1
0
mirror of https://github.com/fadden/6502bench.git synced 2024-12-05 03:49:53 +00:00
Commit Graph

744 Commits

Author SHA1 Message Date
Andy McFadden
88c56616f7 Label rework, part 5
Implemented assembly source generation of non-unique local labels.
The new 2023-non-unique-labels test exercises various edge cases
(though we're still missing local variable interaction).

The format of uniquified labels changed slightly, so the expected
output of 2012-label-localizer needed to be updated.

This changes the "no opcode mnemonics" and "mask leading underscores"
functions into integrated parts of the label localization process.
2019-11-17 16:05:51 -08:00
Andy McFadden
4e08810278 Finish removal of "disable label localizer" feature
The label localizer is now always on.  The regression tests turned
it off by default, but that's no longer allowed, so the generated
output has changed for many of them.  The tests themselves were not
altered.
2019-11-16 17:15:03 -08:00
Andy McFadden
68c324bbe8 Label rework, part 4
Update the symbol lookup in EditInstructionOperand, EditDataOperand,
and GotoBox to correctly deal with non-unique labels.

This is a little awkward because we're doing lookups by name on
a non-unique symbol, and must resolve the ambiguity.  In the case of
an instruction operand that refers to an address this is pretty
straightforward.  For partial bytes (LDA #>:foo) or data directives
(.DD1 :foo) we have to take a guess.  We can probably make a more
informed guess than we currently are, e.g. the LDA case could find
the label that minimizes the adjustment, but I don't want to sink a
lot of time into this until I'm sure it'll be useful.

Data operands with multiple regions are something of a challenge,
but I'm not sure specifying a single symbol for multiple locations
is important.

The "goto" box just finds the match that's closest to the selection.
Unlike "find", it always grabs the closest, not the next one forward.
(Not sure if this is useful or confusing.)
2019-11-16 16:44:08 -08:00
Andy McFadden
8273631917 Label rework, part 3
Added serialization of non-unique labels to project files.

The address labels are stored without the non-unique tag, because we
can get that from the file offset.  (If we stored it, we'd need to
extract the value and verify that it matches the offset.)  Operand
weak references are symbolic, and so do include the tag string.

We weren't validating symbol labels before.  Now we are.

This also adds a "NonU" filter to the Symbols window so the labels
can be shown or hidden as desired.

Also, added source for a first pass at a regression test.
2019-11-16 11:12:32 -08:00
Andy McFadden
be65f280a3 Minor tweaks
- Renamed "strip label prefix/suffix" to "omit label prefix/suffix".

- Changed a Merlin operand workaround so it doesn't apply to code
  that is explicitly not in bank zero.

- Changed {addr}/{const} annotations on project/platform symbol
  equates so they line up a little better on screen and in exported
  sources.
2019-11-15 16:24:07 -08:00
Andy McFadden
5dd7576529 Label rework, part 2
Continue development of non-unique labels.  The actual labels are
still unique, because we append a uniquifier tag, which gets added
and removed behind the scenes.  We're currently using the six-digit
hex file offset because this is only used for internal address
symbols.

The label editor and most of the formatters have been updated.  We
can't yet assemble code that includes non-unique labels, but older
stuff hasn't been broken.

This removes the "disable label localization" property, since that's
fundamentally incompatible with what we're doing, and adds a non-
unique label prefix setting so you can put '@' or ':' in front of
your should-be-local labels.

Also, fixed a field name typo.
2019-11-12 17:44:51 -08:00
Andy McFadden
4d079c8d14 Label rework, part 1
This adds the concept of label annotations.  The primary driver of
the feature is the desire to note that sometimes you know what a
thing is, but sometimes you're just taking an educated guess.
Instead of writing "high_score_maybe", you can now write "high_score?",
which is more compact and consistent.  The annotations are stripped
off when generating source code, making them similar to Notes.

I also created a "Generated" annotation for the labels that are
synthesized by the address table formatter, but don't modify the
label for them, because there's not much need to remind the user
that "T1234" was generated by algorithm.

This also lays some of the groundwork for non-unique labels.
2019-11-08 21:02:15 -08:00
Andy McFadden
b4519b2aa2 Update NES symbols
The PPU registers are mirrored from $2000-3FFF.
2019-11-08 07:53:28 -08:00
Andy McFadden
ceb07c809e Check for bank overrun
If an address map entry wraps around the end of a bank, add a note
to the message log.  This is Error level, since some assemblers
will refuse to handle it.
2019-11-05 13:29:51 -08:00
Andy McFadden
6a8d6950e5 Add change discard confirmation to project properties editor
It's too easy to hit Escape after making a bunch of changes, so
now we ask for confirmation.

(Might make sense to make this strictly an Esc guard, and not
pester the user if they actually hit the Cancel button or close
box.  I'm not convinced though; Esc+Enter isn't terrible.)
2019-11-04 17:41:33 -08:00
Andy McFadden
6411df7ff9 Add toolbar button for "Go to Last Change"
Also, updated a couple of comments.
2019-11-04 15:22:56 -08:00
Andy McFadden
0a0208409a Tweak line folding code
Some style guides say you should only put one space between
sentences, but I and many others still put two.  The line-folding
code was only eating one of them when they straddled the end of the
line, which looked a little funny because the following line was
indented by one space.

This tweaks the code to eat both spaces.  Regression test updated.

Also, nudge some UI elements so they line up.
2019-11-01 19:47:56 -07:00
Andy McFadden
b4213de4c0 Add "Go to Last Change" feature
Jumps to the first offset associated with the change at the top of
the Undo stack.  We generally jump to the code/data offset, not the
specific line affected.  It's possible to do better (and we do, for
Notes), but probably not worthwhile.
2019-11-01 19:11:48 -07:00
Andy McFadden
2ddd3b400d Workaround for GridSplitter lockup
As noted in issue #52, the side panels can't be resized once the
ListView gets focus.  The root of the problem is a workaround for a
selection problem that involves catching the Item Container
Generator's Status Changed event, and setting an item's focus.  It
appears that changing the size of the ListView causes the
StatusChanged event to fire, which cause the handler to grab the
focus, which causes the splitters to stop moving after one step.

This change adds a workaround that prevents the original workaround
from doing anything while a splitter is in the process of being
dragged.  It doesn't solve all problems -- you can't move the
splitters more than one step with the keyboard -- but it allows them
to be dragged around with the mouse.

There's got to be a better way to deal with this.
2019-11-01 11:14:49 -07:00
Andy McFadden
0c02b3ebe3 Version 1.4.0 2019-10-30 15:50:09 -07:00
Andy McFadden
1136759c98 Tweak the tutorial 2019-10-30 15:40:03 -07:00
Andy McFadden
f61f1022e1 Collapse the "offset" line in the info window for non-DEBUG 2019-10-30 14:30:03 -07:00
Andy McFadden
4b46b78e34 Rearrange the tutorial files
Copied the extension script tutorial files out of the Scripts
directory and into the Tutorial directory.  This makes more sense,
and makes it possible to expand the script sample without altering
the tutorial.

Reverted the Scripts sample to be an actual sample, rather than a
tutorial.

Renumbered the last two tutorials and added them to the ToC.  This
gives them actual numbers rather than treating them as add-ons to
the advanced tutorial.

Moved the source files for the tutorial binaries into a subdirectory
to reduce clutter.

This does mean we have two separate copies of the inline string
sample plugins, but that's an artifact of our attempts at security.
2019-10-30 09:49:11 -07:00
Andy McFadden
51081c5db0 Tweak "nearby" label finder
The code that found a nearby data target for an instruction operand
was searching backward but not forward.  We now take one step
forward, so that "LDA TABLE-1,Y" fills in automatically.

This altered 2008-address-changes, which had just this situation.
It didn't alter 2010-target-adjustment, but the existing tests were
insufficient and have been improved.
2019-10-29 18:12:22 -07:00
Andy McFadden
7000a5094e Fix crashing bug in selection management
The fix for Shift+F3 required briefly switching the code list view
to single-select mode.  Unfortunately, while in that mode the
control throws an exception if you touch SelectedItems (plural)
rather than SelectedItem (singular), and in an unusual case the
selection-changed event handler was doing just that.
2019-10-29 18:09:01 -07:00
Andy McFadden
e79064709c Clarify BRK explanation 2019-10-29 10:18:48 -07:00
Andy McFadden
14ecad0849 Improve handling of hidden LV tables
If a local variable table gets buried, it won't appear in the code
list, so most things ignore it.  Unfortunately, the code that adds
new entries and edits tables was finding them, which was causing
variable definitions to appear to fall into a black hole.

This is addressed in two ways.  First, we now add a message to the
log when a hidden table is noticed.  Second, the code that finds
the nearest prior table now keeps track of hidden vs. not hidden.
If a non-hidden table is available, that is returned.  If the only
option is a hidden table, we will return that, because the callers
have already assumed that a table exists by virtue of its presence
in the LvTable list.
2019-10-28 16:54:01 -07:00
Andy McFadden
73f04ef2d2 Update comments, remove unused var 2019-10-27 21:29:44 -07:00
Andy McFadden
b508fa6b4a Version 1.4.0-beta1 2019-10-27 21:17:54 -07:00
Andy McFadden
99d088e800 Mention the message log in the tutorial 2019-10-27 21:17:28 -07:00
Andy McFadden
70353c82e1 Limit value range of project address symbols
Project symbol address values are now limited to positive 24-bit
integers, just as they are for platform symbols.  Constants may
still be 32-bit values.
2019-10-27 15:36:47 -07:00
Andy McFadden
819a363178 Move SPEEDZ from F8-ROM to Applesoft
It's an Applesoft thing.
2019-10-27 11:12:22 -07:00
Andy McFadden
0fc121c9cb Add ability to "erase" previously-defined platform symbols
While disassembling some code I found that I wanted the ROM entry
points, but the zero page usage was significantly different and the
ROM labels were distracting.  Splitting the symbol file in two was
a possibility, but I'm afraid this will lead to a very large
collection of very small files, and we'll lose any sense of relation
between the ROM entry points and the ZP addresses used to pass
arguments.

Platform symbols have the lowest priority when resolving by address,
but using that to hide the unwanted labels requires creating project
symbols or local variables for things that you might not know what
they do yet.  It's possible to hide a platform symbol by adding
another symbol with the same label and an invalid value.

This change formalizes and extends the "hiding" of platform symbols
to full erasure, so that they don't clutter up the symbol table.
This also tightens up the platform symbol parser to only accept
values in the range 0 <= value <= 0x00ffffff (24-bit positive
integers).

An "F8-ROM-nozp" symbol file is now part of the standard set.  A
project can include that to erase the zero-page definitions.

(I'm not entirely convinced this is the right approach, so I'm not
doing this treatment on other symbol files... consider this an
experiment.  Another approach would be some sort of conditional
inclusion, or perhaps erase-by-tag, but that requires some UI work
in the app to define what you want included or excluded.)
2019-10-27 11:09:16 -07:00
Andy McFadden
0709ff94de Fix visibility binding 2019-10-27 10:38:29 -07:00
Andy McFadden
c49888331a Minor tweaks
- Allow user to "unnecessarily" set an address override.  This is
  a handy thing to do when dealing with code that does a lot of
  relocations.

- Moved "save needed" text to the end of the title string.

- Updated F8-ROM syms.

- Added ProDOS 8 error code constants
2019-10-26 23:34:51 -07:00
Andy McFadden
d99eec0d4f Open damaged projects in read-only mode
If we detect a problem that requires intervention during loading,
e.g. we find unknown elements because we're loading a file created
by a newer version, default to read-only mode.

Read only mode (1) refuses to apply changes, (2) refuses to add
changes to the undo/redo list, and (3) disables Save/SaveAs.  The
mode is indicated in the title bar.

Also, flipped the order of items in the title bar so that "6502bench
SourceGen" comes last.  This allows you to read the project name in
short window title snippets.  (Visual Studio, Notepad, and others
do it this way as well.)
2019-10-26 13:20:54 -07:00
Andy McFadden
6344ea59bb Add Apple-1 symbols
From a post on comp.sys.apple2:
https://groups.google.com/d/msg/comp.sys.apple2/SBBdH1ChMwQ/mnygA3WVEAAJ
2019-10-26 08:28:49 -07:00
Andy McFadden
2466090efe Version 1.4.0-alpha1 2019-10-25 18:17:25 -07:00
Andy McFadden
00612097f6 Minor updates
Add a link to the web site in the About box.

Add an up-link to the HTML export template.

Escape '<', '>', and '&' in the Applesoft to HTML converter.
2019-10-25 18:16:09 -07:00
Andy McFadden
8505a5cdd4 Add an Applesoft to HTML converter
This is not an official thing, so I'm hiding it in the DEBUG menu
for now.
2019-10-23 17:49:59 -07:00
Andy McFadden
eb99a716f1 Add scripts and inline data to tutorials
Added an inline data formatting section to Tutorial2.

Added a new section on extension scripts, using Scripts/Sample as
the code.
2019-10-23 14:52:52 -07:00
Andy McFadden
1844fcb8b7 Fix junk alignment check
While adding a message log entry for failing alignment directives,
I noticed that the assembler source generator's test for valid
alignment was allowing some bad alignment values through.

I'm holding off on reporting the message to the log because not all
format changes cause a data-reanalysis, which means the log entry
doesn't always appear and disappear when it should.  If we decide
this is an important message we can add a scan for "softer" errors.
2019-10-23 13:25:50 -07:00
Andy McFadden
4ed2558f9f Experiment with showing character value next to default data
Doesn't seem useful in practice.  If I find a good use case it might
be worth making it an option.
2019-10-23 13:02:31 -07:00
Andy McFadden
9d5f8f8049 Add a blank line between constants and addresses
In the assembler output, add a blank line between the constants
and addresses in the long list of equates.

The earlier change that corrected the BIT instruction caused test
2009-branches-and-banks to fail, because it was relying on the idea
that BIT made the carry flag indeterminate.  Changing a BCC to a
BVS restored the desired behavior.
2019-10-22 22:45:13 -07:00
Andy McFadden
4c43b4063a Tweaks 2019-10-22 22:23:22 -07:00
Andy McFadden
630f7f0f87 Improve the "info" panel
Not a huge improvement, but things are slightly more organized, and
there's a splash of color in the form of a border around the text
describing the format of code and data lines.

Added an "IsConstant" property to Symbol.
2019-10-22 21:27:49 -07:00
Andy McFadden
463fbff368 Show load address in Edit Address dialog
Sometimes code relocates a few bits of itself but not others.  We
don't currently have a way to say, "go back to where we would have
been".  As a cheap alternative, we now show the "load address", i.e.
where we'd be if there were no address map entries after the first.
2019-10-22 13:58:32 -07:00
Andy McFadden
1b0ee7de21 Fix display of instruction attributes
The "affected flags" constants were incorrect for BIT, BRK, COP,
RTI, XCE, and the undocmented instructions ANE, DCP, and SAX.  The
constants are used for the changed-flag summary shown in the info
window and the instruction chart.

Of greater import: the status flag updater for BIT was incorrectly
marking N/V/C as indeterminate instead of N/V/Z.  The undocmented
instructions ANE, DCP, and SAX were also incorrect.

The cycle counts shown in line comments are computed correctly, but
the counts shown in the info window and instruction chart were
displaying the full set of modifiers, ignoring the CPU type.  That's
okay for the info window, which spells the modifiers out, though
it'd be better if the bits were explicitly marked as being applicable
to the current CPU or a different one.
2019-10-22 10:48:02 -07:00
Andy McFadden
7bbccaf71f Minor tweaks
Mark the "info" window as read-only.

When the project closes, clear the contents of the Symbols and
Notes windows.

Clarify some Apple II I/O definitions.
2019-10-21 17:43:40 -07:00
Andy McFadden
bcac8bc6a0 Add instruction chart
This adds a window that displays all of the instructions for a
given CPU in a summary grid.  Undocumented instructions are
included, but shown in grey italics.

Also, tweaked AppSettings to not mark itself as dirty if a "set"
operation doesn't actually change anything.
2019-10-21 15:15:09 -07:00
Andy McFadden
a4f1de1238 Add message list, part 2 (of 2)
Implemented show/hide mechanic, using a button on the right side of
the status bar to show status and to trigger un-hide.

Also, show I/O direction in project symbols editor list.
2019-10-20 18:02:23 -07:00
Andy McFadden
81dbab04ba Add message list, part 1
This converts the "problem list viewer" tool to a grid that appears
below the code list view when non-empty.  Not all messages are
problems, so it's being renamed to "message list".
2019-10-20 14:40:32 -07:00
Andy McFadden
3cc6f32c6f Put some stuff in a "Navigate" menu
Created a Navigate menu, and put the menu items for Find and Go To
in it.  Added menu items for nav-forward and nav-backward, which
until now were only available as toolbar buttons.
2019-10-19 17:46:40 -07:00
Andy McFadden
b6e571afc2 Correctly handle embedded instruction edge case
This began with a change to support "BRK <operand>" in cc65.  The
assembler only supports this for 65816 projects, so we detect that
and enable it when available.

While fiddling with some test code an assertion fired.  This
revealed a minor issue in the code analyzer: when overwriting inline
data with instructions, we weren't resetting the format descriptor.

The code that exercises it, which requires two-byte BRKs and an
inline BRK handler in an extension script, has been added to test
2022-extension-scripts.

The new regression test revealed a flaw in the 64tass code
generator's character encoding scanner that caused it to hang.
Fixed.
2019-10-19 17:28:45 -07:00
Andy McFadden
e5da5ced95 Version 1.4.0-dev2 2019-10-19 11:17:09 -07:00