using frames to encapsulate building blocks, e.g., like macros or functions in
a programming language. Since measurements only know about the frame containing
a vector but not the frames containing that frame, invocations of this frame
from different places can only be distinguished within the min/next/max scheme.
(See the example in README.)
To eliminate this limitation, one needs a way to tell fped to consider a point
only if it has been instantiated through a certain path, e.g., by requiring
some other frames to be visited in its instantiation. This increases the number
of distinct points available for measurements.
The mechanism chosen is to qualify a measurement point with frames that lead to
it. This list of outer frames does not have to include all frames. Without
qualifying, the old behaviour results.
Note that this doesn't cover all possible ways in which a point can appear in
different roles. Multiple frame references can also result from repeating the
same frame reference in the same parent frame. The current qualification
mechanism does not allow such paths to be distinguished. However, one can
always introduce intermediate frames for this purpose.
Furthermore, repetitions create multiple instances of a point, although in what
should be considered the same role.
- fpd.l: make scanner support free-format a little better by switching back to
keyword mode after frame braces. This way, one can write a simple frame in a
single line, which is useful for regression tests.
- fpd.l, fpd.y, README, test/dbg_meas: added %meas directive to print the
result of a measurement
- fpd.y, README: measurements can now be labeled. Note that, due to limitations
of the grammar, the first measurement cannot be labeled.
- error.h, error.c (yywarn): new function for non-fatal diagnostics that always
get reported to standard error
- bitset.h, bitset.c: functions to manipulate variable-size bit sets
- meas.h, fpd.y, README, test/meas_qual: added the means to specify qualifiers
for points used in measurements
- dump.c (print_meas_base, print_meas): dump qualifiers
- delete.c (delete_references, test/del_frame): delete measurements that
reference a frame being deleted in their qualifiers
- obj.h, obj.c (enumerate_frames, instantiate): enumerate all frames so that we
have an index into the bit vector of visited frames
- meas.h, meas.c (reset_samples, meas_post), obj.c (generate_vecs,
generate_frame, instantiate): record the set of frames visited for each
sample
- meas.c (meas_post): only treat two instances of a point as equivalent if the
set of frames visited of one of them is a superset of set of the other. In
this case, keep the union of the two sets.
- meas.h, meas.c (meas_find_min, meas_find_next, meas_find_max),
test/meas_qual: instantiate_meas_pkg only select points for which all frames
in the qualification have been visited
- gui_meas.c (is_min, is_next, is_max, is_a_next): updated for above change
- inst.h, inst.c (curr_frame, propagate_bbox, add_inst, inst_begin_frame,
inst_end_frame, inst_start): renamed curr_frame to frame_instantiating to
avoid clash with curr_frame in fpd.y
- inst.h, inst.c (find_meas_hint): make global
- test/structure, test/del_vec, test/del_frame: fped now warns if a measurement
is in an unlinked frame. Changed regressions tests to avoid this warning.
- test/Common: new function expect_grep to compare only part of the output
git-svn-id: http://svn.openmoko.org/trunk/eda/fped@5967 99fdad57-331a-0410-800a-d7fa5415bdb3
- removed root_frame. "frames" now takes its place.
- removed frame->prev. In those few cases where we need the previous frame (for
deletion and dumping), we walk the list or recurse.
- the list of frames is now in GUI order, not file order.
- when reading the .fpd file, put the root frame first and leave it there.
- instead of walking the frames list and excluding the root frame by testing
frame->name, just start at frames->next
- likewise, instead of testing !frame->name just use frame == frames
git-svn-id: http://svn.openmoko.org/trunk/eda/fped@5948 99fdad57-331a-0410-800a-d7fa5415bdb3
to also delete frames and items in other frames than the current one, and added
the corresponding regression tests.
- fpd.y, README: %del can now also delete frames
- test/del_frame: regression test for frame deletion
- fpd.y: moved all debug items into debug_item, so that they can be invoked
after defining measurements
- README: clarified that object labels aren't entirely hidden in the GUI
- delete.c (delete_references): also delete measurements referencing the frame
(test/del_frame)
- fpd.y (dbg_link_frame): described why we need base_frame in addition to
base_vec
- fpd.y, README: %del can now also reach into frames other than the current one
- delete.c (do_delete_vec): also delete references in the root frame, i.e.,
measurements (tests/del_vec)
git-svn-id: http://svn.openmoko.org/trunk/eda/fped@5947 99fdad57-331a-0410-800a-d7fa5415bdb3
fine. Anyway, here's a new debug construct (%frame) and a bunch of new
regression tests.
- fpd.y, fpd.l, README: added new directive %frame to link frames also to other
frames than the current one (like in the GUI)
- gui_tool.h, gui_tool.c: export is_parent_of
- test/frame_ref: regression tests to ensure that frame order remains valid,
even if we reference late from early frames
git-svn-id: http://svn.openmoko.org/trunk/eda/fped@5945 99fdad57-331a-0410-800a-d7fa5415bdb3
found in the process. Also taught the regression test system a new trick: the
path to "fped" can be passed in the environment variable FPED. E.g.,
FPED=fped.r5943 make test
- fped.c (usage, main): duplicating the -T option produces a dump to stdout
before exiting (like %dump would)
- test/Common: new command fped_dump to invoked fped with a second -T option
- test/Common: if the environment variable FPED is set, use its content to
invoke fped (default is ../fped)
- test/Common: if the environment variable CWD_PREFIX is set, prepend it to
$FPED if the latter is a relative path
- Makefile (test, tests): set CWD_PREFIX to .., so that the path given in FPED
is valid at the point of invocation
- fpd.y: revised grammar to make "package" optional
- fpd.y: measurements were syntactically allowed inside non-root frame
(test/structure)
- test/structure: test various combinations of the grammatical file structure
- test/tsort: removed all the now unnecessary "package" directives
git-svn-id: http://svn.openmoko.org/trunk/eda/fped@5944 99fdad57-331a-0410-800a-d7fa5415bdb3
infrastructure for hole objects. (Postscript and KiCad output are still
missing, and so does sanity checking.)
- gui_tool.c (new_obj_unconnected): set obj->name to NULL
- added hole objects to tool bar, all the various GUI routines, the .fpd
parser, and the .fpd dumper
git-svn-id: http://svn.openmoko.org/trunk/eda/fped@5939 99fdad57-331a-0410-800a-d7fa5415bdb3
they referenced. As a band-aid, we now explicitly keep track of which vectors
have been dumped, and defer objects accordingly. A more correct solution would
be to properly abstract the partial order algorithms (along with the heuristics
for maximizing the number of ".") and to implement it properly.
- fpd.y (debug_item): new rule for %dump and %exit, which can appear also among
measurements
- fpd.y (frame_items, measurements): rearranged grammar to allow debug_item
also in measurements. To avoid ambiguities, the "measurements" section can no
longer be empty, but it can be omitted as a whole.
- obj.h, dump.c (later, recurse_vec, order_frame): vectors now also have a
"dumped" flag which is used in "later" to defer dumping an object until all
the vectors it depends on have been dumped.
git-svn-id: http://svn.openmoko.org/trunk/eda/fped@5928 99fdad57-331a-0410-800a-d7fa5415bdb3
list of frames at the end of parsing. We now tentatively append it each time
a %dump is requested.
- fpd.y: append the root frame before calling "dump"
git-svn-id: http://svn.openmoko.org/trunk/eda/fped@5921 99fdad57-331a-0410-800a-d7fa5415bdb3
of README.
- fpd.l, fpd.y, README: added debugging directives %del, %move, %print, %dump,
and %exit
- obj.h, fpd.y (find_obj, find_label, new_obj): objects can now be labeled
- obj.c (obj_anchors), inst.c (inst_anchors): gathering the list of anchors is
now a per-object function, not an instance "method". inst_anchors implements
the vector vs. object switch.
- inst.h, inst.c: removed all *_op_anchors functions
- expr.c (str_unit): in the past, we returned a malloc'ed string, but these
times are long gone. Thus, don't stralloc("").
git-svn-id: http://svn.openmoko.org/trunk/eda/fped@5919 99fdad57-331a-0410-800a-d7fa5415bdb3
- changed pad classification in instances from bare/other to copper/special
- moved LAYER_* definitions from layer.h to layer.c
- after instantiation, we perform sanity checks on pads and remove layers from
coppery pads that are handled by a special layer
- fped.y: the line number in objects was never set
- overlap.c: fixed overlap calculations
- gui_tool.c: end_new_pad didn't initialize the pad type
git-svn-id: http://svn.openmoko.org/trunk/eda/fped@5635 99fdad57-331a-0410-800a-d7fa5415bdb3
- .fpd file format: new directive "unit" to set the default unit
- new selection was too aggressive - make it only rearrange settings if we also
fail the second vector search
- gui_draw_pad_text: calculation of height vs. width lost too much precision,
causing pad text to be rotated arbitrarily
- drag_new_vec: display distance in mil if unit is mil
- end_new_raw_vec: store distance in mil if unit is mil
- gridify: use a 10 mil grid if unit is mil
- ps_hline: corrected gsave/grestore mismatch
- Makefile: made "all" a prerequisite of "install"
- Postscript output now mentions the default unit (if set)
- ps_package: height and width were swapped, oopsie !
git-svn-id: http://svn.openmoko.org/trunk/eda/fped@5525 99fdad57-331a-0410-800a-d7fa5415bdb3
- README: added build prerequisites
- "part name" is now more correctly called "package name"
- changed keyword "part" to "package"
- removed inst_debug and struct inst_ops.debug functions - never really used
them anyway
- we can now generate multiple packages from a single file
git-svn-id: http://svn.openmoko.org/trunk/eda/fped@5478 99fdad57-331a-0410-800a-d7fa5415bdb3
of successful validation
- free previous expression in set_col_values
- added rapid entry also for table values (fill the row)
git-svn-id: http://svn.openmoko.org/trunk/eda/fped@5460 99fdad57-331a-0410-800a-d7fa5415bdb3
- moved definition of expr_result from expr.c to fpd.y
- new header file fpd.h with all the things fpd.l and fpd.y export
- edit_var already calls edit_nothing, so there's no need to call it before
- added rapid entry option for loops, variables, and columns: var = val, ...
git-svn-id: http://svn.openmoko.org/trunk/eda/fped@5459 99fdad57-331a-0410-800a-d7fa5415bdb3
- removed support for old-style measurements
- new-style measurements are now embedded in "struct obj", so we can dump and
delete them
- "measxy" is now called "meas"
- updated examples to use new-style measurements
git-svn-id: http://svn.openmoko.org/trunk/eda/fped@5412 99fdad57-331a-0410-800a-d7fa5415bdb3
- moved measurement operations from gui_tools.c to new file gui_meas.c
- added support for creating new-style measurements through the GUI
- the offset is optional in new-style expressions
git-svn-id: http://svn.openmoko.org/trunk/eda/fped@5407 99fdad57-331a-0410-800a-d7fa5415bdb3
- columns can now be deleted/undeleted
- reset active row if deleting that row
- grammar now accepts an empty file
- when a non-canvas entry becomes editable, we set the focus to it
- when adding the first reference to a frame, make it active
- when deleting the active reference to a frame, set the pointer to NULL
- measurements didn't have a drag_new action, causing trying to create them to
crash fped
- offset field of measurements was incorrectly labeled "width"
- fixed operator precedence in gridify. Someone needs more sleep :-(
git-svn-id: http://svn.openmoko.org/trunk/eda/fped@5398 99fdad57-331a-0410-800a-d7fa5415bdb3
- added string-valued expressions (on-going)
- loops, tables, and rows can now be deleted/undeleted (still need columns)
git-svn-id: http://svn.openmoko.org/trunk/eda/fped@5397 99fdad57-331a-0410-800a-d7fa5415bdb3
the point being dragged
- added pop-up menus for all frame and variable items (on-going)
- clickable items in the frame/var area now only respond to the left button,
unless they have a pop-up
- removed the frame delete icon - use pop-up instead
git-svn-id: http://svn.openmoko.org/trunk/eda/fped@5396 99fdad57-331a-0410-800a-d7fa5415bdb3
- added temporary solution for having a part name
- added part name to FPD example files
- reduced area in which the frame delete button responds
git-svn-id: http://svn.openmoko.org/trunk/eda/fped@5393 99fdad57-331a-0410-800a-d7fa5415bdb3
operation, not just one object at a time
- vector reference counting is not needed now that we have recursive delete.
Removed it.
- completed recursive delete of vectors
- gridify: give new vectors a minimum length, so that we don't get a weird
display when starting a new footprint.
git-svn-id: http://svn.openmoko.org/trunk/eda/fped@5391 99fdad57-331a-0410-800a-d7fa5415bdb3
- when moving a point, we no longer restrict the area around the original
position once the drag radius has been left
- objects can now be deleted by selecting them and pressing Delete
- deleted objects can be restored by pressing "u"
git-svn-id: http://svn.openmoko.org/trunk/eda/fped@5387 99fdad57-331a-0410-800a-d7fa5415bdb3
- made xpm name manipulation compatible with older versions of ImageMagick
(reported by Alvaro Lopes)
- corrected all lines beginning with five or more spaces
- Makefile
- README: fixed loop example (reported by Joerg Reisenweber)
- moved draw_arc and draw_circle from gui_inst.c to gui_util.c
- added dragging and hovering
- added creation of lines, rectangles, and arcs via GUI
- removed arc tool (use circle instead)
- fixed reference counting in vectors
git-svn-id: http://svn.openmoko.org/trunk/eda/fped@5384 99fdad57-331a-0410-800a-d7fa5415bdb3
- make_var now sets up all back pointers, so we don't crash when deselecting a
variable (this bug was uncovered by cleanup in the previous commit)
- report_parse_error didn't print a newline at the end of the message
- introduced frame reference selection and the data structures needed for it
- bounding box calculation of arcs used x for y
- added frame reference selection example quad.fpd
git-svn-id: http://svn.openmoko.org/trunk/eda/fped@5377 99fdad57-331a-0410-800a-d7fa5415bdb3
* zoom and center to extents
- variables in tables weren't properly initialized, causing stray recursive
evaluation errors
- replaced table->active index with table->active_row pointer
- added row selection
- for fun, added tab.fpd row selection example
git-svn-id: http://svn.openmoko.org/trunk/eda/fped@5375 99fdad57-331a-0410-800a-d7fa5415bdb3