[LLDB] Don't print register fields when asked for a specific format

[DavidSpickett]

Summary

Previously if a register had fields we would always print them after the
value if the register was asked for by name.

(lldb) register read MDCR_EL3
MDCR_EL3 = 0x00000000
         = {
             ETBAD = 0
<...>
             RLTE = 0
           }

This can be quite annoying if there are a whole lot of fields but you
want to see the register in a specific format.

(lldb) register read MDCR_EL3 -f i
MDCR_EL3 = 0x00000000   unknown     udf    #0x0
         = {
             ETBAD = 0
<...lots of fields...>

Since it pushes the interesting bit far up the terminal. To solve this,
don't print fields if the user passes --format. If they're doing that
then I think it's reasonable to assume they know what they want and only
want to see that output.

This also gives users a way to silence fields, but not change the format.
By doing register read foo -f x. In case they are not useful or perhaps
they are trying to work around a crash.

I have customised the help text for --format for register read to explain this:

-f <format> ( --format <format> )
     Specify a format to be used for display. If this is set, register fields will not be dispayed.

[DavidSpickett]

Use the getter for the format value instead of adding a WasSet to the group class.

[DavidSpickett]

One alternative was to come up with a list of formats where printing fields afterwards makes sense. Hex makes sense, instruction does not. However it would be hard to judge and it doesn't solve the main issue where you want to see what you asked for without extras.

This does mean that you will only be able to get the default format if you want fields, but I think that will cover the majority of use cases for now. If it doesn't we could add an option that enables fields which defaults to on when no format is set, and off when one is.

[jasonmolenda]

This seems fine to me; it's nice to have a way to quiet these when they get in the way, even if it's maybe not the first thing you'd look for, it's an uncommon enough use case that I'm not worried about that.

lldb/source/Commands/CommandObjectRegister.cpp

50

displayed

[DavidSpickett]

dispayed -> displayed

lldb/source/Commands/CommandObjectRegister.cpp

public:
   CommandObjectRegisterRead(CommandInterpreter &interpreter)
       : CommandObjectParsed(
             interpreter, "register read",
             "Dump the contents of one or more register values from the current "
             "frame.  If no register is specified, dumps them all.",
             nullptr,
             eCommandRequiresFrame | eCommandRequiresRegContext |
                 eCommandProcessMustBeLaunched | eCommandProcessMustBePaused),
-        m_format_options(eFormatDefault) {
+        m_format_options(eFormatDefault, UINT64_MAX, UINT64_MAX,
+                         {{CommandArgumentType::eArgTypeFormat,
+                           "Specify a format to be used for display. If this "
+                           "is set, register fields will not be displayed."}}) {
     CommandArgumentEntry arg;
     CommandArgumentData register_arg;
 
     // Define the first (and only) variant of this arg.
     register_arg.arg_type = eArgTypeRegisterName;
     register_arg.arg_repetition = eArgRepeatStar;
 
     // There is only one variant this argument could be; put it into the
if (command.GetArgumentCount() == 0) {
           // consistent towards the user and allow them to say reg read $rbx -
           // internally, however, we should be strict and not allow ourselves
           // to call our registers $rbx in our own API
           auto arg_str = entry.ref();
           arg_str.consume_front("$");
 
           if (const RegisterInfo *reg_info =
                   reg_ctx->GetRegisterInfoByName(arg_str)) {
+            // If they have asked for a specific format don't obscure that by
+            // printing flags afterwards.
+            bool print_flags =
+                !m_format_options.GetFormatValue().OptionWasSet();
             if (!DumpRegister(m_exe_ctx, strm, *reg_ctx, *reg_info,
-                              /*print_flags=*/true))
+                              print_flags))
               strm.Printf("%-12s = error: unavailable\n", reg_info->name);
           } else {
             result.AppendErrorWithFormat("Invalid register name '%s'.\n",
                                          arg_str.str().c_str());
           }
         }
       }
     }

lldb/test/API/functionalities/gdb_remote_client/TestXMLRegisterFlags.py

def test_flags_child_limit(self):
           '<field name="field_0" start="0" end="0"/>'
           '<field name="field_1" start="1" end="1"/>'
           '<field name="field_2" start="2" end="2"/>')
 
         self.expect("register read cpsr", substrs=["= (field_2 = 1, field_1 = 1, ...)"])
 
     @skipIfXmlSupportMissing
     @skipIfRemote
+    def test_format_disables_flags(self):
+        # If asked for a specific format, don't print flags after it.
+        self.setup_flags_test('<field name="field_0" start="0" end="0"/>')
+
+        self.expect("register read cpsr --format X", substrs=["cpsr = 0xEEEE7777"])
+        self.expect("register read cpsr --format X", substrs=["field_0"], matching=False)
+
+    @skipIfXmlSupportMissing
+    @skipIfRemote
     def test_xml_includes(self):
         # Certain targets e.g. s390x QEMU split their defintions over multiple
         # files that are included into target.xml.
         self.setup_multidoc_test({
             # The formatting is very specific here. lldb doesn't like leading
             # spaces, and nested tags must be indented more than their parent.
             'target.xml' : dedent("""\
                <?xml version="1.0"?>