Fun with cpp word regex

The cpp word regex driver is a bit too loose and can match too much text where the intent is to match only a number. The first patch fixes that.

The other two patches add support for digit separators and the spaceship operator <=> (generalized comparison operator).

I left out support for hexadecimal floating point constants because that would require to tighten the regex even more to avoid that entire expressions are treated as single tokens.

Johannes Sixt (3):
userdiff: tighten cpp word regex
userdiff: permit the digit-separating single-quote in numbers
userdiff: learn the C++ spaceship operator

 userdiff.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

base-commit: 225bc32a989d7a22fa6addafd4ce7dcd04675dbf
Published-As: https://github.com/gitgitgadget/git/releases/tag/pr-1054%2Fj6t%2Ffun-with-cpp-word-regex-v1
Fetch-It-Via: git fetch https://github.com/gitgitgadget/git pr-1054/j6t/fun-with-cpp-word-regex-v1
Pull-Request: https://github.com/gitgitgadget/git/pull/1054

gitgitgadget

From: Johannes Sixt <[email protected]>

Generally, word regex can be written such that they match tokens liberally and need not model the actual syntax because it can be assumed that the regex will only be applied to syntactically correct text.

The regex for cpp (C/C++) is too liberal, though. It regards these sequences as single tokens:

1+2
1.5-e+2+f

and the following amalgams as one token:

.l as in str.length
.f as in str.find

Tighten the regex in the following way:

- Accept + and - only in one position in the exponent. + and - are no longer regarded as the sign of a number and are treated by the catcher-all that is not visible in the driver's regex.

- Accept a leading decimal point only when it is followed by a digit.

For readability, factor hex- and binary numbers into an own term.

As a drive-by, this fixes that floatingpoint numbers such as 12E5 (with upper-case E) were split into two tokens.

---
 userdiff.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

index d9b2ba752f0..ce2a9230703 100644
@@ -54,8 +54,14 @@ PATTERNS("cpp",
 	 /* functions/methods, variables, and compounds at top level */
 	 "^((::[[:space:]]*)?[A-Za-z_].*)$",
 	 /* -- */
+	 /* identifiers and keywords */
 	 "[a-zA-Z_][a-zA-Z0-9_]*"
-	 "|[-+0-9.e]+[fFlL]?|0[xXbB]?[0-9a-fA-F]+[lLuU]*"
+	 /* decimal and octal integers as well as floatingpoint numbers */
+	 "|[0-9][0-9.]*([Ee][-+]?[0-9]+)?[fFlLuU]*"
+	 /* hexadecimal and binary integers */
+	 "|0[xXbB][0-9a-fA-F]+[lLuU]*"
+	 /* floatingpoint numbers that begin with a decimal point */
+	 "|\\.[0-9]+([Ee][-+]?[0-9]+)?[fFlL]?"
 	 "|[-+*/<>%&^|=!]=|--|\\+\\+|<<=?|>>=?|&&|\\|\\||::|->\\*?|\\.\\*"),
 PATTERNS("csharp",
 	 /* Keywords */

gitgitgadget

From: Johannes Sixt <[email protected]>

Since C++17, the single-quote can be used as digit separator:

3.141'592'654
1'000'000
0xdead'beaf

Make it known to the word regex of the cpp driver, so that numbers are not split into separate tokens at the single-quotes.

---
 userdiff.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

index ce2a9230703..1b640c7df79 100644
@@ -57,11 +57,11 @@ PATTERNS("cpp",
 	 /* identifiers and keywords */
 	 "[a-zA-Z_][a-zA-Z0-9_]*"
 	 /* decimal and octal integers as well as floatingpoint numbers */
-	 "|[0-9][0-9.]*([Ee][-+]?[0-9]+)?[fFlLuU]*"
+	 "|[0-9][0-9.']*([Ee][-+]?[0-9]+)?[fFlLuU]*"
 	 /* hexadecimal and binary integers */
-	 "|0[xXbB][0-9a-fA-F]+[lLuU]*"
+	 "|0[xXbB][0-9a-fA-F']+[lLuU]*"
 	 /* floatingpoint numbers that begin with a decimal point */
-	 "|\\.[0-9]+([Ee][-+]?[0-9]+)?[fFlL]?"
+	 "|\\.[0-9']+([Ee][-+]?[0-9]+)?[fFlL]?"
 	 "|[-+*/<>%&^|=!]=|--|\\+\\+|<<=?|>>=?|&&|\\|\\||::|->\\*?|\\.\\*"),
 PATTERNS("csharp",
 	 /* Keywords */

gitgitgadget

From: Johannes Sixt <[email protected]>

Since C++20, the language has a generalized comparison operator. Teach the cpp driver not to separate it into <= and > tokens.

---
 userdiff.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

index 1b640c7df79..13cec0b48db 100644
@@ -62,7 +62,7 @@ PATTERNS("cpp",
 	 "|0[xXbB][0-9a-fA-F']+[lLuU]*"
 	 /* floatingpoint numbers that begin with a decimal point */
 	 "|\\.[0-9']+([Ee][-+]?[0-9]+)?[fFlL]?"
-	 "|[-+*/<>%&^|=!]=|--|\\+\\+|<<=?|>>=?|&&|\\|\\||::|->\\*?|\\.\\*"),
+	 "|[-+*/<>%&^|=!]=|--|\\+\\+|<<=?|>>=?|&&|\\|\\||::|->\\*?|\\.\\*|<=>"),
 PATTERNS("csharp",
 	 /* Keywords */
 	 "!^[ \t]*(do|while|for|if|else|instanceof|new|return|switch|case|throw|catch|using)\n"

gitgitgadget

I haven't dug into the C++ syntax here, I trust you know what you're doing there.

But some tests in t/t4034/cpp/* would be great, and seem from the files easy to add. E.g. wouldn't changing:

    a<b a<=b a>b a>=b

to:

    a<b a<=b a>b a>=b a<=>b

Give you an updated regression test for your 3/3? Similar changes can be done for 1-2/3.

Am 07.10.21 um 11:14 schrieb Ævar Arnfjörð Bjarmason:

Ah, I didn't notice that we actually have tests for word-diff. I'll look into it and resubmit.

-- Hannes

The cpp word regex driver is a bit too loose and can match too much text where the intent is to match only a number.

The first patch makes the cpp word regex tests more effective.

The second patch adds problematic test cases. The third patch fixes these problems.

The final two patches add support for digit separators and the spaceship operator <=> (generalized comparison operator).

I left out support for hexadecimal floating point constants because that would require to tighten the regex even more to avoid that entire expressions are treated as single tokens.

Changes since V1:

* Tests, tests, tests.
* Polished commit messages.

Johannes Sixt (5):
t4034/cpp: actually test that operator tokens are not split
t4034: add tests showing problematic cpp tokenizations
userdiff-cpp: tighten word regex
userdiff-cpp: permit the digit-separating single-quote in numbers
userdiff-cpp: learn the C++ spaceship operator

 t/t4034/cpp/expect | 63 +++++++++++++++++++++++-----------------------
 t/t4034/cpp/post   | 47 +++++++++++++++++++++-------------
 t/t4034/cpp/pre    | 41 +++++++++++++++++++-----------
 userdiff.c         | 10 ++++++--
 4 files changed, 94 insertions(+), 67 deletions(-)

base-commit: 225bc32a989d7a22fa6addafd4ce7dcd04675dbf
Published-As: https://github.com/gitgitgadget/git/releases/tag/pr-1054%2Fj6t%2Ffun-with-cpp-word-regex-v2
Fetch-It-Via: git fetch https://github.com/gitgitgadget/git pr-1054/j6t/fun-with-cpp-word-regex-v2
Pull-Request: https://github.com/gitgitgadget/git/pull/1054

Range-diff vs v1:

 -:  ----------- > 1:  dd9f82ba712 t4034/cpp: actually test that operator tokens are not split
 -:  ----------- > 2:  5a84fc9cf71 t4034: add tests showing problematic cpp tokenizations
 1:  a47ab9ba20e ! 3:  d4ebe45fddc userdiff: tighten cpp word regex

@@ Metadata
Author: Johannes Sixt <[email protected]>

## Commit message ##
- userdiff: tighten cpp word regex
+ userdiff-cpp: tighten word regex

Generally, word regex can be written such that they match tokens liberally and need not model the actual syntax because it can be assumed
@@ Commit message

.l as in str.length
.f as in str.find
+ .e as in str.erase

Tighten the regex in the following way:

@@ Commit message

For readability, factor hex- and binary numbers into an own term.

- As a drive-by, this fixes that floatingpoint numbers such as 12E5
+ As a drive-by, this fixes that floating point numbers such as 12E5
(with upper-case E) were split into two tokens.

+ ## t/t4034/cpp/expect ##
+@@
+ <BOLD>--- a/pre<RESET>
+ <BOLD>+++ b/post<RESET>
+ <CYAN>@@ -1,30 +1,30 @@<RESET>
+-Foo() : x(0<RED>&&1<RESET><GREEN>&42<RESET>) { <RED>foo0<RESET><GREEN>bar<RESET>(x<RED>.f<RESET><GREEN>.F<RESET>ind); }
++Foo() : x(0<RED>&&1<RESET><GREEN>&42<RESET>) { <RED>foo0<RESET><GREEN>bar<RESET>(x.<RED>find<RESET><GREEN>Find<RESET>); }
+ cout<<"Hello World<RED>!<RESET><GREEN>?<RESET>\n"<<endl;
+-<GREEN>(<RESET>1 <RED>-1e10<RESET><GREEN>+1e10<RESET> 0xabcdef<GREEN>)<RESET> '<RED>x<RESET><GREEN>y<RESET>'
++<GREEN>(<RESET>1 <RED>-<RESET><GREEN>+<RESET>1e10 0xabcdef<GREEN>)<RESET> '<RED>x<RESET><GREEN>y<RESET>'
+ // long double<RESET>
+ <RED>3.141592653e-10l<RESET><GREEN>3.141592654e+10l<RESET>
+ // float<RESET>
+-120<RED>E5f<RESET><GREEN>E6f<RESET>
++<RED>120E5f<RESET><GREEN>120E6f<RESET>
+ // hex<RESET>
+-<RED>0xdeadbeaf+8<RESET><GREEN>0xdeadBeaf+7<RESET>ULL
++<RED>0xdeadbeaf<RESET><GREEN>0xdeadBeaf<RESET>+<RED>8ULL<RESET><GREEN>7ULL<RESET>
+ // octal<RESET>
+ <RED>01234567<RESET><GREEN>01234560<RESET>
+ // binary<RESET>
+ <RED>0b1000<RESET><GREEN>0b1100<RESET>+e1
+ // expression<RESET>
+-<RED>1.5-e+2+f<RESET><GREEN>1.5-e+3+f<RESET>
++1.5-e+<RED>2<RESET><GREEN>3<RESET>+f
+ // another one<RESET>
+-str<RED>.e+65<RESET><GREEN>.e+75<RESET>
+-[a] b<RED>-><RESET><GREEN>->*<RESET>v d<RED>.e<RESET><GREEN>.*e<RESET>
++str.e+<RED>65<RESET><GREEN>75<RESET>
++[a] b<RED>-><RESET><GREEN>->*<RESET>v d<RED>.<RESET><GREEN>.*<RESET>e
+ <GREEN>~<RESET>!a <GREEN>!<RESET>~b c<RED>++<RESET><GREEN>+<RESET> d<RED>--<RESET><GREEN>-<RESET> e*<GREEN>*<RESET>f g<RED>&<RESET><GREEN>&&<RESET>h
+ a<RED>*<RESET><GREEN>*=<RESET>b c<RED>/<RESET><GREEN>/=<RESET>d e<RED>%<RESET><GREEN>%=<RESET>f
+ a<RED>+<RESET><GREEN>++<RESET>b c<RED>-<RESET><GREEN>--<RESET>d
+@@ t/t4034/cpp/expect: a<RED>==<RESET><GREEN>!=<RESET>b c<RED>!=<RESET><GREEN>=<RESET>d
+ a<RED>^<RESET><GREEN>^=<RESET>b c<RED>|<RESET><GREEN>|=<RESET>d e<RED>&&<RESET><GREEN>&=<RESET>f
+ a<RED>||<RESET><GREEN>|<RESET>b
+ a?<GREEN>:<RESET>b
+-a<RED>=<RESET><GREEN>==<RESET>b c<RED>+=<RESET><GREEN>+<RESET>d <RED>e-=f<RESET><GREEN>e-f<RESET> g<RED>*=<RESET><GREEN>*<RESET>h i<RED>/=<RESET><GREEN>/<RESET>j k<RED>%=<RESET><GREEN>%<RESET>l m<RED><<=<RESET><GREEN><<<RESET>n o<RED>>>=<RESET><GREEN>>><RESET>p q<RED>&=<RESET><GREEN>&<RESET>r s<RED>^=<RESET><GREEN>^<RESET>t u<RED>|=<RESET><GREEN>|<RESET>v ++a<RED>=<RESET><GREEN>==<RESET>b c<RED>+=<RESET><GREEN>+<RESET>d e<RED>-=<RESET><GREEN>-<RESET>f g<RED>*=<RESET><GREEN>*<RESET>h i<RED>/=<RESET><GREEN>/<RESET>j k<RED>%=<RESET><GREEN>%<RESET>l m<RED><<=<RESET><GREEN><<<RESET>n o<RED>>>=<RESET><GREEN>>><RESET>p q<RED>&=<RESET><GREEN>&<RESET>r s<RED>^=<RESET><GREEN>^<RESET>t u<RED>|=<RESET><GREEN>|<RESET>v
+ a,b<RESET>
+ a<RED>::<RESET><GREEN>:<RESET>b
+
## userdiff.c ##
@@ userdiff.c: PATTERNS("cpp",
/* functions/methods, variables, and compounds at top level */
2: 9d1c05f5f41 ! 4: dd75d19cee9 userdiff: permit the digit-separating single-quote in numbers
@@ Metadata
Author: Johannes Sixt <[email protected]>

## Commit message ##
- userdiff: permit the digit-separating single-quote in numbers
+ userdiff-cpp: permit the digit-separating single-quote in numbers

Since C++17, the single-quote can be used as digit separator:

@@ Commit message
1'000'000
0xdead'beaf

- Make it known to the word regex of the cpp driver, so that numbers are not
- split into separate tokens at the single-quotes.
+ Make it known to the word regex of the cpp driver, so that numbers are
+ not split into separate tokens at the single-quotes.

+ ## t/t4034/cpp/expect ##
+@@
+ <BOLD>diff --git a/pre b/post<RESET>
+-<BOLD>index 1229cdb..3feae6f 100644<RESET>
++<BOLD>index 60f3640..f6fbf7b 100644<RESET>
+ <BOLD>--- a/pre<RESET>
+ <BOLD>+++ b/post<RESET>
+ <CYAN>@@ -1,30 +1,30 @@<RESET>
+@@ t/t4034/cpp/expect: Foo() : x(0<RED>&&1<RESET><GREEN>&42<RESET>) { <RED>foo0<RESET><GREEN>bar<RESET>
+ cout<<"Hello World<RED>!<RESET><GREEN>?<RESET>\n"<<endl;
+ <GREEN>(<RESET>1 <RED>-<RESET><GREEN>+<RESET>1e10 0xabcdef<GREEN>)<RESET> '<RED>x<RESET><GREEN>y<RESET>'
+ // long double<RESET>
+-<RED>3.141592653e-10l<RESET><GREEN>3.141592654e+10l<RESET> ++<RED>3.141'592'653e-10l<RESET><GREEN>3.141'592'654e+10l<RESET>
+ // float<RESET>
+ <RED>120E5f<RESET><GREEN>120E6f<RESET>
+ // hex<RESET> +-<RED>0xdeadbeaf<RESET><GREEN>0xdeadBeaf<RESET>+<RED>8ULL<RESET><GREEN>7ULL<RESET> ++<RED>0xdead'beaf<RESET><GREEN>0xdead'Beaf<RESET>+<RED>8ULL<RESET><GREEN>7ULL<RESET>
+ // octal<RESET>
+-<RED>01234567<RESET><GREEN>01234560<RESET> ++<RED>0123'4567<RESET><GREEN>0123'4560<RESET>
+ // binary<RESET>
+-<RED>0b1000<RESET><GREEN>0b1100<RESET>+e1 ++<RED>0b10'00<RESET><GREEN>0b11'00<RESET>+e1
+ // expression<RESET>
+ 1.5-e+<RED>2<RESET><GREEN>3<RESET>+f
+ // another one<RESET>
+
+ ## t/t4034/cpp/post ##
+@@ t/t4034/cpp/post: Foo() : x(0&42) { bar(x.Find); }
+ cout<<"Hello World?\n"<<endl;
+ (1 +1e10 0xabcdef) 'y'
+ // long double
+-3.141592654e+10l
++3.141'592'654e+10l
+ // float
+ 120E6f
+ // hex
+-0xdeadBeaf+7ULL
++0xdead'Beaf+7ULL
+ // octal
+-01234560
++0123'4560
+ // binary
+-0b1100+e1
++0b11'00+e1
+ // expression
+ 1.5-e+3+f
+ // another one
+
+ ## t/t4034/cpp/pre ##
+@@ t/t4034/cpp/pre: Foo():x(0&&1){ foo0( x.find); }
+ cout<<"Hello World!\n"<<endl;
+ 1 -1e10 0xabcdef 'x'
+ // long double
+-3.141592653e-10l
++3.141'592'653e-10l
+ // float
+ 120E5f
+ // hex
+-0xdeadbeaf+8ULL
++0xdead'beaf+8ULL
+ // octal
+-01234567
++0123'4567
+ // binary
+-0b1000+e1
++0b10'00+e1
+ // expression
+ 1.5-e+2+f
+ // another one
+
## userdiff.c ##
@@ userdiff.c: PATTERNS("cpp",
/* identifiers and keywords */
3: df66485e7f0 ! 5: 43a701f5ffd userdiff: learn the C++ spaceship operator
@@ Metadata
Author: Johannes Sixt <[email protected]>

## Commit message ##
- userdiff: learn the C++ spaceship operator
+ userdiff-cpp: learn the C++ spaceship operator

- Since C++20, the language has a generalized comparison operator. Teach
- the cpp driver not to separate it into <= and > tokens.
+ Since C++20, the language has a generalized comparison operator <=>.
+ Teach the cpp driver not to separate it into <= and > tokens.

+ ## t/t4034/cpp/expect ##
+@@
+ <BOLD>diff --git a/pre b/post<RESET>
+-<BOLD>index 60f3640..f6fbf7b 100644<RESET>
++<BOLD>index 144cd98..244f79c 100644<RESET>
+ <BOLD>--- a/pre<RESET>
+ <BOLD>+++ b/post<RESET>
+ <CYAN>@@ -1,30 +1,30 @@<RESET>
+@@ t/t4034/cpp/expect: str.e+<RED>65<RESET><GREEN>75<RESET>
+ a<RED>*<RESET><GREEN>*=<RESET>b c<RED>/<RESET><GREEN>/=<RESET>d e<RED>%<RESET><GREEN>%=<RESET>f
+ a<RED>+<RESET><GREEN>++<RESET>b c<RED>-<RESET><GREEN>--<RESET>d
+ a<RED><<<RESET><GREEN><<=<RESET>b c<RED>>><RESET><GREEN>>>=<RESET>d
+-a<RED><<RESET><GREEN><=<RESET>b c<RED><=<RESET><GREEN><<RESET>d e<RED>><RESET><GREEN>>=<RESET>f g<RED>>=<RESET><GREEN>><RESET>h ++a<RED><<RESET><GREEN><=<RESET>b c<RED><=<RESET><GREEN><<RESET>d e<RED>><RESET><GREEN>>=<RESET>f g<RED>>=<RESET><GREEN>><RESET>h i<RED><=<RESET><GREEN><=><RESET>j
+ a<RED>==<RESET><GREEN>!=<RESET>b c<RED>!=<RESET><GREEN>=<RESET>d
+ a<RED>^<RESET><GREEN>^=<RESET>b c<RED>|<RESET><GREEN>|=<RESET>d e<RED>&&<RESET><GREEN>&=<RESET>f
+ a<RED>||<RESET><GREEN>|<RESET>b
+
+ ## t/t4034/cpp/post ##
+@@ t/t4034/cpp/post: str.e+75
+ a*=b c/=d e%=f
+ a++b c--d
+ a<<=b c>>=d
+-a<=b c<d e>=f g>h
++a<=b c<d e>=f g>h i<=>j
+ a!=b c=d
+ a^=b c|=d e&=f
+ a|b
+
+ ## t/t4034/cpp/pre ##
+@@ t/t4034/cpp/pre: str.e+65
+ a*b c/d e%f
+ a+b c-d
+ a<<b c>>d
+-a<b c<=d e>f g>=h
++a<b c<=d e>f g>=h i<=j
+ a==b c!=d
+ a^b c|d e&&f
+ a||b
+
## userdiff.c ##
@@ userdiff.c: PATTERNS("cpp",
"|0[xXbB][0-9a-fA-F']+[lLuU]*"

gitgitgadget

From: Johannes Sixt <[email protected]>

8d96e7288f2b (t4034: bulk verify builtin word regex sanity, 2010-12-18) added many tests with the intent to verify that operators consisting of more than one symbol are kept together. These are tested by probing a transition from, e.g., a!=b to x!=y, which results in the word-diff

[-a-]{+x+}!=[-b-]{+y+}

But that proves only that the letters and operators are separate tokens. To prove that != is an unseparable token, we have to probe a transition from, e.g., a=b to a!=b having a word-diff

a[-=-]{+!=+}b

that proves that the ! is not separate from the =.

In the post-image, add to or remove from operators a character that turns it into another valid operator.

Change the identifiers used around operators such that the diff algorithm does not have an incentive to match, e.g., a<b in one spot in the pre-image with a<b elsewhere in the post-image.

Adjust the expected output to match the new differences. Notice that there are some undesirable tokenizations around e, ., and -. This will be addressed in a later change.

---
 t/t4034/cpp/expect | 45 +++++++++++++++------------------------------
 t/t4034/cpp/post   | 29 +++++++++++++----------------
 t/t4034/cpp/pre    | 25 +++++++++++--------------
 3 files changed, 39 insertions(+), 60 deletions(-)

index 37d1ea25870..41976971b93 100644
@@ -1,36 +1,21 @@
 <BOLD>diff --git a/pre b/post<RESET>
-<BOLD>index 23d5c8a..7e8c026 100644<RESET>
+<BOLD>index c5672a2..4229868 100644<RESET>
 <BOLD>--- a/pre<RESET>
 <BOLD>+++ b/post<RESET>
-<CYAN>@@ -1,19 +1,19 @@<RESET>
+<CYAN>@@ -1,16 +1,16 @@<RESET>
 Foo() : x(0<RED>&&1<RESET><GREEN>&42<RESET>) { <GREEN>bar(x);<RESET> }
 cout<<"Hello World<RED>!<RESET><GREEN>?<RESET>\n"<<endl;
 <GREEN>(<RESET>1<GREEN>) (<RESET>-1e10<GREEN>) (<RESET>0xabcdef<GREEN>)<RESET> '<RED>x<RESET><GREEN>y<RESET>'
-[<RED>a<RESET><GREEN>x<RESET>] <RED>a<RESET><GREEN>x<RESET>-><RED>b a<RESET><GREEN>y x<RESET>.<RED>b<RESET><GREEN>y<RESET>
-!<RED>a<RESET><GREEN>x<RESET> ~<RED>a a<RESET><GREEN>x x<RESET>++ <RED>a<RESET><GREEN>x<RESET>-- <RED>a<RESET><GREEN>x<RESET>*<RED>b a<RESET><GREEN>y x<RESET>&<RED>b<RESET>
-<RED>a<RESET><GREEN>y<RESET>
-<GREEN>x<RESET>*<RED>b a<RESET><GREEN>y x<RESET>/<RED>b a<RESET><GREEN>y x<RESET>%<RED>b<RESET>
-<RED>a<RESET><GREEN>y<RESET>
-<GREEN>x<RESET>+<RED>b a<RESET><GREEN>y x<RESET>-<RED>b<RESET>
-<RED>a<RESET><GREEN>y<RESET>
-<GREEN>x<RESET><<<RED>b a<RESET><GREEN>y x<RESET>>><RED>b<RESET>
-<RED>a<RESET><GREEN>y<RESET>
-<GREEN>x<RESET><<RED>b a<RESET><GREEN>y x<RESET><=<RED>b a<RESET><GREEN>y x<RESET>><RED>b a<RESET><GREEN>y x<RESET>>=<RED>b<RESET>
-<RED>a<RESET><GREEN>y<RESET>
-<GREEN>x<RESET>==<RED>b a<RESET><GREEN>y x<RESET>!=<RED>b<RESET>
-<RED>a<RESET><GREEN>y<RESET>
-<GREEN>x<RESET>&<RED>b<RESET>
-<RED>a<RESET><GREEN>y<RESET>
-<GREEN>x<RESET>^<RED>b<RESET>
-<RED>a<RESET><GREEN>y<RESET>
-<GREEN>x<RESET>|<RED>b<RESET>
-<RED>a<RESET><GREEN>y<RESET>
-<GREEN>x<RESET>&&<RED>b<RESET>
-<RED>a<RESET><GREEN>y<RESET>
-<GREEN>x<RESET>||<RED>b<RESET>
-<RED>a<RESET><GREEN>y<RESET>
-<GREEN>x<RESET>?<RED>b<RESET><GREEN>y<RESET>:z
-<RED>a<RESET><GREEN>x<RESET>=<RED>b a<RESET><GREEN>y x<RESET>+=<RED>b a<RESET><GREEN>y x<RESET>-=<RED>b a<RESET><GREEN>y x<RESET>*=<RED>b a<RESET><GREEN>y x<RESET>/=<RED>b a<RESET><GREEN>y x<RESET>%=<RED>b a<RESET><GREEN>y x<RESET><<=<RED>b a<RESET><GREEN>y x<RESET>>>=<RED>b a<RESET><GREEN>y x<RESET>&=<RED>b a<RESET><GREEN>y x<RESET>^=<RED>b a<RESET><GREEN>y x<RESET>|=<RED>b<RESET>
-<RED>a<RESET><GREEN>y<RESET>
-<GREEN>x<RESET>,y
-<RED>a<RESET><GREEN>x<RESET>::<RED>b<RESET><GREEN>y<RESET>
+[a] b<RED>-><RESET><GREEN>->*<RESET>v d<RED>.e<RESET><GREEN>.*e<RESET>
+<GREEN>~<RESET>!a <GREEN>!<RESET>~b c<RED>++<RESET><GREEN>+<RESET> d<RED>--<RESET><GREEN>-<RESET> e*<GREEN>*<RESET>f g<RED>&<RESET><GREEN>&&<RESET>h
+a<RED>*<RESET><GREEN>*=<RESET>b c<RED>/<RESET><GREEN>/=<RESET>d e<RED>%<RESET><GREEN>%=<RESET>f
+a<RED>+<RESET><GREEN>++<RESET>b c<RED>-<RESET><GREEN>--<RESET>d
+a<RED><<<RESET><GREEN><<=<RESET>b c<RED>>><RESET><GREEN>>>=<RESET>d
+a<RED><<RESET><GREEN><=<RESET>b c<RED><=<RESET><GREEN><<RESET>d e<RED>><RESET><GREEN>>=<RESET>f g<RED>>=<RESET><GREEN>><RESET>h
+a<RED>==<RESET><GREEN>!=<RESET>b c<RED>!=<RESET><GREEN>=<RESET>d
+a<RED>^<RESET><GREEN>^=<RESET>b c<RED>|<RESET><GREEN>|=<RESET>d e<RED>&&<RESET><GREEN>&=<RESET>f
+a<RED>||<RESET><GREEN>|<RESET>b
+a?<GREEN>:<RESET>b
+a<RED>=<RESET><GREEN>==<RESET>b c<RED>+=<RESET><GREEN>+<RESET>d <RED>e-=f<RESET><GREEN>e-f<RESET> g<RED>*=<RESET><GREEN>*<RESET>h i<RED>/=<RESET><GREEN>/<RESET>j k<RED>%=<RESET><GREEN>%<RESET>l m<RED><<=<RESET><GREEN><<<RESET>n o<RED>>>=<RESET><GREEN>>><RESET>p q<RED>&=<RESET><GREEN>&<RESET>r s<RED>^=<RESET><GREEN>^<RESET>t u<RED>|=<RESET><GREEN>|<RESET>v
+a,b<RESET>
+a<RED>::<RESET><GREEN>:<RESET>b

index 7e8c026cefb..4229868ae62 100644
@@ -1,19 +1,16 @@
 Foo() : x(0&42) { bar(x); }
 cout<<"Hello World?\n"<<endl;
 (1) (-1e10) (0xabcdef) 'y'
-[x] x->y x.y
-!x ~x x++ x-- x*y x&y
-x*y x/y x%y
-x+y x-y
-x<<y x>>y
-x<y x<=y x>y x>=y
-x==y x!=y
-x&y
-x^y
-x|y
-x&&y
-x||y
-x?y:z
-x=y x+=y x-=y x*=y x/=y x%=y x<<=y x>>=y x&=y x^=y x|=y
-x,y
-x::y
+[a] b->*v d.*e
+~!a !~b c+ d- e**f g&&h
+a*=b c/=d e%=f
+a++b c--d
+a<<=b c>>=d
+a<=b c<d e>=f g>h
+a!=b c=d
+a^=b c|=d e&=f
+a|b
+a?:b
+a==b c+d e-f g*h i/j k%l m<<n o>>p q&r s^t u|v
+a,b
+a:b

index 23d5c8adf54..c5672a24cfc 100644
@@ -1,19 +1,16 @@
 Foo():x(0&&1){}
 cout<<"Hello World!\n"<<endl;
 1 -1e10 0xabcdef 'x'
-[a] a->b a.b
-!a ~a a++ a-- a*b a&b
-a*b a/b a%b
-a+b a-b
-a<<b a>>b
-a<b a<=b a>b a>=b
-a==b a!=b
-a&b
-a^b
-a|b
-a&&b
+[a] b->v d.e
+!a ~b c++ d-- e*f g&h
+a*b c/d e%f
+a+b c-d
+a<<b c>>d
+a<b c<=d e>f g>=h
+a==b c!=d
+a^b c|d e&&f
 a||b
-a?b:z
-a=b a+=b a-=b a*=b a/=b a%=b a<<=b a>>=b a&=b a^=b a|=b
-a,y
+a?b
+a=b c+=d e-=f g*=h i/=j k%=l m<<=n o>>=p q&=r s^=t u|=v
+a,b
 a::b

gitgitgadget

From: Johannes Sixt <[email protected]>

The word regex is too loose and matches long streaks of characters that should actually be separate tokens. Add these problematic test cases. Separate the lines with text that will remain identical in the pre- and post-image so that the diff algorithm will not lump removals and additions of consecutive lines together. This makes the expected output easier to read.

---
 t/t4034/cpp/expect | 22 ++++++++++++++++++----
 t/t4034/cpp/post   | 18 ++++++++++++++++--
 t/t4034/cpp/pre    | 16 +++++++++++++++-
 3 files changed, 49 insertions(+), 7 deletions(-)

index 41976971b93..63e53a61e62 100644
@@ -1,11 +1,25 @@
 <BOLD>diff --git a/pre b/post<RESET>
-<BOLD>index c5672a2..4229868 100644<RESET>
+<BOLD>index 1229cdb..3feae6f 100644<RESET>
 <BOLD>--- a/pre<RESET>
 <BOLD>+++ b/post<RESET>
-<CYAN>@@ -1,16 +1,16 @@<RESET>
-Foo() : x(0<RED>&&1<RESET><GREEN>&42<RESET>) { <GREEN>bar(x);<RESET> }
+<CYAN>@@ -1,30 +1,30 @@<RESET>
+Foo() : x(0<RED>&&1<RESET><GREEN>&42<RESET>) { <RED>foo0<RESET><GREEN>bar<RESET>(x<RED>.f<RESET><GREEN>.F<RESET>ind); }
 cout<<"Hello World<RED>!<RESET><GREEN>?<RESET>\n"<<endl;
-<GREEN>(<RESET>1<GREEN>) (<RESET>-1e10<GREEN>) (<RESET>0xabcdef<GREEN>)<RESET> '<RED>x<RESET><GREEN>y<RESET>'
+<GREEN>(<RESET>1 <RED>-1e10<RESET><GREEN>+1e10<RESET> 0xabcdef<GREEN>)<RESET> '<RED>x<RESET><GREEN>y<RESET>'
+// long double<RESET>
+<RED>3.141592653e-10l<RESET><GREEN>3.141592654e+10l<RESET>
+// float<RESET>
+120<RED>E5f<RESET><GREEN>E6f<RESET>
+// hex<RESET>
+<RED>0xdeadbeaf+8<RESET><GREEN>0xdeadBeaf+7<RESET>ULL
+// octal<RESET>
+<RED>01234567<RESET><GREEN>01234560<RESET>
+// binary<RESET>
+<RED>0b1000<RESET><GREEN>0b1100<RESET>+e1
+// expression<RESET>
+<RED>1.5-e+2+f<RESET><GREEN>1.5-e+3+f<RESET>
+// another one<RESET>
+str<RED>.e+65<RESET><GREEN>.e+75<RESET>
 [a] b<RED>-><RESET><GREEN>->*<RESET>v d<RED>.e<RESET><GREEN>.*e<RESET>
 <GREEN>~<RESET>!a <GREEN>!<RESET>~b c<RED>++<RESET><GREEN>+<RESET> d<RED>--<RESET><GREEN>-<RESET> e*<GREEN>*<RESET>f g<RED>&<RESET><GREEN>&&<RESET>h
 a<RED>*<RESET><GREEN>*=<RESET>b c<RED>/<RESET><GREEN>/=<RESET>d e<RED>%<RESET><GREEN>%=<RESET>f

index 4229868ae62..3feae6f430f 100644
@@ -1,6 +1,20 @@
-Foo() : x(0&42) { bar(x); }
+Foo() : x(0&42) { bar(x.Find); }
 cout<<"Hello World?\n"<<endl;
-(1) (-1e10) (0xabcdef) 'y'
+(1 +1e10 0xabcdef) 'y'
+// long double
+3.141592654e+10l
+// float
+120E6f
+// hex
+0xdeadBeaf+7ULL
+// octal
+01234560
+// binary
+0b1100+e1
+// expression
+1.5-e+3+f
+// another one
+str.e+75
 [a] b->*v d.*e
 ~!a !~b c+ d- e**f g&&h
 a*=b c/=d e%=f

index c5672a24cfc..1229cdb59d1 100644
@@ -1,6 +1,20 @@
-Foo():x(0&&1){}
+Foo():x(0&&1){ foo0( x.find); }
 cout<<"Hello World!\n"<<endl;
 1 -1e10 0xabcdef 'x'
+// long double
+3.141592653e-10l
+// float
+120E5f
+// hex
+0xdeadbeaf+8ULL
+// octal
+01234567
+// binary
+0b1000+e1
+// expression
+1.5-e+2+f
+// another one
+str.e+65
 [a] b->v d.e
 !a ~b c++ d-- e*f g&h
 a*b c/d e%f

gitgitgadget

From: Johannes Sixt <[email protected]>

Generally, word regex can be written such that they match tokens liberally and need not model the actual syntax because it can be assumed that the regex will only be applied to syntactically correct text.

The regex for cpp (C/C++) is too liberal, though. It regards these sequences as single tokens:

1+2
1.5-e+2+f

and the following amalgams as one token:

   .l      as in str.length
   .f      as in str.find
   .e      as in str.erase

Tighten the regex in the following way:

- Accept + and - only in one position in the exponent. + and - are no longer regarded as the sign of a number and are treated by the catcher-all that is not visible in the driver's regex.

- Accept a leading decimal point only when it is followed by a digit.

For readability, factor hex- and binary numbers into an own term.

As a drive-by, this fixes that floating point numbers such as 12E5 (with upper-case E) were split into two tokens.

---
 t/t4034/cpp/expect | 16 ++++++++--------
 userdiff.c         |  8 +++++++-
 2 files changed, 15 insertions(+), 9 deletions(-)

index 63e53a61e62..46c9460a968 100644
@@ -3,24 +3,24 @@
 <BOLD>--- a/pre<RESET>
 <BOLD>+++ b/post<RESET>
 <CYAN>@@ -1,30 +1,30 @@<RESET>
-Foo() : x(0<RED>&&1<RESET><GREEN>&42<RESET>) { <RED>foo0<RESET><GREEN>bar<RESET>(x<RED>.f<RESET><GREEN>.F<RESET>ind); }
+Foo() : x(0<RED>&&1<RESET><GREEN>&42<RESET>) { <RED>foo0<RESET><GREEN>bar<RESET>(x.<RED>find<RESET><GREEN>Find<RESET>); }
 cout<<"Hello World<RED>!<RESET><GREEN>?<RESET>\n"<<endl;
-<GREEN>(<RESET>1 <RED>-1e10<RESET><GREEN>+1e10<RESET> 0xabcdef<GREEN>)<RESET> '<RED>x<RESET><GREEN>y<RESET>'
+<GREEN>(<RESET>1 <RED>-<RESET><GREEN>+<RESET>1e10 0xabcdef<GREEN>)<RESET> '<RED>x<RESET><GREEN>y<RESET>'
 // long double<RESET>
 <RED>3.141592653e-10l<RESET><GREEN>3.141592654e+10l<RESET>
 // float<RESET>
-120<RED>E5f<RESET><GREEN>E6f<RESET>
+<RED>120E5f<RESET><GREEN>120E6f<RESET>
 // hex<RESET>
-<RED>0xdeadbeaf+8<RESET><GREEN>0xdeadBeaf+7<RESET>ULL
+<RED>0xdeadbeaf<RESET><GREEN>0xdeadBeaf<RESET>+<RED>8ULL<RESET><GREEN>7ULL<RESET>
 // octal<RESET>
 <RED>01234567<RESET><GREEN>01234560<RESET>
 // binary<RESET>
 <RED>0b1000<RESET><GREEN>0b1100<RESET>+e1
 // expression<RESET>
-<RED>1.5-e+2+f<RESET><GREEN>1.5-e+3+f<RESET>
+1.5-e+<RED>2<RESET><GREEN>3<RESET>+f
 // another one<RESET>
-str<RED>.e+65<RESET><GREEN>.e+75<RESET>
-[a] b<RED>-><RESET><GREEN>->*<RESET>v d<RED>.e<RESET><GREEN>.*e<RESET>
+str.e+<RED>65<RESET><GREEN>75<RESET>
+[a] b<RED>-><RESET><GREEN>->*<RESET>v d<RED>.<RESET><GREEN>.*<RESET>e
 <GREEN>~<RESET>!a <GREEN>!<RESET>~b c<RED>++<RESET><GREEN>+<RESET> d<RED>--<RESET><GREEN>-<RESET> e*<GREEN>*<RESET>f g<RED>&<RESET><GREEN>&&<RESET>h
 a<RED>*<RESET><GREEN>*=<RESET>b c<RED>/<RESET><GREEN>/=<RESET>d e<RED>%<RESET><GREEN>%=<RESET>f
 a<RED>+<RESET><GREEN>++<RESET>b c<RED>-<RESET><GREEN>--<RESET>d

@@ -30,6 +30,6 @@ a<RED>==<RESET><GREEN>!=<RESET>b c<RED>!=<RESET><GREEN>=<RESET>d
 a<RED>^<RESET><GREEN>^=<RESET>b c<RED>|<RESET><GREEN>|=<RESET>d e<RED>&&<RESET><GREEN>&=<RESET>f
 a<RED>||<RESET><GREEN>|<RESET>b
 a?<GREEN>:<RESET>b
-a<RED>=<RESET><GREEN>==<RESET>b c<RED>+=<RESET><GREEN>+<RESET>d <RED>e-=f<RESET><GREEN>e-f<RESET> g<RED>*=<RESET><GREEN>*<RESET>h i<RED>/=<RESET><GREEN>/<RESET>j k<RED>%=<RESET><GREEN>%<RESET>l m<RED><<=<RESET><GREEN><<<RESET>n o<RED>>>=<RESET><GREEN>>><RESET>p q<RED>&=<RESET><GREEN>&<RESET>r s<RED>^=<RESET><GREEN>^<RESET>t u<RED>|=<RESET><GREEN>|<RESET>v
+a<RED>=<RESET><GREEN>==<RESET>b c<RED>+=<RESET><GREEN>+<RESET>d e<RED>-=<RESET><GREEN>-<RESET>f g<RED>*=<RESET><GREEN>*<RESET>h i<RED>/=<RESET><GREEN>/<RESET>j k<RED>%=<RESET><GREEN>%<RESET>l m<RED><<=<RESET><GREEN><<<RESET>n o<RED>>>=<RESET><GREEN>>><RESET>p q<RED>&=<RESET><GREEN>&<RESET>r s<RED>^=<RESET><GREEN>^<RESET>t u<RED>|=<RESET><GREEN>|<RESET>v
 a,b<RESET>
 a<RED>::<RESET><GREEN>:<RESET>b

index d9b2ba752f0..ce2a9230703 100644
@@ -54,8 +54,14 @@ PATTERNS("cpp",
 	 /* functions/methods, variables, and compounds at top level */
 	 "^((::[[:space:]]*)?[A-Za-z_].*)$",
 	 /* -- */
+	 /* identifiers and keywords */
 	 "[a-zA-Z_][a-zA-Z0-9_]*"
-	 "|[-+0-9.e]+[fFlL]?|0[xXbB]?[0-9a-fA-F]+[lLuU]*"
+	 /* decimal and octal integers as well as floatingpoint numbers */
+	 "|[0-9][0-9.]*([Ee][-+]?[0-9]+)?[fFlLuU]*"
+	 /* hexadecimal and binary integers */
+	 "|0[xXbB][0-9a-fA-F]+[lLuU]*"
+	 /* floatingpoint numbers that begin with a decimal point */
+	 "|\\.[0-9]+([Ee][-+]?[0-9]+)?[fFlL]?"
 	 "|[-+*/<>%&^|=!]=|--|\\+\\+|<<=?|>>=?|&&|\\|\\||::|->\\*?|\\.\\*"),
 PATTERNS("csharp",
 	 /* Keywords */

gitgitgadget

From: Johannes Sixt <[email protected]>

Since C++17, the single-quote can be used as digit separator:

3.141'592'654
1'000'000
0xdead'beaf

Make it known to the word regex of the cpp driver, so that numbers are not split into separate tokens at the single-quotes.

---
 t/t4034/cpp/expect | 10 +++++-----
 t/t4034/cpp/post   |  8 ++++----
 t/t4034/cpp/pre    |  8 ++++----
 userdiff.c         |  6 +++---
 4 files changed, 16 insertions(+), 16 deletions(-)

index 46c9460a968..a3a234f5461 100644
@@ -1,5 +1,5 @@
 <BOLD>diff --git a/pre b/post<RESET>
-<BOLD>index 1229cdb..3feae6f 100644<RESET>
+<BOLD>index 60f3640..f6fbf7b 100644<RESET>
 <BOLD>--- a/pre<RESET>
 <BOLD>+++ b/post<RESET>
 <CYAN>@@ -1,30 +1,30 @@<RESET>

@@ -7,15 +7,15 @@ Foo() : x(0<RED>&&1<RESET><GREEN>&42<RESET>) { <RED>foo0<RESET><GREEN>bar<RESET>
 cout<<"Hello World<RED>!<RESET><GREEN>?<RESET>\n"<<endl;
 <GREEN>(<RESET>1 <RED>-<RESET><GREEN>+<RESET>1e10 0xabcdef<GREEN>)<RESET> '<RED>x<RESET><GREEN>y<RESET>'
 // long double<RESET>
-<RED>3.141592653e-10l<RESET><GREEN>3.141592654e+10l<RESET>
+<RED>3.141'592'653e-10l<RESET><GREEN>3.141'592'654e+10l<RESET>
 // float<RESET>
 <RED>120E5f<RESET><GREEN>120E6f<RESET>
 // hex<RESET>
-<RED>0xdeadbeaf<RESET><GREEN>0xdeadBeaf<RESET>+<RED>8ULL<RESET><GREEN>7ULL<RESET>
+<RED>0xdead'beaf<RESET><GREEN>0xdead'Beaf<RESET>+<RED>8ULL<RESET><GREEN>7ULL<RESET>
 // octal<RESET>
-<RED>01234567<RESET><GREEN>01234560<RESET>
+<RED>0123'4567<RESET><GREEN>0123'4560<RESET>
 // binary<RESET>
-<RED>0b1000<RESET><GREEN>0b1100<RESET>+e1
+<RED>0b10'00<RESET><GREEN>0b11'00<RESET>+e1
 // expression<RESET>
 1.5-e+<RED>2<RESET><GREEN>3<RESET>+f
 // another one<RESET>

index 3feae6f430f..f6fbf7bc04c 100644
@@ -2,15 +2,15 @@ Foo() : x(0&42) { bar(x.Find); }
 cout<<"Hello World?\n"<<endl;
 (1 +1e10 0xabcdef) 'y'
 // long double
-3.141592654e+10l
+3.141'592'654e+10l
 // float
 120E6f
 // hex
-0xdeadBeaf+7ULL
+0xdead'Beaf+7ULL
 // octal
-01234560
+0123'4560
 // binary
-0b1100+e1
+0b11'00+e1
 // expression
 1.5-e+3+f
 // another one

index 1229cdb59d1..60f3640d773 100644
@@ -2,15 +2,15 @@ Foo():x(0&&1){ foo0( x.find); }
 cout<<"Hello World!\n"<<endl;
 1 -1e10 0xabcdef 'x'
 // long double
-3.141592653e-10l
+3.141'592'653e-10l
 // float
 120E5f
 // hex
-0xdeadbeaf+8ULL
+0xdead'beaf+8ULL
 // octal
-01234567
+0123'4567
 // binary
-0b1000+e1
+0b10'00+e1
 // expression
 1.5-e+2+f
 // another one

index ce2a9230703..1b640c7df79 100644
@@ -57,11 +57,11 @@ PATTERNS("cpp",
 	 /* identifiers and keywords */
 	 "[a-zA-Z_][a-zA-Z0-9_]*"
 	 /* decimal and octal integers as well as floatingpoint numbers */
-	 "|[0-9][0-9.]*([Ee][-+]?[0-9]+)?[fFlLuU]*"
+	 "|[0-9][0-9.']*([Ee][-+]?[0-9]+)?[fFlLuU]*"
 	 /* hexadecimal and binary integers */
-	 "|0[xXbB][0-9a-fA-F]+[lLuU]*"
+	 "|0[xXbB][0-9a-fA-F']+[lLuU]*"
 	 /* floatingpoint numbers that begin with a decimal point */
-	 "|\\.[0-9]+([Ee][-+]?[0-9]+)?[fFlL]?"
+	 "|\\.[0-9']+([Ee][-+]?[0-9]+)?[fFlL]?"
 	 "|[-+*/<>%&^|=!]=|--|\\+\\+|<<=?|>>=?|&&|\\|\\||::|->\\*?|\\.\\*"),
 PATTERNS("csharp",
 	 /* Keywords */

gitgitgadget

From: Johannes Sixt <[email protected]>

Since C++20, the language has a generalized comparison operator <=>. Teach the cpp driver not to separate it into <= and > tokens.

---
 t/t4034/cpp/expect | 4 ++--
 t/t4034/cpp/post   | 2 +-
 t/t4034/cpp/pre    | 2 +-
 userdiff.c         | 2 +-
 4 files changed, 5 insertions(+), 5 deletions(-)

index a3a234f5461..bf3cd2abc74 100644
@@ -1,5 +1,5 @@
 <BOLD>diff --git a/pre b/post<RESET>
-<BOLD>index 60f3640..f6fbf7b 100644<RESET>
+<BOLD>index 144cd98..244f79c 100644<RESET>
 <BOLD>--- a/pre<RESET>
 <BOLD>+++ b/post<RESET>
 <CYAN>@@ -1,30 +1,30 @@<RESET>

@@ -25,7 +25,7 @@ str.e+<RED>65<RESET><GREEN>75<RESET>
 a<RED>*<RESET><GREEN>*=<RESET>b c<RED>/<RESET><GREEN>/=<RESET>d e<RED>%<RESET><GREEN>%=<RESET>f
 a<RED>+<RESET><GREEN>++<RESET>b c<RED>-<RESET><GREEN>--<RESET>d
 a<RED><<<RESET><GREEN><<=<RESET>b c<RED>>><RESET><GREEN>>>=<RESET>d
-a<RED><<RESET><GREEN><=<RESET>b c<RED><=<RESET><GREEN><<RESET>d e<RED>><RESET><GREEN>>=<RESET>f g<RED>>=<RESET><GREEN>><RESET>h
+a<RED><<RESET><GREEN><=<RESET>b c<RED><=<RESET><GREEN><<RESET>d e<RED>><RESET><GREEN>>=<RESET>f g<RED>>=<RESET><GREEN>><RESET>h i<RED><=<RESET><GREEN><=><RESET>j
 a<RED>==<RESET><GREEN>!=<RESET>b c<RED>!=<RESET><GREEN>=<RESET>d
 a<RED>^<RESET><GREEN>^=<RESET>b c<RED>|<RESET><GREEN>|=<RESET>d e<RED>&&<RESET><GREEN>&=<RESET>f
 a<RED>||<RESET><GREEN>|<RESET>b

index f6fbf7bc04c..244f79c9900 100644
@@ -20,7 +20,7 @@ str.e+75
 a*=b c/=d e%=f
 a++b c--d
 a<<=b c>>=d
-a<=b c<d e>=f g>h
+a<=b c<d e>=f g>h i<=>j
 a!=b c=d
 a^=b c|=d e&=f
 a|b

index 60f3640d773..144cd980d6b 100644
@@ -20,7 +20,7 @@ str.e+65
 a*b c/d e%f
 a+b c-d
 a<<b c>>d
-a<b c<=d e>f g>=h
+a<b c<=d e>f g>=h i<=j
 a==b c!=d
 a^b c|d e&&f
 a||b

index 1b640c7df79..13cec0b48db 100644
@@ -62,7 +62,7 @@ PATTERNS("cpp",
 	 "|0[xXbB][0-9a-fA-F']+[lLuU]*"
 	 /* floatingpoint numbers that begin with a decimal point */
 	 "|\\.[0-9']+([Ee][-+]?[0-9]+)?[fFlL]?"
-	 "|[-+*/<>%&^|=!]=|--|\\+\\+|<<=?|>>=?|&&|\\|\\||::|->\\*?|\\.\\*"),
+	 "|[-+*/<>%&^|=!]=|--|\\+\\+|<<=?|>>=?|&&|\\|\\||::|->\\*?|\\.\\*|<=>"),
 PATTERNS("csharp",
 	 /* Keywords */
 	 "!^[ \t]*(do|while|for|if|else|instanceof|new|return|switch|case|throw|catch|using)\n"

gitgitgadget

I've read this much improved v2 over, thanks a lot. This looks much better.

Just some general comments, in my mind this is more than good enough already and doesn't require a re-roll, just food for thought:

* I wonder if it isn't time to split up "cpp" into a "c" driver, e.g. git.git's .gitattributes has "cpp" for *.[ch] files, but as C++ adds more syntax sugar.

So e.g. if you use "<=>" after this series we'll tokenize it differently in *.c files, but it's a C++-only operator, on the other hand probably nobody cares that much...

* I found myself back-porting some of your tests (manually mostly), maybe you disagree, but in cases like 123'123, <=> etc. I'd find it easier to follow if we first added the test data, and then the changed behavior.

Because after all, we're going to change how we highlight existing data, so testing for that would be informative.

* This pre-dates your much improved tests, but these test files could really use some test comments, as in:

/* Now that we're going to understand the "'" character somehow, will any of this change? */ /* We haven't written code like this since the 1960's ... */
/* Run & free */

I.e. we don't just highlight code the compiler likes to eat, but also comments. So particularly for smaller tokens that also occur in natural language like "'" and "&" are we getting expected results?

It looked like it from skimming your changes, i.e. the ' rule is checked by surrounding digits, just might be handy to test it ... :)

Am 08.10.21 um 22:07 schrieb Ævar Arnfjörð Bjarmason:

Yes, it is that: <=> won't appear in a correct C file (outside of comments), so no-one will care. As far as tokenization is concerned, C is a subset of C++. I don't think we need to separate the drivers.

Good point. I'll work a bit more on that.

Comments are free text. Anything can happen. There is no such thing as "correct tokenization" in comments. Not interested.

Thank you for the review.
-- Hannes

[re-arranged a bit]

Great!

..mmm..

Sure there is, just because the problem is fuzzy doesn't mean there aren't more and less correct things to do.

But most importantly the output of "git diff" is made for human consumption, people who use --word-diff are going to be looking at code that contains comments, embedded natural language in C-strings etc.

I've got no reason to think that your changes here make it worse, but just as a general matter we absolutely should consider that one of the top priorities when it comes to these language drivers.

E.g. in some languages (like CJK) it's common for characters or words not to have any unicode whitespace between them, so even a word-diff mode for C can benefit from recognizing those character ranges and splitting them appropriately, or at least trying.

So to take a comment of yours where you changed a comment at random:

    $ git log -U0 --oneline -1 --word-diff -p af920e36977 --word-diff-regex='[ a-zA-Z]*'
    [...]
    [- Note that this seemingly redundant second declaration is required-]{+ Note that this redundant forward declaration is required+}

Don't you think that would suck v.s. the now-behavior of:

[...]
Note that this[-seemingly-] redundant [-second-]{+forward+} declaration is required

The former is exactly the sort of thing you'd get in CJK languages with a word-diff driver thought the line we should stop at was the same as a comiler tokenizer.

Anyway, the cpp driver seems to do just fine on CJK.

I'm just saying that as a general thing it's definitely a priority for these drivers to not only handle the narrow cases of tokens a compiler would know about. Text that people commonly use should be presented in some way that isn't line noise.

For an example of something we do a bit badly with the cpp driver is parts of my 66f5f6dca95 (C style: use standard style for "TRANSLATORS" comments, 2017-05-11).

I.e. there I was changing a comment format, and added a full stop to a sentence, the word-diff is:

        /*
         {+*+} TRANSLATORS: here is a comment that explains the string to
         {+*+} be translated, that follows immediately after [-it-]{+it.+}
         */

Even though it has nothing to do with C syntax per-se that would be much more useful as:

        /*
         {+*+} TRANSLATORS: here is a comment that explains the string to
         {+*+} be translated, that follows immediately after it{+.+}
         */

I.e. treating a "." at the end of a word specially isn't C or C++ syntax, but it's absolutely input that the cpp driver *is* getting and should be if possible be handling well.

I just did that by experimenting with --word-diff-regex='([A-Za-z:]*|\*|\.)', that example is unchanged with your series, but maybe low-hanging fruit....

Thanks for working on this, just an unsolicited braindump :)

The cpp word regex driver is a bit too loose and can match too much text where the intent is to match only a number.

The first patch makes the cpp word regex tests more effective.

The second patch adds problematic test cases. The third patch fixes these problems.

The remaining three patches add support for digit separators and the spaceship operator <=> (generalized comparison operator).

I left out support for hexadecimal floating point constants because that would require to tighten the regex even more to avoid that entire expressions are treated as single tokens.

Changes since V2:

* Add test cases for the new features in a separate commit so that the new behavior is better visible.
* Don't treat .' as in '.' as a token.

Changes since V1:

* Tests, tests, tests.
* Polished commit messages.

Johannes Sixt (6):
t4034/cpp: actually test that operator tokens are not split
t4034: add tests showing problematic cpp tokenizations
userdiff-cpp: tighten word regex
userdiff-cpp: prepare test cases with yet unsupported features
userdiff-cpp: permit the digit-separating single-quote in numbers
userdiff-cpp: learn the C++ spaceship operator

 t/t4034/cpp/expect | 63 +++++++++++++++++++++++-----------------------
 t/t4034/cpp/post   | 47 +++++++++++++++++++++-------------
 t/t4034/cpp/pre    | 41 +++++++++++++++++++-----------
 userdiff.c         | 10 ++++++--
 4 files changed, 94 insertions(+), 67 deletions(-)

base-commit: 225bc32a989d7a22fa6addafd4ce7dcd04675dbf
Published-As: https://github.com/gitgitgadget/git/releases/tag/pr-1054%2Fj6t%2Ffun-with-cpp-word-regex-v3
Fetch-It-Via: git fetch https://github.com/gitgitgadget/git pr-1054/j6t/fun-with-cpp-word-regex-v3
Pull-Request: https://github.com/gitgitgadget/git/pull/1054

Range-diff vs v2:

 1:  dd9f82ba712 = 1:  dd9f82ba712 t4034/cpp: actually test that operator tokens are not split
 2:  5a84fc9cf71 = 2:  5a84fc9cf71 t4034: add tests showing problematic cpp tokenizations
 3:  d4ebe45fddc = 3:  d4ebe45fddc userdiff-cpp: tighten word regex
 4:  dd75d19cee9 ! 4:  c9f58b5e82f userdiff-cpp: permit the digit-separating single-quote in numbers
     @@ Metadata
      Author: Johannes Sixt <[email protected]>
      
       ## Commit message ##
     -    userdiff-cpp: permit the digit-separating single-quote in numbers
     +    userdiff-cpp: prepare test cases with yet unsupported features
      
     -    Since C++17, the single-quote can be used as digit separator:
     -
     -       3.141'592'654
     -       1'000'000
     -       0xdead'beaf
     -
     -    Make it known to the word regex of the cpp driver, so that numbers are
     -    not split into separate tokens at the single-quotes.
     +    We are going to add support for C++'s digit-separating single-quote and
     +    the spaceship operator. By adding the test cases in this separate
     +    commit, the effect on the word highlighting will become more obvious
     +    as the features are implemented and the file cpp/expect is updated.
      
          Signed-off-by: Johannes Sixt <[email protected]>
      
     @@ t/t4034/cpp/expect
      @@
       <BOLD>diff --git a/pre b/post<RESET>
      -<BOLD>index 1229cdb..3feae6f 100644<RESET>
     -+<BOLD>index 60f3640..f6fbf7b 100644<RESET>
     ++<BOLD>index 144cd98..64e78af 100644<RESET>
       <BOLD>--- a/pre<RESET>
       <BOLD>+++ b/post<RESET>
       <CYAN>@@ -1,30 +1,30 @@<RESET>
     -@@ t/t4034/cpp/expect: Foo() : x(0<RED>&&1<RESET><GREEN>&42<RESET>) { <RED>foo0<RESET><GREEN>bar<RESET>
     + Foo() : x(0<RED>&&1<RESET><GREEN>&42<RESET>) { <RED>foo0<RESET><GREEN>bar<RESET>(x.<RED>find<RESET><GREEN>Find<RESET>); }
       cout<<"Hello World<RED>!<RESET><GREEN>?<RESET>\n"<<endl;
     - <GREEN>(<RESET>1 <RED>-<RESET><GREEN>+<RESET>1e10 0xabcdef<GREEN>)<RESET> '<RED>x<RESET><GREEN>y<RESET>'
     +-<GREEN>(<RESET>1 <RED>-<RESET><GREEN>+<RESET>1e10 0xabcdef<GREEN>)<RESET> '<RED>x<RESET><GREEN>y<RESET>'
     ++<GREEN>(<RESET>1 <RED>-<RESET><GREEN>+<RESET>1e10 0xabcdef<GREEN>)<RESET> '<RED>x<RESET><GREEN>.<RESET>'
       // long double<RESET>
      -<RED>3.141592653e-10l<RESET><GREEN>3.141592654e+10l<RESET>
     -+<RED>3.141'592'653e-10l<RESET><GREEN>3.141'592'654e+10l<RESET>
     ++3.141'592'<RED>653e-10l<RESET><GREEN>654e+10l<RESET>
       // float<RESET>
       <RED>120E5f<RESET><GREEN>120E6f<RESET>
       // hex<RESET>
      -<RED>0xdeadbeaf<RESET><GREEN>0xdeadBeaf<RESET>+<RED>8ULL<RESET><GREEN>7ULL<RESET>
     -+<RED>0xdead'beaf<RESET><GREEN>0xdead'Beaf<RESET>+<RED>8ULL<RESET><GREEN>7ULL<RESET>
     ++0xdead'<RED>beaf<RESET><GREEN>Beaf<RESET>+<RED>8ULL<RESET><GREEN>7ULL<RESET>
       // octal<RESET>
      -<RED>01234567<RESET><GREEN>01234560<RESET>
     -+<RED>0123'4567<RESET><GREEN>0123'4560<RESET>
     ++0123'<RED>4567<RESET><GREEN>4560<RESET>
       // binary<RESET>
      -<RED>0b1000<RESET><GREEN>0b1100<RESET>+e1
     -+<RED>0b10'00<RESET><GREEN>0b11'00<RESET>+e1
     ++<RED>0b10<RESET><GREEN>0b11<RESET>'00+e1
       // expression<RESET>
       1.5-e+<RED>2<RESET><GREEN>3<RESET>+f
       // another one<RESET>
     +@@ t/t4034/cpp/expect: str.e+<RED>65<RESET><GREEN>75<RESET>
     + a<RED>*<RESET><GREEN>*=<RESET>b c<RED>/<RESET><GREEN>/=<RESET>d e<RED>%<RESET><GREEN>%=<RESET>f
     + a<RED>+<RESET><GREEN>++<RESET>b c<RED>-<RESET><GREEN>--<RESET>d
     + a<RED><<<RESET><GREEN><<=<RESET>b c<RED>>><RESET><GREEN>>>=<RESET>d
     +-a<RED><<RESET><GREEN><=<RESET>b c<RED><=<RESET><GREEN><<RESET>d e<RED>><RESET><GREEN>>=<RESET>f g<RED>>=<RESET><GREEN>><RESET>h
     ++a<RED><<RESET><GREEN><=<RESET>b c<RED><=<RESET><GREEN><<RESET>d e<RED>><RESET><GREEN>>=<RESET>f g<RED>>=<RESET><GREEN>><RESET>h i<=<GREEN>><RESET>j
     + a<RED>==<RESET><GREEN>!=<RESET>b c<RED>!=<RESET><GREEN>=<RESET>d
     + a<RED>^<RESET><GREEN>^=<RESET>b c<RED>|<RESET><GREEN>|=<RESET>d e<RED>&&<RESET><GREEN>&=<RESET>f
     + a<RED>||<RESET><GREEN>|<RESET>b
      
       ## t/t4034/cpp/post ##
     -@@ t/t4034/cpp/post: Foo() : x(0&42) { bar(x.Find); }
     +@@
     + Foo() : x(0&42) { bar(x.Find); }
       cout<<"Hello World?\n"<<endl;
     - (1 +1e10 0xabcdef) 'y'
     +-(1 +1e10 0xabcdef) 'y'
     ++(1 +1e10 0xabcdef) '.'
       // long double
      -3.141592654e+10l
      +3.141'592'654e+10l
     @@ t/t4034/cpp/post: Foo() : x(0&42) { bar(x.Find); }
       // expression
       1.5-e+3+f
       // another one
     +@@ t/t4034/cpp/post: str.e+75
     + a*=b c/=d e%=f
     + a++b c--d
     + a<<=b c>>=d
     +-a<=b c<d e>=f g>h
     ++a<=b c<d e>=f g>h i<=>j
     + a!=b c=d
     + a^=b c|=d e&=f
     + a|b
      
       ## t/t4034/cpp/pre ##
      @@ t/t4034/cpp/pre: Foo():x(0&&1){ foo0( x.find); }
     @@ t/t4034/cpp/pre: Foo():x(0&&1){ foo0( x.find); }
       // expression
       1.5-e+2+f
       // another one
     -
     - ## userdiff.c ##
     -@@ userdiff.c: PATTERNS("cpp",
     - 	 /* identifiers and keywords */
     - 	 "[a-zA-Z_][a-zA-Z0-9_]*"
     - 	 /* decimal and octal integers as well as floatingpoint numbers */
     --	 "|[0-9][0-9.]*([Ee][-+]?[0-9]+)?[fFlLuU]*"
     -+	 "|[0-9][0-9.']*([Ee][-+]?[0-9]+)?[fFlLuU]*"
     - 	 /* hexadecimal and binary integers */
     --	 "|0[xXbB][0-9a-fA-F]+[lLuU]*"
     -+	 "|0[xXbB][0-9a-fA-F']+[lLuU]*"
     - 	 /* floatingpoint numbers that begin with a decimal point */
     --	 "|\\.[0-9]+([Ee][-+]?[0-9]+)?[fFlL]?"
     -+	 "|\\.[0-9']+([Ee][-+]?[0-9]+)?[fFlL]?"
     - 	 "|[-+*/<>%&^|=!]=|--|\\+\\+|<<=?|>>=?|&&|\\|\\||::|->\\*?|\\.\\*"),
     - PATTERNS("csharp",
     - 	 /* Keywords */
     +@@ t/t4034/cpp/pre: str.e+65
     + a*b c/d e%f
     + a+b c-d
     + a<<b c>>d
     +-a<b c<=d e>f g>=h
     ++a<b c<=d e>f g>=h i<=j
     + a==b c!=d
     + a^b c|d e&&f
     + a||b
 -:  ----------- > 5:  037c743d9e3 userdiff-cpp: permit the digit-separating single-quote in numbers
 5:  43a701f5ffd ! 6:  cc9dc967f10 userdiff-cpp: learn the C++ spaceship operator
     @@ Commit message
          Signed-off-by: Johannes Sixt <[email protected]>
      
       ## t/t4034/cpp/expect ##
     -@@
     - <BOLD>diff --git a/pre b/post<RESET>
     --<BOLD>index 60f3640..f6fbf7b 100644<RESET>
     -+<BOLD>index 144cd98..244f79c 100644<RESET>
     - <BOLD>--- a/pre<RESET>
     - <BOLD>+++ b/post<RESET>
     - <CYAN>@@ -1,30 +1,30 @@<RESET>
      @@ t/t4034/cpp/expect: str.e+<RED>65<RESET><GREEN>75<RESET>
       a<RED>*<RESET><GREEN>*=<RESET>b c<RED>/<RESET><GREEN>/=<RESET>d e<RED>%<RESET><GREEN>%=<RESET>f
       a<RED>+<RESET><GREEN>++<RESET>b c<RED>-<RESET><GREEN>--<RESET>d
       a<RED><<<RESET><GREEN><<=<RESET>b c<RED>>><RESET><GREEN>>>=<RESET>d
     --a<RED><<RESET><GREEN><=<RESET>b c<RED><=<RESET><GREEN><<RESET>d e<RED>><RESET><GREEN>>=<RESET>f g<RED>>=<RESET><GREEN>><RESET>h
     +-a<RED><<RESET><GREEN><=<RESET>b c<RED><=<RESET><GREEN><<RESET>d e<RED>><RESET><GREEN>>=<RESET>f g<RED>>=<RESET><GREEN>><RESET>h i<=<GREEN>><RESET>j
      +a<RED><<RESET><GREEN><=<RESET>b c<RED><=<RESET><GREEN><<RESET>d e<RED>><RESET><GREEN>>=<RESET>f g<RED>>=<RESET><GREEN>><RESET>h i<RED><=<RESET><GREEN><=><RESET>j
       a<RED>==<RESET><GREEN>!=<RESET>b c<RED>!=<RESET><GREEN>=<RESET>d
       a<RED>^<RESET><GREEN>^=<RESET>b c<RED>|<RESET><GREEN>|=<RESET>d e<RED>&&<RESET><GREEN>&=<RESET>f
       a<RED>||<RESET><GREEN>|<RESET>b
      
     - ## t/t4034/cpp/post ##
     -@@ t/t4034/cpp/post: str.e+75
     - a*=b c/=d e%=f
     - a++b c--d
     - a<<=b c>>=d
     --a<=b c<d e>=f g>h
     -+a<=b c<d e>=f g>h i<=>j
     - a!=b c=d
     - a^=b c|=d e&=f
     - a|b
     -
     - ## t/t4034/cpp/pre ##
     -@@ t/t4034/cpp/pre: str.e+65
     - a*b c/d e%f
     - a+b c-d
     - a<<b c>>d
     --a<b c<=d e>f g>=h
     -+a<b c<=d e>f g>=h i<=j
     - a==b c!=d
     - a^b c|d e&&f
     - a||b
     -
       ## userdiff.c ##
      @@ userdiff.c: PATTERNS("cpp",
       	 "|0[xXbB][0-9a-fA-F']+[lLuU]*"
       	 /* floatingpoint numbers that begin with a decimal point */
     - 	 "|\\.[0-9']+([Ee][-+]?[0-9]+)?[fFlL]?"
     + 	 "|\\.[0-9][0-9']*([Ee][-+]?[0-9]+)?[fFlL]?"
      -	 "|[-+*/<>%&^|=!]=|--|\\+\\+|<<=?|>>=?|&&|\\|\\||::|->\\*?|\\.\\*"),
      +	 "|[-+*/<>%&^|=!]=|--|\\+\\+|<<=?|>>=?|&&|\\|\\||::|->\\*?|\\.\\*|<=>"),
       PATTERNS("csharp",

gitgitgadget

From: Johannes Sixt <[email protected]>

8d96e7288f2b (t4034: bulk verify builtin word regex sanity, 2010-12-18) added many tests with the intent to verify that operators consisting of more than one symbol are kept together. These are tested by probing a transition from, e.g., a!=b to x!=y, which results in the word-diff

[-a-]{+x+}!=[-b-]{+y+}

But that proves only that the letters and operators are separate tokens. To prove that != is an unseparable token, we have to probe a transition from, e.g., a=b to a!=b having a word-diff

a[-=-]{+!=+}b

that proves that the ! is not separate from the =.

In the post-image, add to or remove from operators a character that turns it into another valid operator.

Change the identifiers used around operators such that the diff algorithm does not have an incentive to match, e.g., a<b in one spot in the pre-image with a<b elsewhere in the post-image.

Adjust the expected output to match the new differences. Notice that there are some undesirable tokenizations around e, ., and -. This will be addressed in a later change.

---
 t/t4034/cpp/expect | 45 +++++++++++++++------------------------------
 t/t4034/cpp/post   | 29 +++++++++++++----------------
 t/t4034/cpp/pre    | 25 +++++++++++--------------
 3 files changed, 39 insertions(+), 60 deletions(-)

index 37d1ea25870..41976971b93 100644
@@ -1,36 +1,21 @@
 <BOLD>diff --git a/pre b/post<RESET>
-<BOLD>index 23d5c8a..7e8c026 100644<RESET>
+<BOLD>index c5672a2..4229868 100644<RESET>
 <BOLD>--- a/pre<RESET>
 <BOLD>+++ b/post<RESET>
-<CYAN>@@ -1,19 +1,19 @@<RESET>
+<CYAN>@@ -1,16 +1,16 @@<RESET>
 Foo() : x(0<RED>&&1<RESET><GREEN>&42<RESET>) { <GREEN>bar(x);<RESET> }
 cout<<"Hello World<RED>!<RESET><GREEN>?<RESET>\n"<<endl;
 <GREEN>(<RESET>1<GREEN>) (<RESET>-1e10<GREEN>) (<RESET>0xabcdef<GREEN>)<RESET> '<RED>x<RESET><GREEN>y<RESET>'
-[<RED>a<RESET><GREEN>x<RESET>] <RED>a<RESET><GREEN>x<RESET>-><RED>b a<RESET><GREEN>y x<RESET>.<RED>b<RESET><GREEN>y<RESET>
-!<RED>a<RESET><GREEN>x<RESET> ~<RED>a a<RESET><GREEN>x x<RESET>++ <RED>a<RESET><GREEN>x<RESET>-- <RED>a<RESET><GREEN>x<RESET>*<RED>b a<RESET><GREEN>y x<RESET>&<RED>b<RESET>
-<RED>a<RESET><GREEN>y<RESET>
-<GREEN>x<RESET>*<RED>b a<RESET><GREEN>y x<RESET>/<RED>b a<RESET><GREEN>y x<RESET>%<RED>b<RESET>
-<RED>a<RESET><GREEN>y<RESET>
-<GREEN>x<RESET>+<RED>b a<RESET><GREEN>y x<RESET>-<RED>b<RESET>
-<RED>a<RESET><GREEN>y<RESET>
-<GREEN>x<RESET><<<RED>b a<RESET><GREEN>y x<RESET>>><RED>b<RESET>
-<RED>a<RESET><GREEN>y<RESET>
-<GREEN>x<RESET><<RED>b a<RESET><GREEN>y x<RESET><=<RED>b a<RESET><GREEN>y x<RESET>><RED>b a<RESET><GREEN>y x<RESET>>=<RED>b<RESET>
-<RED>a<RESET><GREEN>y<RESET>
-<GREEN>x<RESET>==<RED>b a<RESET><GREEN>y x<RESET>!=<RED>b<RESET>
-<RED>a<RESET><GREEN>y<RESET>
-<GREEN>x<RESET>&<RED>b<RESET>
-<RED>a<RESET><GREEN>y<RESET>
-<GREEN>x<RESET>^<RED>b<RESET>
-<RED>a<RESET><GREEN>y<RESET>
-<GREEN>x<RESET>|<RED>b<RESET>
-<RED>a<RESET><GREEN>y<RESET>
-<GREEN>x<RESET>&&<RED>b<RESET>
-<RED>a<RESET><GREEN>y<RESET>
-<GREEN>x<RESET>||<RED>b<RESET>
-<RED>a<RESET><GREEN>y<RESET>
-<GREEN>x<RESET>?<RED>b<RESET><GREEN>y<RESET>:z
-<RED>a<RESET><GREEN>x<RESET>=<RED>b a<RESET><GREEN>y x<RESET>+=<RED>b a<RESET><GREEN>y x<RESET>-=<RED>b a<RESET><GREEN>y x<RESET>*=<RED>b a<RESET><GREEN>y x<RESET>/=<RED>b a<RESET><GREEN>y x<RESET>%=<RED>b a<RESET><GREEN>y x<RESET><<=<RED>b a<RESET><GREEN>y x<RESET>>>=<RED>b a<RESET><GREEN>y x<RESET>&=<RED>b a<RESET><GREEN>y x<RESET>^=<RED>b a<RESET><GREEN>y x<RESET>|=<RED>b<RESET>
-<RED>a<RESET><GREEN>y<RESET>
-<GREEN>x<RESET>,y
-<RED>a<RESET><GREEN>x<RESET>::<RED>b<RESET><GREEN>y<RESET>
+[a] b<RED>-><RESET><GREEN>->*<RESET>v d<RED>.e<RESET><GREEN>.*e<RESET>
+<GREEN>~<RESET>!a <GREEN>!<RESET>~b c<RED>++<RESET><GREEN>+<RESET> d<RED>--<RESET><GREEN>-<RESET> e*<GREEN>*<RESET>f g<RED>&<RESET><GREEN>&&<RESET>h
+a<RED>*<RESET><GREEN>*=<RESET>b c<RED>/<RESET><GREEN>/=<RESET>d e<RED>%<RESET><GREEN>%=<RESET>f
+a<RED>+<RESET><GREEN>++<RESET>b c<RED>-<RESET><GREEN>--<RESET>d
+a<RED><<<RESET><GREEN><<=<RESET>b c<RED>>><RESET><GREEN>>>=<RESET>d
+a<RED><<RESET><GREEN><=<RESET>b c<RED><=<RESET><GREEN><<RESET>d e<RED>><RESET><GREEN>>=<RESET>f g<RED>>=<RESET><GREEN>><RESET>h
+a<RED>==<RESET><GREEN>!=<RESET>b c<RED>!=<RESET><GREEN>=<RESET>d
+a<RED>^<RESET><GREEN>^=<RESET>b c<RED>|<RESET><GREEN>|=<RESET>d e<RED>&&<RESET><GREEN>&=<RESET>f
+a<RED>||<RESET><GREEN>|<RESET>b
+a?<GREEN>:<RESET>b
+a<RED>=<RESET><GREEN>==<RESET>b c<RED>+=<RESET><GREEN>+<RESET>d <RED>e-=f<RESET><GREEN>e-f<RESET> g<RED>*=<RESET><GREEN>*<RESET>h i<RED>/=<RESET><GREEN>/<RESET>j k<RED>%=<RESET><GREEN>%<RESET>l m<RED><<=<RESET><GREEN><<<RESET>n o<RED>>>=<RESET><GREEN>>><RESET>p q<RED>&=<RESET><GREEN>&<RESET>r s<RED>^=<RESET><GREEN>^<RESET>t u<RED>|=<RESET><GREEN>|<RESET>v
+a,b<RESET>
+a<RED>::<RESET><GREEN>:<RESET>b

index 7e8c026cefb..4229868ae62 100644
@@ -1,19 +1,16 @@
 Foo() : x(0&42) { bar(x); }
 cout<<"Hello World?\n"<<endl;
 (1) (-1e10) (0xabcdef) 'y'
-[x] x->y x.y
-!x ~x x++ x-- x*y x&y
-x*y x/y x%y
-x+y x-y
-x<<y x>>y
-x<y x<=y x>y x>=y
-x==y x!=y
-x&y
-x^y
-x|y
-x&&y
-x||y
-x?y:z
-x=y x+=y x-=y x*=y x/=y x%=y x<<=y x>>=y x&=y x^=y x|=y
-x,y
-x::y
+[a] b->*v d.*e
+~!a !~b c+ d- e**f g&&h
+a*=b c/=d e%=f
+a++b c--d
+a<<=b c>>=d
+a<=b c<d e>=f g>h
+a!=b c=d
+a^=b c|=d e&=f
+a|b
+a?:b
+a==b c+d e-f g*h i/j k%l m<<n o>>p q&r s^t u|v
+a,b
+a:b

index 23d5c8adf54..c5672a24cfc 100644
@@ -1,19 +1,16 @@
 Foo():x(0&&1){}
 cout<<"Hello World!\n"<<endl;
 1 -1e10 0xabcdef 'x'
-[a] a->b a.b
-!a ~a a++ a-- a*b a&b
-a*b a/b a%b
-a+b a-b
-a<<b a>>b
-a<b a<=b a>b a>=b
-a==b a!=b
-a&b
-a^b
-a|b
-a&&b
+[a] b->v d.e
+!a ~b c++ d-- e*f g&h
+a*b c/d e%f
+a+b c-d
+a<<b c>>d
+a<b c<=d e>f g>=h
+a==b c!=d
+a^b c|d e&&f
 a||b
-a?b:z
-a=b a+=b a-=b a*=b a/=b a%=b a<<=b a>>=b a&=b a^=b a|=b
-a,y
+a?b
+a=b c+=d e-=f g*=h i/=j k%=l m<<=n o>>=p q&=r s^=t u|=v
+a,b
 a::b

gitgitgadget

From: Johannes Sixt <[email protected]>

The word regex is too loose and matches long streaks of characters that should actually be separate tokens. Add these problematic test cases. Separate the lines with text that will remain identical in the pre- and post-image so that the diff algorithm will not lump removals and additions of consecutive lines together. This makes the expected output easier to read.

---
 t/t4034/cpp/expect | 22 ++++++++++++++++++----
 t/t4034/cpp/post   | 18 ++++++++++++++++--
 t/t4034/cpp/pre    | 16 +++++++++++++++-
 3 files changed, 49 insertions(+), 7 deletions(-)

index 41976971b93..63e53a61e62 100644
@@ -1,11 +1,25 @@
 <BOLD>diff --git a/pre b/post<RESET>
-<BOLD>index c5672a2..4229868 100644<RESET>
+<BOLD>index 1229cdb..3feae6f 100644<RESET>
 <BOLD>--- a/pre<RESET>
 <BOLD>+++ b/post<RESET>
-<CYAN>@@ -1,16 +1,16 @@<RESET>
-Foo() : x(0<RED>&&1<RESET><GREEN>&42<RESET>) { <GREEN>bar(x);<RESET> }
+<CYAN>@@ -1,30 +1,30 @@<RESET>
+Foo() : x(0<RED>&&1<RESET><GREEN>&42<RESET>) { <RED>foo0<RESET><GREEN>bar<RESET>(x<RED>.f<RESET><GREEN>.F<RESET>ind); }
 cout<<"Hello World<RED>!<RESET><GREEN>?<RESET>\n"<<endl;
-<GREEN>(<RESET>1<GREEN>) (<RESET>-1e10<GREEN>) (<RESET>0xabcdef<GREEN>)<RESET> '<RED>x<RESET><GREEN>y<RESET>'
+<GREEN>(<RESET>1 <RED>-1e10<RESET><GREEN>+1e10<RESET> 0xabcdef<GREEN>)<RESET> '<RED>x<RESET><GREEN>y<RESET>'
+// long double<RESET>
+<RED>3.141592653e-10l<RESET><GREEN>3.141592654e+10l<RESET>
+// float<RESET>
+120<RED>E5f<RESET><GREEN>E6f<RESET>
+// hex<RESET>
+<RED>0xdeadbeaf+8<RESET><GREEN>0xdeadBeaf+7<RESET>ULL
+// octal<RESET>
+<RED>01234567<RESET><GREEN>01234560<RESET>
+// binary<RESET>
+<RED>0b1000<RESET><GREEN>0b1100<RESET>+e1
+// expression<RESET>
+<RED>1.5-e+2+f<RESET><GREEN>1.5-e+3+f<RESET>
+// another one<RESET>
+str<RED>.e+65<RESET><GREEN>.e+75<RESET>
 [a] b<RED>-><RESET><GREEN>->*<RESET>v d<RED>.e<RESET><GREEN>.*e<RESET>
 <GREEN>~<RESET>!a <GREEN>!<RESET>~b c<RED>++<RESET><GREEN>+<RESET> d<RED>--<RESET><GREEN>-<RESET> e*<GREEN>*<RESET>f g<RED>&<RESET><GREEN>&&<RESET>h
 a<RED>*<RESET><GREEN>*=<RESET>b c<RED>/<RESET><GREEN>/=<RESET>d e<RED>%<RESET><GREEN>%=<RESET>f

index 4229868ae62..3feae6f430f 100644
@@ -1,6 +1,20 @@
-Foo() : x(0&42) { bar(x); }
+Foo() : x(0&42) { bar(x.Find); }
 cout<<"Hello World?\n"<<endl;
-(1) (-1e10) (0xabcdef) 'y'
+(1 +1e10 0xabcdef) 'y'
+// long double
+3.141592654e+10l
+// float
+120E6f
+// hex
+0xdeadBeaf+7ULL
+// octal
+01234560
+// binary
+0b1100+e1
+// expression
+1.5-e+3+f
+// another one
+str.e+75
 [a] b->*v d.*e
 ~!a !~b c+ d- e**f g&&h
 a*=b c/=d e%=f

index c5672a24cfc..1229cdb59d1 100644
@@ -1,6 +1,20 @@
-Foo():x(0&&1){}
+Foo():x(0&&1){ foo0( x.find); }
 cout<<"Hello World!\n"<<endl;
 1 -1e10 0xabcdef 'x'
+// long double
+3.141592653e-10l
+// float
+120E5f
+// hex
+0xdeadbeaf+8ULL
+// octal
+01234567
+// binary
+0b1000+e1
+// expression
+1.5-e+2+f
+// another one
+str.e+65
 [a] b->v d.e
 !a ~b c++ d-- e*f g&h
 a*b c/d e%f

gitgitgadget

From: Johannes Sixt <[email protected]>

Generally, word regex can be written such that they match tokens liberally and need not model the actual syntax because it can be assumed that the regex will only be applied to syntactically correct text.

The regex for cpp (C/C++) is too liberal, though. It regards these sequences as single tokens:

1+2
1.5-e+2+f

and the following amalgams as one token:

   .l      as in str.length
   .f      as in str.find
   .e      as in str.erase

Tighten the regex in the following way:

- Accept + and - only in one position in the exponent. + and - are no longer regarded as the sign of a number and are treated by the catcher-all that is not visible in the driver's regex.

- Accept a leading decimal point only when it is followed by a digit.

For readability, factor hex- and binary numbers into an own term.

As a drive-by, this fixes that floating point numbers such as 12E5 (with upper-case E) were split into two tokens.

---
 t/t4034/cpp/expect | 16 ++++++++--------
 userdiff.c         |  8 +++++++-
 2 files changed, 15 insertions(+), 9 deletions(-)

index 63e53a61e62..46c9460a968 100644
@@ -3,24 +3,24 @@
 <BOLD>--- a/pre<RESET>
 <BOLD>+++ b/post<RESET>
 <CYAN>@@ -1,30 +1,30 @@<RESET>
-Foo() : x(0<RED>&&1<RESET><GREEN>&42<RESET>) { <RED>foo0<RESET><GREEN>bar<RESET>(x<RED>.f<RESET><GREEN>.F<RESET>ind); }
+Foo() : x(0<RED>&&1<RESET><GREEN>&42<RESET>) { <RED>foo0<RESET><GREEN>bar<RESET>(x.<RED>find<RESET><GREEN>Find<RESET>); }
 cout<<"Hello World<RED>!<RESET><GREEN>?<RESET>\n"<<endl;
-<GREEN>(<RESET>1 <RED>-1e10<RESET><GREEN>+1e10<RESET> 0xabcdef<GREEN>)<RESET> '<RED>x<RESET><GREEN>y<RESET>'
+<GREEN>(<RESET>1 <RED>-<RESET><GREEN>+<RESET>1e10 0xabcdef<GREEN>)<RESET> '<RED>x<RESET><GREEN>y<RESET>'
 // long double<RESET>
 <RED>3.141592653e-10l<RESET><GREEN>3.141592654e+10l<RESET>
 // float<RESET>
-120<RED>E5f<RESET><GREEN>E6f<RESET>
+<RED>120E5f<RESET><GREEN>120E6f<RESET>
 // hex<RESET>
-<RED>0xdeadbeaf+8<RESET><GREEN>0xdeadBeaf+7<RESET>ULL
+<RED>0xdeadbeaf<RESET><GREEN>0xdeadBeaf<RESET>+<RED>8ULL<RESET><GREEN>7ULL<RESET>
 // octal<RESET>
 <RED>01234567<RESET><GREEN>01234560<RESET>
 // binary<RESET>
 <RED>0b1000<RESET><GREEN>0b1100<RESET>+e1
 // expression<RESET>
-<RED>1.5-e+2+f<RESET><GREEN>1.5-e+3+f<RESET>
+1.5-e+<RED>2<RESET><GREEN>3<RESET>+f
 // another one<RESET>
-str<RED>.e+65<RESET><GREEN>.e+75<RESET>
-[a] b<RED>-><RESET><GREEN>->*<RESET>v d<RED>.e<RESET><GREEN>.*e<RESET>
+str.e+<RED>65<RESET><GREEN>75<RESET>
+[a] b<RED>-><RESET><GREEN>->*<RESET>v d<RED>.<RESET><GREEN>.*<RESET>e
 <GREEN>~<RESET>!a <GREEN>!<RESET>~b c<RED>++<RESET><GREEN>+<RESET> d<RED>--<RESET><GREEN>-<RESET> e*<GREEN>*<RESET>f g<RED>&<RESET><GREEN>&&<RESET>h
 a<RED>*<RESET><GREEN>*=<RESET>b c<RED>/<RESET><GREEN>/=<RESET>d e<RED>%<RESET><GREEN>%=<RESET>f
 a<RED>+<RESET><GREEN>++<RESET>b c<RED>-<RESET><GREEN>--<RESET>d

@@ -30,6 +30,6 @@ a<RED>==<RESET><GREEN>!=<RESET>b c<RED>!=<RESET><GREEN>=<RESET>d
 a<RED>^<RESET><GREEN>^=<RESET>b c<RED>|<RESET><GREEN>|=<RESET>d e<RED>&&<RESET><GREEN>&=<RESET>f
 a<RED>||<RESET><GREEN>|<RESET>b
 a?<GREEN>:<RESET>b
-a<RED>=<RESET><GREEN>==<RESET>b c<RED>+=<RESET><GREEN>+<RESET>d <RED>e-=f<RESET><GREEN>e-f<RESET> g<RED>*=<RESET><GREEN>*<RESET>h i<RED>/=<RESET><GREEN>/<RESET>j k<RED>%=<RESET><GREEN>%<RESET>l m<RED><<=<RESET><GREEN><<<RESET>n o<RED>>>=<RESET><GREEN>>><RESET>p q<RED>&=<RESET><GREEN>&<RESET>r s<RED>^=<RESET><GREEN>^<RESET>t u<RED>|=<RESET><GREEN>|<RESET>v
+a<RED>=<RESET><GREEN>==<RESET>b c<RED>+=<RESET><GREEN>+<RESET>d e<RED>-=<RESET><GREEN>-<RESET>f g<RED>*=<RESET><GREEN>*<RESET>h i<RED>/=<RESET><GREEN>/<RESET>j k<RED>%=<RESET><GREEN>%<RESET>l m<RED><<=<RESET><GREEN><<<RESET>n o<RED>>>=<RESET><GREEN>>><RESET>p q<RED>&=<RESET><GREEN>&<RESET>r s<RED>^=<RESET><GREEN>^<RESET>t u<RED>|=<RESET><GREEN>|<RESET>v
 a,b<RESET>
 a<RED>::<RESET><GREEN>:<RESET>b

index d9b2ba752f0..ce2a9230703 100644
@@ -54,8 +54,14 @@ PATTERNS("cpp",
 	 /* functions/methods, variables, and compounds at top level */
 	 "^((::[[:space:]]*)?[A-Za-z_].*)$",
 	 /* -- */
+	 /* identifiers and keywords */
 	 "[a-zA-Z_][a-zA-Z0-9_]*"
-	 "|[-+0-9.e]+[fFlL]?|0[xXbB]?[0-9a-fA-F]+[lLuU]*"
+	 /* decimal and octal integers as well as floatingpoint numbers */
+	 "|[0-9][0-9.]*([Ee][-+]?[0-9]+)?[fFlLuU]*"
+	 /* hexadecimal and binary integers */
+	 "|0[xXbB][0-9a-fA-F]+[lLuU]*"
+	 /* floatingpoint numbers that begin with a decimal point */
+	 "|\\.[0-9]+([Ee][-+]?[0-9]+)?[fFlL]?"
 	 "|[-+*/<>%&^|=!]=|--|\\+\\+|<<=?|>>=?|&&|\\|\\||::|->\\*?|\\.\\*"),
 PATTERNS("csharp",
 	 /* Keywords */

gitgitgadget

From: Johannes Sixt <[email protected]>

We are going to add support for C++'s digit-separating single-quote and the spaceship operator. By adding the test cases in this separate commit, the effect on the word highlighting will become more obvious as the features are implemented and the file cpp/expect is updated.

---
 t/t4034/cpp/expect | 14 +++++++-------
 t/t4034/cpp/post   | 12 ++++++------
 t/t4034/cpp/pre    | 10 +++++-----
 3 files changed, 18 insertions(+), 18 deletions(-)

index 46c9460a968..3d37ddac42c 100644
@@ -1,21 +1,21 @@
 <BOLD>diff --git a/pre b/post<RESET>
-<BOLD>index 1229cdb..3feae6f 100644<RESET>
+<BOLD>index 144cd98..64e78af 100644<RESET>
 <BOLD>--- a/pre<RESET>
 <BOLD>+++ b/post<RESET>
 <CYAN>@@ -1,30 +1,30 @@<RESET>
 Foo() : x(0<RED>&&1<RESET><GREEN>&42<RESET>) { <RED>foo0<RESET><GREEN>bar<RESET>(x.<RED>find<RESET><GREEN>Find<RESET>); }
 cout<<"Hello World<RED>!<RESET><GREEN>?<RESET>\n"<<endl;
-<GREEN>(<RESET>1 <RED>-<RESET><GREEN>+<RESET>1e10 0xabcdef<GREEN>)<RESET> '<RED>x<RESET><GREEN>y<RESET>'
+<GREEN>(<RESET>1 <RED>-<RESET><GREEN>+<RESET>1e10 0xabcdef<GREEN>)<RESET> '<RED>x<RESET><GREEN>.<RESET>'
 // long double<RESET>
-<RED>3.141592653e-10l<RESET><GREEN>3.141592654e+10l<RESET>
+3.141'592'<RED>653e-10l<RESET><GREEN>654e+10l<RESET>
 // float<RESET>
 <RED>120E5f<RESET><GREEN>120E6f<RESET>
 // hex<RESET>
-<RED>0xdeadbeaf<RESET><GREEN>0xdeadBeaf<RESET>+<RED>8ULL<RESET><GREEN>7ULL<RESET>
+0xdead'<RED>beaf<RESET><GREEN>Beaf<RESET>+<RED>8ULL<RESET><GREEN>7ULL<RESET>
 // octal<RESET>
-<RED>01234567<RESET><GREEN>01234560<RESET>
+0123'<RED>4567<RESET><GREEN>4560<RESET>
 // binary<RESET>
-<RED>0b1000<RESET><GREEN>0b1100<RESET>+e1
+<RED>0b10<RESET><GREEN>0b11<RESET>'00+e1
 // expression<RESET>
 1.5-e+<RED>2<RESET><GREEN>3<RESET>+f
 // another one<RESET>

@@ -25,7 +25,7 @@ str.e+<RED>65<RESET><GREEN>75<RESET>
 a<RED>*<RESET><GREEN>*=<RESET>b c<RED>/<RESET><GREEN>/=<RESET>d e<RED>%<RESET><GREEN>%=<RESET>f
 a<RED>+<RESET><GREEN>++<RESET>b c<RED>-<RESET><GREEN>--<RESET>d
 a<RED><<<RESET><GREEN><<=<RESET>b c<RED>>><RESET><GREEN>>>=<RESET>d
-a<RED><<RESET><GREEN><=<RESET>b c<RED><=<RESET><GREEN><<RESET>d e<RED>><RESET><GREEN>>=<RESET>f g<RED>>=<RESET><GREEN>><RESET>h
+a<RED><<RESET><GREEN><=<RESET>b c<RED><=<RESET><GREEN><<RESET>d e<RED>><RESET><GREEN>>=<RESET>f g<RED>>=<RESET><GREEN>><RESET>h i<=<GREEN>><RESET>j
 a<RED>==<RESET><GREEN>!=<RESET>b c<RED>!=<RESET><GREEN>=<RESET>d
 a<RED>^<RESET><GREEN>^=<RESET>b c<RED>|<RESET><GREEN>|=<RESET>d e<RED>&&<RESET><GREEN>&=<RESET>f
 a<RED>||<RESET><GREEN>|<RESET>b

index 3feae6f430f..64e78afbfb5 100644
@@ -1,16 +1,16 @@
 Foo() : x(0&42) { bar(x.Find); }
 cout<<"Hello World?\n"<<endl;
-(1 +1e10 0xabcdef) 'y'
+(1 +1e10 0xabcdef) '.'
 // long double
-3.141592654e+10l
+3.141'592'654e+10l
 // float
 120E6f
 // hex
-0xdeadBeaf+7ULL
+0xdead'Beaf+7ULL
 // octal
-01234560
+0123'4560
 // binary
-0b1100+e1
+0b11'00+e1
 // expression
 1.5-e+3+f
 // another one

@@ -20,7 +20,7 @@ str.e+75
 a*=b c/=d e%=f
 a++b c--d
 a<<=b c>>=d
-a<=b c<d e>=f g>h
+a<=b c<d e>=f g>h i<=>j
 a!=b c=d
 a^=b c|=d e&=f
 a|b

index 1229cdb59d1..144cd980d6b 100644
@@ -2,15 +2,15 @@ Foo():x(0&&1){ foo0( x.find); }
 cout<<"Hello World!\n"<<endl;
 1 -1e10 0xabcdef 'x'
 // long double
-3.141592653e-10l
+3.141'592'653e-10l
 // float
 120E5f
 // hex
-0xdeadbeaf+8ULL
+0xdead'beaf+8ULL
 // octal
-01234567
+0123'4567
 // binary
-0b1000+e1
+0b10'00+e1
 // expression
 1.5-e+2+f
 // another one

@@ -20,7 +20,7 @@ str.e+65
 a*b c/d e%f
 a+b c-d
 a<<b c>>d
-a<b c<=d e>f g>=h
+a<b c<=d e>f g>=h i<=j
 a==b c!=d
 a^b c|d e&&f
 a||b

gitgitgadget

From: Johannes Sixt <[email protected]>

Since C++17, the single-quote can be used as digit separator:

3.141'592'654
1'000'000
0xdead'beaf

Make it known to the word regex of the cpp driver, so that numbers are not split into separate tokens at the single-quotes.

---
 t/t4034/cpp/expect | 8 ++++----
 userdiff.c         | 6 +++---
 2 files changed, 7 insertions(+), 7 deletions(-)

index 3d37ddac42c..b90b3f207bf 100644
@@ -7,15 +7,15 @@ Foo() : x(0<RED>&&1<RESET><GREEN>&42<RESET>) { <RED>foo0<RESET><GREEN>bar<RESET>
 cout<<"Hello World<RED>!<RESET><GREEN>?<RESET>\n"<<endl;
 <GREEN>(<RESET>1 <RED>-<RESET><GREEN>+<RESET>1e10 0xabcdef<GREEN>)<RESET> '<RED>x<RESET><GREEN>.<RESET>'
 // long double<RESET>
-3.141'592'<RED>653e-10l<RESET><GREEN>654e+10l<RESET>
+<RED>3.141'592'653e-10l<RESET><GREEN>3.141'592'654e+10l<RESET>
 // float<RESET>
 <RED>120E5f<RESET><GREEN>120E6f<RESET>
 // hex<RESET>
-0xdead'<RED>beaf<RESET><GREEN>Beaf<RESET>+<RED>8ULL<RESET><GREEN>7ULL<RESET>
+<RED>0xdead'beaf<RESET><GREEN>0xdead'Beaf<RESET>+<RED>8ULL<RESET><GREEN>7ULL<RESET>
 // octal<RESET>
-0123'<RED>4567<RESET><GREEN>4560<RESET>
+<RED>0123'4567<RESET><GREEN>0123'4560<RESET>
 // binary<RESET>
-<RED>0b10<RESET><GREEN>0b11<RESET>'00+e1
+<RED>0b10'00<RESET><GREEN>0b11'00<RESET>+e1
 // expression<RESET>
 1.5-e+<RED>2<RESET><GREEN>3<RESET>+f
 // another one<RESET>

index ce2a9230703..5072d12e51e 100644
@@ -57,11 +57,11 @@ PATTERNS("cpp",
 	 /* identifiers and keywords */
 	 "[a-zA-Z_][a-zA-Z0-9_]*"
 	 /* decimal and octal integers as well as floatingpoint numbers */
-	 "|[0-9][0-9.]*([Ee][-+]?[0-9]+)?[fFlLuU]*"
+	 "|[0-9][0-9.']*([Ee][-+]?[0-9]+)?[fFlLuU]*"
 	 /* hexadecimal and binary integers */
-	 "|0[xXbB][0-9a-fA-F]+[lLuU]*"
+	 "|0[xXbB][0-9a-fA-F']+[lLuU]*"
 	 /* floatingpoint numbers that begin with a decimal point */
-	 "|\\.[0-9]+([Ee][-+]?[0-9]+)?[fFlL]?"
+	 "|\\.[0-9][0-9']*([Ee][-+]?[0-9]+)?[fFlL]?"
 	 "|[-+*/<>%&^|=!]=|--|\\+\\+|<<=?|>>=?|&&|\\|\\||::|->\\*?|\\.\\*"),
 PATTERNS("csharp",
 	 /* Keywords */

gitgitgadget

From: Johannes Sixt <[email protected]>

Since C++20, the language has a generalized comparison operator <=>. Teach the cpp driver not to separate it into <= and > tokens.

---
 t/t4034/cpp/expect | 2 +-
 userdiff.c         | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

index b90b3f207bf..5ff4ce477b4 100644
@@ -25,7 +25,7 @@ str.e+<RED>65<RESET><GREEN>75<RESET>
 a<RED>*<RESET><GREEN>*=<RESET>b c<RED>/<RESET><GREEN>/=<RESET>d e<RED>%<RESET><GREEN>%=<RESET>f
 a<RED>+<RESET><GREEN>++<RESET>b c<RED>-<RESET><GREEN>--<RESET>d
 a<RED><<<RESET><GREEN><<=<RESET>b c<RED>>><RESET><GREEN>>>=<RESET>d
-a<RED><<RESET><GREEN><=<RESET>b c<RED><=<RESET><GREEN><<RESET>d e<RED>><RESET><GREEN>>=<RESET>f g<RED>>=<RESET><GREEN>><RESET>h i<=<GREEN>><RESET>j
+a<RED><<RESET><GREEN><=<RESET>b c<RED><=<RESET><GREEN><<RESET>d e<RED>><RESET><GREEN>>=<RESET>f g<RED>>=<RESET><GREEN>><RESET>h i<RED><=<RESET><GREEN><=><RESET>j
 a<RED>==<RESET><GREEN>!=<RESET>b c<RED>!=<RESET><GREEN>=<RESET>d
 a<RED>^<RESET><GREEN>^=<RESET>b c<RED>|<RESET><GREEN>|=<RESET>d e<RED>&&<RESET><GREEN>&=<RESET>f
 a<RED>||<RESET><GREEN>|<RESET>b

index 5072d12e51e..96adddd6f9a 100644
@@ -62,7 +62,7 @@ PATTERNS("cpp",
 	 "|0[xXbB][0-9a-fA-F']+[lLuU]*"
 	 /* floatingpoint numbers that begin with a decimal point */
 	 "|\\.[0-9][0-9']*([Ee][-+]?[0-9]+)?[fFlL]?"
-	 "|[-+*/<>%&^|=!]=|--|\\+\\+|<<=?|>>=?|&&|\\|\\||::|->\\*?|\\.\\*"),
+	 "|[-+*/<>%&^|=!]=|--|\\+\\+|<<=?|>>=?|&&|\\|\\||::|->\\*?|\\.\\*|<=>"),
 PATTERNS("csharp",
 	 /* Keywords */
 	 "!^[ \t]*(do|while|for|if|else|instanceof|new|return|switch|case|throw|catch|using)\n"

gitgitgadget

Am 09.10.21 um 02:00 schrieb Ævar Arnfjörð Bjarmason:

FWIW, I wondered why the cpp driver should not handle this case as you expect it, and the answer is that it actually does when this comment appears in a C file. This particular word-diff occurs in Documentation/CodingGuidelines and that is not covered by the cpp driver.

-- Hannes

The implementation of digit-separating single-quotes introduced a note-worthy regression: the change of a character literal with a digit would splice the digit and the closing single-quote. For example, the change from 'a' to '2' is now tokenized as '[-a'-]{+2'+} instead of '[-a-]{+2+}'.

The options to fix the regression are:

- Tighten the regular expression such that the single-quote can only occur between digits (that would match the official syntax).

- Remove support for digit separators.

I chose to remove support, because

- I have not seen a lot of code make use of digit separators.

- If code does use digit separators, then the numbers are typically long. If a change in one of the segments occurs, it is actually better visible if only that segment is highlighted as the word that changed instead of the whole long number.

This choice does introduce another minor regression, though, which is highlighted in the test case: when a change occurs in the second or later segment of a hexadecimal number where the segment begins with a digit, but also has letters, the segment is mistaken as consisting of a number and an identifier. I can live with that.

---
 t/t4034/cpp/expect | 12 ++++++------
 t/t4034/cpp/post   | 10 +++++-----
 t/t4034/cpp/pre    |  8 ++++----
 userdiff.c         |  6 +++---
 4 files changed, 18 insertions(+), 18 deletions(-)

index 5ff4ce477b..dc500ae092 100644
@@ -1,21 +1,21 @@
 <BOLD>diff --git a/pre b/post<RESET>
-<BOLD>index 144cd98..64e78af 100644<RESET>
+<BOLD>index a1a09b7..f1b6f3c 100644<RESET>
 <BOLD>--- a/pre<RESET>
 <BOLD>+++ b/post<RESET>
 <CYAN>@@ -1,30 +1,30 @@<RESET>
 Foo() : x(0<RED>&&1<RESET><GREEN>&42<RESET>) { <RED>foo0<RESET><GREEN>bar<RESET>(x.<RED>find<RESET><GREEN>Find<RESET>); }
 cout<<"Hello World<RED>!<RESET><GREEN>?<RESET>\n"<<endl;
-<GREEN>(<RESET>1 <RED>-<RESET><GREEN>+<RESET>1e10 0xabcdef<GREEN>)<RESET> '<RED>x<RESET><GREEN>.<RESET>'
+<GREEN>(<RESET>1 <RED>-<RESET><GREEN>+<RESET>1e10 0xabcdef<GREEN>)<RESET> '<RED>x<RESET><GREEN>2<RESET>'
 // long double<RESET>
-<RED>3.141'592'653e-10l<RESET><GREEN>3.141'592'654e+10l<RESET>
+<RED>3.141592653e-10l<RESET><GREEN>3.141592654e+10l<RESET>
 // float<RESET>
 <RED>120E5f<RESET><GREEN>120E6f<RESET>
 // hex<RESET>
-<RED>0xdead'beaf<RESET><GREEN>0xdead'Beaf<RESET>+<RED>8ULL<RESET><GREEN>7ULL<RESET>
+<RED>0xdead<RESET><GREEN>0xdeaf<RESET>'1<RED>eaF<RESET><GREEN>eaf<RESET>+<RED>8ULL<RESET><GREEN>7ULL<RESET>
 // octal<RESET>
-<RED>0123'4567<RESET><GREEN>0123'4560<RESET>
+<RED>01234567<RESET><GREEN>01234560<RESET>
 // binary<RESET>
-<RED>0b10'00<RESET><GREEN>0b11'00<RESET>+e1
+<RED>0b1000<RESET><GREEN>0b1100<RESET>+e1
 // expression<RESET>
 1.5-e+<RED>2<RESET><GREEN>3<RESET>+f
 // another one<RESET>

index 64e78afbfb..f1b6f3c228 100644
@@ -1,16 +1,16 @@
 Foo() : x(0&42) { bar(x.Find); }
 cout<<"Hello World?\n"<<endl;
-(1 +1e10 0xabcdef) '.'
+(1 +1e10 0xabcdef) '2'
 // long double
-3.141'592'654e+10l
+3.141592654e+10l
 // float
 120E6f
 // hex
-0xdead'Beaf+7ULL
+0xdeaf'1eaf+7ULL
 // octal
-0123'4560
+01234560
 // binary
-0b11'00+e1
+0b1100+e1
 // expression
 1.5-e+3+f
 // another one

index 144cd980d6..a1a09b7712 100644
@@ -2,15 +2,15 @@ Foo():x(0&&1){ foo0( x.find); }
 cout<<"Hello World!\n"<<endl;
 1 -1e10 0xabcdef 'x'
 // long double
-3.141'592'653e-10l
+3.141592653e-10l
 // float
 120E5f
 // hex
-0xdead'beaf+8ULL
+0xdead'1eaF+8ULL
 // octal
-0123'4567
+01234567
 // binary
-0b10'00+e1
+0b1000+e1
 // expression
 1.5-e+2+f
 // another one

index 7b143ef36b..8578cb0d12 100644
@@ -67,11 +67,11 @@
 	 /* identifiers and keywords */
 	 "[a-zA-Z_][a-zA-Z0-9_]*"
 	 /* decimal and octal integers as well as floatingpoint numbers */
-	 "|[0-9][0-9.']*([Ee][-+]?[0-9]+)?[fFlLuU]*"
+	 "|[0-9][0-9.]*([Ee][-+]?[0-9]+)?[fFlLuU]*"
 	 /* hexadecimal and binary integers */
-	 "|0[xXbB][0-9a-fA-F']+[lLuU]*"
+	 "|0[xXbB][0-9a-fA-F]+[lLuU]*"
 	 /* floatingpoint numbers that begin with a decimal point */
-	 "|\\.[0-9][0-9']*([Ee][-+]?[0-9]+)?[fFlL]?"
+	 "|\\.[0-9][0-9]*([Ee][-+]?[0-9]+)?[fFlL]?"
 	 "|[-+*/<>%&^|=!]=|--|\\+\\+|<<=?|>>=?|&&|\\|\\||::|->\\*?|\\.\\*|<=>"),
 PATTERNS("csharp",
 	 /* Keywords */

2.33.0.129.g739793498e