Bitcoin-Qt: disable addressbook button when an addr is valid #2219

pull Diapolo wants to merge 1 commits into bitcoin:master from Diapolo:Qt_addrbutton changing 3 files +32 −0

Diapolo commented at 1:47 PM on January 26, 2013: none
- this change is active for sendcoinsentry and signverifymessagedialog
- it disables the addressbook button, when an entered address is a valid Bitcoin address (it gets re-enabled when that case changes)
Partly related to #2218.
BitcoinPullTester commented at 2:19 PM on January 26, 2013: none

Automatic sanity-testing: PASSED, see http://jenkins.bluematt.me/pull-tester/8257c7e3e845a455606373b480154877274bbc6f for binaries and test log.
laanwj commented at 4:36 PM on January 26, 2013: member

I personally don't really like buttons appearing/disappearing while typing (causing layout changes). It might be better to use enable/disable?
schildbach commented at 5:03 PM on January 26, 2013: contributor

Let's take this even one step further. If the user has already selected an address, why would she want to type another one? So I'd suggest disabling the address field (meaning it is only used for display, editing is disabled) and remove the addressbook button in that case.

No interactive layout changes necessary any more...
laanwj commented at 6:16 PM on January 26, 2013: member

They may have selected the wrong one and want to change it...
schildbach commented at 6:23 PM on January 26, 2013: contributor

In this case, go back and select the right one. Also, do you think a lot of people will even notice their mistake? They would need compare the hash value against a memorized value.
Diapolo commented at 10:29 AM on January 27, 2013: none

This discussion seems rather controversial, so @laanwj would you ACK this, if I just disable the button on valid address entered? I'm fine with such a solution.

Edit: Changed to enable/disable instead of show/hide.
BitcoinPullTester commented at 11:07 AM on January 27, 2013: none

Automatic sanity-testing: PASSED, see http://jenkins.bluematt.me/pull-tester/90ca3d8ff0f8e4e91c13ed1ed801bda7aa08ea49 for binaries and test log.
laanwj commented at 2:40 PM on January 28, 2013: member

@philip yes, disabling is fine
luke-jr commented at 6:16 PM on January 30, 2013: member

I disagree with this change. I often sign a message, then want to change to another address later. It'd be annoying to have to edit the address to make it invalid just to change it...
Diapolo commented at 6:17 PM on January 30, 2013: none

You are right, but you could also use the clear button ^^.

Bitcoin-Qt: disable addressbook button when an addr is valid

- this change is active for sendcoinsentry and signverifymessagedialog
- it disables the addressbook button, when an entered address is a valid
  Bitcoin address (it gets re-enabled when that case changes)

177783b285

BitcoinPullTester commented at 7:28 AM on March 18, 2013: none

Automatic sanity-testing: PASSED, see http://jenkins.bluematt.me/pull-tester/177783b285c6a421a2c02f7434f058bb5f843143 for binaries and test log.
Diapolo commented at 6:58 AM on July 23, 2013: none

@laanwj @luke-jr Is there any interest in this? If no I'm fine with just closing this pull.
laanwj commented at 5:54 AM on July 26, 2013: member

I'm not really convinced this is an useful or expected behavior, to be honest. It arbitrary prevents the user from doing things (ie, select another address) without being clear what can be done to do override that. He'd have to jump through a hoop by deleting a character from the address. Usually I don't like when GUIs do that.
Diapolo commented at 6:30 AM on July 26, 2013: none

That's fine with me, closing this.
Diapolo closed this on Jul 26, 2013
DrahtBot locked this on Sep 8, 2021

Contributors

Linked (view graph)

#2218 "Verify message" button doesn't work when invoked recursively