| Type | Wish | Status | waiting | Date | 30-Oct-2010 22:04 |
|---|---|---|---|---|---|
| Version | alpha 109 | Category | Mezzanine | Submitted by | BrianH |
| Platform | All | Severity | minor | Priority | normal |
| Summary | KEEP-EACH function |
|---|---|
| Description |
KEEP-EACH would be the inverse of REMOVE-EACH: Values would be kept if the condition evaluates to true. As such, it could be a simple mezzanine wrapper for REMOVE-EACH. Suggested by Gregg in a #1718 comment. Code below. The #539 problem applies. |
| Example code |
keep-each: func [
"Removes values for each block that returns FALSE. Returns remove count."
'word [get-word! word! block!] "Word or block of words to set each time (local)"
data [series!] "Series to traverse (modified)"
body [block!] "Block to evaluate (return TRUE to keep)"
][
remove-each (word) data reduce [:not :do body]
]
|
| Assigned to | n/a | Fixed in | - | Last Update | 2-Nov-2010 22:43 |
|---|
| Comments | |
|---|---|
|
(0002714)
Carl 1-Nov-2010 00:06 |
Do we think it's worth-while adding?
It's the same as: remove-each word series [not condition] That's not a big difference. Really it comes down to if we think the abstraction of the KEEP word is of benefit. I'm borderline on it. |
|
(0002760)
Gregg 2-Nov-2010 18:35 |
I find it useful, though I won't say it's necessary. It makes the intent clearer than the double negative remove+not approach.
Why did you choose to have a negative filter option initially? |
| Date | User | Field | Action | Change |
|---|---|---|---|---|
| 2-Nov-2010 22:43 | BrianH | Description | Modified | - |
| 2-Nov-2010 18:35 | Gregg | Comment : 0002760 | Added | - |
| 1-Nov-2010 00:06 | carl | Comment : 0002714 | Added | - |
| 1-Nov-2010 00:03 | carl | Status | Modified | submitted => waiting |
| 30-Oct-2010 22:20 | BrianH | Code | Modified | - |
| 30-Oct-2010 22:06 | BrianH | Description | Modified | - |
| 30-Oct-2010 22:06 | BrianH | Code | Modified | - |
| 30-Oct-2010 22:04 | BrianH | Ticket | Added | - |