Even more CFGAS_FormatString tests.

Change-Id: Ied2a169a3e7e9b178d746b04c67286d7a51fa635
Reviewed-on: https://pdfium-review.googlesource.com/6330
Commit-Queue: dsinclair <dsinclair@chromium.org>
Reviewed-by: Nicolás Peña <npm@chromium.org>

3 files changed, 265 insertions, 78 deletions

diff --git a/xfa/fgas/crt/cfgas_formatstring.cpp b/xfa/fgas/crt/cfgas_formatstring.cpp
index 5c15135d9a..6e2952a7cc 100644
--- a/xfa/fgas/crt/cfgas_formatstring.cpp
+++ b/xfa/fgas/crt/cfgas_formatstring.cpp
@@ -1040,21 +1040,11 @@ bool FX_TimeFromCanonical(const CFX_WideStringC& wsTime,
         }
       }
 
+      // Skip until we find a + or - for the time zone.
       while (cc < len) {
-        // Skip until we find a + or - for the time zone.
-        if (str[cc] != '+' && str[cc] != '-') {
-          ++cc;
-          continue;
-        }
-
-        FX_TIMEZONE tzDiff;
-        tzDiff.tzHour = 0;
-        tzDiff.tzMinute = 0;
-        if (str[cc] != 'Z')
-          cc += ParseTimeZone(str + cc, len - cc, &tzDiff);
-
-        ResolveZone(tzDiff, pLocale, &hour, &minute);
-        break;
+        if (str[cc] == '+' || str[cc] == '-')
+          break;
+        ++cc;
       }
 
       if (cc < len) {
@@ -1183,9 +1173,9 @@ CFX_WideString CFGAS_FormatString::GetTextFormat(
       while (ccf < iLenf) {
         if (pStr[ccf] == '(') {
           ccf++;
-          CFX_WideString wsLCID;
+          // Skip over the encoding name.
           while (ccf < iLenf && pStr[ccf] != ')')
-            wsLCID += pStr[ccf++];
+            ccf++;
         } else if (pStr[ccf] == '{') {
           bBrackOpen = true;
           break;
@@ -1410,11 +1400,22 @@ bool CFGAS_FormatString::ParseNum(const CFX_WideString& wsSrcNum,
   bool bNeg = false;
   bool bReverseParse = false;
   int32_t dot_index = 0;
+
+  // If we're looking for a '.', 'V' or 'v' and the input string does not
+  // have a dot index for one of those, then we disable parsing the decimal.
   if (!GetNumericDotIndex(wsSrcNum, wsDotSymbol, &dot_index) &&
       (dwFormatStyle & FX_NUMSTYLE_DotVorv)) {
     bReverseParse = true;
   }
-  bReverseParse = false;
+
+  // This parse is broken into two parts based on the '.' in the number
+  // (or 'V' or 'v'). |dot_index_f| is the location of the dot in the format and
+  // |dot_index| is the location of the dot in the number.
+  //
+  // This first while() starts at the '.' and walks backwards to the start of
+  // the number. The second while() walks from the dot forwards to the end of
+  // the decimal.
+
   int ccf = dot_index_f - 1;
   int cc = dot_index - 1;
   while (ccf >= 0 && cc >= 0) {
@@ -1483,9 +1484,6 @@ bool CFGAS_FormatString::ParseNum(const CFX_WideString& wsSrcNum,
         ccf--;
         break;
       case 'E': {
-        if (cc >= dot_index)
-          return false;
-
         bool bExpSign = false;
         while (cc >= 0) {
           if (str[cc] == 'E' || str[cc] == 'e')
@@ -1539,7 +1537,7 @@ bool CFGAS_FormatString::ParseNum(const CFX_WideString& wsSrcNum,
       case 'R':
         if (ccf - 1 >= 0 && strf[ccf - 1] == 'C') {
           if (str[cc] == ' ') {
-            cc++;
+            cc -= 2;
           } else if (str[cc] == 'R' && cc - 1 >= 0 && str[cc - 1] == 'C') {
             bNeg = true;
             cc -= 2;
@@ -1563,7 +1561,7 @@ bool CFGAS_FormatString::ParseNum(const CFX_WideString& wsSrcNum,
       case 'B':
         if (ccf - 1 >= 0 && strf[ccf - 1] == 'D') {
           if (str[cc] == ' ') {
-            cc++;
+            cc -= 2;
           } else if (str[cc] == 'B' && cc - 1 >= 0 && str[cc - 1] == 'D') {
             bNeg = true;
             cc -= 2;
diff --git a/xfa/fgas/crt/cfgas_formatstring.h b/xfa/fgas/crt/cfgas_formatstring.h
index 87252f472e..c91f43598e 100644
--- a/xfa/fgas/crt/cfgas_formatstring.h
+++ b/xfa/fgas/crt/cfgas_formatstring.h
@@ -25,6 +25,7 @@ class CFGAS_FormatString {
   void SplitFormatString(const CFX_WideString& wsFormatString,
                          std::vector<CFX_WideString>* wsPatterns);
   FX_LOCALECATEGORY GetCategory(const CFX_WideString& wsPattern);
+
   bool ParseText(const CFX_WideString& wsSrcText,
                  const CFX_WideString& wsPattern,
                  CFX_WideString* wsValue);
@@ -39,6 +40,7 @@ class CFGAS_FormatString {
                  const CFX_WideString& wsPattern);
   bool ParseNull(const CFX_WideString& wsSrcText,
                  const CFX_WideString& wsPattern);
+
   bool FormatText(const CFX_WideString& wsSrcText,
                   const CFX_WideString& wsPattern,
                   CFX_WideString* wsOutput);
diff --git a/xfa/fgas/crt/cfgas_formatstring_unittest.cpp b/xfa/fgas/crt/cfgas_formatstring_unittest.cpp
index 4cf2a89025..37fb3e62a4 100644
--- a/xfa/fgas/crt/cfgas_formatstring_unittest.cpp
+++ b/xfa/fgas/crt/cfgas_formatstring_unittest.cpp
@@ -72,12 +72,14 @@ TEST_F(CFGAS_FormatStringTest, DateFormat) {
       {L"en", L"20040722", L"YYYY-'W'WW-e", L"2004-W30-4"},
       {L"en", L"20040722", L"E 'days after Saturday'",
        L"5 days after Saturday"},
+      {L"en", L"2000-01-01", L"EEE, 'the' D 'of' MMMM, YYYY",
+       L"Sat, the 1 of January, 2000"},
       {L"en", L"2000-01-01", L"EEEE, 'the' D 'of' MMMM, YYYY",
        L"Saturday, the 1 of January, 2000"},
       {L"en", L"19991202", L"MM/D/YY", L"12/2/99"},
       {L"en", L"19990110", L"MMM D, YYYY", L"Jan 10, 1999"},
-      {L"en", L"19990102", L"J", L"2"},
-      {L"en", L"19990102", L"JJJ", L"002"},
+      {L"en", L"19990202", L"J", L"33"},
+      {L"en", L"19990202", L"JJJ", L"033"},
       {L"en", L"19990102", L"M", L"1"},
       {L"en", L"19990102", L"MMM", L"Jan"},
       {L"en", L"19990102", L"YYYY G", L"1999 AD"},
@@ -118,14 +120,18 @@ TEST_F(CFGAS_FormatStringTest, TimeFormat) {
     const wchar_t* pattern;
     const wchar_t* output;
   } tests[] = {{L"en", L"01:01:11", L"h:M A", L"1:1 AM"},
+               {L"en", L"13:01:11", L"h:M A", L"1:1 PM"},
                {L"en", L"01:01:11", L"hh:MM:SS A", L"01:01:11 AM"},
+               {L"en", L"13:01:11", L"hh:MM:SS A", L"01:01:11 PM"},
                {L"en", L"01:01:11", L"hh:MM:SS A Z", L"01:01:11 AM GMT-02:00"},
                {L"en", L"01:01:11", L"hh:MM:SS A z", L"01:01:11 AM -02:00"},
                // {L"en", L"01:01:11", L"hh:MM:SS A zz", L"01:01:11 AM GMT"},
                // Should change ?*+ into ' ' when formatting.
                // {L"en", L"01:01:11", L"hh:MM:SS?*+A", L"01:01:11   AM"},
                {L"en", L"12:01:01", L"k:MM:SS", L"12:01:01"},
+               {L"en", L"14:01:01", L"k:MM:SS", L"2:01:01"},
                {L"en", L"12:01:11", L"kk:MM", L"12:01"},
+               {L"en", L"14:01:11", L"kk:MM", L"02:01"},
                {L"en", L"12:01:11 +04:30", L"kk:MM", L"05:31"},
                {L"en", L"12:01:11", L"kk:MM A", L"12:01 PM"},
                {L"en", L"00:01:01", L"H:M:S", L"0:1:1"},
@@ -191,16 +197,39 @@ TEST_F(CFGAS_FormatStringTest, DateParse) {
     CFX_DateTime output;
   } tests[] = {
       {L"en", L"12/2/99", L"MM/D/YY", CFX_DateTime(1999, 12, 2, 0, 0, 0, 0)},
+      {L"en", L"2/2/99", L"M/D/YY", CFX_DateTime(1999, 2, 2, 0, 0, 0, 0)},
+      {L"en", L"2/2/10", L"M/D/YY", CFX_DateTime(2010, 2, 2, 0, 0, 0, 0)},
       {L"en", L"Jan 10, 1999", L"MMM D, YYYY",
        CFX_DateTime(1999, 1, 10, 0, 0, 0, 0)},
+      {L"en", L"Jan 10, 1999 AD", L"MMM D, YYYY G",
+       CFX_DateTime(1999, 1, 10, 0, 0, 0, 0)},
+      // TODO(dsinclair): Should this be -2 instead of 2?
+      {L"en", L"Jan 10, 0002 BC", L"MMM D, YYYY G",
+       CFX_DateTime(2, 1, 10, 0, 0, 0, 0)},
       {L"en", L"October 25, 2002", L"MMMM DD, YYYY",
        CFX_DateTime(2002, 10, 25, 0, 0, 0, 0)},
+      // TODO(dsinclair): The J and JJJ are ignored during parsing when they
+      // could be turned back into a date.
+      {L"en", L"1999-33", L"YYYY-J", CFX_DateTime(1999, 1, 1, 0, 0, 0, 0)},
+      {L"en", L"1999-033", L"YYYY-JJJ", CFX_DateTime(1999, 1, 1, 0, 0, 0, 0)},
       {L"de_CH", L"30. Oktober 2004", L"D. MMMM YYYY",
        CFX_DateTime(2004, 10, 30, 0, 0, 0, 0)},
       {L"fr_CA", L"30 octobre 2004", L"D MMMM YYYY",
        CFX_DateTime(2004, 10, 30, 0, 0, 0, 0)},
       {L"en", L"Saturday, the 1 of January, 2000",
        L"EEEE, 'the' D 'of' MMMM, YYYY", CFX_DateTime(2000, 1, 1, 0, 0, 0, 0)},
+      {L"en", L"Sat, the 1 of January, 2000", L"EEE, 'the' D 'of' MMMM, YYYY",
+       CFX_DateTime(2000, 1, 1, 0, 0, 0, 0)},
+      {L"en", L"7, the 1 of January, 2000",  // 7 == Saturday as 1 == Sunday
+       L"E, 'the' D 'of' MMMM, YYYY", CFX_DateTime(2000, 1, 1, 0, 0, 0, 0)},
+      {L"en", L"6, the 1 of January, 2000",  // 6 == Saturday as 1 == Monday
+       L"e, 'the' D 'of' MMMM, YYYY", CFX_DateTime(2000, 1, 1, 0, 0, 0, 0)},
+      {L"en", L"2004-07-22 Week of the month is 3",
+       L"YYYY-MM-DD 'Week of the month is' w",
+       CFX_DateTime(2004, 7, 22, 0, 0, 0, 0)},
+      {L"en", L"2004-07-22 Week of the year is 03",
+       L"YYYY-MM-DD 'Week of the year is' WW",
+       CFX_DateTime(2004, 7, 22, 0, 0, 0, 0)}
       // {L"ja", L"H15/11/3", L"gY/M/D", CFX_DateTime(2003, 11, 3, 0, 0, 0, 0)},
       // {L"ja", L"\u5e731-1-8", L"ggY-M-D", CFX_DateTime(1989, 1, 8, 0, 0, 0,
       // 0)}, {L"ja", L"\u5e73\u621089/11/03", L"gggYY/MM/DD",
@@ -270,24 +299,117 @@ TEST_F(CFGAS_FormatStringTest, NumParse) {
     const wchar_t* pattern;
     const wchar_t* output;
   } tests[] = {
-      // TODO(dsinclair): Numeric parsing fails when encountering a .
       // {L"en", L"€100.00", L"num(en_GB){$z,zz9.99}", L"100"},
       // {L"en", L"1050", L"99V99", L"10.50"},
       // {L"en", L"3125", L"99V99", L"31.25"},
-      // {L"en", L"12.345E3", L"99.999E", L"12345"},
-      // {L"en", L"12.345E-2", L"99.999E", L".12345"},
+      {L"en", L"12.345e3", L"99.999E", L"12345.000000"},
+      {L"en", L"12.345e+3", L"99.999E", L"12345.000000"},
+      {L"en", L"12.345E-2", L"99.999E", L"0.123450"},
+      // TODO(dsinclair): Returns 0.000?
+      // {L"en", L"12e-2", L"99E", L"0.12"},
       {L"en", L"150", L"z999", L"150"},
-      // {L"en", L"0150", L"z999", L"150"},
+      {L"en", L"150.50$", L"zzz.zz$", L"150.50"},
+      {L"en", L"0150", L"z999", L"0150"},
+      {L"en", L"123CR", L"999cr", L"-123"},
+      {L"en", L"123", L"999cr", L"123"},
+      {L"en", L"123CR", L"999CR", L"-123"},
+      {L"en", L"123  ", L"999CR", L"123"},
+      {L"en", L"123DB", L"999db", L"-123"},
+      {L"en", L"123", L"999db", L"123"},
+      {L"en", L"123DB", L"999DB", L"-123"},
+      {L"en", L"123  ", L"999DB", L"123"},
+      {L"en", L"123.5CR", L"999.9cr", L"-123.5"},
+      {L"en", L"123.5", L"999.9cr", L"123.5"},
+      {L"en", L"123.5CR", L"999.9CR", L"-123.5"},
+      // {L"en", L"123.5  ", L"999.9CR", L"123.5"},
+      {L"en", L"123.5DB", L"999.9db", L"-123.5"},
+      {L"en", L"123.5", L"999.9db", L"123.5"},
+      {L"en", L"123.5DB", L"999.9DB", L"-123.5"},
+      // {L"en", L"123.5  ", L"999.9DB", L"123.5"},
       {L"en", L"10.50", L"z,zz9.99", L"10.50"},
       {L"en", L"3,125.00", L"z,zz9.99", L"3125.00"},
       {L"en", L"$1,234.00", L"$z,zz9.99DB", L"1234.00"},
-      // {L"en", L"$,1234.00DB", L"$z,zz9.99DB", L"-1234.00"}
+      // TODO(dsinclair): Comes out as 1234 instead of -1234.
+      // {L"en", L"$,1234.00DB", L"$z,zz9.99DB", L"-1234.00"},
+      {L"en", L"1.234", L"zz9.zzz", L"1.234"},
+      {L"en", L"1 text", L"num{z 'text'}", L"1"},
+      {L"en", L"1.234 text", L"z.zzz 'text'", L"1.234"},
+      {L"en", L"  1.234", L"ZZ9.ZZZ", L"1.234"},
+      {L"en", L"12.345", L"zz9.zzz", L"12.345"},
+      {L"en", L" 12.345", L"ZZ9.ZZZ", L"12.345"},
+      {L"en", L"123.456", L"zz9.zzz", L"123.456"},
+      {L"en", L"123.456", L"ZZ9.ZZZ", L"123.456"},
+      {L"en", L"123.456-", L"ZZ9.ZZZS", L"-123.456"},
+      {L"en", L"123.456+", L"ZZ9.ZZZS", L"123.456"},
+      {L"en", L"123.456 ", L"ZZ9.ZZZS", L"123.456"},
+      {L"en", L"123.456-", L"ZZ9.ZZZS", L"-123.456"},
+      {L"en", L"123.456+", L"ZZ9.ZZZS", L"123.456"},
+      {L"en", L"123", L"zz9.zzz", L"123"},
+      {L"en", L"123.", L"ZZ9.ZZZ", L"123."},
+      {L"en", L"123.", L"zz9.zzz", L"123."},
+      {L"en", L"123.", L"ZZ9.ZZZ", L"123."},
+      {L"en", L"123.0", L"zz9.zzz", L"123.0"},
+      {L"en", L"123.0", L"ZZ9.ZZZ", L"123.0"},
+      {L"en", L"123.000", L"zz9.zzz", L"123.000"},
+      {L"en", L"123.000", L"ZZ9.ZZZ", L"123.000"},
+      {L"en", L"12,345.67", L"zzz,zz9.88888888", L"12345.67"},
+      {L"en", L"12,345.0000", L"zzz,zz9.88888888", L"12345.0000"},
+      {L"en", L"12,345.6789", L"zzz,zz9.8", L"12345.6789"},
+      {L"en", L"12,345.", L"zzz,zz9.8", L"12345."},
+      {L"en", L"123,456.000", L"zzz,zz9.8888", L"123456.000"},
+      {L"en", L"123,456.0", L"zzz,zz9.8888", L"123456.0"},
+      {L"en", L"123,456", L"zzz,zz9.8888", L"123456"},
+      {L"en", L"123,456", L"ZZZ,ZZ9.88", L"123456"},
+      {L"en", L"12,345.67", L"zzz,zz9.88888888", L"12345.67"},
+      {L"en", L"12,345.0000", L"zzz,zz9.88888888", L"12345.0000"},
+      {L"en", L"12,345.6789", L"zzz,zz9.8", L"12345.6789"},
+      {L"en", L"12,345.", L"zzz,zz9.8", L"12345."},
+      // TODO(dsinclair): Parses to 0
+      // {L"en", L"12%", L"zz9.%%", L".12"},
+      {L"en", L"1,234.50%", L"zzz,zz9.99%%", L"12.345"},
+      // {L"en", L"-00123", L"S999v99", L"-1.23"},
+      {L"en", L" 001.23", L"S999V99", L"001.23"},
+      // {L"en", L" 123.00", L"S999V99", L"123"},
+      {L"en", L"  12.30", L"SZZ9.99", L"12.30"},
+      {L"en", L"- 12.30", L"SZ99.99", L"-12.30"},
+      {L"en", L"123.00", L"szz9.99", L"123.00"},
+      {L"en", L"-123.00", L"szz9.99", L"-123.00"},
+      // {L"en", L"$  1,234.00  ", L"$ZZ,ZZ9.99CR", L"1234"},
+      // {L"en", L"$  1,234.00CR", L"$ZZ,ZZ9.99CR", L"-1234"},
+      // {L"en", L"$1,23400", L"$z,zz9.99DB", L"1234"},
+      {L"en", L"$1,234.00DB", L"$z,zz9.99DB", L"-1234.00"},
+      {L"en",
+       L"1\xA0"
+       L"234",
+       L"num(fr){z,zzz}", L"1234"},
+      // TODO(dsinclair): Parses to blank
+      // {L"en", L"1,234%", L"num.percent{}", L"12.34"},
+      // {L"en", L"1\xA0" L"234%%", L"num(fr).percent{}", L"12.34"},
+      // TODO(dsinclair): Parses to blank
+      // {L"en", L"1,234%", L"num{9,999%%}", L"12.34"},
+      {L"fr",
+       L"123\xA0"
+       L"456",
+       L"zzz,zzz", L"123456"},
+      {L"en", L"12%", L"zz%", L"0.12"},
+      {L"en", L"(123", L"(zzz", L"-123"},
+      {L"en", L"123)", L"zzz)", L"-123"},
+      {L"en", L"(123)", L"(zzz)", L"-123"},
+      {L"en", L"123 ", L"zzz)", L"123"},
+      {L"en", L" 123", L"(zzz", L"123"},
+      {L"en", L" 123 ", L"(zzz)", L"123"},
+      {L"en", L"123.5(", L"zzz.z(", L"-123.5"},
+      {L"en", L"123.5)", L"zzz.z)", L"-123.5"},
+      {L"en", L"123.5 ", L"zzz.z)", L"123.5"},
+      {L"en", L"123.5 ", L"zzz.z(", L"123.5"},
+      {L"en", L"123.545,4", L"zzz.zzz,z", L"123.5454"},
   };
 
   for (size_t i = 0; i < FX_ArraySize(tests); ++i) {
     CFX_WideString result;
     EXPECT_TRUE(fmt(tests[i].locale)
-                    ->ParseNum(tests[i].input, tests[i].pattern, &result));
+                    ->ParseNum(tests[i].input, tests[i].pattern, &result))
+        << " TEST: " << i;
     EXPECT_STREQ(tests[i].output, result.c_str()) << " TEST: " << i;
   }
 }
@@ -298,52 +420,92 @@ TEST_F(CFGAS_FormatStringTest, NumFormat) {
     const wchar_t* input;
     const wchar_t* pattern;
     const wchar_t* output;
-  } tests[] = {{L"en", L"1.234", L"zz9.zzz", L"1.234"},
-               {L"en", L"1.234", L"ZZ9.ZZZ", L"  1.234"},
-               {L"en", L"12.345", L"zz9.zzz", L"12.345"},
-               {L"en", L"12.345", L"ZZ9.ZZZ", L" 12.345"},
-               {L"en", L"123.456", L"zz9.zzz", L"123.456"},
-               {L"en", L"123.456", L"ZZ9.ZZZ", L"123.456"},
-               {L"en", L"123", L"zz9.zzz", L"123"},
-               // {L"en", L"123", L"ZZ9.ZZZ", L"123."},
-               {L"en", L"123.", L"zz9.zzz", L"123."},
-               // {L"en", L"123.", L"ZZ9.ZZZ", L"123."},
-               // {L"en", L"123.0", L"zz9.zzz", L"123.0"},
-               // {L"en", L"123.0", L"ZZ9.ZZZ", L"123.0"},
-               // {L"en", L"123.000", L"zz9.zzz", L"123.000"},
-               {L"en", L"123.000", L"ZZ9.ZZZ", L"123.000"},
-               // {L"en", L"12345.67", L"zzz,zz9.88888888", L"12,345.67"},
-               // {L"en", L"12345.0000", L"zzz,zz9.88888888", L"12,345.0000"},
-               // {L"en", L"12345.6789", L"zzz,zz9.8", L"12,345.6789"},
-               // {L"en", L"12345.", L"zzz,zz9.8", L"12,345"},
-               // {L"en", L"123456.000", L"zzz,zz9.8888", L"123,456.000"},
-               // {L"en", L"123456.0", L"zzz,zz9.8888", L"123,456.0"},
-               {L"en", L"123456", L"zzz,zz9.8888", L"123,456"},
-               {L"en", L"123456", L"ZZZ,ZZ9.88", L"123,456"},
-               // {L"en", L"12345.67", L"zzz,zz9.88888888", L"12,345.67"},
-               // {L"en", L"12345.0000", L"zzz,zz9.88888888", L"12,345.0000"},
-               // {L"en", L"12345.6789", L"zzz,zz9.8", L"12,345.6789"},
-               // {L"en", L"12345.", L"zzz,zz9.8", L"12,345"},
-               // {L"en", L"12%%", L"zz9.%%", L"12%%"},
-               // {L"en", L"1,234.5%%", L"zzz,zz9.99%%", L"1,234.50%%"},
-               {L"en", L"-1.23", L"S999v99", L"-00123"},
-               {L"en", L"1.23", L"S999V99", L" 001.23"},
-               {L"en", L"123", L"S999V99", L" 123.00"},
-               {L"en", L"12.3", L"SZZ9.99", L"  12.30"},
-               {L"en", L"-12.3", L"SZ99.99", L"- 12.30"},
-               {L"en", L"123", L"szz9.99", L"123.00"},
-               {L"en", L"-123", L"szz9.99", L"-123.00"},
-               // {L"en", L"1234", L"$ZZ,ZZ9.99CR", L"$  1,234.00  "},
-               // {L"en", L"-1234", L"$ZZ,ZZ9.99CR", L"$  1,234.00CR"},
-               // {L"en", L"1234", L"$z,zz9.99DB", L"$1,234.00"},
-               {L"en", L"-1234", L"$z,zz9.99DB", L"$1,234.00DB"},
-               {L"en", L"12345", L"99.999E", L"12.345E+3"},
-               {L"en", L".12345", L"99.999E", L"12.345E-2"}};
+  } tests[] = {
+      {L"en", L"1.234", L"zz9.zzz", L"1.234"},
+      {L"en", L"1", L"num{z 'text'}", L"1 text"},
+      {L"en", L"1", L"num{'text' z}", L"text 1"},
+      {L"en", L"1.234", L"ZZ9.ZZZ", L"  1.234"},
+      {L"en", L"12.345", L"zz9.zzz", L"12.345"},
+      {L"en", L"12.345", L"ZZ9.ZZZ", L" 12.345"},
+      {L"en", L"123.456", L"zz9.zzz", L"123.456"},
+      {L"en", L"123.456", L"ZZ9.ZZZ", L"123.456"},
+      {L"en", L"123", L"zz9.zzz", L"123"},
+      {L"en", L"123", L"ZZ9.ZZZ", L"123.000"},
+      {L"en", L"123.", L"zz9.zzz", L"123."},
+      {L"en", L"123.", L"ZZ9.ZZZ", L"123.000"},
+      {L"en", L"123.0", L"zz9.zzz", L"123"},
+      {L"en", L"123.0", L"ZZ9.ZZZ", L"123.000"},
+      {L"en", L"123.000", L"zz9.zzz", L"123"},
+      {L"en", L"123.000", L"ZZ9.ZZZ", L"123.000"},
+      // {L"en", L"12345.67", L"zzz,zz9.88888888", L"12,345.67"},
+      // {L"en", L"12345.0000", L"zzz,zz9.88888888", L"12,345.0000"},
+      // {L"en", L"12345.6789", L"zzz,zz9.8", L"12,345.6789"},
+      // {L"en", L"12345.", L"zzz,zz9.8", L"12,345"},
+      // {L"en", L"123456.000", L"zzz,zz9.8888", L"123,456.000"},
+      // {L"en", L"123456.0", L"zzz,zz9.8888", L"123,456.0"},
+      {L"en", L"123456", L"zzz,zz9.8888", L"123,456"},
+      {L"en", L"123456", L"ZZZ,ZZ9.88", L"123,456"},
+      // {L"en", L"12345.67", L"zzz,zz9.88888888", L"12,345.67"},
+      // {L"en", L"12345.0000", L"zzz,zz9.88888888", L"12,345.0000"},
+      // {L"en", L"12345.6789", L"zzz,zz9.8", L"12,345.6789"},
+      // {L"en", L"12345.", L"zzz,zz9.8", L"12,345"},
+      // {L"en", L"12%%", L"zz9.%%", L"12%%"},
+      // {L"en", L"1,234.5%%", L"zzz,zz9.99%%", L"1,234.50%%"},
+      {L"en", L"-1.23", L"S999v99", L"-00123"},
+      {L"en", L"1.23", L"S999V99", L" 001.23"},
+      {L"en", L"123", L"S999V99", L" 123.00"},
+      {L"en", L"12.3", L"SZZ9.99", L"  12.30"},
+      {L"en", L"-12.3", L"SZ99.99", L"- 12.30"},
+      {L"en", L"123", L"szz9.99", L"123.00"},
+      {L"en", L"-123", L"szz9.99", L"-123.00"},
+      // {L"en", L"1234", L"$ZZ,ZZ9.99CR", L"$  1,234.00  "},
+      // {L"en", L"-1234", L"$ZZ,ZZ9.99CR", L"$  1,234.00CR"},
+      // {L"en", L"1234", L"$z,zz9.99DB", L"$1,234.00"},
+      {L"en", L"-1234", L"$z,zz9.99DB", L"$1,234.00DB"},
+      {L"en", L"12345", L"99.999E", L"12.345E+3"},
+      {L"en", L"12345", L"99999E", L"12345E+0"},
+      {L"en", L".12345", L"99.999E", L"12.345E-2"},
+      {L"en", L"12345", L"99,999", L"12,345"},
+      {L"en", L"1234", L"num(fr){z,zzz}",
+       L"1\xA0"
+       L"234"},
+      {L"en", L"12.34", L"num.percent{}", L"1,234%"},
+      {L"en", L"12.34", L"num(fr).percent{}",
+       L"1\xA0"
+       L"234%"},
+      // {L"en", L"12.34", L"num{9,999%%}", L"1,234%"},
+      {L"en", L"-123", L"zzzCR", L"123CR"},
+      {L"en", L"123", L"zzzCR", L"123  "},
+      {L"en", L"-123", L"zzzcr", L"123CR"},
+      {L"en", L"123", L"zzzcr", L"123"},
+      {L"en", L"123", L"zzz$", L"123$"},
+      {L"en", L"-123.5", L"zzz.zCR", L"123.5CR"},
+      {L"en", L"123.5", L"zzz.zCR", L"123.5  "},
+      {L"en", L"-123.5", L"zzz.zcr", L"123.5CR"},
+      {L"en", L"123.5", L"zzz.zcr", L"123.5"},
+
+      {L"en", L"-123.5", L"999.9db", L"123.5db"},
+      {L"en", L"123.5", L"999.9db", L"123.5"},
+      {L"en", L"-123.5", L"999.9DB", L"123.5DB"},
+      {L"en", L"123.5", L"999.9DB", L"123.5  "},
+
+      {L"en", L"-123", L"(zzz", L"(123"},
+      // {L"en", L"-123", L"zzz)", L"123)"},
+      {L"en", L"-123", L"(zzz)", L"(123)"},
+      {L"en", L"123", L"zzz)", L"123 "},
+      {L"en", L"123", L"(zzz", L" 123"},
+      {L"en", L"123", L"(zzz)", L" 123 "},
+      {L"en", L"-123.5", L"zzz.z(", L"123.5("},
+      // {L"en", L"-123.5", L"zzz.z)", L"123.5)"},
+      {L"en", L"123.5", L"zzz.z)", L"123.5 "},
+      {L"en", L"123.5", L"zzz.z(", L"123.5 "},
+  };
 
   for (size_t i = 0; i < FX_ArraySize(tests); ++i) {
     CFX_WideString result;
     EXPECT_TRUE(fmt(tests[i].locale)
-                    ->FormatNum(tests[i].input, tests[i].pattern, &result));
+                    ->FormatNum(tests[i].input, tests[i].pattern, &result))
+        << " TEST: " << i;
     EXPECT_STREQ(tests[i].output, result.c_str()) << " TEST: " << i;
   }
 }
@@ -360,7 +522,10 @@ TEST_F(CFGAS_FormatStringTest, TextParse) {
                //  + - one or more whitespace
                // {L"en", L"555-1212", L"text(th_TH){999*9999}", L"5551212"},
                {L"en", L"ABC-1234-5", L"AAA-9999-X", L"ABC12345"},
-               {L"en", L"ABC-1234-D", L"AAA-9999-X", L"ABC1234D"}};
+               {L"en", L"ABC-1234-D", L"AAA-9999-X", L"ABC1234D"},
+               {L"en", L"A1C-1234-D", L"OOO-9999-X", L"A1C1234D"},
+               {L"en", L"A1C-1234-D", L"000-9999-X", L"A1C1234D"},
+               {L"en", L"A1C-1234-D text", L"000-9999-X 'text'", L"A1C1234D"}};
 
   for (size_t i = 0; i < FX_ArraySize(tests); ++i) {
     CFX_WideString result;
@@ -384,8 +549,13 @@ TEST_F(CFGAS_FormatStringTest, TextFormat) {
     const wchar_t* output;
   } tests[] = {
       {L"en", L"K1S5K2", L"A9A 9A9", L"K1S 5K2"},
-      {L"en", L"6135551212", L"'+1 ('999') '999-9999", L"+1 (613) 555-1212"},
-      {L"en", L"6135551212", L"999.999.9999", L"613.555.1212"}};
+      {L"en", L"K1S5K2", L"text(fr){A9A 9A9}", L"K1S 5K2"},
+      {L"en", L"6135551212", L"'+1 ('9\u002399') '999-9999",
+       L"+1 (6#13) 555-1212"},
+      {L"en", L"6135551212", L"999.999.9999", L"613.555.1212"},
+      {L"en", L"6135551212", L"999\u0023999\u002A9999", L"613#555*1212"},
+      {L"en", L"K1#5K2", L"00X OO9", L"K1# 5K2"},
+  };
 
   for (size_t i = 0; i < FX_ArraySize(tests); ++i) {
     CFX_WideString result;
@@ -430,9 +600,9 @@ TEST_F(CFGAS_FormatStringTest, ZeroParse) {
     const wchar_t* locale;
     const wchar_t* input;
     const wchar_t* pattern;
-  } tests[] = {
-      {L"en", L"", L"zero{}"}, {L"en", L"9", L"zero{9}"},
-  };
+  } tests[] = {{L"en", L"", L"zero{}"},
+               {L"en", L"9", L"zero{9}"},
+               {L"en", L"a", L"zero{'a'}"}};
 
   for (size_t i = 0; i < FX_ArraySize(tests); ++i) {
     EXPECT_TRUE(
@@ -462,3 +632,20 @@ TEST_F(CFGAS_FormatStringTest, ZeroFormat) {
     EXPECT_STREQ(tests[i].output, result.c_str()) << " TEST: " << i;
   }
 }
+
+TEST_F(CFGAS_FormatStringTest, GetCategory) {
+  CFGAS_FormatString* f = fmt(L"en");
+
+  EXPECT_EQ(FX_LOCALECATEGORY_Unknown, f->GetCategory(L"'just text'"));
+  EXPECT_EQ(FX_LOCALECATEGORY_Null, f->GetCategory(L"null{}"));
+  EXPECT_EQ(FX_LOCALECATEGORY_Zero, f->GetCategory(L"zero{}"));
+  EXPECT_EQ(FX_LOCALECATEGORY_Num, f->GetCategory(L"num{}"));
+  EXPECT_EQ(FX_LOCALECATEGORY_Text, f->GetCategory(L"text{}"));
+  EXPECT_EQ(FX_LOCALECATEGORY_DateTime, f->GetCategory(L"datetime{}"));
+  EXPECT_EQ(FX_LOCALECATEGORY_Time, f->GetCategory(L"time{}"));
+  EXPECT_EQ(FX_LOCALECATEGORY_Date, f->GetCategory(L"date{}"));
+  EXPECT_EQ(FX_LOCALECATEGORY_DateTime, f->GetCategory(L"time{} date{}"));
+  EXPECT_EQ(FX_LOCALECATEGORY_DateTime, f->GetCategory(L"date{} time{}"));
+  EXPECT_EQ(FX_LOCALECATEGORY_Num, f->GetCategory(L"num(en_GB){}"));
+  EXPECT_EQ(FX_LOCALECATEGORY_Date, f->GetCategory(L"date.long{}"));
+}