Truncate strings to maximum length
显示 更早的评论
I have an array of strings (new strings, not old character arrays) of varying lengths, and I would like to truncate them to below some maximum length. I thought that extractBefore would be helpful, but it fails for strings shorter than the truncation length.
>> str = ["ab", "cdefg"];
>> extractBefore(str, 3) % this works fine
ans =
1×2 string array
"ab" "cd"
>> extractBefore(str, 4) % this fails
Error using extractBefore
Numeric value exceeds the number of characters in element 1.
This is my current solution:
>> arrayfun(@(s)extractBefore(s, min(4, s.strlength())+1), str)
ans =
1×2 string array
"ab" "cdef"
However, this is awkward and difficult to read.
Is there no ready-made functionality for doing this?
1 个评论
dpb
2024-10-21
"Is there no ready-made functionality for doing this?"
Amazingly, no...with everything they did add, the equivalent of MID$, LEFT$, RIGHT$ are not available in a directly-callable functional form.
回答（2 个）
Omit arrayfun:
str = ["ab", "cdefg"];
extractBefore(str, min(4, str.strlength())+1)
If this still looks to clumsy, write your own function:
limitWidth(str, 4)
function s = limitWidth(s, n)
s = extractBefore(s, min(n, s.strlength()) + 1);
end
str = ["", "ab", "1234", "cdefgh"];
out = regexprep(str,'^(.{1,4}).*?$','$1') % LEFT
out = regexprep(str,'^.*?(.{1,4})$','$1') % RIGHT
or (but note the behavior of the empty string):
out = regexp(str,'^.{1,4}','match','once') % LEFT
out = regexp(str,'.{1,4}$','match','once') % RIGHT
4 个评论
dpb
2024-10-29
str = ["", "ab", "1234", "cdefgh"];
out = regexprep(str,'^(.*?)(.{1,4})(??@sprintf(".{%u}",numel($1)))$','$2') % MID
Note that MID is a somewhat ill-defined task: for odd vs even numbers of characters there could easily be quite different expected outputs (I can think of three reasonable interpretations). The above returns up to the maximum requested characters, ensuring that the same number of characters are trimmed from both ends.
@Stephen23, by writing the function names with the $ sign, I was presuming it would be clear the intent was to mimic BASIC (and VBA) which originally required the trailing $ on string variables. Current dialects have dropped the $ convention, of course, but the functionality is retained.
The definition of MID$
res=mid(str,start,[len])
to pick up to len characters from str beginning at start position in str. If omitted, the rest of the original string is returned, in which case it mimics RIGHT$.
It doesn't mean to try to split the string midway so there's no dependency on whether the string length is even/odd nor any ambiguity in the result.
Perhaps:
str = ["", "ab", "1234", "cdefgh"];
out = regexprep(str,'^.{0,2}(.{0,4}).*$','$1') % MID(START,LEN)
类别
在 帮助中心 和 File Exchange 中查找有关 Characters and Strings 的更多信息
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!
Translated by 
