previous sort used QString::toInt() to order terminal names,
which returns 0 for any string containing a non-numeric prefix
(e.g. "R1", "R2", "L1", "L2"...). This caused undefined sort order
and incorrect pole pairing in the Xref contact mirror.
Example: a 4-pole NC power contact with terminals R1..R8 was
displaying R1/R3, R5/R7, R2/R4, R6/R8 instead of the correct
R1/R2, R3/R4, R5/R6, R7/R8.
Fix: extract the trailing numeric part of each terminal name and
compare prefixes separately. If both names share the same prefix
and both have a trailing number, sort numerically on that number;
otherwise fall back to full string comparison.
This covers all naming conventions: "1"/"2"/"3", "R1"/"R2"/"R3",
"L1"/"L2"/"L3", etc.
Applied in both drawContact() and setUpCrossBoundingRect().
Three bugs were causing a crash (SIGSEGV in QRegion::begin) when
a power NC slave element was placed on a folio, even before linking
it to a master element.
1. m_drawing (QPicture) was never reset between updateLabel() calls.
QPicture accumulates paint commands — calling qp.begin() on an
existing QPicture appends to it rather than replacing its content.
After several updates (load, move, hover...) the picture became
corrupted and crashed on play().
Fix: reset m_drawing = QPicture() at the start of updateLabel().
2. m_drawed_contacts was only initialized to 0 in drawAsContacts(),
but not in drawAsCross(). When drawing in Cross mode, fillCrossRef()
called drawContact() with an uninitialized m_drawed_contacts value,
producing a garbage offset. The NC contact symbol uses drawPolyline()
with a sub-pixel Y coordinate (offset+2.5); with a random offset Qt
generated an invalid QRegion and crashed.
This explains why NC contacts crashed but NO contacts did not: the
NO symbol only uses drawLine() which is more tolerant of bad coords.
Fix: add m_drawed_contacts = 0 at the start of drawAsCross().
3. setUpCrossBoundingRect() used QRectF() (null rect) as the reference
rect for painter.boundingRect(), which can return invalid dimensions.
Additionally, height was accumulated incorrectly: united() + setHeight()
doubled the height at each iteration, causing an exponentially growing
bounding rect with multiple contacts.
Fix: use QRectF(0, 0, 500, 20) as reference rect and accumulate
height and width independently.
Extend TerminalData::Type enum with three new semantic values:
- No : Normally Open terminal of a switch (SW) contact
- Nc : Normally Closed terminal of a switch (SW) contact
- Common : Common terminal of a switch (SW) contact
Update typeToString() and typeFromString() accordingly.
Fully backward compatible: existing Generic/Inner/Outer types
are unchanged. Elements without typed terminals fall back
to the previous behavior (first 2 named terminals).
terminal: expose terminalType() as public accessor
Add Terminal::terminalType() returning the TerminalData::Type
of this terminal. This allows crossrefitem and other consumers
to filter terminals by semantic role (No, Nc, Common) without
accessing TerminalData internals directly.
terminaleditor: add No, Nc, Common entries to type combobox
Expose the three new TerminalData types (No, Nc, Common) in
the element editor UI so users can assign a semantic role to
each terminal of a SW contact element.
Also fix a pre-existing bug in updateForm() where m_type_cb
was incorrectly using m_orientation_cb->findData() instead
of m_type_cb->findData(), preventing the type from being
restored correctly when selecting a terminal.
terminaleditor: add No, Nc, Common entries to type combobox
Expose the three new TerminalData types (No, Nc, Common) in
the element editor UI so users can assign a semantic role to
each terminal of a SW contact element.
Also fix a pre-existing bug in updateForm() where m_type_cb
was incorrectly using m_orientation_cb->findData() instead
of m_type_cb->findData(), preventing the type from being
restored correctly when selecting a terminal.
Shane Ringrose