Much better! I'll change it soon. BTW, I'm sorry about *R>* for "Roll Down". The original code used *R<* and *R>*, apparently because we don't have up and down arrows on a normal keyboard (well, speaking for US English keyboards, anyway). I changed the one to a caret (*R^*) but there's no similar down symbol one can type. I could have used a unicode arrow but it can't be typed conveniently. The way the code works, if you type the string on a button (some of them, anyway) it activates the same command as if you had clicked on its button.
On Monday, May 1, 2023 at 2:22:05 PM UTC-4 jkn wrote: > Shurely that should be called >CLIP ? ;-) > > On Monday, May 1, 2023 at 5:46:45 PM UTC+1 [email protected] wrote: > >> Devel now contains one more change. I've changed the *EXIT* key (which >> isn't needed in the Leo tab version of the calculator) to *TOCLIP*. It >> copies the "X" register - the calculation result - to the system clipboard. >> >> On Monday, May 1, 2023 at 9:31:00 AM UTC-4 Thomas Passin wrote: >> >>> When I was using TurboPascal and doing a lot of numerical 2-D >>> integrations with complex numbers, I actually wrote a little library module >>> to calculate with complex numbers as if I was using an RPN calculator. So >>> you could push a complex number on the stack, pop it off, multiply or add >>> the two numbers on the stack bottom, etc. At that time TurboPascal did not >>> have complex numbers of its own, IIRC. If N1 and N2 were two complex >>> numbers you could write, for example (based on hazy memories from long ago): >>> >>> push(N1) >>> push(N2) >>> CMul() >>> { and so forth, pun intended } >>> >>> I enjoyed using the library because it was so easy for me to write and >>> debug calculations. I just pictured how I would do the calculation on my >>> HP calculator and walked through the steps. I timed it once, and the extra >>> overhead of using the stack library compared with a hand-crafted sequence >>> of operations was about 25% (I'm sure my implementation could have been >>> improved, it was pretty brute-force). But the ease of writing the >>> calculation and debugging it - the RPN library won hands down. >>> >>> On Monday, May 1, 2023 at 9:02:49 AM UTC-4 jkn wrote: >>> >>>> I got to play with a then- just out Hewlett Packard HP-67 RPN >>>> calculator at the age of around 14. It blew my mind ... and may well have >>>> directly led to me doing what I do to this day. >>>> >>>> J^n >>>> >>>> On Sunday, April 30, 2023 at 5:59:34 PM UTC+1 [email protected] wrote: >>>> >>>>> RPCalc is a recent discovery for me. As originally written, it runs as >>>>> a standalone program, and requires Qt5. You don't need to use the >>>>> installer package for Windows. Just download the Linux tarball, >>>>> decompress >>>>> it, and navigate to the "source" directory. The file to run is - >>>>> surprise! >>>>> - rpcalc.py. It seems to do everything I want from an RPN calculator, >>>>> except that copying the stack bottom is awkward. >>>>> >>>>> To adapt it for Leo, one change was to combine all the source files >>>>> into one Leo @file tree. Another was to change the imports to use leoQt, >>>>> which makes it easier to adapt to Qt5 vs Qt6, and anyway is essential if >>>>> the program is to run in a Leo frame. I'm still finding little things >>>>> that >>>>> aren't working for both Qt5 and Qt6 - mostly enums and flags - but I'm >>>>> making progress. But overall, most of the functionality works and the >>>>> thing >>>>> is usable as it stands. I'll post an updated outline soon, and after >>>>> some >>>>> more work it should be ready to appear in the Leo repo. >>>>> >>>>> On Sunday, April 30, 2023 at 11:55:06 AM UTC-4 jkn wrote: >>>>> >>>>> I have wondered about suggesting something like this for a while, so >>>>> thank you Thomas. My 'main' editor has a simple HP calculator built into >>>>> it >>>>> and it was an easy step to consider one for Leo. >>>>> >>>>> I didn't know about RPNCalc (I have some Android RPN apps on my phone, >>>>> as well as a real HP-35s), but it sounds like a good choice. >>>>> >>>>> >>>>> I've used HP RPN calculators since way back in HP-45 days. I >>>>> liked the HP-25C even better, and finally ended up using an HP-15C. Mine >>>>> still works though it's slightly misplaced just now. On my computer I've >>>>> been using Free42, which seems to me to be a good balance between >>>>> readability, complexity, and capability. Now it looks like RPCalc will >>>>> be >>>>> taking over from Free42. >>>>> >>>>> >>>>> I will take a look at this shortly - thanks. >>>>> >>>>> J^n >>>>> >>>>> >>>>> On Sunday, April 30, 2023 at 12:03:14 PM UTC+1 Edward K. Ream wrote: >>>>> >>>>> On Sat, Apr 29, 2023 at 12:42 PM Thomas Passin <[email protected]> >>>>> wrote: >>>>> >>>>> I have adapted the open-source *RPCalc* calculator to run in a tab in >>>>> the Leo log frame. This calculator is a Reverse Polish Notation (RPN) >>>>> style calculator, which IMHO is much better than the algebraic-entry >>>>> type. It is the type of calculator that Hewlett-Packard made famous. >>>>> >>>>> >>>>> Thanks for this work, Thomas. The calculator appears as expected for >>>>> me. >>>>> >>>>> PR #3301 <https://github.com/leo-editor/leo-editor/pull/3301> is a >>>>> draft containing the files you mention. It's a good start. The PR lists >>>>> three problems. >>>>> >>>>> Edward >>>>> >>>>> -- You received this message because you are subscribed to the Google Groups "leo-editor" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/leo-editor/3ba71d1a-611f-41d1-917a-3cd2fdb45514n%40googlegroups.com.
