RGBCube/serenity
1
Fork 0
mirror of https://github.com/RGBCube/serenity synced 2025-07-25 15:17:36 +00:00
Code Issues Activity Actions

LibVT: Implement new ANSI escape sequence parser

Browse source 
This commit replaces the former, hand-written parser with a new one that
can be generated automatically according to a state change diagram.

The new `EscapeSequenceParser` class provides a more ergonomic interface
to dealing with escape sequences. This interface has been inspired by
Alacritty's [vte library](https://github.com/alacritty/vte/).

I tried to avoid changing the application logic inside the `Terminal`
class. While this code has not been thoroughly tested, I can't find
regressions in the basic command line utilities or `vttest`.

`Terminal` now displays nicer debug messages when it encounters an
unknown escape sequence. Defensive programming and bounds checks have
been added where we access parameters, and as a result, we can now
endure 4-5 seconds of `cat /dev/urandom`. :D

We generate EscapeSequenceStateMachine.h when building the in-kernel
LibVT, and we assume that the file is already in place when the userland
library is being built. This will probably cause problems later on, but
I can't find a way to do it nicely.
This commit is contained in:
Daniel Bertalan 2021-05-08 20:37:43 +02:00 • committed by Andreas Kling
parent 1b347298f1
commit be519022c3
11 changed files with 707 additions and 571 deletions
Download patch file Download diff file Expand all files Collapse all files

162
Userland/Libraries/LibVT/EscapeSequenceParser.cpp Normal file
View file

@ -0,0 +1,162 @@
/*
* Copyright (c) 2021, the SerenityOS developers.
*
* SPDX-License-Identifier: BSD-2-Clause
*/
#include <AK/Format.h>
#include <AK/Types.h>
#include <LibVT/EscapeSequenceParser.h>
#include <LibVT/EscapeSequenceStateMachine.h>
namespace VT {
EscapeSequenceParser::EscapeSequenceParser(EscapeSequenceExecutor& executor)
: m_executor(executor)
, m_state_machine([this](auto action, auto byte) { perform_action(action, byte); })
{
}
EscapeSequenceParser::~EscapeSequenceParser()
{
}
Vector<EscapeSequenceParser::OscParameter> EscapeSequenceParser::osc_parameters() const
{
VERIFY(m_osc_raw.size() >= m_osc_parameter_indexes.last());
Vector<EscapeSequenceParser::OscParameter> params;
size_t prev_idx = 0;
for (auto end_idx : m_osc_parameter_indexes) {
// If the parameter is empty, we take an out of bounds index as the beginning of the Span.
// This should not be a problem as we won't dereference the 0-length Span that's created.
// Using &m_osc_raw[prev_idx] to get the start pointer checks whether we're out of bounds,
// so we would crash.
params.append({ m_osc_raw.data() + prev_idx, end_idx - prev_idx });
prev_idx = end_idx;
}
return params;
}
void EscapeSequenceParser::perform_action(EscapeSequenceStateMachine::Action action, u8 byte)
{
auto advance_utf8 = [&](u8 byte) {
u32 new_codepoint = m_code_point;
new_codepoint <<= 6;
new_codepoint |= byte & 0x3f;
return new_codepoint;
};
switch (action) {
case EscapeSequenceStateMachine::Action::_Ignore:
break;
case EscapeSequenceStateMachine::Action::Print:
m_executor.emit_code_point((u32)byte);
break;
case EscapeSequenceStateMachine::Action::PrintUTF8:
m_executor.emit_code_point(advance_utf8(byte));
break;
case EscapeSequenceStateMachine::Action::Execute:
m_executor.execute_control_code(byte);
break;
case EscapeSequenceStateMachine::Action::Hook:
if (m_param_vector.size() == MAX_PARAMETERS)
m_ignoring = true;
else
m_param_vector.append(m_param);
m_executor.dcs_hook(m_param_vector, intermediates(), m_ignoring, byte);
break;
case EscapeSequenceStateMachine::Action::Put:
m_executor.receive_dcs_char(byte);
break;
case EscapeSequenceStateMachine::Action::BeginUTF8:
if ((byte & 0xe0) == 0xc0) {
m_code_point = byte & 0x1f;
} else if ((byte & 0xf0) == 0xe0) {
m_code_point = byte & 0x0f;
} else if ((byte & 0xf8) == 0xf0) {
m_code_point = byte & 0x07;
} else {
dbgln("Invalid character was parsed as UTF-8 initial byte {:02x}", byte);
VERIFY_NOT_REACHED();
}
break;
case EscapeSequenceStateMachine::Action::AdvanceUTF8:
VERIFY((byte & 0xc0) == 0x80);
m_code_point = advance_utf8(byte);
break;
case EscapeSequenceStateMachine::Action::FailUTF8:
m_executor.emit_code_point(U'<EFBFBD>');
break;
case EscapeSequenceStateMachine::Action::OscStart:
m_osc_raw.clear();
m_osc_parameter_indexes.clear();
break;
case EscapeSequenceStateMachine::Action::OscPut:
if (byte == ';') {
if (m_osc_parameter_indexes.size() == MAX_OSC_PARAMETERS) {
dbgln("EscapeSequenceParser::perform_action: shenanigans! OSC sequence has too many parameters");
} else {
m_osc_parameter_indexes.append(m_osc_raw.size());
}
} else {
m_osc_raw.append(byte);
}
break;
case EscapeSequenceStateMachine::Action::OscEnd:
if (m_osc_parameter_indexes.size() == MAX_OSC_PARAMETERS) {
dbgln("EscapeSequenceParser::perform_action: shenanigans! OSC sequence has too many parameters");
} else {
m_osc_parameter_indexes.append(m_osc_raw.size());
}
m_executor.execute_osc_sequence(osc_parameters(), byte);
break;
case EscapeSequenceStateMachine::Action::Unhook:
m_executor.execute_dcs_sequence();
break;
case EscapeSequenceStateMachine::Action::CsiDispatch:
if (m_param_vector.size() > MAX_PARAMETERS) {
dbgln("EscapeSequenceParser::perform_action: shenanigans! CSI sequence has too many parameters");
m_ignoring = true;
} else {
m_param_vector.append(m_param);
}
m_executor.execute_csi_sequence(m_param_vector, intermediates(), m_ignoring, byte);
break;
case EscapeSequenceStateMachine::Action::EscDispatch:
m_executor.execute_escape_sequence(intermediates(), m_ignoring, byte);
break;
case EscapeSequenceStateMachine::Action::Collect:
if (m_intermediate_idx == MAX_INTERMEDIATES) {
dbgln("EscapeSequenceParser::perform_action: shenanigans! escape sequence has too many intermediates");
m_ignoring = true;
} else {
m_intermediates[m_intermediate_idx++] = byte;
}
break;
case EscapeSequenceStateMachine::Action::Param:
if (m_param_vector.size() == MAX_PARAMETERS) {
dbgln("EscapeSequenceParser::perform_action: shenanigans! escape sequence has too many parameters");
m_ignoring = true;
} else {
if (byte == ';') {
m_param_vector.append(m_param);
m_param = 0;
} else if (byte == ':') {
dbgln("EscapeSequenceParser::perform_action: subparameters are not yet implemented");
} else {
m_param *= 10;
m_param += (byte - '0');
}
}
break;
case EscapeSequenceStateMachine::Action::Clear:
m_intermediate_idx = 0;
m_ignoring = false;
m_param = 0;
m_param_vector.clear_with_capacity();
break;
}
}
}