Reverted the addition of a NORMALIZE_NUMBERS option, per Tim Peter's

request.  Tim says that "correct 'fuzzy' comparison of floats cannot
be automated."  (The motivation behind adding the new option
was verifying interactive examples in Python's latex documentation;
several such examples use numbers that don't print consistently on
different platforms.)

1 files changed, 1 insertions, 19 deletions

diff --git a/Doc/lib/libdoctest.tex b/Doc/lib/libdoctest.tex
index a30a432..2635486 100644
--- a/Doc/lib/libdoctest.tex
+++ b/Doc/lib/libdoctest.tex
@@ -581,17 +581,6 @@ TypeError: object doesn't support item assignment
 
 \end{datadesc}
 
-\begin{datadesc}{NORMALIZE_NUMBERS}
-    When specified, number literals in the expected output will match
-    corresponding number literals in the actual output if their values
-    are equal (to ten digits of precision).  For example, \code{1.1}
-    will match \code{1.1000000000000001}; and \code{1L} will match
-    \code{1} and \code{1.0}.  Currently, \constant{NORMALIZE_NUMBERS}
-    can fail to normalize numbers when used in conjunction with
-    ellipsis.  In particular, if an ellipsis marker matches one or
-    more numbers, then number normalization is not supported.
-\end{datadesc}
-
 \begin{datadesc}{COMPARISON_FLAGS}
     A bitmask or'ing together all the comparison flags above.
 \end{datadesc}
@@ -713,7 +702,7 @@ can be useful.
 
 \versionchanged[Constants \constant{DONT_ACCEPT_BLANKLINE},
     \constant{NORMALIZE_WHITESPACE}, \constant{ELLIPSIS},
-    \constant{IGNORE_EXCEPTION_DETAIL}, \constant{NORMALIZE_NUMBERS},
+    \constant{IGNORE_EXCEPTION_DETAIL},
     \constant{REPORT_UDIFF}, \constant{REPORT_CDIFF},
     \constant{REPORT_NDIFF}, \constant{REPORT_ONLY_FIRST_FAILURE},
     \constant{COMPARISON_FLAGS} and \constant{REPORTING_FLAGS}
@@ -751,7 +740,6 @@ in any particular order, so a test like
 
 % Hey! What happened to Monty Python examples?
 % Tim: ask Guido -- it's his example!
-% doctest: ignore
 \begin{verbatim}
 >>> foo()
 {"Hermione": "hippogryph", "Harry": "broomstick"}
@@ -759,7 +747,6 @@ in any particular order, so a test like
 
 is vulnerable!  One workaround is to do
 
-% doctest: ignore
 \begin{verbatim}
 >>> foo() == {"Hermione": "hippogryph", "Harry": "broomstick"}
 True
@@ -767,7 +754,6 @@ True
 
 instead.  Another is to do
 
-% doctest: ignore
 \begin{verbatim}
 >>> d = foo().items()
 >>> d.sort()
@@ -779,7 +765,6 @@ There are others, but you get the idea.
 
 Another bad idea is to print things that embed an object address, like
 
-% doctest: ignore
 \begin{verbatim}
 >>> id(1.0) # certain to fail some of the time
 7948648
@@ -791,7 +776,6 @@ Another bad idea is to print things that embed an object address, like
 The \constant{ELLIPSIS} directive gives a nice approach for the last
 example:
 
-% doctest: ignore
 \begin{verbatim}
 >>> C() #doctest: +ELLIPSIS
 <__main__.C instance at 0x...>
@@ -801,7 +785,6 @@ Floating-point numbers are also subject to small output variations across
 platforms, because Python defers to the platform C library for float
 formatting, and C libraries vary widely in quality here.
 
-% doctest: ignore
 \begin{verbatim}
 >>> 1./7  # risky
 0.14285714285714285
@@ -1635,7 +1618,6 @@ Doctest provides several mechanisms for debugging doctest examples:
 
       Then an interactive Python session may look like this:
 
-% doctest: ignore
 \begin{verbatim}
 >>> import a, doctest
 >>> doctest.testmod(a)