Home • About Us • A-Z Index • Search * Contact Us • Register • Login • Press • Shop

The Open Brand -- Problem Reporting and Interpretations System

Problem Report 0725 Details

Help Show help | Quick Search | Submit a Test Suite Support Request | Click here to view your privileges

This page provides all information on Problem Report 0725.

Report 0725 Actions

    Problem Report Number 0725
    Submitter's Classification Specification problem
    State Resolved
    Resolution Withdrawn (WDN)
    Problem Resolution ID WDN.X.0022
    Raised 1970-01-01 07:00
    Updated 2005-07-22 00:47
    Published 2005-07-22 00:47
    Product Standard Commands and Utilities V2 (UNIX 95)
    Certification Program The Open Brand certification program
    Test Suite VSC version 4.1.5
    Test Identification POSIX.cmd/pr 18,20
    Specification Commands and Utilities Issue 4 Version 2
    Location in Spec See Problem Text
    Problem Summary TIN4C.00045 This test may fail on implementations because of a grey area
    in the treatment of adjacent space.
    Problem Text This test may fail on implementations because of a grey area in the
    treatment of adjacent space.

    The assertions in question test the semantics of the -i[char][gap]
    option to the pr utility. Both of these tests were changed from
    VSC4.1.4 to VSC4.1.5 based on a waiver request that was accepted as
    TSD4C.00086. In our view, that request was clearly in error and
    should
    not have been granted. This TSD says, in part:

    For the case of 6*gap+1, which is position 25, and 15*gap+1,
    which is position 61, the test expects space replacement
    to occur
    at these positions. In the case of tp20 it is an X and
    in the
    case of tp18 it is a tab character. It should not
    expect any
    space replacement at these points because there is only one
    ^^^^^^^^^^^^^^^^^
    adjacent space to these column positions in both cases.
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

    This is simply not true. There is only one adjacent space to
    the left
    of the specified column, but the standard makes no distinction
    in this
    regard. The test in VSC4.1.4 expected tab replacement for the five
    adjacent spaces at 7*gap+1 and the nine adjacent spaces at 17*gap+1.
    (See TSD4C.00086 for a description of the input files.)
    Incidentally,
    this is also the historical behavior of the pr utility.

    We have submitted a request for a POSIX.2 interpretation on this
    issue. The text of this interpretation request is attached, and
    gives our reasoning. Until this interpretation is available, we
    request a waiver on the grounds that this is a grey area in the
    specification.

    ------------------start of POSIX.2 interpretation
    request---------------

    Dear Interpretations Committee,

    I would like to request an official binding interpretation of IEEE
    Std 1003.2-1992 (POSIX.2). This request deals with the semantics
    of the "-i" option to the "pr" utility. These semantics are
    specified
    in POSIX.2 subclause 4.49.3 lines 8037-8043, which state (in part):

    -i[char][gap]
    In output, replace multiple <space>s with <tab>s
    wherever two or more adjacent <space>s reach
    column positions gap + 1, 2 x gap + 1, 3 x gap + 1,
    etc. If gap is zero or is omitted, default <tab>
    settings at every eighth column position shall be
    assumed.

    The issue that I would like addressed is the following: If an
    input line contains a space at position (n x gap + 1) followed by
    n consecutive spaces, should the output file contain a single
    space at position (n x gap + 1) followed by a single tab, or
    should it contain two tabs? An example may make this clearer.
    Suppose that "inputfile" contains the following two lines

    1234567890 This line shows column positions
    123 90

    (five consecutive spaces, no tabs, in the second line) and
    suppose that the command

    pr -i4 inputfile

    is issued. Should the output generated by the second line
    be

    123 ^I90
    or
    123^I^I90

    Here ^I stands for a tab.

    I contend that the second output is correct (or that, at best,
    the standard is ambiguous). To support this, note that the five
    consecutive spaces reach column position 9 = (2 x 4 + 1), and
    therefore (according to the specification) should all be replaced
    by tabs. The fact that the first of these spaces is at a tab
    position does not seem relevant and is not addressed in the
    standard.

    Thank you for your attention to this matter.

    -------------------end of POSIX.2 interpretation
    request----------------

    Test Output -----------------start of test output------------------

    520|244 1 9964 1 1|Assertion #18 (A): Test the -i[gap] option
    520|244 1 9964 1 11|Command failed: 'cmp -s pr_eso_18_1 pr_eso_18_2'
    220|244 1 1 16:03:21|FAIL
    410|244 18 1 16:03:21|IC End

    520|244 1 9964 1 1|Assertion #20 (A): Test the -i[char][gap] option
    520|244 1 9964 1 11|Command failed: 'cmp -s pr_eso_20_1 pr_out_20_2'
    220|244 1 1 16:03:24|FAIL
    410|244 20 1 16:03:24|IC End

    ------------------end of test output-------------------

    Review Information

    Review Type TSMA Review
    Start Date null
    Completed null
    Status Complete
    Review Recommendation No Resolution Given
    Review Response The standard is not clear whether the two or more adjacent
    spaces rule applies to the greatest number of contiguous spaces
    prior to a defined column position (x*gap+1), or only spaces between
    each defined column position (x*gap+1 and (x+1)*gap+1):

    In output, replace multiple <space>s with <tab>s
    wherever two or more adjacent <space>s reach
    column positions gap + 1, 2 x gap + 1, 3 x gap + 1,
    etc. If gap is zero or is omitted, default <tab>

    Consider the example again.

    1234567890 This line shows column positions
    123 90

    Either the rule applies to the single space at column 4 and columns
    5-8 separately, or the rule applies to all the spaces between 4
    and 8.

    In the former case, the single space remains and the second block is
    turned into a tab. In the latter case, the entire block of spaces
    is turned into two tabs, one for each column position (x*gap+1)
    reached.

    Pending the outcome of the POSIX.2 interpretation filed, a temporary
    interpretation is recommended.

    Review Type SA Review
    Start Date null
    Completed null
    Status Complete
    Review Resolution Temporary Interpretation (TIN)
    Review Conclusion A Temporary Interpretation is granted.

    Problem Reporting System Options:

     

    Back   

Contact the Certification Authority