Fix doc on ranges and locales again. Add test.

9 files changed, 154 insertions, 65 deletions

diff --git a/doc/ChangeLog b/doc/ChangeLog
index 88743adc..f279986c 100644
--- a/doc/ChangeLog
+++ b/doc/ChangeLog
@@ -1,3 +1,8 @@
+2012-08-12         Arnold D. Robbins     <arnold@skeeve.com>
+
+	* gawk.texi (Ranges and Locales): Clarified ranges and
+	locales. Again.
+
 2012-08-05         Arnold D. Robbins     <arnold@skeeve.com>
 
 	* gawk.texi (PC Binary Installation): Document Eli Zaretskii's
diff --git a/doc/gawk.info b/doc/gawk.info
index 91c9d516..275188ef 100644
--- a/doc/gawk.info
+++ b/doc/gawk.info
@@ -20186,8 +20186,7 @@ outside those locales, the ordering was defined to be based on
    In many locales, `A' and `a' are both less than `B'.  In other
 words, these locales sort characters in dictionary order, and
 `[a-dx-z]' is typically not equivalent to `[abcdxyz]'; instead it might
-be equivalent to `[aBbCcDdXxYyZz]', for example.  (And to make things
-worse, on other systems, it might be equivalent to `[aAbBcCdDxXyYz]'.)
+be equivalent to `[ABCXYabcdxyz]', for example.
 
    This point needs to be emphasized: Much literature teaches that you
 should use `[a-z]' to match a lowercase character.  But on systems with
@@ -27834,63 +27833,63 @@ Node: BTL798431
 Node: POSIX/GNU799165
 Node: Common Extensions804316
 Node: Ranges and Locales805423
-Ref: Ranges and Locales-Footnote-1810134
-Ref: Ranges and Locales-Footnote-2810161
-Ref: Ranges and Locales-Footnote-3810421
-Node: Contributors810642
-Node: Installation814904
-Node: Gawk Distribution815798
-Node: Getting816282
-Node: Extracting817108
-Node: Distribution contents818800
-Node: Unix Installation824022
-Node: Quick Installation824639
-Node: Additional Configuration Options826601
-Node: Configuration Philosophy828078
-Node: Non-Unix Installation830420
-Node: PC Installation830878
-Node: PC Binary Installation832177
-Node: PC Compiling834192
-Node: PC Testing837136
-Node: PC Using838312
-Node: Cygwin842497
-Node: MSYS843497
-Node: VMS Installation844011
-Node: VMS Compilation844614
-Ref: VMS Compilation-Footnote-1845621
-Node: VMS Installation Details845679
-Node: VMS Running847314
-Node: VMS Old Gawk848921
-Node: Bugs849395
-Node: Other Versions853247
-Node: Notes858528
-Node: Compatibility Mode859220
-Node: Additions860003
-Node: Accessing The Source860815
-Node: Adding Code862240
-Node: New Ports868207
-Node: Dynamic Extensions872320
-Node: Internals873696
-Node: Plugin License882799
-Node: Sample Library883433
-Node: Internal File Description884119
-Node: Internal File Ops887834
-Ref: Internal File Ops-Footnote-1892615
-Node: Using Internal File Ops892755
-Node: Future Extensions895132
-Node: Basic Concepts897636
-Node: Basic High Level898393
-Ref: Basic High Level-Footnote-1902428
-Node: Basic Data Typing902613
-Node: Floating Point Issues907138
-Node: String Conversion Precision908221
-Ref: String Conversion Precision-Footnote-1909921
-Node: Unexpected Results910030
-Node: POSIX Floating Point Problems911856
-Ref: POSIX Floating Point Problems-Footnote-1915561
-Node: Glossary915599
-Node: Copying940575
-Node: GNU Free Documentation License978132
-Node: Index1003269
+Ref: Ranges and Locales-Footnote-1810041
+Ref: Ranges and Locales-Footnote-2810068
+Ref: Ranges and Locales-Footnote-3810328
+Node: Contributors810549
+Node: Installation814811
+Node: Gawk Distribution815705
+Node: Getting816189
+Node: Extracting817015
+Node: Distribution contents818707
+Node: Unix Installation823929
+Node: Quick Installation824546
+Node: Additional Configuration Options826508
+Node: Configuration Philosophy827985
+Node: Non-Unix Installation830327
+Node: PC Installation830785
+Node: PC Binary Installation832084
+Node: PC Compiling834099
+Node: PC Testing837043
+Node: PC Using838219
+Node: Cygwin842404
+Node: MSYS843404
+Node: VMS Installation843918
+Node: VMS Compilation844521
+Ref: VMS Compilation-Footnote-1845528
+Node: VMS Installation Details845586
+Node: VMS Running847221
+Node: VMS Old Gawk848828
+Node: Bugs849302
+Node: Other Versions853154
+Node: Notes858435
+Node: Compatibility Mode859127
+Node: Additions859910
+Node: Accessing The Source860722
+Node: Adding Code862147
+Node: New Ports868114
+Node: Dynamic Extensions872227
+Node: Internals873603
+Node: Plugin License882706
+Node: Sample Library883340
+Node: Internal File Description884026
+Node: Internal File Ops887741
+Ref: Internal File Ops-Footnote-1892522
+Node: Using Internal File Ops892662
+Node: Future Extensions895039
+Node: Basic Concepts897543
+Node: Basic High Level898300
+Ref: Basic High Level-Footnote-1902335
+Node: Basic Data Typing902520
+Node: Floating Point Issues907045
+Node: String Conversion Precision908128
+Ref: String Conversion Precision-Footnote-1909828
+Node: Unexpected Results909937
+Node: POSIX Floating Point Problems911763
+Ref: POSIX Floating Point Problems-Footnote-1915468
+Node: Glossary915506
+Node: Copying940482
+Node: GNU Free Documentation License978039
+Node: Index1003176
 
 End Tag Table
diff --git a/doc/gawk.texi b/doc/gawk.texi
index b2044649..1ff75230 100644
--- a/doc/gawk.texi
+++ b/doc/gawk.texi
@@ -27095,9 +27095,7 @@ But outside those locales, the ordering was defined to be based on
 In many locales, @samp{A} and @samp{a} are both less than @samp{B}.
 In other words, these locales sort characters in dictionary order,
 and @samp{[a-dx-z]} is typically not equivalent to @samp{[abcdxyz]};
-instead it might be equivalent to @samp{[aBbCcDdXxYyZz]}, for example.
-(And to make things worse, on other systems, it might be equivalent to
-@samp{[aAbBcCdDxXyYz]}.)
+instead it might be equivalent to @samp{[ABCXYabcdxyz]}, for example.
 
 This point needs to be emphasized: Much literature teaches that you should
 use @samp{[a-z]} to match a lowercase character.  But on systems with
diff --git a/test/ChangeLog b/test/ChangeLog
index 219396cf..75eb5ec9 100644
--- a/test/ChangeLog
+++ b/test/ChangeLog
@@ -1,3 +1,8 @@
+2012-08-12         Arnold D. Robbins     <arnold@skeeve.com>
+
+	* Makefile.am (regexprange): New test.
+	* regexprange.awk, regexprange.ok: New files.
+
 2012-08-05         Arnold D. Robbins     <arnold@skeeve.com>
 
 	New test from Nelson Beebe.
diff --git a/test/Makefile.am b/test/Makefile.am
index 9c96c386..592cf144 100644
--- a/test/Makefile.am
+++ b/test/Makefile.am
@@ -615,6 +615,8 @@ EXTRA_DIST = \
 	regeq.awk \
 	regeq.in \
 	regeq.ok \
+	regexprange.awk \
+	regexprange.ok \
 	regrange.awk \
 	regrange.ok \
 	regtest.sh \
@@ -823,7 +825,8 @@ BASIC_TESTS = \
 	paramdup paramres paramtyp parse1 parsefld parseme pcntplus \
 	posix2008sub prdupval prec printf0 printf1 prmarscl prmreuse \
 	prt1eval prtoeval \
-	rand range1 rebt8b1 redfilnm regeq regrange reindops reparse \
+	rand range1 rebt8b1 redfilnm regeq regexprange regrange \
+	reindops reparse \
 	resplit rri1 rs rsnul1nl rsnulbig rsnulbig2 rstest1 rstest2 \
 	rstest3 rstest4 rstest5 rswhite \
 	scalar sclforin sclifin sortempty splitargv splitarr splitdef \
diff --git a/test/Makefile.in b/test/Makefile.in
index 56c07305..d76d7387 100644
--- a/test/Makefile.in
+++ b/test/Makefile.in
@@ -820,6 +820,8 @@ EXTRA_DIST = \
 	regeq.awk \
 	regeq.in \
 	regeq.ok \
+	regexprange.awk \
+	regexprange.ok \
 	regrange.awk \
 	regrange.ok \
 	regtest.sh \
@@ -1028,7 +1030,8 @@ BASIC_TESTS = \
 	paramdup paramres paramtyp parse1 parsefld parseme pcntplus \
 	posix2008sub prdupval prec printf0 printf1 prmarscl prmreuse \
 	prt1eval prtoeval \
-	rand range1 rebt8b1 redfilnm regeq regrange reindops reparse \
+	rand range1 rebt8b1 redfilnm regeq regexprange regrange \
+	reindops reparse \
 	resplit rri1 rs rsnul1nl rsnulbig rsnulbig2 rstest1 rstest2 \
 	rstest3 rstest4 rstest5 rswhite \
 	scalar sclforin sclifin sortempty splitargv splitarr splitdef \
@@ -2523,6 +2526,11 @@ regeq:
 	@AWKPATH=$(srcdir) $(AWK) -f $@.awk  < $(srcdir)/$@.in >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
 	@-$(CMP) $(srcdir)/$@.ok _$@ && rm -f _$@
 
+regexprange:
+	@echo regexprange
+	@AWKPATH=$(srcdir) $(AWK) -f $@.awk  >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
+	@-$(CMP) $(srcdir)/$@.ok _$@ && rm -f _$@
+
 regrange:
 	@echo regrange
 	@AWKPATH=$(srcdir) $(AWK) -f $@.awk  >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
diff --git a/test/Maketests b/test/Maketests
index 12d7e35c..df52108d 100644
--- a/test/Maketests
+++ b/test/Maketests
@@ -670,6 +670,11 @@ regeq:
 	@AWKPATH=$(srcdir) $(AWK) -f $@.awk  < $(srcdir)/$@.in >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
 	@-$(CMP) $(srcdir)/$@.ok _$@ && rm -f _$@
 
+regexprange:
+	@echo regexprange
+	@AWKPATH=$(srcdir) $(AWK) -f $@.awk  >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
+	@-$(CMP) $(srcdir)/$@.ok _$@ && rm -f _$@
+
 regrange:
 	@echo regrange
 	@AWKPATH=$(srcdir) $(AWK) -f $@.awk  >_$@ 2>&1 || echo EXIT CODE: $$? >>_$@
diff --git a/test/regexprange.awk b/test/regexprange.awk
new file mode 100644
index 00000000..861e5ee5
--- /dev/null
+++ b/test/regexprange.awk
@@ -0,0 +1,14 @@
+BEGIN {
+	range = "[a-dx-z]"
+
+	split("ABCDEFGHIJKLMNOPQRSTUVWXYZ", upper, "")
+	split("abcdefghijklmnopqrstuvwxyz", lower, "")
+
+	for (i = 1; i in upper; i++)
+		printf("%s ~ %s ---> %s\n",
+			upper[i], range, (upper[i] ~ range) ? "true" : "false")
+
+	for (i = 1; i in lower; i++)
+		printf("%s ~ %s ---> %s\n",
+			lower[i], range, (lower[i] ~ range) ? "true" : "false")
+}
diff --git a/test/regexprange.ok b/test/regexprange.ok
new file mode 100644
index 00000000..dbf5d35a
--- /dev/null
+++ b/test/regexprange.ok
@@ -0,0 +1,52 @@
+A ~ [a-dx-z] ---> false
+B ~ [a-dx-z] ---> false
+C ~ [a-dx-z] ---> false
+D ~ [a-dx-z] ---> false
+E ~ [a-dx-z] ---> false
+F ~ [a-dx-z] ---> false
+G ~ [a-dx-z] ---> false
+H ~ [a-dx-z] ---> false
+I ~ [a-dx-z] ---> false
+J ~ [a-dx-z] ---> false
+K ~ [a-dx-z] ---> false
+L ~ [a-dx-z] ---> false
+M ~ [a-dx-z] ---> false
+N ~ [a-dx-z] ---> false
+O ~ [a-dx-z] ---> false
+P ~ [a-dx-z] ---> false
+Q ~ [a-dx-z] ---> false
+R ~ [a-dx-z] ---> false
+S ~ [a-dx-z] ---> false
+T ~ [a-dx-z] ---> false
+U ~ [a-dx-z] ---> false
+V ~ [a-dx-z] ---> false
+W ~ [a-dx-z] ---> false
+X ~ [a-dx-z] ---> false
+Y ~ [a-dx-z] ---> false
+Z ~ [a-dx-z] ---> false
+a ~ [a-dx-z] ---> true
+b ~ [a-dx-z] ---> true
+c ~ [a-dx-z] ---> true
+d ~ [a-dx-z] ---> true
+e ~ [a-dx-z] ---> false
+f ~ [a-dx-z] ---> false
+g ~ [a-dx-z] ---> false
+h ~ [a-dx-z] ---> false
+i ~ [a-dx-z] ---> false
+j ~ [a-dx-z] ---> false
+k ~ [a-dx-z] ---> false
+l ~ [a-dx-z] ---> false
+m ~ [a-dx-z] ---> false
+n ~ [a-dx-z] ---> false
+o ~ [a-dx-z] ---> false
+p ~ [a-dx-z] ---> false
+q ~ [a-dx-z] ---> false
+r ~ [a-dx-z] ---> false
+s ~ [a-dx-z] ---> false
+t ~ [a-dx-z] ---> false
+u ~ [a-dx-z] ---> false
+v ~ [a-dx-z] ---> false
+w ~ [a-dx-z] ---> false
+x ~ [a-dx-z] ---> true
+y ~ [a-dx-z] ---> true
+z ~ [a-dx-z] ---> true