I have a data string that is defined as having two characters for the ID value, six characters for longitude, etc. "Fixed-length string" does not convey the requirement for internal bits of the string to be fixed-length as well.

  • Apologies if this belongs in a computer-science-oriented community instead. Commented Apr 25, 2012 at 6:38
  • Indeed, it's only fixed-length overall because the constituents are fixed-length. And if the overall length is fixed, the constituents are usually also.
    – Kaz
    Commented Apr 25, 2012 at 7:47
  • The string itself cannot be "called" something just because its constituents obey a pattern. You have to call it a standardized string perhaps, and explain 'standardized'. I'm not posting this as an answer, hoping someone will come up with a better one, if there is any.
    – Kris
    Commented Apr 25, 2012 at 8:01
  • When I saw the title to this question, my first thought was "chain". :-)
    – Pitarou
    Commented Apr 25, 2012 at 11:54

6 Answers 6


The concept comes from punch cards and mechanical readers where field locations had to be set in hardware. It is called a:

fixed length record

A straightforward google search for fixed length record will find you lots of similar definitions.


I would call it "a data string made of fixed-length fields". A bit verbose, but precise enough, I think.

  • Not sure if, at a time where such data was more common (because it was easy to parse), there was a specific term. Now, such format is mostly seen as outdated, being quite rigid, preferred format is CSV, or more structured data like Json, Yaml, XML... I suppose you got it from some old software or device, or some protocol / file format.
    – PhiLho
    Commented Apr 25, 2012 at 10:15

You could call it a data structure.

Or, to be more precise, a data structure that happens to reside within a fixed length string.

  • You're almost but not quite there. Afterall, a data structure is not necessarily, or even really, a string.
    – Kris
    Commented Apr 25, 2012 at 8:07
  • I've expanded my answer. Do you still think I'm not there?
    – Pitarou
    Commented Apr 25, 2012 at 11:50

Fixed-format string comes closest to the idea of a string whose costituents are all of fixed length.

Borrowing fron the legacy concept of a fixed-format record where fields are always of the same length, this term however, avoids all references to the file-record system. So much so, I do not want to mention even 'data structure'.


This is squarely in my field, yet nothing that specific comes to mind. The length of the constituents of this data unit being fixed is not really its most important attribute. Once upon a time, if you used the word record that would instantly conjure up an image of some fixed length unit with fixed fields. One word which still carries those connotations might be block.

Today, even if you are dealing with such a string, you're likely to make it "future proof" by writing the code so that it actually parses the pieces as if they were not fixed length (unless it is impossible to avoid relying on the fixed lengths due to some inherent ambiguity and lack of delimiting).

Good question for programmers.stackexchange.com.

  • I agree that this is a pretty outdated concept. However, the format I am working with--International Marine Meteorological Archive (IMMA) format--is a string with fixed-length portions with the attendant danger of a missed character meaning all following portions will be incorrect. I am creating an XML schema to ease data input with the hopes that someone will write a parser to take an XML file and turn it into this unwieldy string. In the documentation I'm writing, it is this rigid format that I am trying to emphasize, so to me it is the most important attribute. Commented Apr 25, 2012 at 20:42

You can call it a position-delimited string. Just a few days ago someone used this term, and I thought of it being exactly what you are describing despite having never heard it before. When I later asked what it really is, it turned out to be correct. As such, this term is both technically correct and intuitive to understand.

  • I don't think delimited means what you think it means. Delimited means “surrounded by”; it doesn’t mean “determined by”. Parenthetical statements are delimited by parentheses, while quotations are delimited by quotation marks. But fixed-width components are determined by their position, offset, or width. There is no “delimiting” involved here at all. There are no delimiters.
    – tchrist
    Commented May 8, 2017 at 0:50
  • @tchrist the Merriam-webster doctionary states that the definition of delimit is "to fix or define the limits of". In this case the limits of each field are determined by position. As such, it is a position-delimited string. (merriam-webster.com/dictionary/delimit)
    – john01dav
    Commented May 8, 2017 at 1:00
  • That's nice. However, in computer science, delimit and delimiter are technical jargon, terms of art well outside the compass of mere collegiate dictionaries. A more relevant published definition from a standard reference book in the associated specialist domain involved here gives delimiter: A character or string that sets bounds to an arbitrarily sized textual object, not to be confused with a separator or terminator. "To delimit" really just means "to surround" or "to enclose" (like these parentheses are doing).
    – tchrist
    Commented May 8, 2017 at 1:03
  • @tchrist where is that definition from?
    – john01dav
    Commented May 8, 2017 at 1:09
  • From here :).
    – tchrist
    Commented May 8, 2017 at 1:10

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.