I find that trunk+2.71 won't build as bigide, since PascalScript has no sparc.inc and it attempts to build for x86. I've raided this as http://mantis.freepascal.org/view.php?id=25709 and suggest that for the moment PscalScript is excluded for that processor. It's OK on PPC, I've not checked others.

Looking at the IDE, I find the "new" bar on the right of SynEdit useful but I wonder whether the developers would consider putting a digit marker rather than a rectangle to mark the position of a bookmark? That would allow a quick keyboard jump, and would also give an easy indication of what bookmark was on which page when multiple windows were open.

Still on the IDE, I remember that editors such as MultiEdit had a command to reflow a paragraph to a predefined width. Is anything like that available in Lazarus, specifically to reformat comment blocks?

Nice idea.

The "overview gutter is a big piece of unfinished work. All those thinks should be configurable. But I hawe not found the time to continue on it for a long time now. In is still on my list though.

The other possible enhancement would be putting bookmark markers on the page tags, but I'm assuming that implementation of those is buried deep in each widget set. Might be feasible to use some of the special circled-digit Unicode characters: not as pretty as the coloured markers, but adequately functional.

The "overview gutter is a big piece of unfinished work. All those thinks should be configurable. But I hawe not found the time to continue on it for a long time now. In is still on my list though.

Running Lazarus tunnelled over SSH I'm finding that scrollbar operations tend to "run away" with unintentional repeats on occasion, which means that an overview of where bookmarks and recent changes are works out quite nicely :-)

yes.
Eventually this may become a custom drawn pagecontrol....

What I had thought to be useful (just a draft of the idea, details still may ripen):
- A window that list s all the bookmarks, also shows the line of code, and if available the name of the enclosing function.
- This window could provide room for more than 10 "remembered" locations (10 can have a shortcut, and are bookmarks)

The part that I like really is to see where the implementation section starts. That is often where some important constants or similar are.

On a tablet sitting next to the keyboard :-)

I must admit that I only noticed that yesterday: my screens are fairly large but elderly so the contrast isn't as good as new ones.

There's still the question of how this sort of thing works best with a split window (i.e. two parts of the same file shown on the same form).

2 parts/ same file, 1 large overview gutter. (gutter is NOT split)

other scenarios, virtual files:
- limiting the edit to one procedure
~ overview gutter also shows this procedure only
~ line number gutter starts at the real number (e.g. top line = 933)

- include file displayed inlined
~ overview gutter also shows both files

The problem with that is that you're going to need a scrollbar for each pane, so you're heading towards a situation where you either have to have [multiple panes]|[single shared gutter]|[multiple scroll bars] or a non-standard [multiple panes]|[multiple scroll bars]|[single shared gutter] i.e. the scroll bar is no longer at the extreme right of the window where expected.

Just tested on ARM (ARMHF, with FPC trunk), Laz trunk. Noticed Martin's fix that still allows compilation on ARM.
Compiling the IDE went ok.
Starting it showed an error dialog:
The package EditorMacroScript (pascalscript macros) has detected a problem and was deactivated.
The package failed its selftest with the message "Failed getstr2 [Expr: "456" / Got: "abc"]".

I wonder whether disabling the package on ARM CPUs as well as other non x86/x64 CPUs would be a good solution?

Again IMHO 2 issues.

1) Sparc:
there is nothing there yet, so yes, an IFDEF in EditorMacroScript would be in order to avoid fact I wonder why it does not happen. There already is:

{$if defined(cpupowerpc)}
   {$ifNdef darwin}  {$DEFINE PasScriptNotAvail } {$ifend}
   {$ifNdef cpu32}  {$DEFINE PasScriptNotAvail } {$ifend}
{$ifend}
{$if defined(cpusparc) }  {$DEFINE PasScriptNotAvail } {$ifend}
const
   EMSSupported = {$IFDEF PasScriptNotAvail} False {$ELSE} True {$ENDIF} ;

- ppc only on 32 bit darwin
- no sparc

2)ARM:
pascal script has some support, it may be fixable, but I can not test...

Sorry, perhaps I wasn't clear: I wasn't talking about SPARC; you & Mark covered that AFAIU?

2)ARM: pascal script has some support, it may be fixable, but I can not test...

TBH, I'm not interested in testing it either. I'll just keep it disabled for all platforms in my default fpcup settings then.

Thanks,
Reinier

I look at disabling it, (if I remember) but it may be back some day.

Also, just to be sure. This should have been a one off (per install) message? It should not come back. It should stay disabled, even if you upgrade. (Except, when I change the selftest version in the code, but then again it will be a once off)

See below: no need to do that on my account ;)

Also, just to be sure. This should have been a one off (per install) message? It should not come back. It should stay disabled, even if you upgrade. (Except, when I change the selftest version in the code, but then again it will be a once off)

Yes, it appears only once after an installation; restarting Lazarus or rebuilding Lazarus doesn't lead to it showing again. I didn't know that, sorry - that makes it much, much more acceptable, thanks.