Each folder now holds icons from one source: qet/ for QET's own SVGs,
breeze/ for KDE Breeze (with breeze/added/), flags/, colors/, and
themes/ for script output (with themes/generated/). The unused
128x128/terminalstrip.svg is removed. The generator, qrc files,
qeticons.cpp and ico/copyright follow the new paths. ico/README.md
lists each folder's source, script and license.
Part of #1269.
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>
startEdition() records the bounding rect, but the aligned point was only
restored in endEdition(), so a right/centre/bottom-aligned text grew from
its old left/top edge while being typed into and jumped into place when
the edition ended. Re-anchor after every content change during inline
edition.
endEdition() now always clears previous_text, which marks an edition in
progress. It used to stay set after an edition that left the text
unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
itemsBoundingRect() also counts hidden items. The text of a single-line
wire, and the wire texts hidden by "one text per potential", are never
positioned again and can sit far outside the drawing, so Zoom to content
zoomed out to include them. On photovoltaique.qet 14 such texts reach
x = -729 and the zoom covered 2421x924 instead of 1135x648.
Add Diagram::visibleItemsBoundingRect() and use it for Zoom to content,
the scene rect and the borderless export. Zoom to content now goes through
zoomToRect(), which re-centres after the scroll bars appear; before, the
view could stay scrolled to the top-left with part of the content cut off.
Fixes#1281
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>
With "snap to bottom", each master's cross was centred under it at the
bottom of the folio on its own, so two coils in one column drew their
crosses on the same spot and the references printed over each other.
The crosses of a folio are now placed together: when two would overlap,
the one of the higher element goes above the other, so a column of
coils gets its crosses in the same order. Crosses that do not overlap
stay where they were. Moving an element sideways now re-places them
too, since it can move a cross into or out of another's column.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
With SOURCE_DATE_EPOCH set, the document id came from the project's uuid
and a hash of the .qet file. A project generated again from the same
data has new uuids, so its PDF differed in the id alone (XMP DocumentID
and trailer /ID, 120 bytes) although every page was the same.
Qt now writes a fixed placeholder id, and once the file is written the
placeholder is replaced with an id derived from the bytes of the PDF:
the same pages give the same id. Same length, so no offset changes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
lupdate (run on qet_en.ts only) found strings added since the last
update that the English interface showed in French:
- wires per terminal (#1272, #1273): the rules in Settings > General
and Project properties, the experimental-feature warning, and the
tooltip on a terminal that refuses a conductor (both plural forms);
- Display > Show (#1268): the item kinds and the "%n kinds of items
hidden" status message;
- the 3D mouse settings page.
Terms follow the existing English: conductor (conducteur), terminal
(borne), sheet (folio), sheet reference (renvoi de folio). The menu
"Affichage" is "Display", so its "Afficher" submenu is "Show" rather
than lupdate's guess "Display". lrelease reports 3573 finished,
0 unfinished. Other languages are untouched.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
qet_element_search matched every word as a substring of the names and
the path, so "NC contact" offered a remanence coil ("nc" in "remanence",
"contact" in its folder) and an outlet before any auxiliary contact,
"normally closed contact" found nothing, and "emergency stop" ranked a
maker's part and two assembly-plan drawings above the push button.
- A word of one or two letters must now be a whole word. Longer words,
and short ones with a digit, still match inside a word, so German
compounds ("schutz", "klemme") and "3p" find what they found before;
on 37 sample queries no result is lost.
- NO, N/O, "normally open", "normalement ouvert" are one word, and NC,
N/C, NF, "normally closed", "normalement fermé" another.
- Symbols under *_graphics, *_manufacturers_articles and
*_miscellaneous_unsorted come after every other match.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Circle and ellipse had a blue square handle and the arc none; all three
now have hollow 3 px handles and 1 px lines, like draw-line. Breeze draws
transform-crop as a crop mark at 16 px and a dashed box at 22, so a 16 px
dashed box in ico/breeze-added/16/ replaces Breeze's 16 and 32 px art.
The test set the elements folder in a settings file under XDG_CONFIG_HOME,
which macOS does not read, so common:// paths did not resolve and the
first placement failed. It now passes --common-elements-dir.
The icon section was one Breeze (LGPL) paragraph that listed GPL and
CC BY-ND icons in its text, so a DEP-5 reader saw every icon as LGPL.
Each license now has its own Files paragraph and License text.
The search panel's close button and the five KDE Breeze icons that sat in
ico/breeze-icons/ now come from ico/breeze/, with 16, 22 and 32 px art.
André's three SVGs moved from ico/generated/ to ico/scalable/. Removed the
18 .svgz files and the QETIcon and QETOxygenLogo icons nothing used.
Schemas now get saved with a UUID and do not have to be looked up according to the form. They can be renamed and it gets reflected in existing elements. The GUI for
editing them is now much easier, e.g. forms can be reshuffled. A formula can be applied. Many options now also for conductors and sheets.
Review of #1272: scorpio810 asked for a warning box for users; the inline
box added in 9a55939bb is replaced by a pop-up, as the element editor
gives for unnamed terminals. When OK in Settings > General or in Project
properties turns a rule on (a limit where there was none, or one wire
per folio report), a warning says the rules are experimental, may change
with settings to redo if wires and conductors become separate objects,
and can all be turned off in Settings > General. OK goes on. Changing a
limit that was already on, or turning rules off, shows nothing.
tst_wiringrules: when the warning is due (WiringRules::turnsRuleOn()).
Checked in the GUI from both pages, and that a plain OK shows nothing.
English for translators: "Warning"; "The wires-per-terminal rules are an
experimental feature. They count wires as QElectroTech represents them
today. They could change, and your settings need redoing, if wires and
conductors become separate objects in a future version. All these rules
can be turned off in Configure QElectroTech > General."
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_015FPuYPS4T7QuEwjNu22rXD
The DXF export warns that pictures become outlines. Pictures hidden by
View > Show are left out of the export, so they no longer trigger it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Review of #1272 (scorpio810, after arummler's note that the rules would
need redesigning if wires and conductors become separate objects): a
warning box, with the standard warning icon, at the top of the rules in
Settings > General and in Project properties. It says the feature is
experimental, counts wires as QElectroTech models them today, and that
the rules, and the user's settings, could change if wires and conductors
are separated.
English for translators: "Experimental feature. These rules count wires
as QElectroTech represents them today; they could change, and your
settings need redoing, if wires and conductors become separate objects
in a future version."
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_015FPuYPS4T7QuEwjNu22rXD
The project's CI builds QElectroTech without scripting, where --run is
not an option and the process never exits, so the test waited out its
timeout and failed (#1272's stack, CI run 37118548599). The --run checks
now skip there, as the script tests are left out of such builds; the
rules and the save round trip still run.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_015FPuYPS4T7QuEwjNu22rXD
With "one text per potential", deleting the wire that carries the text
moves it to another wire of the potential, but only if the text
isVisible(). While wire texts were hidden by View > Show none was, so the
text was never moved and the potential came back with no text.
ShownKinds::wantsVisible() also counts a text hidden only by its kind.
Also drop the always-true/false enum range checks clang warned about.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Jeff Patterson