Rewriting history

[Trevoke]

Bringing this issue over from google code.

History rewriting can be a very useful feature. It means going back to the
previous calculation, change something, and the subsequent calculations
should be adjusted to the change.

Example: "x=1+2" followed by "x*2" yields 6. If the user goes back to the
first expression and change it to "x=1+10", now the end result should be 22.

command	result
x = 1 + 2	3
x * 2	6

Change first expression ....

command	result
x = 1 + 10	11
x * 2	22

I would like not only rewriting, but REMOVAL of SPECIFIC old entries as well. Considering that the program has a limited number of entries it displays in history, it's really annoying if I accidentally hold down enter (even for a fraction of a second due to my key delay and repeat delay settings) and I lose a whole bunch of expressions in the history due to it.

[frodohakr]

I agree, these features would be useful.

• Frodo Baggins: Hobbit, Wikipedia editor (blackhat999
  https://en.wikipedia.org/wiki/User:Blackhat999), Linux freak, and hacker
  extraordinaire.

On Mon, May 11, 2015 at 2:04 PM, Aldric Giacomoni [email protected]
wrote:

Bringing this issue over from google code
https://code.google.com/p/speedcrunch/issues/detail?id=204.

History rewriting can be a very useful feature. It means going back to the
previous calculation, change something, and the subsequent calculations
should be adjusted to the change.

Example: "x=1+2" followed by "x*2" yields 6. If the user goes back to the
first expression and change it to "x=1+10", now the end result should be
22.

command result x = 1 + 2 3 x * 2 6

Change first expression ....
command result x = 1 + 10 11 x * 2 22

I would like not only rewriting, but REMOVAL of SPECIFIC old entries as
well. Considering that the program has a limited number of entries it
displays in history, it's really annoying if I accidentally hold down enter
(even for a fraction of a second due to my key delay and repeat delay
settings) and I lose a whole bunch of expressions in the history due to it.

—
Reply to this email directly or view it on GitHub
#57.

[heldercorreia]

Thanks for doing this. With the announced death of Google Code I'm still trying to decide whether to use this issue tracker or Bitbucket's though. Feedback welcomed.

[Trevoke]

I don't know which one is your master repository; you should use whichever issue tracker is most comfortable for YOU, since you will be using it way more than most people who come in with a single issue.

[AlexDaniel]

I'd say that a comfortable issue tracker for users is very important, otherwise some people might refrain from submitting bug reports if it takes a bit more time than they expect. Somehow everything is on GitHub nowadays, so it is a good choice. At the same time, I'm really worried about the fact that GitHub is actually a commercial organization that has nothing to do with free software, so a monopoly like this might not be the best thing for the community.

Also, git is distributed. That is, nothing should stop you from using as many git hostings as you want. You can keep both bitbucket and github. Torture the maintainer on behalf of the user, hehe.

[heldercorreia]

@AlexDaniel You are spot on about the users. About the distributed Git nature - I've been making sure in the last few years that SpeedCrunch's code is absolutely replicated here, on Bitbucket, Gitorious (now dead), and Google Code (dying). That's not the issue, but rather which issue tracker to use. I think that 1) keeping a multiple (unsynchronized and decentralized) tracker system does not feel right, and 2) keeping multiple synchronized trackers also sounds awful (from a coordinator's perspective, at least). I truly feel I should pick one and stick to it officially. So far, from my professional experience, I like Bitbucket's flow and interface better. But I know that Github is widely popular and I don't want to rush the decision until Google Code effectively dies. That's why I asked for feedback.

[AlexDaniel]

@heldercorreia which makes me think that git should actually have its own built-in issue tracker... what a crazy idea, but just imagine how many problems will actually be solved after this is implemented...

[aaaxx]

👍 I cannot upvote this history feature highly enough (including removal of entries)!