Element::linkedElements() sorted the element's own list of linked
elements by position before returning it, so any caller reordered what
is saved. #1441 restores the file's order after linking; #1442 fills the
link table through linkedElements(), which sorted it again, so on master
a coil's contacts are saved in position order and
tst_resaveunchanged::savedLinkOrderKept fails.
linkedElements() now returns a sorted copy. Callers still get the list
by position; the saved order is left alone.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019Wga9B92BNurDEtxxtCDxf
A summary or parts-list table split over several folios saves its data
with its first part only; each later part names the part before it.
Deleting a folio keeps its tables in memory for undo, still in the
chain, so saving then named the deleted part as the previous table of
the next one. On reopening, every part after the gap had no data, and
closing the project crashed: the first orphaned part, destroyed, handed
its missing data on to the next part through a null model.
toXml() now names the nearest previous part still in the project, or
saves the data if there is none. setPreviousTable() no longer copies a
null model, so files already saved this way open and close.
Tests (tst_resaveunchanged): missingPreviousTableOpens,
deletedFolioKeepsTableChain; both fail without the fix.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019Wga9B92BNurDEtxxtCDxf
On opening a project, a coil and its contacts are linked in the order
the folios' items are visited. When contacts are on an earlier folio
than the coil they link first, so the coil's list followed the folio
order instead of the file's, and contacts on one folio swapped places
from run to run. Saving the same project twice could then give two
different files.
After linking, Element::initLink() now puts the linked elements back in
the order the file listed them. Elements linked but not in that list
keep their place after it.
Test: tst_resaveunchanged::savedLinkOrderKept reverses a coil's two
contacts in m_000.qet and checks a resave keeps that order; it fails
without the fix.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_019Wga9B92BNurDEtxxtCDxf
DiagramImageItem::fromXml() read the size attribute with a plain
toDouble(). A picture whose size was missing or not a number got size
0, so it was invisible on the folio, left out of the SVG export, and
a resave wrote size="0" back, which then stayed 0 on every load. "nan"
was kept and written back as such. The <transform> child, which newer
files carry, already defaults its scale to 1.
Read the size with an ok flag and use 1, the size a picture is added
with, when it is missing, not a number, not finite, or 0.
No file-format change for usable sizes. A picture that an earlier save
left at size="0" shows again, at size 1, after this change.
Tests: tst_imagemissingsize runs the binary's --resave and --export-svg
on a fixture with six small pictures: size missing, "0", "nan", "big",
"0.5" and "2". The first four fail on master (saved as 0, 0, nan, 0)
and the export draws only 3 of the 6 pictures there.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Signed-off-by: Beat Hangartner <beat@hangartners.ch>
"nan", "inf" and "-inf" parse as numbers, and IndependentTextItem::
fromXml() used them as they came for x, y and rotation. A project with
such a value kept it: --resave wrote x="nan" back, and the SVG export
drew the text at matrix(1,0,0,1,nan,95), which is not valid SVG.
Conductors and element texts already load a non-finite position as 0.
Load a non-finite x, y or rotation of a free text as 0, the same way.
No file-format change: finite values load and save as before.
Tests: tst_freetextnonfinite runs the binary's --resave and --export-svg
on a fixture with x="nan", y="inf", x="-inf", rotation="nan" and a
finite text (61.7, 61.3, 37 degrees, kept as they are). The four
non-finite cases and the export check fail on master.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Signed-off-by: Beat Hangartner <beat@hangartners.ch>
Symbols copied in old versions can share one uuid on a folio. Up to
0.100 that was harmless: a wire on a terminal without uuid was saved by
terminal number. Since #1118 every terminal has a uuid, so every wire was
saved by symbol uuid and terminal uuid, and on reopening findTerminal()
took the first symbol carrying it: all the copies' wires landed on it.
- Save: a wire end on a symbol whose uuid another symbol of the folio
also carries is written by terminal number again.
- Load: when a wire's uuid names several symbols, pickEnds() takes the
pair whose distance matches the wire's saved path, then the symbols
whose labels match the ones saved with the wire, then two terminals no
wire joins yet; otherwise the first, as before. This repairs a file a
build with #1118 saved once.
tst_sharedsymboluuid gives the three lamps of
Habitat-Schemas_developpes.qet one uuid: master loses a net after a
save (8 of 9) and reopens the file it saved with 8; with this, 9 both
times. The 25 examples save the same as before.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A shape remembers whether its pivot was placed by hand: a hand-placed
pivot is saved and no longer follows the shape's centre when the shape
is resized. Dragging the pivot handle marked it as hand-placed, but the
undo step restored only the pivot and the position. After Ctrl+Z the
pivot was back at the centre and still marked as hand-placed, so it
stayed behind on the next resize and was saved as a <transform>.
pivotIsCustom becomes a property and goes into the pivot drag's undo
step, the same fix as #1367 for pictures.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Pressing Escape while a handle of a picture was dragged went to
DiagramView, which clears the selection. That removed the handles under
the mouse, so the drag stopped where it was: the picture kept the new
size, angle or pivot, and since the release never arrived, no undo step
was recorded and the change could not be undone.
Diagram gets a keyboardItem: an item in a state Escape must end. While
a resize, rotate, skew or pivot handle is dragged, the picture is that
item; DiagramView sends each key to it first, and only keys it does not
take reach the view's own shortcuts. Escape puts the picture back as it
was when the handle was pressed (position, transform, and whether the
pivot was placed by hand) and ignores the rest of the gesture, so no
undo step is recorded. The drag also ends when the handles go away.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Generated by tools/qet-en-source/regenerate.sh, not edited by hand. The
tree is the fresh conversion; this commit only joins it to the PR's
history so the branch is never force-pushed.
Two position comparisons handed to std::sort were not strict weak
orderings, which std::sort requires; with the wrong kind of comparator
the sort is undefined behaviour (libstdc++ can read past the range,
MSVC debug builds assert "invalid comparator").
- comparPos(), used when renumbering the elements of a project, ended
with "<=" on x and y, so two elements at the same position (pasted at
the origin, placed by a script, stacked symbols) were each "before"
the other.
- The terminal numbering dialog compared positions with a 1 px
tolerance ("within 1 px counts as aligned, then compare the other
axis"), which is not transitive: terminals at x = 2, 1.1 and 0.2 give
a < b, b < c and c < a.
Move the comparisons into positionorder.h, a header-only helper:
xThenY()/yThenX() with "<", and roundedXThenY()/roundedYThenX(), which
round the positions to whole pixels first so that items a fraction of a
pixel apart still count as aligned, as the tolerance meant to, while
staying transitive. comparPos() keeps its folio and row-letter stages
and calls xThenY() for the last one.
No file-format change. Elements at distinct positions sort exactly as
before; the terminal order changes only for terminals less than a pixel
apart that straddle a half-pixel boundary.
Tests: tst_positionorder checks each order on a grid of awkward
positions by brute force (irreflexive, asymmetric, transitive, with a
transitive equivalence), the three-terminal cycle, twenty items at one
position, and that std::sort leaves the list sorted and intact. The
helper is new, so the test cannot fail on master; the call sites are
the two replacements in the diff.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Signed-off-by: Beat Hangartner <beat@hangartners.ch>
Every tr() source string in the code and the forms is now the English
text, and French is a translation (qet_fr.ts) like the other languages.
Generated, not hand-edited: master ef795a21e5 converted by
qet-en-source (harness 4c4ddc9).
3144 messages get an English source, 665 already had one, 0 left French; 8 shared-wording pairs disambiguated.
C++: 2904 literals in 233 files. Forms: 809 strings in 65 files. All 34 .ts files re-keyed; qet_fr.ts filled.
lupdate check: PASS. What each of 38 languages shows, old vs new: same 118266, English instead of French 38433, comment fallback 51, differences 0.
Before the <sequentialNumbers> element, an element's or conductor's
sequential numbers were saved as the attributes sequ_1, sequf_1,
seqt_1, seqtf_1, seqh_1 and seqhf_1. The readers check for one of them
to take the old route, and all three lists had the same slip: sequf_1
twice, seqhf_1 never (Element::fromXml, Conductor::fromXml, and
readSequence() in the project database). A file whose only old
sequence was the hundred-folio one took the new route, found no
<sequentialNumbers>, and lost it; the database built its labels from
an empty sequence instead of refusing the fast path as it does for the
other five attributes.
Name seqhf_1 in the three lists.
No file-format change: nothing is written differently, and a file with
any of the other five attributes loads exactly as before.
Tests: tst_legacysequentialattributes runs the binary's --resave on a
fixture whose element and conductor carry only seqhf_1 (and a second
pair carrying sequ_1 as a control) and reads the <sequentialNumbers>
written back. The two seqhf_1 cases fail on master.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Signed-off-by: Beat Hangartner <beat@hangartners.ch>
Before #1310, undoing a crop left the crop rectangle in place, so a
project saved afterwards has a <crop> but shows the uncropped picture.
Since #1368 the picture shown is computed from the original and the
crop, so the first mirror or colour key of such a picture applied the
old crop again, and undoing that edit showed the cropped picture.
The picture shown is always exactly the size of the crop. fromXml() now
drops a crop that does not match it: the crop becomes the whole
original, and when the shown picture is not the original's size either,
the shown picture becomes the original. Valid files load and save as
before.
New tst_imagestalecrop: a valid crop is kept; a stale one is dropped
for a shown picture of the original's size and of another size, where a
later crop cuts the picture that was shown.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A picture remembers whether its pivot was placed by hand: a hand-placed
pivot is saved and kept through a resize, a default one is not saved
and recentres after a resize. Two actions changed that flag outside
their undo step:
- applyCrop() marked the pivot as default after moving it to the centre
of the kept region. After Ctrl+Z the pivot was back where the user
had put it, but marked as default, so the next save dropped it.
- dragging the pivot handle marked it as hand-placed. After Ctrl+Z the
pivot was back at the centre but still marked as hand-placed, so a
later resize left it at the handle's anchor corner.
pivotIsCustom becomes a property, and both actions put it into their
undo command next to rawPivot. tst_imagecropundo checks that a project
saved after undoing a crop keeps its hand-placed pivot.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Brings in master's DXF blocks (#1339) through #1354. The upright texts
now live in drawSymbol() too, and a turned symbol whose texts stay
horizontal is drawn in full like a mirrored one: an INSERT would turn
its texts.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Conflicts with the DXF blocks export (#1350) and the picture factory
leak fix. The mirror now lives in drawSymbol(); a mirrored symbol is
drawn in full rather than as an INSERT of its block, since a negative
INSERT scale would mirror its texts too.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The displayed pixmap and imageSource (original, crop rectangle,
transparent colours) were two undo values that had to change together,
although the pixmap follows from the source. An action that updated
one and not the other would bring back the bug fixed in #1310.
setImageSource() now recomputes the displayed pixmap, and crop, colour
key, mirror and replace push one undo step on imageSource alone. A
plain QUndoCommand holds the property change, so two identical actions
in a row stay two steps. Loading a project still shows the saved pixmap
as before.
tst_imagecropundo also checks the size of the saved picture after undo
and after redo.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
A symbol text on a folio with its width on "Auto" kept every line of a
multi-line text on the left, whatever its alignment: a QTextDocument
with no text width has no width to centre or right-align lines in. It
only worked once the user gave the text a width wide enough.
With the width on Auto and a centre or right alignment, give the
document the width of its longest line. The size of the text does not
change, only where the shorter lines sit. The fit is redone when the
text, the alignment, the width or the font changes; on a font change
the text's anchor is kept, as the font change itself does.
textWidth() now returns the width the user set (-1 for Auto), the value
that is saved, rather than the document's, as IndependentTextItem
already does. Readers (properties panel, resize-handle undo, script
API, project database) see -1 for Auto as before.
Checked: every folio of the 25 example projects renders identically to
master (136 PNGs); a centred and a right-aligned two-line text now
render aligned; changing their font size leaves the same box and anchor
as master, and without the font-change refit the lines wrap.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
itemChange() removed the corner handles when the text left the scene
but did not bring them back when it returned still selected, which Qt
allows (undoing a scripted delete does it) without a selection change.
Refresh on every scene change. The two handle helpers are protected:
nothing outside the class calls them.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
A project can now keep the texts drawn in its symbols, and the names of
their terminals, horizontal when a symbol is turned: Project properties >
General, "Garder horizontaux les textes des symboles pivotés". The box of
each text turns with the symbol; the text does not, and reads as it does
in the symbol itself.
It is a project setting, saved as <symbol_texts upright="true"/> and
only when on. A new project starts with it on; a project saved without
it (every existing one) reads with it off, so it looks and saves exactly
as before, and looks the same on every computer. The MCP server's new
projects start with it on too.
It builds on the mirror of #1354, which already redraws the texts of a
mirrored symbol readable: what the symbol does to its texts is now its
mirrors and, with the setting on, its turn (Element::symbolTextsTransform()).
ElementPictureFactory caches one drawing per such transform, terminal
names undo it the same way, and the DXF export places the texts alike.
The fields of a symbol (label, comment...) already keep their angle with
"Garder la rotation visuelle" and are left as they are.
Known limit: two texts stacked in a symbol end up side by side when it is
turned, and can overlap when kept horizontal; the setting can be turned
off for such a project.
Test: tst_uprightsymboltexts turns the symbols of a folio and checks the
angle of each motor's "M" in the DXF, with the setting off and on.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Found by review of #1354:
- Terminal names (shown on the folio) read backwards on a mirrored
symbol: Terminal::paint() now mirrors them back about the centre of
their box, in the frame they are drawn in.
- "Reload the drawings of the symbols" left mirrored symbols with the old
drawing: reloadPicture() refreshes the mirrored pictures too.
- Turning a group of texts on a mirrored symbol left the group's
readable mirror about the centre of its old box until the next layout:
ElementTextItemGroup::setRotation() reapplies it, as texts already do.
- The mirrored build of a symbol's drawing leaked the texts of its
primitives, which only the drawing without mirror keeps.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Edit > "Miroir horizontal" (M) and "Miroir vertical" (F), also in the
folio's right-click menu, mirror the selected symbols in place. The keys
are the ones the element editor uses for the same two actions.
An element keeps two mirrors about its own axes, applied before its
rotation, and saves them as mirror="horizontal|vertical|both" on its
<element> (written only when set, so other projects save byte for byte
as before). On a symbol turned by 90 or 270 degrees, a mirror of the
folio is the other mirror of the symbol itself, so the rotation never
changes: a label kept upright does not swing round, and "Pivoter" still
turns a mirrored symbol clockwise.
- Terminals face the mirrored way (Terminal::orientation()), so wires
follow.
- The symbol stays where it was: its centre is kept, on the grid, since
the hotspot is often a corner.
- Texts read normally. The element's texts, text groups and cross
reference are mirrored a second time about the centre of their own box
(Element::keepReadable()), and ElementPictureFactory draws the texts
of the symbol itself the same way, in a cached picture per mirror.
Groups and cross references held at the bottom of the folio stay
centred under their element.
- DXF export mirrors the symbol's lines, arcs and texts.
- Scripting: qet.mirrorElement(folio, uuid, vertical) and
qet.elementMirror(folio, uuid); live mode may run both menu commands;
the MCP server gets a mirror_element op and qet_diff reports mirrors.
Not done: the parts of a PLC table drawn at run time
(Element::drawPlcTable()) are not kept readable on a mirrored PLC; the
project database has no column for it, as it has none for the rotation.
Test: tst_scriptmirror mirrors a symbol through --run, checks every
terminal's side and facing, the round trip, undo, a save and reload,
and a symbol turned by 90 degrees.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A selected free text gets the corner handles of the previous commit:
dragging one sets the width the text wraps to, keeping the opposite
corner in place; double-clicking a handle goes back to the automatic
width. One undo step per drag, on the folio's undo stack. The handles
are hidden while the text is typed in.
DiagramTextItem shows the handles for a subclass that asks for them
(textResizeHandlesWanted()); only free texts do for now.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- setTextWidth() returns without a signal when the width does not
change, which is also the case on every load of a text without one.
- Say why IndependentTextItem::textWidth()/setTextWidth() hide the
non-virtual QGraphicsTextItem functions, and what that means for code
holding a base pointer.
- The fixture no longer carries a /tmp path from the machine it was
made on.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A free text (IndependentTextItem) gets a textWidth property: -1 is the
automatic width (one line, or lines broken by hand, as today); a
positive width makes the text wrap between words. Its top-left corner
stays in place. The property is what a spin box and resize handles will
set in follow-up changes.
File format: the width is saved as text_width on <input>, only when one
is set, so a project without widths is saved byte-identical
(tst_resaveunchanged). An older QElectroTech ignores the attribute and
shows the text on one line, with no load error.
DiagramTextItem::setHtml() no longer widens a centred or right-aligned
rich text to idealWidth() + 40 when the text has a width of its own;
with the automatic width it still does.
Test: tst_freetextwidth (--resave keeps the width only where set,
--export-svg draws the text wrapped).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- replaceImage(), mirror() and setTransparentColor() no longer add a
dummy QUndoCommand child: the imageSource child already keeps
QPropertyUndoCommand::mergeWith() from merging two of them.
- The explanation of the position and pivot maths moves from crop() to
applyCrop(), where that code now lives; the stale older doc block of
crop() goes.
- applyCrop() returns false when nothing was cropped, and
qet.cropImage() passes that on. tst_imagecropundo checks it for the
current crop, an empty rectangle and one outside the picture.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The stacking added in #1287 tested the crosses' bounding rects, so
crosses whose frames or empty margins touched were raised although no
text was hidden (the list display case found in review). The texts each
cross really draws are now recorded while its geometry is built, and a
cross is raised only when one of its texts would be drawn over a text of
another one.
A new "Empiler les références dont les textes se chevauchent" checkbox
in the cross reference properties, per type, turns the stacking off;
the crosses of that type are then only centred, as before #1287. It is
saved in the project (xref attribute "stackoverlapping") and in the
settings, and is on when absent.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Cropping, keying out a colour, mirroring and replacing a picture set
its crop rectangle, transparent colours and original directly and put
only the displayed pixmap into the undo command. After Ctrl+Z the
picture looked right, but a save still wrote the undone crop and
colours, and the picture came back cropped once reopened; a later
crop or colour dialog also started from the undone values. The three
are now one property, imageSource, changed in the same undo step as
the pixmap.
The crop itself moves out of the dialog into applyCrop(), so that it
can be applied without one.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Cause: in list mode `drawAsContacts()` adds a margin (`adjust(-30, -4, 4, 4)`) to `boundingRect()`, and the horizontal test in `stackAtBottom()` uses that rect. Two lists overlapping only inside the empty margin (1.4 units in my log) count as overlapping. `gap` only changes how far the raised ones move, not which ones.
Fix: test the content only in list mode (`rect.adjust(30, 0, -4, 0)`), with the two margins as constants shared with `drawAsContacts()`.
Result: all references stay at the bottom (980 -> 980), none raised by mistake, also when moving an element sideways.
Checked: [to complete: same-column coils in list mode still stack, cross mode unchanged]
A text group kept at the bottom of the page, on a master element,
connects to its project's XRefPropertiesChanged with a lambda that has
no context object. The connection belongs to the project, so it
outlives the group: once the group is deleted (paste a coil, undo,
then make any other edit), changing a cross-reference setting in the
project properties calls QTimer::singleShot() on the freed group and
QElectroTech crashes.
Pass the group as the context of that connection and of the
linkedElementChanged one beside it, so Qt removes both when the group
goes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
A conductor draws a junction dot at each of its bends that lies on
another conductor. It only looked at the conductors on its own two
terminals, so a bend lying on a conductor of the same potential that
shares no terminal with it got no dot. Dragging the horizontal parts of
a chain of conductors onto one line, as in the report, does exactly that.
Look at every conductor of the potential on the folio instead
(relatedPotentialConductors(false)). The folio scene uses NoIndex, so a
scene lookup at each bend scans every item and made export of the
366-conductor Polonez example 60% slower; walking the potential costs
nothing measurable there (2.08 s vs 2.11 s) or on a 153-conductor
potential (0.97 s both).
The test for one other conductor moves into bendMakesJunction(). It
also stops a point being appended once per matching segment: master drew
the dot above K2 in the fixture ten times.
Across the 23 examples (133 folios) exported to SVG, 6 dots are added,
all at T-joints that had none, and none removed.
tst_junctiondots exports fixtures/junction_dot_shared_potential.qet and
checks for both dots; it fails on master with the dot above K3 missing.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
ispyisail